iñigo gonzález ricoy

CONSTITUTIONALISM, THE WORKPLACE, AND THE SCOPE OF DEMOCRACY

Universitat de Barcelona, Facultat de Filosoﬁa Departament de Filosoﬁa Teorètica i Pràctica Programa de doctorat: Filosoﬁa, Ética i Política, Bienni 2004-2006 Doctorand: Iñigo González Ricoy Directors: João Manuel Cardoso Rosas i José Manuel Bermudo Ávila

AVISO IMPORTANTE

La presente tesis doctoral está acogida a la Licencia Creative Commons Reconocimiento –
No Comercial (by-nc)

“Se permite la generación de obras derivadas siempre que no se haga un uso comercial. Tampoco se puede utilizar la obra original con finalidades comerciales”

a
A mi madre Andrea

Abstract/resumen
The dissertation analyzes the scope and limits of democracy both historically and normatively. It has two main goals. First, it historically explains how a number of institutional devices that had been traditionally seen as limits on the scope of democracy turned out to be seen not only as consistent with democracy but actually as necessary for any good-working democratic system. Second, it normatively analyzes two outstanding cases in which the scope of democracy is heavily limited—constitutional constraints on legislatures and decision-making within the ﬁrm, a domain that is usually taken to be beyond the scope of democracy. Part I historically analyzes a two-fold shift in the concept of democracy in the Early American Republic. First, the increasing acceptance of the idea of democracy—along with the institutional devices traditionally attached to it—due to the acceptance of disagreement as a legitimate feature of the American society and the subsequent crisis of the theory of virtual representation, according to which the society is an organic entity with a shared set of interests that ought to be insulated from democratic struggle and partisanship. Second, the shift in the very concept of democracy, by which many of the institutional devices that had been traditionally seen as constraints on democracy turned out to be seen not only as consistent with democracy but also as necessary for any good-working democratic system. In a nutshell, democracy gained acceptability as long as the scope of democratic decision-making was downsized. The remainder of the dissertation normatively analyzes two outstanding cases in which the scope of democracy is clearly constrained—namely, constitutionalism and decision-making within ﬁrms. Part II addresses several normative theories of constitutionalism—and, notably, of constitutional rigidity and judicial review—and shows that they all fail to justify constitutional constraints from a democratic standpoint. Three prominent types of theories are analyzed—pure instrumentalist, precommitment-based, and proceduralist. Even though a number of problems are identiﬁed, a common and central problem is that they all fail to address adequately the fact of disagreement and thus fall into new forms of organicism. Finally, Part III analyzes another sphere that is usually taken to be beyond the scope of democracy—namely, the workplace. It addresses the core arguments for and against extending democratic decision-making to the workplace and develops a novel, republican case for workplace democracy based on incomplete (labor) contract theory. Finally, bargaining power asymmetries and moral hazard problems arising

6

Abstract/resumen

from the formal separation of ownership and control rights in democratic ﬁrms are analyzed. Resumen breve1 La tesis analiza el alcance y los límites de la democracia de forma histórica y normativa. Tiene dos objetivos principales. Primero, explicar históricamente cómo una serie de instituciones que habían sido tradicionalmente consideradas como límites a la democracia fueron incorporadas al concepto de democracia, hasta el punto de ser consideradas necesarias para su correcto funcionamiento. Segundo, analizar dos casos especialmente relevantes en los que el alcance de la democracia está fuertemente limitado: el constitutionalismo y la toma de decisiones en las empresas. La Primera Parte analiza históricamente el concepto de democracia, y su alcance, en los Estados Unidos de América en el periodo revolucionario y post-revolucionario. Se halla un doble desplazamiento. Primero, un incremento en la aceptación de la idea de democracia debido a la aceptación del desacuerdo político como un fenómeno legítimo y la consiguiente crisis de la teoría de la representación virtual, según la cual la sociedad es una entidad orgánica con un conjunto de intereses compartidos que deben ser aislados de la política democrática y el partidismo. Segundo, un desplazamiento del concepto de democracia, según el cual muchos de los mecanismos institucionales que habían sido tradicionalmente vistos como límites al alcance de la democracia pasaron a ser considerados no sólo consistentes con la democracia sino de hecho necesarios para su correcto funcionamiento. En resumen, la democracia ganó aceptación en la medida en que su alcance quedó reducido. El resto de la tesis analiza normativamente los límites al alcance de la democracia en relación a dos casos especialmente relevantes: el constitucionalismo y la toma de decisiones en la empresa. La Segunda Parte analiza las principales teorías normativas del constitucionalismo —y, concretamente, de la rigidez constitucional y la revisión judicial de las leyes— y muestra que todas ellas resultan insuﬁcientes para justiﬁcar los límites constitucionales desde un punto de vista democrático. Se analizan tres tipos de teorías: instrumentalistas, basadas en precompromisos y procedimentales. Aunque se identiﬁca toda una serie de problemas, es común a todas ellas su incapacidad para acomodar adecuadamente el desacuerdo político en materia constitucional, incurriendo así en diversas formas de organicismo. Finalmente, la Tercera Parte analiza la toma de decisiones en las empresas, un ámbito que es generalmente considerado ajeno al alcance de la democracia. Se analizan los principales modelos de democracia en la empresa —cooperativismo y codeterminación—, así como los principales argumentos a favor y en contra de la extensión de la toma de decisiones democrática a las empresas. Por último, se presenta un argumento original, de raíz republicana y apoyado en la teoría de los contratos (laborales) incompletos, a favor de la democracia en las empresas. Asímismo, se analizan las asimetrías negociadoras y los problemas de riesgo moral derivados de la separación formal entre control y propiedad en las empresas democráticas.
1

Véase un resumen extenso en castellano en las pp. 275-295.

Acknowledgments
This dissertation has been possible thanks to a number of institutions and individuals. The Catalan Agency for Management of University and Research Grants, the Spanish Ministry of Education and Science, the Spanish Ministry of Science and Innovation, and the Universitat de Barcelona have provided research funding. I am also grateful to those institutions in which I have worked on the dissertation. I am especially grateful to the Department of Theoretical and Practical Philosophy at the Universitat de Barcelona, where I have developed my research and teaching over the last few years. But also to those institutions that I have visited as a visiting research student—the Centro de Estudos Humanísticos at the Universidade do Minho (Portugal), the Chaire Hoover d’étique economique et sociale at the Université Catholique de Louvain (Belgium), the Department of Philosophy at New York University (USA), and the School of Public Policy at the University College London (UK). The main reason why the dissertation is in English is that most of it was written while I was visiting NYU ﬁrst and UCL next. As for the individuals, ﬁrst of all, I wish to thank my two advisors—José Manuel Bermudo and João Cardoso Rosas. Professor Bermudo invited me to join his group while I was still an undergraduate student, and since then he has been as generous in his support as critical in his supervision. Professor Rosas welcomed me in Braga in 2006 with open arms and has supported and supervised me in ways that extend well beyond this dissertation ever since. I should also wish to thank the members of the Political Philosophy Seminar at the Universitat de Barcelona, with whom I have been discussing on a regular basis since 2005 and with whom I have embarked in not few academic adventures. Over these years I have become close friends with many of them, especially with Güido Sender, Joan Lara, Lluís Pla, and Martha Palacio. My department and oﬃce colleagues, as well as the administrative staﬀ of the institutions in which I have worked, have also contributed to make this dissertation possible, making my daily work easy and enjoyable. I am especially indebted to Begoña Román, who has always been available to me and has guided me in my ﬁrst steps as a lecturer. Thank you very much to all of them, as well as to all the people at the universities that I have visited, specially to those who invited and hosted me during my research stays—João Rosas and Acílio Rocha in Braga, Axel Gosseries in Louvain, Jeremy Waldron in New York, and Richard Bellamy in London. I have presented parts of the dissertation in a number of conferences and seminars

8

Acknowledgments

in SUNY at Albany (USA), Universidad Nacional del Sur (Argentina), Universitat de Barcelona, Universidade do Minho (Portugal), Universidad de Murcia, New York University (USA), Euskal Herriko Unibertsitatea, Princeton University (USA), Universidade de Santiago de Compostela, Universidad de Salamanca, University College London (UK), and the Instituto de Filosofía at the CSIC. Unfortunately, it is not possible to mention all the people with whom I have discussed about this or that point of the dissertation. However, I wish to thank, in addition to my advisors, Antonio Alegre, Sara Amighetti, Daniel Attas, Guy Atchinson, Richard Bellamy, David Corrales, Andrés de Francisco, Colin Donnaruma, Axel Gosseries, Russell Hardin, Volkan Gul, Oscar Horta, Nicholas James, Sharon Krause, Joan Lara, Sebastián Linares, Joseph Mazor, Víctor Méndez, Josep Monserrat, Jeﬀrey Moriarty, Liam Murphy, Martha Palacio, Begoña Román, Genevieve Rouseliere, Ramón Ruiz Ruiz, Joan Vergés, Jeremy Waldron, and Albert Weale for discussions and comments on parts of the dissertation. Part III of the dissertation began as a collaboration with Lluís Pla, to whom I am grateful for endless discussions on this an other topics of the dissertation. In addition, Sara Amigheti, Nina Eschke, and Peter López Stiedl have provided linguistic advice. Dejo para el ﬁnal a quienes sin embargo ocupan el primer lugar en mi vida. Mi madre Andrea, mi hermana Andere y mi padre Joako, siempre disponibles y dispuestos a despejar de un plumazo cualquier duda que pudiera surgir. Mis amigos —Alberto, Coqué, Itxaso, Majo, Nacho, Pablo, Sara—, que no pierden la paciencia cuando me lanzo cuesta abajo y sin frenos a hablar de la tesis. Y por último, Teresa primero y Helena después, por su amor durante estos años.

Introduction and summary
Over the last decades we have witnessed two apparently paradoxical political trends. First, the world-wide extension of democratic institutions and democratic ideology and, second, the global proliferation of non-majoritarian bodies, such as independent central banks and constitutional courts, with the consequent downsizing of the scope of democratic decision-making. In a nutshell, democratic institutions have recently expanded as much as the scope of democratic decision-making has been reduced. These two trends have gone hand-in-hand with two parallel yet more philosophical and long-term trends. First, the crisis of the metaphysical conception of the society as an organic entity with a “general will,” which has lead to the acceptance of pluralism and partisanship as legitimate features of politics and, second, the increasingly inﬂuential view that there are a number of issues that ought to be insulated from ordinary politics and left to non-elected experts either because they are beyond dispute (e.g. human rights), or because they are purely technical (e.g. monetary policy), or yet because they are private (e.g. the family) and should not be submitted to majority rule. Consider the ﬁrst two trends. On the one hand, democracy has rapidly expanded in the West and elsewhere ever since World War II, as ﬁgure 1 shows regarding the proportion of countries with fully universal suﬀrage. While in 1900 only 6 out of 43 existing states were minimally democratic, and only one had fully universal suﬀrage, in 1946 20 out of 71 independent states were already democratic. And in the early 1990s the number increased from 48 in 1989 to 77 in 1994, a ﬁgure that grew up to 92 in 2009 and is likely to continue to grow after the so-called Arab Spring.2 Democratic ideology has also expanded very fast. To the point that blatantly undemocratic governments have often opportunistically referred to themselves as “democratic,” the most obvious instances being the Democratic Republic of Germany, the Democratic People’s Republic of Korea, Franco’s “organic democracy,” or Pinochet’s attempt to “give to democracy an authentically representative dimension.” 3 On the other hand, the scope of democratic decisions has been reduced as much as democracy has been extended. Not only due to the powerful and raising de facto
The ﬁgures come from R. A. Dahl, On Democracy (New Haven: Yale University Press, 1998), p. 14; D. Potter, “Explaining Democratization,” in Democratization, ed. D. Potter, D. Goldblatt, M. Kiloh and P. Lewis (Cambridge: Polity Press, 1997); and M, G. Marshall and B. R. Cole, Global Report 2009: Conﬂict, Governance, and State Fragility (Vienna: Center for Systemic Peace, 2009), pp. 10-11. 3 A. Pinochet, “Chile should Not Fall into the Vices of the Past,” in The Politics of Antipolitics, B. Loveman and D. Thomas, eds. (Wilmington, DE: Scholarly Resources, 1997), pp. 248-249.
2

10

Introduction and summary

presence of international actors, such as transnational corporations or the “gnomes of Zurich,” as the former British Prime Minister Harold Wilson referred to ﬁnancial speculators. But also because of the de iure transfer of power from parliaments to non-majoritarian institutions (i.e. institutions that are neither directly elected by the people, nor directly managed by elected oﬃcials), such as the European Central Bank at the supranational level, or constitutional courts at the domestic level. As Mark Tatcher and Alec Stone Sweet have put it regarding European countries, In the ﬁelds of utility regulation, telecommunications, antitrust, and media pluralism, and even in the provision of health and welfare beneﬁts, myriad independent regulatory bodies have been created and become the loci for making new rules, or applying existing ones to new situations, at the national level. At the supranational level, central bankers, insulated from direct political control, set monetary policy. In Brussels, European Commission oﬃcials propose legislation and enforce ever wider European Union regulation. In Luxembourg, the Court of Justice controls member state compliance with European law, reviewing the lawfulness of activities of national parliaments, governments, and administrators.4

Figure 1. Proportion of countries with universal suﬀrage, by year.5

These two trends have gone hand-in-hand with two parallel yet more philosophical and long-term trends. On the one hand, the acceptance of disagreement and partisanship as legitimate political features vis-à-vis the classic conception of the society as an organic entity with common interests that parties and partisanship could only jeopardize. As we shall see in chapter 1, the organic conception of the society was hegemonic until the late-eighteenth century. This conception can be divided into two core ideas. The ﬁrst idea is that the society is an organic body with a common
4 M. Tatcher and A. Stone Sweet, “Theory and Practice of Delegation to Non-Majoritarian Institutions,” West European Politics 25 (2002): 1–22, p. 1. 5 A. Przeworski, “Conquered or Granted? A History of Suﬀrage Extensions,” British Journal of Political Science 39 (2008): 291-321, p. 292.

11 set of interests and values that are beyond particular interests and factions. The analogy with the body appeared in the late medieval period, but even when the contractual perspective became dominant, parties to a covenant were seen as parts of a whole. It is not by chance that “the people” has always appeared in the singular, as le peuple, el pueblo, das Volk, el poble, herria, and so on—as an organic entity with a uniﬁed will.6 The second idea is that social and political groups advance particular interests and thus threaten the social unity. As Rousseau clearly put it, “[w]hen particular interests begin to make themselves felt and sectional societies begin to exert an inﬂuence ... the common interest becomes corrupted.” 7 In the organicist view democracy is ruled out because it implies political struggle and partisanship and these, in turn, are taken to be inimical to the pre-political and organic interests of the society. As Hans Kelsen would rather critically put it in 1929, [T]he ideal of a general interest superior and transcending interests of groups, thus parties, the ideal of solidarity of interests of all members of the collectivity without distinction of religion, of nationality, of class, etc. is a metaphysical, more exactly, a metapolitical illusion, habitually expressed by speaking ... of an ’organic’ collective or ’organic’ structure.8 Now, as the nineteenth century progressed, the organicist conception gradually gave way to an ever-growing acceptance of the fact of disagreement at the levels of both interests and values, and thus to a more sympathetic view toward political struggle and partisanship—as inevitable and legitimate means to advance such interests and values. Of course, the acknowledgment of the existence of disagreement and partisanship was as old as Aristotle, who stated that “the rich and the poor, are held to be parts of a city-state. Besides ... among the parts of the city-state these two are opposites.” 9 However, political opposition and partisanship would not be taken to be legitimate—rather than as inimical to the social unity—until much more recently. Even Marx and Engels, who famously deﬁned “the history of all hitherto existing society [as] the history of class struggles,” still believed in “the foundation of a new society without classes.” 10 Now, for reasons that will become clear below (§ 1.3.1), pluralism and disagreement would become more pressing as the nineteenth century progressed, and the organicist conception could no longer be credibly sustained. As Schumpeter would conclude in the mid-twentieth century, “[t]here is no such thing as a uniquely determined common good that all people could agree on or be made
See A. Przeworski, “Consensus and Conﬂict in Western Thought on Representative Government,” Procedia Social and Behavioral Sciences 2 (2010): 7042–7055, p. 7046; and Democracy and the Limits of Self-Government (New York: Cambridge University Press, 2010), p. 19. 7 J.-J. Rousseau, Du contrat social, in Oeuvres complètes, vol. 3, B. Gagnebin and M. Raymond, eds. (Paris: Gallimard, 1961). Quoted from J.-J. Rousseau, The Social Contract, M. Cranston, trans. (New York: Penguin, 1968), p. 99. 8 H. Kelsen, Vom Wesen und Wert der Demokratie (Mohr: Tübingen 1920). Quoted from H. Kelsen, La Démocratie. Sa Nature-Sa Valeur (Paris: Economica, 1988), pp. 32-33. 9 Aristotle, Politics, C. D. C. Reeve, trans. (Indianapolis: Hackett, 1948), 1291b. 10 K. Marx and F. Engels, “Manifest der Kommunistischen Partei,” in Werke, vol. 4 (Berlin: Dietz Verlag, 1971). Quoted from K. Marx and F. engels, The Communist Manifesto, S. Moore and F. Engels, trans. (London: Pluto Press, 2008), p. 33; and “Rules of the Communist League,” art. 1.
6

12

Introduction and summary

to agree on by the force of rational argument.” 11 Accordingly, political struggle and partisanship would gain legitimacy, and democracy, traditionally despised by philosophers of all sorts, would eventually obtain a respectable philosophical pedigree. As Habermas would put it more recently, under “conditions of postmetaphysical thinking” no prepolitical substantive goals can be worked out—they can only be the upshot of the democratic process, which is the only source of legitimacy.12 On the other hand, as the organic conception was left behind and disagreement and partisanship characteristic of democratic politics gained respectability, an apparently paradoxical idea emerged—namely, that there are a number of issues that should not be submitted to majority rule, either because they are beyond dispute, because they are merely technical, or yet because they are purely private and ought to be protected from the “tyranny of the majority,” in J. S. Mill’s famous formulation.13 Consider property rights as an instance of this shift. Even though “Democracies [had] ever been found incompatible with personal security, or the rights of property,” as Madison put it, the entrenchment of property rights is nowadays taken not only to be consistent with democracy, but often to be necessary for any good-working democratic system.14 As John Dunn has described it, [n]o one at all in 1750 either did or could have seen democracy as a natural name or an apt institutional form for the eﬀective protection of productive wealth. But today we know better. In the teeth of ex ante perceived probability, that is exactly what representative democracy has in the long run proved.” 15 And, as we shall see in chapter 2, this shift in the scope and limits of democratic decision-making applied to many other issues, including representation, constitutionalism, the division of powers or the employment relationship. In a nutshell, the more legitimacy democracy gained and the more democratic institutions expanded, the more constrained the scope of democratic decision-making turned out to be. The upshot of these two trends, both at the political and philosophical levels, is what we usually label as liberal democracy. A political system in which sovereignty rests on the people and is exercised by the people, either directly or through their elected representatives, yet the scope of democratic decisions is heavily constrained by a number of issues that are removed from majority rule and protected by non-majoritarian bodies, such as constitutional courts and independent central banks. The two main goals of this dissertation are to historically explain the emergence of these two features of liberal democracy in the case of the Early American Republic (Part I) and to more normatively analyze two outstanding sorts of liberal constraints on the scope of democracy—namely, constitutional constraints on demoJ. A. Schumpeter, Capitalism, Socialism, and Democracy (New York: Harper & Brothers, 1942), p. 250. 12 See J. Habermas, Faktizität und Geltung. Beiträge zur Diskurstheorie des Rechtes und des demokratischen Rechtsstaats (Frankfurt am Main: Suhrkamp, 1992). 13 J. S. Mill, “On Liberty,” in Collected Works of John Stuart Mill, vol. 18, J. Robson, ed. (Toronto: University of Toronto Press). 14 J. Madison, “The Federalist no. 10,” in The Federalist, G. W. Carey and J. McCleallan, eds. (Indianapolis: Liberty Fund, 2001), p. 67. 15 J. Dunn, “Democracy before the Age of the Democratic Revolution,” Paper delivered at Columbia University, 2003, p. 10.
11

13 cratic legislatures (Part II) and decision-making in the workplace, a sphere that is taken to be beyond the scope of democracy (Part III). Part I focuses on the Early American Republic, a period in which the two above-mentioned trends are outstandingly acute and which world-wide political inﬂuence can be hardly exaggerated. It has two main goals, to which chapter 1 and chapter 2 are respectively devoted. First, to show how democracy, a form of government that had been despised for more than two thousand years, gained legitimacy and was increasingly demanded as the fact of disagreement was accepted as an unavoidable feature of the American society due, among other causes, to the development of the market economy and the emergence of the ﬁrst and second party systems. Second, to show that the idea of democracy also went through a dramatic conceptual shift that made it include a number of elements that had been traditionally seen as constraints on democracy. This shift helps explain a well-know yet still striking phenomenon. Namely, that the founders of the American constitutional system openly referred to it as a non-democratic one and that, only half a century later, virtually everyone—the heirs of the Federalists included—turned out to refer to very much the same system as a fully democratic one. In a nutshell, what the Founders meant by “democracy” in 1787 was not what the term would turn out to mean half a century later. And the shift also helps explain why a number of limits on the scope of democracy that will be analyzed in Part II and Part III, and that had been traditionally seen to be at odds with democracy, turned out to be seen (and are seen nowadays) not only as consistent with democracy but, in fact, as necessary for any good-working democratic system. Part II and Part III more normatively analyze two outstanding instances of such limits, and show that there are good grounds to call into question the justiﬁcation of those limits from a democratic standpoint. Even though liberal democracies are characterized by a large number of spheres that are insulated from the scope of political decision-making (e.g. religion, sexuality, monetary policy, property rights), two especially relevant cases are addressed in Part II and Part III—ﬁrst, constitutional constraints on legislative action and, second, decision-making in the workplace. These two cases are relevant under present circumstances because both constitutional rigidity and judicial review, on the one hand, and the undemocratic and unpolitical nature of decision-making in the workplace, on the other, have turned out to be seen as perfectly natural and legitimate features of our liberal democracies. And they are relevant to this dissertation because they are both closely related to the period of the American history studied in Part I. On the one hand, judicial review of legislative action originated in the Early American Republic—after the Marbury v. Madison Supreme Court decision of 1803—as a check on democracy that only later turned out to be considered democratic, as we shall see in section 1.2.2.2. On the other hand, the employment relationship, which had been traditionally seen as a form of “limited slavery,” as Aristotle put it, turned out be seen as a free and voluntary exchange among

14

Introduction and summary

juridical equals as the market economy expanded in nineteenth-century America and elsewhere, as we shall see in section 1.3.3.16

Figure 2. Approximate number of countries with a written constitution (A)

compared with the approximate number of constitutional democracies (B).17

Part II will analyze the justiﬁcation of constitutional constraints on legislative action—and, more precisely, constitutional rigidity and judicial review—from a democratic point of view. This is a very pressing issue, given the recent world-wide expansion of constitutionalism, which has expanded as much as democracy (and actually more than democracy, as Figure 2 above shows). and that on the reasons for entrenching institutional arrangements and substantive rights “constitutional theory remains in a surprisingly primitive state,” as Cass Sunstein has put it.18 Three prominent attempts to justify constitutional rigidity and judicial review on democratic grounds are addressed—pure instrumentalist theories, precommitment-based theories, and proceduralist theories. According to these theories, constitutionalism need not be at odds with democracy, and is in fact necessary for a good-working democratic system, because constitutional constraints are necessary to ensure that democratic procedures bring just outcomes about (chapter 4), or because they are the upshot of democratic precommitments made by the people themselves (chapter 5), or yet because they set up the procedural rules and preconditions required for the proper functioning of the democratic system (chapter 6). As Cherie Booth, the barrister wife of the former British Prime Minister Tony Blair, put it, [R]esponsibility for a value-based substantive commitment to democracy rests in large part on judges ... [J]udges in constitutional democracies are set
Aristotle, Politics, 1260a-b. D. S. Lutz, Principles of Constitutional Design (Cambridge: Cambridge University Press, 2006), p. 4. 18 C. Sunstein, Designing Democracy. What Constitutions Do (Oxford: Oxford University Press, 2001), p. 97.
17 16

15 aside as the guardians of individual rights ... [and] aﬀorded the opportunity and duty to do justice for all citizens by reliance on universal standards of decency and humaneness ... in a way that teaches citizens and government about the ethical responsibilities of being participants in a true democracy.19 Even though a number of problems of each of these theories will be raised, a problem common to all of them is that they do not address adequately the fact of disagreement that, as it should become clear in Part I regarding the American case, turns out to be a necessary condition of democratic legitimacy. They all assume or end up assuming that there are a number of political issues that are beyond dispute and that can be entrenched in a constitutional text and advanced by constitutional means. However, as it is carefully shown throughout Part II, even though constitutions tend to be presented as complete contracts which content is beyond dispute, they turn out to be inevitably pervaded by the fact of disagreement across the board—e.g. at the drafting stage, as well as during their ratiﬁcation, interpretation, and enforcement. Even though this problem does not rule out the possibility of these theories being justiﬁed by other, non-democratic, means, it shows that they cannot be justiﬁed on democratic grounds. Part III of the dissertation addresses another outstanding instance in which the scope of democracy is heavily constrained—namely, decision-making in the workplace. The possibility of extending democracy to the workplace is especially relevant to this dissertation because in liberal democracies the workplace is considered a sphere of voluntary relations among free and equal individuals that falls beyond the scope of politics. Further, this conception of the workplace as a sphere in which power relations are absent and employers and employees freely agree to a voluntary exchange did not emerge until the market economy expanded in nineteenth-century America and elsewhere, as we shall see in section 1.3.3, turning into dominant ever since. Hence, in Alchian and Demsetz’s famous description of the employment relationship, It is common to see the ﬁrm characterized by the power to settle issues by ﬁat, by authority, or by disciplinary action superior to that available in the conventional market. This is delusion ... The ﬁrm ... has no power of ﬁat, no authority, no disciplinary action any diﬀerent in the slightest degree from ordinary market contracting between any two people ... Telling an employee to type this letter rather than to ﬁle that document is like my telling a grocer to sell me this brand of tuna rather than that brand of bread.20 The most prominent arguments for and against workplace democracy will be critically addressed in chapter 7 of this Part. However, the main contribution of Part III is to develop a rather original, republican argument for workplace democracy in chapC. Booth, “The Role of the Judge in a Human Rights World,” Speech to the Malaysian Bar Association, July 26, 2005. Quoted in R. Bellamy, Political Constitutionalim. A Republican Defence of the Constitutionality of Democracy (Cambridge: Cambridge University Press, 2007), p. 1. 20 Alchian and Demsetz, “Production, Information Costs, and Economic Organization,” American Economic Review 62 (1972): 777-795, p. 777.
19

16

Introduction and summary

ter 8. Since republicanism was the mainstream political philosophy during the Early American Republic (even though it goes back to Aristotle, Cicero, Machiavelli, and Harrington, among many other philosophers), this chapter will draw heavily upon Part I of the dissertation in deﬁning the republican notion of freedom as immunity from arbitrary interference. The main research question in this chapter is whether workplace democracy turns out to be a necessary condition if immunity from arbitrary interference is to ensured within the ﬁrm. Borrowing ideas from incomplete contract theory, it will be shown that labor contracts, as well as public attempts to regulate the workplace by means of labor legislation, are incomplete and thus open the door to arbitrary interference. Put simply, since it would be impossible or prohibitively costly for the parties of the employment relationship to make their contract complete, and for public agencies to completely regulate the relationship, and since some ﬂexibility is also desirable as to adequately address day-to-day unforeseeable eventualities, discretion in the workplace is not only unavoidable but also desirable. Now, the existence of discretion in the enforcement of employment contracts and workplace regulation opens the door to disagreement among parties and to the possibility of arbitrary interference taking place. Thus, if immunity from arbitrary interference is to be ensured in the workplace, workers are to be given a say not only at the outset of the employment relationship or over the elaboration of labor legislation, but also on a daily basis, as the employment relationship goes along. Summarizing, as it should be apparent by now, even though the main goal of the dissertation is to analyze the scope and limits of democracy both historically and normatively, the three parts in which it is divided are intended to be self-standing and independent from each other. However, the fact of disagreement and its relationship with democratic decision-making and democratic legitimacy is common to all of them and runs through the whole dissertation. In the following sense. Part I of the dissertation shows how the fact of disagreement came to be accepted in nineteenth-century America. And it shows how this acceptance gave way to the crisis of the theory of virtual representation and to an increasing acceptance of democracy as a means to make decisions in the face deep disagreements on which course of governmental action ought to be taken. Put diﬀerently, it shows that the demand for democracy only triggered once the fact of disagreement had turned to be seen as an unavoidable and legitimate feature of the American society. As it will be put more analytically in Part II (sections 4.3.1 and 4.3.2), the fact of disagreement turns out to be a necessary condition of democratic authority being called for, for if all citizens agreed on what ought to be done across the board, no authority would have to set up in the ﬁrst place. Thus Part II focuses on how constitutions, which are often presented as contracts which content is beyond dispute, are unavoidably pervaded by the fact of disagreement across the board. Accordingly, it argues that there is no reason to insulate constitutional rules from democratic decision-making by ordinary citizens and their representatives by means of rigidity and judicial review. Finally, Part III shows that the workplace, a sphere that is usually taken to be beyond the scope of

17 politics, is equally pervaded by the fact of disagreement and provides an argument for its democratization based on the republican notion of freedom as immunity from arbitrary interference. In the remainder of the Introduction, I summarize the main arguments of the three parts of the dissertation in some detail. Part I: Democracy Part I of the dissertation analyzes the use of the term “democracy” in the Early American Republic from its rather derogatory sense in the constitutional period to its widely-assumed positive one in the 1830s. It is divided into two chapters, which respectively account for the two above-mentioned trends. Chapter 1 provides an explanation for the increasing support of the term, along with the institutional devices traditionally attached to it. Chapter 2 shows that, besides the increasing acceptance of the term, the very concept of democracy and its scope also shifted—and reconstructs the early and later concepts of democracy. A striking feature of the American constitutional system is that its framers openly referred to it as a non-democratic one. The Federalists not only blamed “democracy—a Beast without a head ... the Mob ruling.” They also conceived the Federal Constitution as a means “to restrain” and “to keep down the turbulence of democracy.” 21 Not less striking is that, only a few decades after the ratiﬁcation of the Federal Constitution, virtually everyone—the heirs of the Federalists included—turned out to refer to very much the same system as a fully democratic one. Thus Jeﬀerson would conclude in 1826 that “We of the United States ... are constitutionally and conscientiously democrats.” 22 How could it be that a constitutional system formerly known as nondemocratic had turned in less than half a century into the democratic system par excellence? This part of the dissertation shows that, once we realize of the commonly derogatory sense of the term “democracy” in the 1780s, it should not come with surprise that the Federalists hesitated to use the term to label the political system they were putting forward. Likewise, we should not be surprised by the fact that virtually everyone referred to very much the same system as a democracy ﬁfty years later, when the term gained a widespread positive sense and was used as a battle-cry by Andrew Jackson ﬁrst and the Whig Party next. Thus Part I focuses on the two main trends behind this shift, which are analyzed in detail in chapter 1 and chapter 2. Chapter 1 will show that, despite its rather derogatory sense in the 1780s, the term “democracy”—along with some of the institutional devices traditionally attached to it—gained increasing support as soon as ordinary people became increasingly aware of the plurality of interests and values across the American society, organized themselves to advance such interests and the theory of virtual representation used by the Federalists in the 1780s gradually lost its political eﬃcacy. According to
W. Hooper, quoted in F. Dupuis-Deri, The Political Power of Words: The Birth of Pro-Democratic Discourse in the 19th century in the United States and France (PhD diss., University of British Columbia, 2001); R. Morris, in M. Farrand, ed., The Records of the Federal Convention of 1787, (New Haven: Yale University Press, 1911), vol. 1, pp. 58 and 517. 22 Quoted in Dupuis-Deri, The Political Power of Words, p. 227.
21

18

Introduction and summary

the theory of virtual representation, the society was an organic body with a common set of interests and values that were beyond particular interests and factions. Thus, however excluded from actual political decision-making, all members of the political community were represented—if not actually, virtually—by their taking part in such shared interests and values of the community. Now, as the nineteenth century wore on, ordinary people became more aware of the plurality of interests and values in the American society, organized themselves to advance such interests and values, and virtual representation lost great part of its credibility. Accordingly, actual representation through suﬀrage extension and other devices traditionally attached to democracy were increasingly demanded, and “democracy” lost part of its derogatory sense and gained wider support. The chapter is in two sections. Section 1.2 analyzes in detail the theory virtual of representation, which was hegemonic during the revolution and its aftermath. The theory can be divided into three tenets—the organicist tenet, the ruling elite tenet and the independence tenet—, which are analyzed in turn. First, the organicist tenet is analyzed and its pervasive use is showed during the revolutionary war, during the so-called “critical period,” and during the Federal convention of 1787. Second, the ruling elite tenet is analyzed, according to which political decisions should be made only by those with the expertise and impartiality required to let aside their factional interests and to represent the organic interests of the community. The alleged reasons behind the ruling elite tenet are analyzed, along with the main institutional means put forward so as to enact it. Third, the independence tenet is analyzed, according to which only those who are materially independent (which was identiﬁed with owning real estate) are impartial enough to represent the organic interests of the community rather than their particular ones. The three main reasons used to justify this requirement are analyzed—put brieﬂy, freedom from inﬂuence by those on whom otherwise it was likely to depend, having the leisure time required to get informed about and take part in politics, and having a “stake in society” so as to have proper incentives to take part in politics responsibly. As Kant would put it by the same time, “[t]he only justiﬁcation required by a citizen ... is that he must be his own master (sui iuris), and must have some property ... to support himself.” 23 Section 1.3, in turn, analyzes the crisis of the three tenets of the theory of virtual representation and how it made “democracy” lose part of its derogatory sense. First, three causes are found behind the crisis of the organicist tenet—the development of the market economy, the inﬂuence of the French revolution and the subsequent creation of democratic societies across the country, and the emergence of the ﬁrst and second party systems. Second, the crisis of the ruling elite tenet was to great extent substituted by the idea that ordinary people rather than impartial experts ought to be elected, and that institutional devices ought to be enacted so as to ensure that
I. Kant, “Über den Gemeinspruch: das mag in der Theorie richtig sein, taugt aber nicht für die Praxis,” in Kant’s gesammelte Schriften, vol. VIII, Königlichen Preußischen Akademie der Wissenschaften, ed. (Walter De Gruyter: Berlin, 1923). Quoted from I. Kant, “On the Common Saying: ’This May Be True in Theory, but It Does Not Apply in Practice’,” Political Writings, H. S. Reiss, ed. (Cambridge: Cambridge University Press, 1970), p. 78.
23

19 representatives were held responsive to the interests of their constituencies. Finally, two causes are found behind the crisis of the independence tenet. First, that wage laborers—which wages had been traditionally taken to be “a pledge of the their slavery,” as Cicero had put it—were now taken to be materially independent, and thus as worth being granted full citizenship, because they were considered to be owners—owners of their own labor force, which they were able to freely exchange in the market.24 Second, because it was increasingly believed that all members of the political community were worth being granted full citizenship, despite them being materially independent or not. In short, the combination of all these factors made the theory of virtual representation lose credibility and democracy, in turn, gain respectability. Chapter 2 will show that, besides its increasing acceptance, “democracy” also went through a notable semantic shift during the decades after the ratiﬁcation of the Federal Constitution. Although some features of the concept of democracy—and some of the institutional devices associated to it—remained the same, some others shifted. This helps explain why the Federalists believed that the representative government they were putting forward was not only diﬀerent from democracy but in fact a very check against it. As Bernard Manin puts it, “for Madison, representative government was not one kind of democracy; it was an essentially diﬀerent and furthermore preferable form of government.” 25 And it also helps explain why, only some decades later, virtually everyone referred to very much the same system as an openly democratic one. In a nutshell, what the Federalists meant by “democracy” in the 1780s is not exactly what the term turned out to mean half a century later. Likewise, the institutional devices attached to the former concept were not the same as those attached to the latter concept and, in fact, many of the institutional devices that had been traditionally seen as constraints on democracy turned out to be seen not only as consistent with democracy but also as necessary for any democratic system proper. The chapter is in two sections. Section 2.2 brieﬂy traces the evolution of the concept from its rather derogatory sense in the 1780s to its virtually undisputed positive one ﬁfty years later. It does so by analyzing three key steps. First, the inﬂuence of the classic republicanism of Aristotle, Polybius, and Cicero—on which the Founders had been intellectually raised—over their conception of democracy. Second, Jeﬀerson’s presidency and the explicit debate among Pennsylvania Jeﬀersonians over the meaning and implications of democracy. And, ﬁnally, Jacksonian democracy and the second party system, in which democracy was used as a battle-cry for the very ﬁrst time. Next, section 2.3 compares the early and the later concepts as for ﬁve outstanding issues—representation and popular sovereignty, feasibility, property rights, stability and the division of powers. As we shall see, the shift in the concept of democracy regarding these issues helps explain why only a few decades
24

Cicero, De Oﬁciis, W. Miller, trans. (Cambridge, MA: Harvard University Press, 1913), I.

XLII.

B. Manin, The Principles of Representative Government (Cambridge: Cambridge University Press, 1996), p. 2.

25

20

Introduction and summary

after the ratiﬁcation of the Federal Constitution virtually everyone could refer to very much the same constitutional order as an openly democratic one. As William Duane wrote to Jeﬀerson in 1824, “Democratic government is no longer Jacobinism; and those who formerly reprobated now use the language and profess the doctrine they reviled twenty four years ago.” A “revolution in speech,” as Duane labeled it, had been produced.26 Part II: Constitutionalism Part I of the dissertation shows, among other things, how some of the devices that were originally designed as checks on democracy turned out to be considered not only as consistent with democracy, but also as necessary for any good-working political system. Part II focuses on two of the above-mentioned devices—constitutional rigidity and judicial review—and on the very notion of constitutionalism, which recent world-wide expansion can be hardly exaggerated. In 1802 no constitutional system proper existed in the world, and in 1978 only 26 percent of constitutions provided for a constitutional court. By contrast, 101 out of the 106 constitutions ratiﬁed between 1985 and 2008 have included constitutional courts with ﬁnal powers of review of the legislative action.27 This part is divided into four chapters, which analyze—and criticize—in detail the most salient normative theories of constitutional democracy. According to these theories, constitutionalism—including constitutional rigidity and judicial review of legislation—need not be at odds with democracy, either because constitutional constraints are necessary for democratic procedures to bring about just outcomes, such as the protection of human rights or the improvement of the economic output (chapter 4), or because they are the upshot of commitments that actual citizens have made democratically (chapter 5), or yet because they set up the procedural rules and preconditions required for the proper functioning of the democratic system (chapter 6). Before turning to analyze these theories, chapter 3 provides the analytical background used throughout Part II. Constitutionalism is deﬁned as the limitation of the powers of the legislative body entrenched in a rigid document and enacted through judicial review. Section 3.2 brieﬂy presents the history of constitutionalism, compares it with absolutist constitutionalism, the Westminster model of parliamentary sovereignty, and the so-called “new commonwealth” model of constitutionalism, and describes the two main sorts of constraining provisions of democratic decision-making—substantive provisions and procedural ones. In turn, section 3.3 brieﬂy analyzes the diﬀerence between written constitutions and constitutional conventions, along with the role of the latter in constitutionalism.
William Duane, “Letter to Jeﬀerson, October 19, 1924,” Proceedings of the Massachusetts Historical Society 20 (1906-1907), 381-383, p. 382. 27 The ﬁgures come from D. L. Horowitz, “Constitutional Courts: A Primer for Decision Makers,” Journal of Democracy 17 (2006): 125-137; and A. Stone Sweet, “Constitutionalism, Rights, and Judicial Power,” in Comparative Politics, D. Caramani, ed. (Oxford: Oxford University Press, 2008).
26

21 The remainder of the chapter analyzes the two main institutional devices of constitutionalism—constitutional rigidity and judicial review. Section 3.4 deﬁnes constitutional rigidity as the greater stringency of the constitutional amendment procedure in contrast to the ordinary legislative procedure, being this stringency a necessary condition for the normative supremacy of the constitution and, thus, for constitutionalism. It presents the speciﬁc devices used to enact rigidity and discusses in some detail the classic philosophical problem of whether a sovereign can really bind itself for, as Hobbes put it, “he that can bind can release.” 28 Put simply, it is argued that under non-ideal circumstances, constitutional constraints do pose costs and constraints on ordinary decisions made by the people and the state, and that it is naive to believe that they do not impose costs at all—just as it is to believe that constitutions are all-powerful institutional devices. Finally, section 3.5 analyzes judicial review, which is deﬁned as the authority of courts to assess the constitutionality of legislation, including the ability to decline to apply statutes, to modify the eﬀect of their application so as to make them compatible with the constitution, and, in the limit, to strike them down altogether. The section distinguishes strong from weak judicial review, American-style from European-style judicial review, and analyzes its eﬀectiveness in the light of the empirical evidence that shows that, on average, constitutional courts tend not to make decisions against the public opinion. Chapter 4 addresses instrumentalist theories of democratic legitimacy and their justiﬁcation of substantive constitutional constraints. According to these theories, that update the organicist view analyzed in Part I, democratic decision-making has legitimate authority if and only if it brings about a set of substantive and pre-political goals, such as the protection of fundamental rights, economic growth, or the stability of the political system. Thus a trade-oﬀ between constitutionalism and democratic legitimacy need not arise, for the legitimacy of democracy rests exclusively in its ability to achieve some substantive goals that are entrenched in the constitution and thus advanced by constitutional means. The chapter opens by setting up the structure of instrumentalism in section 4.2, and turns next to two prominent instrumentalist accounts of the legitimacy of constitutionalism—namely, that it helps democratic decision-making protect fundamental rights and that it improves the economic output of the system. Finally, the following three problems of instrumentalism are presented. Section 4.3.1 argues that instrumentalism is unable to address Wollheim’s paradox—the puzzling fact that many citizens whole-heartedly observe the existing legal order while at the same time considering that many of its laws are blatantly unjust. Put more formally, the paradox presents a citizen X who both believes that Policy A ought to be enacted (because A is her preferred option) and that policy B, which is incompatible with A, ought to be enacted (for she is a democrat and B is the option backed by the majority). Two previous solutions to the paradox are critically analyzed—the Rousseauian solution and the Hobbesian solution—and a novel, two-level solution is put forward. According to this solution the above-mentioned
28

T. Hobbes, Leviathan, R. Tuck, ed. (Cambridge: Cambridge University Press, 1996), p. 184.

22

Introduction and summary

beliefs can be consistently held because they provide an answer to two very diﬀerent normative questions—namely, to “what ought to be done” and to “what ought to be done provided that we disagree on what ought to be done and that, nevertheless, we have to take a common course of action.” Next, it is shown that, once we accept the two-level solution, pure democratic instrumentalism turns out to be an incomplete normative account of political authority, for it is unable to address why citizens often consistently hold not only the former belief but also the latter one. Section 4.3.2 elaborates on the criticism presented at the end of the previous section—that pure instrumentalist theories of democratic legitimacy turn out to be self-defeating because they do not address adequately the existing disagreements on which outcomes are to be achieved through democratic procedures. It is divided into ﬁve subsections. The ﬁrst four ones deﬁne the circumstances or politics—which arise when there is a need for a binding collective agreement on a uniﬁed course of action yet the parties disagree on which precise course of action ought to be taken—, along with the scope, causes and implications of such circumstances. The last subsection explains why pure instrumentalism turns out to be self-defeating under the circumstances of politics. As Waldron has put it, “any theory that makes authority depend on the goodness of political outcomes is self-defeating, for it is precisely because people disagree about the goodness of outcomes that they need to set up and recognize an authority.” 29 Finally, section 4.3.3 shows that pure instrumentalism turns out to be too radical an account of political legitimacy when we consider it from the point of view of actual citizens, who tend to take into account not only the outcomes of political decisions but also the fairness of the procedures employed to bring such outcomes about. The section reviews the social psychology literature on the topic, which appears to converge on the centrality of procedural fairness for the perceived legitimacy of the decisions made. Chapter 5 analyzes precommitment-based theories of constitutional constraints, which are justiﬁed for being the upshot of an actual decision made by citizens themselves. Thus from a precommitment-based standpoint, constitutional constraints need not be at odds with democracy when they are adopted—through the use of equal voting and (perhaps qualiﬁed) majority rule—by those citizens against whom such constraints are to be enforced, and only by them. As Hayek put it, “Only a demagogue can represent as ’antidemocratic’ the limitations which long-term decisions and the general principles held by the people impose upon the power of temporary majorities.” 30 The chapter is divided into three sections. First, the structure of democratic precommitments is outlined. Second, Ackerman’s account of dualist democracy is presented as the most elaborated and inﬂuential instance of

J. Waldron, Law and Disagreement (Oxford: Oxford University Press, 1999), p. 253. F. Hayek, The Constitution of Liberty (Chicago: University of Chicago Press, 1960), pp. 106-107.
30

29

23 a precommitment-based approach.31 Finally, ﬁve diﬃculties of this approach are analyzed. Section 5.4.1 argues that precommitment-based theories fall into a categorial confusion, for political arrangements—including constitutional constraints—do not turn into any more democratic just because they have been chosen by democratic means, for a majority of citizens could democratically pass blatantly undemocratic constitutional provisions without such provisions turning into democratic for that reason. Section 5.4.2 analyzes the much-cited analogy between collective commitments to constitutional rules and individual commitments to personal rules. This is a very powerful analogy, for self-constraining decisions tend to be seen as more autonomous than unconstrained ones. As Montesquieu famously put it, “political liberty does not consist of doing what one wants ... liberty cannot be other thing but to be able to do what one ought to want.” 32 However, the analogy does not hold because, unlike individual commitments, collective commitments made in constituent moments are pervaded by the fact of disagreement, at least, for three reasons. First, because if a consensus existed qualiﬁed majorities could be used for the ratiﬁcation of constitutions, yet simple majority tends to be used, which shows that constitutional commitments may not always enjoy a wider consensus than ordinary statutes during ordinary politics. Second, because the circumstances under which constitutions are adopted or amended (e.g. transition from autocratic regimes, economic crisis, civil wars) tend not to be the better ones for citizens to leave such disagreements aside. And, third, because the stakes are too high—due to the immeasurable impact and hardly amendable nature of constitutions—for parties to let aside their particular interests. Rather than giving way to wider consensus in which the people speak with a uniﬁed voice, constitutional moments are better depicted as bargaining processes and constitutions as equilibria. Section 5.4.3 shows that, even if fragile equilibria among disagreeing parties, constitutions can still be democratic commitments as long as the resulting equilibria are adopted by a majority of citizens. However, it is shown that failures to meet these criteria are usual both along synchronic and diachronic lines. Three types of failures are analyzed among the former are analyzed in this section. First, when a minority of citizens imposes their interests in the constitution against a majority of their fellow citizens. Second, when a majority of citizens, even a qualiﬁed one, wants to amend the constitution yet is unable to do so because the amendment is never carried out or is blocked by the political elites. And third, when an external agent is able to impose their interests in the constitution against those of the citizenry as a whole, as it happened to Japan in the aftermath of World War II. In turn, section 5.4.4 analyzes additional failures to meet the above-mentioned criteria—only that now across generations. As Thomas Paine pointed out, “Every
31

B. Ackerman, We the People. Foundations (Cambridge, MA: Harvard University Press, C.-L. Montesquieu, De l’esprit des lois, Laurent Versini, ed. (Paris: Gallimard, 1995), p.

1991).
32

325.

24

Introduction and summary

age and generation must be free to act for itself, in all cases, as the ages and generations which preceded it.” 33 Accordingly, for a constitution to be fully democratic across generations founding and amendment costs ought to be symmetric. That is, the existing provisions ought to be amended by the subsequent generations provided that the amendment enjoys the support of a majority. However, there is an asymmetry between adopting and amending a constitution, both formally and informally. Formally, because, as a general rule, constitutions only require simple majorities for their ratiﬁcation yet qualiﬁed majorities for their amendment. Informally, because constitutions tend to be much more resilient to change than what is formally speciﬁed, among other phenomena that are analyzed, due to path dependence, status quo biases, and the expressive function of law. Further, even though it may be just for subsequent generations to assume the costs of the burdens assumed by the founding generation as much as they assume their beneﬁts, from a democratic point of view what counts is whether such generations can amend the constitution by means that are not more stringent than those of the adoption process. However, this symmetry does not apply. Finally, chapter 6 analyzes proceduralist theories of constitutional constraints. According to these theories, democratic institutions cannot be said to be really democratic in the absence of a set of constitutive conditions (e.g. “general elections, freedom of the press, freedom of assembly, and freedom of speech,” as Rosa Luxemburg put it criticizing the suspension of civil liberties by the Russian Bolsheviks) and preconditions (e.g. “decent wages and universal reading,” as John S. Mill put it).34 Further, it is also argued that, unless entrenched in a rigid constitution and enacted by a constitutional court, such conditions and preconditions can be jeopardized by temporary parliamentary majorities that can gerrymander, disenfranchise minorities, or ban opposing parties and media. In short, unless constitutionally constrained, ephemeral parliamentary majorities can downsize or even destroy democracy from within. The chapter is divided into four sections. Section 6.2 brieﬂy analyzes the much cited instance of a democratic system allegedly destroyed by democratic means—namely, the fall of the Weimar Republic in the hands of the Nazis. As Goebbles himself liked to tell, “[i]t will always be one of the best jokes of democracy that it gives its deadly enemies the means to destroy it.” 35 Section 6.3 analyzes preconditions-based theories of constitutionalism, more speciﬁcally, the rather standard account of Holmes and Sunstein and Habermas’ more peculiar one, according to which “[a] constitutional court guided by a proceduralist understanding of the constitution does not have to overdraw on its legitimation credit.” 36 Section 6.4 presents John H. Ely’s (and
T. Paine, “The Rights of Man,” Political Writings, B. Kuklick, ed. (Cambridge: Cambridge University Press, 2000), p. 63. 34 Luxemburg is quoted in C. E. Schorske, German Social Democracy, 1905-1917: The Development of the Great Schism (Cambridge, MA: Harvard University Press, 1955), p. 324. Mill is quoted in Przeworski, Democracy and the Limits of Self-Government, p. xiii. 35 J. Goebbles, quoted in A. S. Kirshner, “Proceduralism and Popular Threats to Democracy,” Journal of Political Philosophy 18 (2010): 405-424, p. 405. 36 J. Habermas, Faktizität und Geltung. Quoted from J. Habermas, Between Facts and Norms, W. Rehg, trans. (Cambridge, Ma.: MIT Press, 1996), p. 279.
33

25 Stephen Holmes’, though more brieﬂy) “participation-oriented, representation-reinforcing” approach to judicial review as the most inﬂuential instance of a purely procedural account of constitutionalism.37 Ely’s main argument is that courts are provided with a better institutional position vis-à-vis legislatures to ensure the fairness of the political process because elected oﬃcials are judges in their own cause when having to ensure that the political market remains fair and competitive, for they have a clear incentive to use their position while they are in oﬃce to increase their political power. By contrast, appointed judges are less prone to conﬂicts of interests of this sort because their permanence in oﬃce does not depend on the political cycle (even more when they enjoy life-long tenure). Four criticisms of proceduralist theories are raised next. Section 6.5.1 argues that preconditions-based accounts fall into the paradox of preconditions pointed out by Carlos Nino.38 Put brieﬂy, the more issues are constitutionally entrenched, the better democracy functions, yet the less issues are left for citizens to decide. And, conversely, the more issues are left for citizens to decide, the less issues are constitutionally entrenched and the worse democracy functions. Section 6.5.2 analyzes the view according to which, although disagreements are inherent to substantive political issues, they need not pervade the decision-making procedures used to deal with such disagreements. However, it shown that such disagreements do pervade procedures at the levels of preconditions, procedures, and values behind procedures. Section 6.5.3 analyzes Stephen Holmes’ conception of constitutions as constitutive rules. In this view, while substantive rules are regulative and may constrain the exercise of the democratic power, procedural rules are constitutive—they create and organize it. We cannot make democratic decisions without democratic rules, just as we cannot play chess without the rules of chess. However, Holmes’ argument falls into a non sequitur. Put simply, even though procedural rules are certainly necessary (say, electoral rules) to arrive at democratic decisions, there is no reason why such rules ought to be entrenched in a rigid constitution, enacted by courts and be amendable only by very cumbersome means. Section 6.5.4 closes the chapter by analyzing Ely’s argument about the more adequate institutional position of courts vis-à-vis legislatures. The US Supreme Court decision in Bush v. Gore is analyzed and it is conluded that, as part of the political system, Supreme Court justices are aﬀected as much as legislatures by the way the political system is organized and how power is allocated—let alone the fact that they tend to disagree with each other as much as MPs and citizens do, as their use of majority rule to reach decisions shows. Part III: The workplace Part III of the dissertation analyzes the extension of democratic decision-making to the workplace, a sphere that is usually taken to be beyond the scope of democracy. It is divided into two chapters. Chapter 7 introduces and deﬁnes the main forms
J. H. Ely, Democracy and Distrust. A Theory of Judicial Review (Cambridge, MA.: Harvard University Press, 1980); S. Holmes, “Precommitment and the Paradox of Democracy,” in Constitutionalism and Democracy, J. Elster and R. Slagstad, eds. (Cambridge: Cambridge University Press, 1988). 38 C. S. Nino, La constitución de la democracia deliberativa (Barcelona: Gedisa, 1997).
37

26

Introduction and summary

of workplace democracy (WD), and reviews the core arguments for and against this form of managerial organization. Chapter 8, in turn, puts forward an original, republican case for WD. Since republicanism was the mainstream political philosophy during the Early American Republic (even though it goes back to Aristotle, Cicero, Machiavelli, and Harrington, among many other philosophers), this chapter heavily draws upon Part I of the dissertation. Chapter 7 provides the analytical and normative framework used throughout this part. Section 7.2 deﬁnes WD as a form of managerial organization in which a majority of workers, though not necessarily only them, have equal control rights over the management of the ﬁrm, and distinguishes between control rights by capital or by labor and between public and private ownership of the ﬁrm. In the remainder of the section I focus on private democratic ﬁrms, which are controlled by workers (rather than by capital) and are privately (rather than publicly) owned. Next, a further distinction is made based on whether democratic ﬁrms are owned by external investors or by workers themselves (who may ﬁnance collective asset ownership through debt contracting or by drawing upon their own savings). The German co-determination system and cooperativism are taken as the most outstanding instances of the former and the latter alternatives, and are carefully deﬁned in sections 7.2.1 and 7.2.2. Section 7.3 reviews the main arguments for and against WD in some detail. Section 7.3.1 tackles its feasibility, considering that the proposal is not in the agenda, that it does not seem to be supported by workers themselves, and that there may be more pressing issues right now. Section 7.3.2 addresses a number of eﬃciency-based arguments for WD (greater eﬀort and information-sharing, lower unemployment output, and better vertical and horizontal monitoring), as well as against WD (greater decision-making costs, diﬃculties to attract external capital and to diversify investment, and likely free-riding due to common ownership in large democratic ﬁrms). In addition, three potential eﬀects of WD are analyzed—over workplace security, over workers’ political apathy, and over the raising political inﬂuence of corporations. Sections 7.3.3 and 7.3.4 address in some detail the so-called “epistemic” and “propertarian” arguments against WD, respectively. According to the former, why should workers be granted a say when they often lack the expertise required to make informed decisions about institutions as complex as ﬁrms? According to the latter, why should workers be granted control rights over the ﬁrm when they are not its owners? Finally, section 7.3.5 critically analyzes Bowles and Gintis’ theory of contested exchange, which they use to make a case for WD. Chapter 8 focuses exclusively on the republican case for WD. Section 8.2 deﬁnes the republican notion of freedom in contrast to its liberal counterpart—namely, as immunity against arbitrary interference. Next, section 8.3 shows the forms of arbitrary interference that are likely to appear in the ﬁrm and the problems of two competing views to WD in addressing them—ﬁrst, that the availability of an exit option alone ensures workers’ freedom from arbitrary interference and, second, that ensuring this sort of freedom requires workplace regulation but not WD. Section 8.3.1 analyzes the exit option argument. According to this argument,

27 in free and competitive markets (and with a proper welfare system, would more socialdemocratic versions add) workers’ right to quit serves as an ever-present threat against any ex post managerial decision that could lead to arbitrary interference and abuse, ruling out the possibility of arbitrary interference taking place. Three objections are raised. First, both imperfect and perfect labor markets have involuntary unemployment—an “industrial reserve army,” to use Marx’s famous expression—, which in turn makes it costly for workers to quit (since they would not be in a position to ﬁnd another job easily). Second, even if labor markets cleared, there are additional exit costs, such as workers’ investment in developing ﬁrm-speciﬁc human capital, searching and transition costs, and the psychological costs in quitting work altogether—work being a relevant source of self-respect in itself in capitalist societies. Third, even if exit were completely costless, job alternatives could still be as despotic, unregulated and arbitrary as their current job, in which case the exit option would be of little value. In turn, section 8.3.2 analyzes the so-called workplace constitutionalism (WC). Unlike the exit option argument, WC does not attempt to grant workers the right to costlessly quit, but is rather aimed at regulating the workplace and at setting clear and speciﬁc standards to which managers have to conform in the exercise of their authority, thus reducing their discretion and ability to interfere at will and turning WD into unnecessary. Two concerns are raised. First, WC does not prevent workplace regulation from being arbitrary itself, for it does not guarantee that workers’ interests are taken into account in its drafting and passing. Second, WC does not prevent workplace regulation from being arbitrarily enforced to particular cases and unforeseeable contingencies typical of the economic activity in a complex economy, since it does not guarantee that workers’ interests are taken into account in the process of enforcement. As the Spanish Count of Romanones, an early-twentieth century liberal Minister, put it, “let others do the law, and I will write the rules.” Section 8.4 puts forward the republican case for WD as straightforward as possible. From a civic republican standpoint, citizens’ active and continued political participation is required not because the political sphere “is the realm where freedom is a worldly reality,” as Hannah Arendt (and more recently, many communitarian political philosophers) put it,39 but in order to ensure their own immunity from arbitrary interference. This is so for at least two reasons—to ensure that law does not turn into arbitrary itself and to ensure that it is not arbitrarily enforced. Both reasons apply to labor law. Thus workplace regulation can turn into arbitrary itself unless those against whom it is enforced also enjoy and exercise the right to inﬂuence its drafting. Likewise, workplace regulation can be arbitrarily enforced unless those against whom it is enforced also enjoy and exercise the right to inﬂuence its enforcement. Put simply, since it would be impossible or prohibitively costly for the parties of the employment relationship to make their contract complete and some ﬂexibility is also desirable as to adequately address day-to-day unforeseeable eventualities, WC
H. Arendt, “What Is Freedom?,” in Between Past and Future (New York: Penguin Books, 1993), p. 154.
39

28

Introduction and summary

turns out to be insuﬃcient. Workers are thus to be given a say not only at the outset of the employment relationship but also as the relationship goes along. The section closes by discussing two potential objection to WD—namely, that WD adds little to WC, for in WC workers already enjoy control rights at the political level and that institutional devices other than WD, such as judicial litigation or unionism, may often better ensure workers’ immunity from arbitrary interference. Section 8.5 closes the chapter by analyzing the separation of control rights and ownership of the ﬁrm from a republican standpoint. The section goes back to the relationship between ownership and full citizenship presented in Part I, section 1.2.3 of the dissertation. It is conluded that WD requires not only workers’ control but also workers’ ownership of the ﬁrm due to bargaining power asymmetries and moral hazard problems arising from the formal separation of ownership and control rights. Accordingly, it is argued that cooperativism is ceteris paribus preferable to other forms of WD in which ownership and control rights are separated, such as co-determination.

Part I

Democracy

Chapter 1

From demophobia to demophilia
Part I analyzes the shift in the use of the term “democracy” in the early American Republic—from its rather derogatory sense in the constitutional period to its widely assumed positive one in the 1830s. The present chapter provides an explanation for the increasing support of the term, along with the institutional devices attached to it. In turn, chapter 2 shows that, besides the increasing acceptance of the term, its very meaning also shifted—and reconstructs the early concept and the later concept. It shows that the later concept turned out to include a number of institutional devices that had been traditionally considered to be constraints on democracy and only later came to be seen not only as consistent with democracy but actually as necessary for any good-working democratic system. This helps explain why the founders of the American constitutional system openly referred to it as a non democratic one and why, by 1830, virtually everyone turned out to refer to very much the same system as a fully democratic one. In a nutshell, what the Founders meant by “democracy” in 1787 was not identical to what the term turned out to mean half a century later.

1.1. Introduction
A well-known but still striking feature of the history of the American constitutional system is that its framers openly referred to it as a non-democratic one. The Federalists did not only blame “the turbulence and follies of democracy,” which “have ever been spectacles of turbulence and contention; have ever been found incompatible with personal security or the rights of property; and have in general been as short in their lives as they have been violent in their deaths.” 1 But they also conceived the Constitution as a means “to restrain” and “to keep down the turbulency of democracy.” 2 Not less striking is that, only a few decades after the ratiﬁcation of the Constitution, virtually everyone—the heirs of the Federalists included—turned out to refer to very much the same system as a fully democratic one. “Democracy ... more especially designates our government than republic,” stated the Jeﬀersonian Aurora in 1802.3 And by 1830, as Tocqueville reported, most Americans claimed that their country
J. Madison, in M. Farrand, ed., The Records of the Federal Convention of 1787, (New Haven: Yale University Press, 1911), vol. 1, p. 51 and “The Federalist no. 10,” in The Federalist, ed. G. W. Carey and J. McCleallan (Indianapolis: Liberty Fund, 2001), p. 46. 2 R. Morris, in Farrand, ed., The Records of the Federal Convention of 1787, vol. 1, pp. 58 and 517. 3 Aurora, September 18, 1802.
1

32

Chapter 1. From demophobia to demophilia

was a democracy.4 In a letter written right before passing away in 1826, Jeﬀerson declared that “We of the United States ... are constitutionally and conscientiously democrats.” 5 Even the Whig Party celebrated the American democracy and called themselves the “true democrats” in 1840.6 How could it be that a constitutional system formerly known as non-democratic had turned in less than half a century into the democratic system par excellence?7 Part I of the dissertation argues that, once we realize of the derogatory sense of the idea of democracy—along with the institutional devices commonly attached to it—in the 1780s, it should not come with surprise that the Federalists hesitated to use the term to label the political system they were putting forward. It is obviously the case that most Federalists were not democrats even according to our current standards. But it is diﬃcult to understand why they were so categorically opposed to democracy until we understand what “democracy” meant at their time.8 Likewise, we should not be surprised by the fact that virtually everyone referred to very much the same system as a democracy ﬁfty years later, when the term turned out to have a positive sense and was used as a battle-cry by Andrew Jackson ﬁrst and the Whig Party next. Such an increasing acceptance of the term is also showed by the use of the term in the press, which increased dramatically during this period. As Morantz has shown, no newspaper had “democrat,” “democratic” or “democratic-republican” in its title before 1800. By contrast, no less than 202 newspapers used the term “democrat” in their titles between 1820 and 1850.9 Of course, this shift still leaves open why the term had a derogatory sense in the 1780s and why it became so positive just a few decades later. A number of political, social, ideological and rhetorical reasons lie behind the shift. This Part of the dissertation focuses on two of them, which are analyzed in detail in this chapter and the next one. The present chapter will show that, despite its rather derogatory sense in the 1780s, the idea of democracy, along with the institutional devices traditionally attached to it, gained increasing support as soon as ordinary Americans accepted the fact of disagreement at the levels of both interests and values, organized themselves to advance such interests and values, and the theory of virtual representation used by the Federalists in the 1780s gradually lost its political eﬃcacy.10 AccordA. de Tocqueville, Democracy in America (Chicago, The University of Chicago Press, 2002). Quoted in F. Dupui-Deri, The Political Power of Words: The Birth of Pro-Democratic Discourse in the 19th century in the United States and France (PhD diss., University of British Columbia, 2001), p. 227. 6 J. Giddings, quoted in Regina Ann Morkell Morantz, “Democracy” and “Republic” in American Ideology, 1787-1840 (PhD diss., Columbia University, 1971), p. 259. 7 Exceptions to this consensus existed indeed—just as they do nowadays. See e.g. Robert A. Dahl, How Democratic is the American Constitution? (New Haven and London: Yale University Press, 2001). 8 Throughout the chapter, “democracy” in inverted commas refers to the term; without inverted commas, in turn, to the institution and/or concept. 9 Morantz, “Democracy" and "Republic" in American Ideology, pp. 164-165. 10 The expression “the fact of disagreement” comes from Jeremy Waldron, and is part of the so-called “circumstances of politics,” which are two. First, the need for a binding collective agreement on a uniﬁed course of action due to the fact that all aﬀected parties would be better-oﬀ than if no agreement was reached. And, second, the existence of a persistent disagreement on which precise course of action ought to be taken, i.e. the fact of disagreement. See J. Waldron, Law and Disagreement (Oxford: Oxford University Press), pp. 101-106. The relationship between the
5 4

1.1. Introduction

33

ing to the theory of virtual representation, however excluded from actual political decision-making, all members of the political community were represented—if not actually, virtually—by their taking part in the shared interests of the community, which was seen as an organic unity. So, as the market economy evolved, ordinary people became aware of their idiosyncratic and plural interests and values, started organizing themselves to advance such interests, and virtual representation became less credible. In turn, as pluralism of interests and values and political partisanship were accepted as inevitable and legitimate, representation through suﬀrage extensions and other institutional devices traditionally attached to democracy were increasingly demanded, and “democracy” lost part of its derogatory sense and gained wider support.11 Even though most Americans—notably, women, blacks, American natives and paupers—would have to wait long for the franchise, the state constitutional conventions called during the 1820s, ’30s and ’40s partially answered to this claims. In addition, chapter 2 will show that, besides its increasing acceptance, the term also went through a notable semantic shift during the decades after the ratiﬁcation of the Constitution. In the colonial period the term had been rarely used, and had a mainly technical sense inherited from the undemocratic branch of the classical republican tradition of political thought in which the revolutionary gentry had been intellectually raised. So, when the term began to be more widely used in the 1780s, this classic sense still had a great inﬂuence, and the Federalists were very successful in using it to label and criticize the institutional devices of some of the state constitutions of the 1770s, to warn against the events occurred during the so-called critical period of the 1780s and to discredit the position of the Antifederalists regarding the ratiﬁcation of the Federal Constitution. This use would go through a dramatic change in the subsequent decades, not only because of the increasing acceptance of the idea of democracy, but also due to the shift of the very idea of democracy. This shift occurred, in turn, due to both its adaptation to changing circumstances and its rhetorical uses (and abuses) by diverse political actors, as we shall see in section 2.2. As William Duane would write to Jeﬀerson in 1824, “Democratic government is no longer Jacobinism; and those who formerly reprobated now use the language and profess the doctrine they reviled twenty four years ago.” A “revolution in speech,” as Duane labeled it, had been produced.12 Although some features of the concept of democracy—and some of the institutional devices associated to it—remained the same, some others shifted. This shift helps to explain why the Federalists believed that the representative government
fact of disagreement and democratic legitimacy is more analytically analyzed in section 4.3.2 of the dissertation. 11 As Robert Dahl points out, no president of the U.S. referred to the American political system as a democracy in his innaugural address until William Henry Harrison in 1841—they used “republic” instead. By contrast, in his innaugural addresses, Franklin Delano Roosevelt used the terms “democracy” or “democratic” twnety times and “republic” not once. Reagan and the ﬁrst Bush used both terms equally, while Clinton and the second Bush used only “democracy.” R. Dahl, “James Madison: Republican or Democrat?,” Perspectives in Politics 3 (2005): 446-447. 12 William Duane, “Letter to Jeﬀerson, October 19, 1924,” Proceedings of the Massachusetts Historical Society 20 (1906-1907), 381-383, p. 382.

34

Chapter 1. From demophobia to demophilia

they put forward was not only diﬀerent from democracy but in fact a check against it, and why, only half a century later, virtually everyone turned out to refer to very much the same system as an openly democratic one. As Bernard Manin puts it, “for Madison, representative government was not one kind of democracy; it was an essentially diﬀerent and furthermore preferable form of government.” 13 In short, what the Federalists meant by democracy was not exactly what the term turned out to mean half a century later. Likewise, the shift helps to explain why a number of limits on the scope of democracy that will be analyzed in Part II and Part III, and that had been traditionally seen to be at odds with democracy, turned out to be seen (and are seen nowadays) not only as consistent with democracy but, in fact, as necessary for any good-working democratic system. This chapter and the next one analyze these two phenomena respectively—namely, the increasing acceptance of the idea of democracy, on the one hand, and the shift in the very idea of democracy, on the other, along with the main events behind the two shifts. After this introduction, section 1.2 reconstructs the theory of virtual representation and its relationship with the anti-democratic rhetoric during the 1780s. Section 1.3, in turn, shows the challenges to this theory and the reasons why actual representation obtained more support as the nineteenth century progressed and thus “democracy” turned from being despised to being cherished.

1.2. Virtual representation
This section reconstructs the theory of virtual representation used by many Federalists to support the central and strong Federal government endorsed in the Constitution and to justify some of its institutional provisions. It also states the relationship between this theory and the anti- and pro-democratic rhetoric during the 1780s. In its bare bones, the theory of virtual representation can be outlined in the following steps: (A1) (A2) Organicist tenet: A political community is an organic entity with a common set of interests. Ruling elite tenet: Political decisions should be made only by those with the expertise and impartiality required to let aside their private, factional interests and to represent the shared interests of the community. (A3) Independence tenet: Not all members of the community have this ability—notably, those who are materially dependent do not. Nevertheless, even though they are excluded from actual political decision-making, the latter are also represented—if not actually, virtually—by their taking part in the shared interests of their community. As we shall see in Part II of the dissertation, the two ﬁrst tenets of the theory still have considerable appealing nowadays, notably, when a political issue is claimed to be beyond dispute and is insulated from ordinary parliamentary decision-making and
B. Manin, The Principles of Representative Government (Cambridge: Cambridge University Press, 1996), p. 2.
13

1.2. Virtual representation

35

is entrenched in a rigid constitution that is enforced by judges with ﬁnal powers of review. The section reconstructs these steps in turn. 1.2.1. Organic society The theory of virtual representation was much older than the American revolution, as we have seen above in the Introduction. But it gained a renewed inﬂuence when the Stamp Act triggered colonial opposition in 1765 and several English pamphleteers and politicians set out to justify taxation of the colonies. Their aim was to reply the colonial claim that taxation was unfair unless consented through parliamentary representation—“no taxation without representation,” in the famous phrase attributed to James Otis. One line of argumentation stood out. According to it, the colonies were already represented, if not actually, virtually, by their being part of the British empire, and thus sharing a common interest with the rest of it. As John P. Reid has stated, “virtual representation [would be], in the ﬁrst part of the revolutionary controversy, the chief legal argument made by imperialists contending not only that the colonies were bound but also that they were represented in Parliament.” 14 Most British citizens were not actually represented in the parliament either, or so the argument went. But that was because they were already represented virtually, as the colonists were. The former included not only women and children, but also an overwhelming majority of adult men. These included the owners of land held in copyhold rather than freehold, the inhabitants of large boroughs such as Leeds, Manchester, Halifax and Birmingham that had not been granted franchise, and the owners of personal property rather than real estate, as well as merchants, manufacturers, and other individuals engaged in commerce who might be rich but did not qualify as electors.15 In an attempt to reduce ad absurdum the claims of the colonies, Thomas Whateley claimed that the inhabitants of Leeds, of Halifax, of Birmingham, and of Manchester, Towns that are each of them larger than the Largest in the Plantations ... none of them chuse their representatives; and yet are they not represented in the Parliament? Is their vast Property subject to Taxes without their Consent?16 According to Watheley, The Colonies are in exactly the same Situation ... none are actually, all are virtually represented; for every Member of Parliament sits in the House, not as representative of his own Constituents, but as one of that august Assembly by which all the Commons of Great Britain are represented.17
J. P. Reid, The Concept of Representation in the Age of the American Revolution (Chicago: The University of Chicago Press, 1989), p. 50. 15 See Reid, The Concept of Representation in the Age of the American Revolution, p. 52. 16 Thomas Whateley, The regulations Lately Made concerning the Colonies and the Taxes Imposed upon Them, considered (London: J. Wilkie, 1765), p. 107. 17 Ibíd., p. 109
14

36

Chapter 1. From demophobia to demophilia

Note that Watheley’s argument has two diﬀerent parts. First, the function of the parliamentary representatives was not to represent the factional interests of the diﬀerent subjects under the Crown, but to represent their common interests. Second, American colonists shared such interests with the rest of the English subjects. Conclusion: No particular subject needed to be actually represented because all were virtually so already. Demanding actual seats in the Parliament and voting rights was therefore irrelevant. Thus, it is not surprising that, as Wood puts it, “the English defenders of virtual representation should have denigrated the electoral process by which members were sent to Parliament. Election in and by itself was not what gave the member his representative power.” 18 The colonists immediately and forcefully rejected the British claim. Some revolutionary leaders even rejected the idea of virtual representation altogether by arguing that representation ought to be “direct and actual,” as it was often the case in colonial assemblies.19 However, not all of them rejected the theory altogether. On the contrary, many did accept that, in a country with a common set of interests, the function of the assemblies was to represent these common interests. That is, they accepted the ﬁrst part of Whateley’s argument for virtual representation. And they also assumed that American colonies shared a set of common interests with each other. What they challenged is the second part of Whateley’s argument, that is, whether these common interests were coincident with the British ones—which, as the colonial rebellion progressed, they increasingly thought not to be the case. Once the British had imposed tax burdens over the colonies that had not imposed over their own territories, it became clear that the allegedly common interests were often no more than the particular interests of the British territories. Hence, it was easy for the rebellious leaders to argue that, even though the organic society was valid, the tenet did not apply to the relationship between Britain and the colonies and, accordingly, that the colonies were neither actually nor virtually represented in the British parliament any longer. As the voters of Salem, Massachusetts resolved, “if in any sense we are supposed to be represented [in the British parliament] most certainly it is by such only as have an interest in laying burthens upon us for their own relief.” 20 In short, the uprising against the Crown did not only leave the organicist tenet
18 G. S. Wood, The Creation of the American republic, 1776-1787 (Chapel Hill, NC: The University of North Carolina Press, 1969), p. 175. 19 Baylin also recalls that “there was no theoretical basis for such direct and actual representation. It had been created and was continuously reinforced by the pressure of local politics in the colonies and by the political circum- stances in England, to which the colonists had found it necessary to send closely instructed, paid representatives-agents, so called-from the very beginning.” Yet some other revolutionaries tried to block the analogy between non-represented British counties and the colonies by arguing that, unlike the latter, the former had at least the chance to inﬂuence the decisions made in the House of the Commons and to have actual representatives once their estate was above the statutory qualiﬁcation—i.e., that it was not the case that non-elected British counties were not or could not be to some extent actually represented. See B. Bailyn, “Political Experience and Enlightenment Ideas in Eighteenth-Century America,” The American Historical Review 67 (1962): 339-351, p. 347; and Reid, The Concept of Representation in the Age of the American Revolution, pp. 53-62. 20 Quoted in Reid, The Concept of Representation in the Age of the American Revolution, p. 62.

1.2. Virtual representation

37

unchallenged in the colonies, but actually reinforced the sense of unity among them. To be sure, the increasing appeal to some allegedly common interests of the colonies was part of the revolutionary rhetoric aimed at mobilizing the people against a common enemy. But it helped to justify the existence of a strong set of shared interests—beyond merely having a common enemy. Likewise, it helped to justify the sharp distinction between representatives and their constituencies on the basis that the role of the former was to represent the common interest of the confederation rather than the particular interests of each constituency. Among the events that triggered the renewed use of the theory—now, by the former colonists—three stand out: The so-called critical period, the democratization of the legislatures during the 1770s and 1780s, and the call for a Federal constitutional convention. They are analyzed in the remainder of the section in turn. 1.2.1.1. The critical period Even though it might have been partially exaggerated, the so-called critical period of the American history had two salient eﬀects for present purposes.21 First, it brought back the use of the term “democracy” to the daily political jargon—through the continuous condemns of the democratic excesses of the period—and, second, it served as a rationale for the call for a constitutional convention in which the delegates would pervasively appeal to the organicist tenet—along with the more broad theory of virtual representation. A number of economic and political incidents characterized the period, such as the instability of the state laws, the weakness of the foreign trade of the Confederation, the increasing conﬂicts among the states, or the printing of paper money in many states. As Madison wrote to Jeﬀerson, “the evils issuing from these sources contributed more to that uneasiness which produced the Convention, and prepared the public mind for a general reform” 22 For the moment, let us focus on the issue of the paper money schemes. The issue had triggered in the mid-1780s, when a large number of farmer yeomen who had got themselves into debt were prosecuted and jailed for default in payment. This was not a new phenomenon indeed, but its frequency raised dramatically in this period. As Roberto Gargarella reports, “just in the county of Hampshire (between 1784 and 1786), 2,977 yeomen were brought to trial by their creditors (which represented an increase of more than 200 percent of the trials happened in a similar period, between 1770 and 1772). Even worse, in Worcester 4,000 trials occured just in 1785.” 23
21 The label comes from J. Fiske, The Critical Period of American History (Boston and New York, The Riverside Press, 1888). Fiske’s account has not gone without critiques though. According to Charles Beard, for example, the critical period was nothing “but a phantom of the imagination produced by some undoubted evils which could have been remedied without a political revolution”. However, one need not go so far to accept, as Wood does, that it is “diﬃcult to look back at the period and not feel that the pessimism and apprehension so widely expressed did not in some way exaggerate the real problems of the 1780’s.” See C. A. Beard, An Economic Interpretation of the Constitution of the United States (New York: MacMillan, 1913) and Wood, The Creation of the American Republic, p. 395. 22 James Madison to Thomas Jeﬀerson, October 24, 1787, in Ralph Ketchman, ed., The Selected Writings of James Madison (Indianapolis, Heckett, 1996), p. 80. 23 R. Gargarella, La justicia frente al gobierno. Sobre el carácter contramayoritario del poder

38

Chapter 1. From demophobia to demophilia

The situation was dramatic for many ordinary people who had fought in the war, partly encouraged by promises of future prosperity, and now were seeing themselves in a critical economic position—if not in jail. Hence, many of them began asking the state legislatures to print paper money to alleviate their condition. At ﬁrst, most state assemblies hesitated to do so and, as a consequence, many petty debtors organized themselves and eventually some rebellions triggered in Northampton, Tauton, Concord and elsewhere—being Shays’ rebellion, which was leaded by a former colonel in the revolutionary army, the most outstanding one. But, eventually, some state assemblies—namely, those of Pennsylvania, South and North Carolina, New York, Georgia, New jersey and Gerogia—began paying attention to the demands of these groups of debtors, in part to avoid new rebellions, and passed laws for the printing of paper money. Furthermore, in Rhode Island Jonathan Hazard, who had served as the spokesperson of the debtors, was appointed as state governor. In short, the state legislative assemblies were serving as an eﬀective means for the interests of small debtors—often even with them being elected and having a seat in such assemblies. As Wood points out, now “it was the very force of the laws of the states, not anarchy or the absence of law,” which was making the demands of debtors be met.24 As of mid-1770s the state legislatures had been gradually democratized. As Main and Bailyn have pointed out, during the colonial period even ordinary people believed that representatives ought to be “men of birth and fortune, [who] should be invested with power, and enjoy higher honours than the people,” as a Williamsburg radical newspaper stated in 1768.25 Not only did an overwhelming majority of the representatives (at least eighty ﬁve percent of them) belong to the wealthiest ten percent of the population (those whose property was valued at more than £ 5,000). But also, when given the right to vote, ordinary people also tended to elect members of the wealthy elite.26 This had already changed dramatically by 1774, when popular participation increased and was channeled through the various provincial congresses and other extralegal organizations.27 In turn, this shift had endogenous eﬀects, for the more ordinary people began taking part in public aﬀairs, the more they wanted to do it and the better organized they were. Hence, when thousands of debtors
judicial (Barcelona: Ariel, 1997), p. 21. The references to the critical period in this section rely heavily on his account. 24 Wood, The Creation of the American Republic, p. 406. 25 Even if inmediately before had been said that laws were “for the people, and not the people for the laws.” Quoted in J. T. Main, “Government by the People: The American Revolution and the Democratization of the Legislatures”, The William and Mary Quarterly 23 (1966): 391-407, p. 392. 26 See Main, “Government by the People”, pp. 393 and 397. The degree of enfranchisement had been much larger in the colonies than in England (and larger in New England than in the southern colonies). By the 1760s an average of 60 percent of adult white men could vote in the colonies. See A. Keyssar, The Right to Vote. The Contested History of Democracy in the United States (Philadelphia: Basic Books, 2009), p. 7. 27 According to Main, several reasons explain this shift. First, legal prohibitions on artisans and farmers sending representatives to the colonial assemblies did not apply to the extralegal congresses. Second, these congresses were much larger. Third, the interior areas had no real upper class, so they had no choice but to send men of moderate property to these assemblies. And ﬁnally, as the war went on, many ordinary people got involved in politics and gained fame. “Voters were ceasing to elect only men of wealth and family [and] more and more often chose ordinary yeomen or artisans.” Main, “Government by the People,” p. 404.

1.2. Virtual representation

39

were prosecuted and jailed, many ordinary people began putting pressure on the legislative assemblies, and the latter eventually agreed to their demands and began issuing paper money. Consider now the relationship between these events and the derogatory use of the idea of democracy. The increasing participation of ordinary people in the legislatures during the revolution had been a frightening experience “not only to the conservative-minded leaders of the colonies, but to many of the popular leaders as well.” 28 However, when this participation gave way to conﬁscations of property, suspensions of the ordinary means for the recovery of debts, and paper money schemes—with the skyrocketing inﬂation produced as a result—, the revolutionary gentry began talking about a new form of despotism—“democratic despotism,” in John Adams famous expression. For many, as for James McHenry of Maryland in 1787, America was suﬀering the “disorders of a democracy.” 29 Among those who denounced the democratic disorders of the 1780s, many believed that they were not only due to the increasing participation of ordinary people in the legislatures but also because of the very structure of the revolutionary state constitutions, which provided insuﬃcient checks for the legislative assemblies. As a general rule, they referred to the constitutions of Pennsylvania, Georgia, Vermont, New Hampshire and Rhode Island, which were considered the more democratic ones. “Look at the legislature of Rhode Island!,” claimed a politician of that period. “What is it but the perfect picture of a mob.” 30 Yet only in Georgia, Vermont and Pennsylvania would prevail the unicameral legislative branch, which was seen as the democratic and thus threatening part of the state constitutions. According to McHenry, Our chief danger arises from the democratic parts of our constitutions. It is a maxim which I hold incontrovertible, that the powers of government exercised by the people swallows up the other branches. None of the constitutions have provided suﬃcient checks against the democracy.31 For present purposes, it is important to note that democracy was identiﬁed solely with the legislative branch. Hence, the more radical state constitutions fueled the anti-democratic concerns because they were said to have all-powerful legislatures. Among these, Pennsylvania was the obvious reference, since its constitution was not only unicameral but also lacked the checks required to control its democratic part.32 Parliaments were taken to be democratic because they were very represenM. Jensen, “Democracy and the American Revolution”, The Huntington Library Quarterly 20 (1957): 321-341, p. 328. 29 Quoted in Dupuis-Deri, The Political Power of Words, p. 82. 30 Quoted in Dupuis-Deri, The Political Power of Words, p. 88. 31 Quoted in Farrand, The Records of the Federal Convention of 1787, vol. 1, pp. 26-27. 32 Apart from having a one-house legislature to be appointed annually, bills could not come into force until, after having publicized throughout the colony, was discussed and voted in the following session. The executive was plural, formed of a twelve-person Council, elected for three-year terms, with one-third rotating every yearand rotation required; no one could serve more than four years in seven.A Council of Censors popularly elected, and with a seven-year term, was to review the constitutionality of laws. Other oﬃcials such as justices of the peace, sherrifs and militia oﬃcers were popularly elected every year. There were no property qualiﬁcations and suﬀrage was open to all male taxpayers.
28

40

Chapter 1. From demophobia to demophilia

tative of the factional interests of ordinary citizens due to their very nature—they included hundreds of members and were designed to address disagreement and diversity, ideological confrontation and conﬂict of interests. As Walter Bagehot described the House of Commons in 1867, “Here are 658 persons, collected from all parts of England [sic], diﬀerent in nature, diﬀerent in look and language.” 33 By contrast, courts or unipersonal executives were considered to be more independent from such interests and, accordingly, more suitable to represent the common interests of the country. For that reason—rather than because democracy was associated with other parts of the constitutions or with the constitutions as a whole (see infra, §2.3.5)—, William Hooper of North Carolina could state that Pennsylvania was a “execrable democracy—a Beast without a head ... the Mob ruling.” 34 The problem with these constitutions was not that they lacked representativeness, but rather the contrary—that they were too representative of the particular interests of ordinary people. For those who assumed that the American society was an organic entity with a common set of interests and that the function of the legislative assemblies was to pursue these interests, rather than the particular interests of the diverse factions of the society, it was the case that “a spirit of locality” was destroying “the aggregate interests of the community.” 35 The “excess of democracy,” as Elbridge Gerry from Massachusetts would put it in the Philadelphia Convention, was seriously jeopardizing the organicist tenet.36 The excessive factionalism of local assemblies and state parliaments was the problem, for which the strong and centralized government of the Federal Constitution would be the solution for the Federalists. Hence, in a letter to his friend Lafayette in May 10, 1786, Washington would write that “It is one of the evils of democratic governments, that the people, not always seeing and frequently misled, must often feel before they can act right ... [but] I am not without hopes, that matters will take a more favorable turn in the foederal Constitution.” 37 1.2.1.2. The Federal Constitution The Federalists were very concerned about the increasing presence of factions—which Madison deﬁned as a “number of citizens, whether amounting to a majority or minority of the whole, who are united and actuated by some common impulse of passion, or of interest, adverse to the rights of other citizens, or to the permanent and aggregate interests of the community.” 38 These concerns are clear in Madison’s frequent references in the Philadelphia Convention to the plurality of interests between the states, which “have diﬀerent interests and are ignorant of each other’s interests,” between “the Northern and Southern,” between “the Aristocratic and other,” between “the landed interest, and the commercial including the stockholders” or, as famously
33 W. Bagehot, quoted in J. Waldron, The Dignity of Legislation (Cambridge: Cambridge University Press, 1999), p. 34. 34 Quoted in Dupuis-Deri, The Political Power of Words, p. 89. 35 G. S. Wood, The Radicalism of the American Revolution (New York: Vintage Books, 1991), p. 251. 36 Farrand, The Records of the Federal Convention of 1787, vol 1., p. 48. 37 Quoted in Dupuis-Deri, The Political Power of Words, p. 91. 38 Madison, “The Federalist no. 10,” p. 44.

1.2. Virtual representation

41

stated in his Federalist no. 10, between “those who hold, and those who are without property, [who] have ever formed distinct interests in society.” 39 But the position of the Federalists in the Philadelphia Convention and the ratiﬁcation process was that, despite the increasing factionalism, the states shared a common set of interests diﬀerent from their factional ones and that a properly designed constitution could overcome the latter by setting a central and autonomous public authority strictly aimed at the pursue of this common interest. As Wood points out, contrary to what is sometimes claimed, “Madison and the Federalists ... were not modern-day pluralists. They still clung to the republican ideal of an autonomous public authority that was diﬀerent from the many interests of the society.” 40 Perhaps no document other than Hamilton’s Federalist no. 35 expresses better the organic tenet that the Federalists embraced. According to Hamilton, the interdependency of the parts of the society made absurd the idea of actual representation—in which each faction was to appoint and instruct its own representatives (see infra §1.3). This would be so even from the point of view of the members of the factions, which would be perfectly well represented through virtual representation. As an instance of this, he wrote, mechanics and manufacturers will always be inclined, with few exceptions, to give their votes to merchants, in preference to persons of their own professions or trades. Those discerning citizens are well aware, that the mechanic and manufacturing arts furnish the materials of mercantile enterprise and industry. Many of them, indeed, are immediately connected with the operations of commerce. They know that the merchant is their natural patron and friend; and they are aware, that however great the conﬁdence they may justly feel in their own good sense, their interests can be more eﬀectually promoted by the merchant than by themselves ... We must therefore consider merchants as the natural representatives of all these classes of the community.41 And the same applied to the case of great and petty landowners, whom Hamilton took to be perfectly united, from the wealthiest landlord, down to the poorest tenant. No tax can be laid on land which will not aﬀect the proprietor of thousands of acres, as well as the proprietor of a single acre. Every landholder will therefore have a common interest to keep the taxes on land as low as possible; and common interest may always be reckoned upon as the surest bond of sympathy.42 As it has been admitted before, the Federalists used the organicist tenet strategically to deny the Antifederalist demands for actual representation and to advance the
Farrand, The Records of the Federal Convention of 1787, vol. 1, pp. 166, 486, 461 and 152; Madison, “The Federalist no. 10,” p. 44 40 Wood, The Radicalism of the American Revolution, p. 253. 41 A. Hamilton, “The Federalist no. 35,” in The Federalist, p. 170. 42 Hamilton, “The Federalist no. 35,” in The Federalist, p. 170.
39

42

Chapter 1. From demophobia to demophilia

centralized government and the sharp separation between representatives and their constituents they were for.43 But, however opportunistically used in part, the open, intensive and very successful uses of the tenet show its widespread acceptance and normative appeal at that time—an acceptance and appeal that, for reasons exposed below (infra, §1.3.1), would decrease dramatically in just a few decades. True, many of the institutional devices proposed—notably, in the Virgina plan—and some of the ﬁnally adopted ones did not diﬀer much from the ones included in some of the state constitutions—notably, in the Constitution of Massachusetts drafted by John adams and approved in 1780. The main innovation introduced by the Federal Constitution was, rather, that it would reduce dramatically the power of the state assemblies by transferring a substantial part of it to the federal level. That, in turn, would make much more diﬃcult for local demands to exercise their inﬂuence over the political decision-making process. In addition, the amendment the constitution would be made very cumbersome. Hence, the Constitution would both transfer key powers to the Federal level and forbid the states to take actions in a number of fundamental issues. On the one hand, exclusive power was given to the Congress regarding monetary and ﬁscal policies, the declaration of war or the raising of national military forces. On the other hand, states were explicitly prohibited from printing bills of credit, coining anything other than “gold or silver coin” legal tender for payment of debts, passing laws aﬀecting the obligation of contracts, imposing taxes on imports or exports or entering into bilateral agreements (either with each other or with other countries).44 These were issues over which the democratic excesses of the critical period had been said to have a larger inﬂuence, and they were now under the sole authority of the Federal government. Interestingly, the Federalists saw this transfer as a key move away from democracy toward republicanism. “Pure democracy,” Madison famously warned, “can admit of no cure for the mischiefs of faction.” 45 By transferring these issues from the state level to the federal level, decisions were now guarded from both “factious leaders” and “the confusion of a multitude.” 46 The question now was, of course, who would qualify to make those decisions instead of the “multitude.” 1.2.2. Ruling elitism Once the existence of a common set of interests among the states and the advancement of such interests as the proper function of the government had been assumed, it was required that decisions be made by those with the ability to put aside their factional interests and pursue such common interests. In Madison’s words, the Federal government had to be ruled by “some disinterested and dispassionate umpire,” who could overcome the “disputes between diﬀerent passions and interests in the State”
In the case of Hamilton this is especially likely, for he embraced a much more realistic social ontology than, say, Madison. See T. P. Govan, “The Rich, the Well-born, and Alexander Hamilton,” The Mississippi Valley Historical Review 36 (1950), pp. 675-680. 44 On the the system of check and balances, the relationship between representatives of the various branches and the people and the devices to ensure accountability, see infra §1.2.2. 45 Madison, “The Federalist no. 10,” p. 46-47. 46 Madison, “The Federalist no. 10,” p. 46-47.
43

1.2. Virtual representation

43

and could replace the “speciuous, interested designing men,” “men, respectable neither for their property, their virtue, nor their abilities” that were standing in public aﬀairs.47 These men would not only represent the organic interests of the nation. They would also represent such interests better than citizens themselves, for if given the opportunity the latter would be driven by factional interests and particular values. Hence, in contrast to a democratic government, a republican government would not be deﬁned so much by “the total exclusion of the representatives of the people from the administration of the former,” but rather by “the total exclusion of the people, in their collective capacity, from any share in the latter ” 48 Obviously, the ruling elite tenet did not go without challenge, as it will become clear below in section1.3.2. But for now, let us make clear the case put forward by the Federalists, along with the institutional devices aimed at ensuring the rule, independence, and control of this elite. 1.2.2.1. The case for a ruling elite The factionalism originated by the “excess of democracy” of the 1770s and 1780s, as Gouverneur Morris referred to it, served as the main rationale for the demand of qualiﬁed and unbiased representatives put forward in the Philadelphia Convention and the ratiﬁcation process. Many were the issues that the Federalists associated to this factionalism. Among them, the commercial weakness abroad and the insecurity of Western states at home due to the military powerlessness of the Confederation, the skyrocketing inﬂation produced by the paper money schemes, the threats to private property due to the weak authority of the central government, the pervasive vetoing of the amendments of the Articles of the Confederation by the states, and so on. The Federalists believed that this factionalism was due to a number of reasons that could be eradicated through the limitation of the exercise of government to a ruling elite of impartial and expert representatives. Three of these reasons stand out. A. Reasons relative to the representatives themselves. As we have already seen (supra, §1.2.1.1), more and more local and plain men had been seating in the state assemblies from the mid-seventies onward. These men, or so the argument went, were advancing the local and short-term interests of their constituencies—and thus pervading the state assemblies with a “spirit of factionalism” that was contrary to the national public interest. In order to prevent this factionalism and to “reﬁne and enlarge the public view,” it was required to pass these interests, as Madison famously put it, “through the medium of a chosen body of citizens, whose wisdom may best discern the true interest of their country, and whose patriotism and love of justice, will be least likely to sacriﬁce it to temporary or partial considerations.” 49 For two reasons. First, a dispositional reason—to have representatives with interests diﬀerent from those of their constituencies. Second, an epistemic one—to have representatives with the country-scale and long-term political knowledge required to take into account the
47 48 49

Quoted in Wood, The Creation of the American republic, pp. 474 and 477. Madison, “The Federalist no. 63,” in The Federalist, p. 329, italics in the original. Madison, “The Federalist no. 10,” in The Federalist, p. 46.

44 wider interests of the country.

Chapter 1. From demophobia to demophilia

B. Reasons relative to the relationship between representatives and their constituencies. According to the Federalists, the state legislatures were too dependent on their constituents, and thus unable to pursue interests other than local interests. Even if the proper men—i.e. men with interests diﬀerent from their constituencies and with a country-scale political view—were often elected, their impartiality and independence were undermined by a set of institutional arrangements that were in need of reform. (“Had every Athenian citizen been a Socrates, every Athenian assembly would still have been a mob,” had written Madison.)50 Consider just three among them. First, the proximity between state legislatures and ordinary people who, as Elbridge Gerry of Massachusetts declared, “[did] not want virtue.” 51 . Second, the large number of seats in the legislatures—for “the more you enlarge the body, the greater chance there is, of introducing weak and unqualiﬁed men,” as it was addressed to Connecticut.52 Or, as Madison put it, “the more multitudinous a representative assembly may be rendered, the more it will partake of the inﬁrmities incident to collective meetings of the people.” 53 Third, the pervasive use of instructions by many local assemblies, which had been a rather common practice since the colonial period.54 . C. Reasons relative to the relationship between legislatures and other branches of government. In addition to the former two problems, the legislatures were believed to be too unchecked in many states—most notably, in Pennsylvania (see supra, §1.2.1.1). And so they were the representatives seating on them who had way too many incentives to exceed their authority and control other branches of the government. As Gouverneur Morris claimed, “The ﬁrst branch, originating from the people, will ever be subject to precipitancy, changeability, and excess. Experience evinces the truth of this remark without having recourse to reading.” 55 Through which precise devices did the Federalists propose to eradicate this phenomena it will become clear in the section immediately below. For the moment, it is enough to point out that they could openly and successfully put forward the idea of a virtuous elite of experts as the general solution for the problem of locality and factionalism. And that this openness is diﬃcult to understand—moreover, in an era of democratic revolutions, as Robert Palmer called it,56 in which the old social hierarchies were supposedly vanishing—without taking into account the sense of social deference still dominant in the immediate aftermath of the revolution (on this, see infra §1.3.2.3).
Madison, “The Federalist no. 55,” in The Federalist, p. 288. Farrand, The Records of the Federal Convention of 1787, vo. 1, p. 48. 52 Quoted in Wood, The Creation of the American Republic, p. 512. 53 J. Madison, quoted in J. Waldron, The Dignity of Legislation (Cambridge: Cambridge University Press, 1999), p. 32. 54 See Reid, The Concept of Representation in the Age of the American Revolution, ch. 8. 55 Quoted in Farrand, The Records of the Federal Convention, vol. 1, p. 517. 56 R. R. Palmer, The Age of Democratic Revolution, 2 vols. (Princeton: Princeton University Press, 1959-1964).
51 50

1.2. Virtual representation

45

To be certain, the Federalists enjoyed a number of organizational and economic resources that the Antifederalists did not. For example, they controlled most newspapers, of which only one dozen supported the anti-Federalist cause.57 This asymmetry of resources dramatically improved the prospects of the Federalists in the ratiﬁcation process, as wells as their ability to advance their agenda. However, even though the Revolution had a profound impact on the understanding of social hierarchy, three facts show how powerful the demand for impartial and qualiﬁed representatives was. Firstly, the Federalists could openly set out their ruling elitism, which would have been strategically unwise if this elitism had a bad press. Moreover, they could do it even though the ratiﬁcation of the Constitution needed the support of nine out of thirteenth state legislatures, and these legislatures were supposedly formed by the plain people that the Federalists had denounced for being inexpert and locally biased. Finally, as it will become clear below in section1.3.2, the same elitist language could no longer be used just a few decades later, which shows that, regardless of how successful the elitist tenet was in absolute terms, it was very successful in relative terms. 1.2.2.2. Institutional devices The dispositional account put forward by the Federalists was not a static one. They did not defend that members of the elite were virtuous ex hypothesi. Quite on the contrary, according to Madison, they were motivated in variable degrees by ambition, personal interest and the public good—of which “the two ﬁrst” were unhappily “proved by experience to be most prevalent.” 58 Thus, the Federalists were well aware of the ever-present threat of corruption, as Pocock has pointed out.59 However, they also believed that a properly designed constitution would reduce the inﬂuence of the two ﬁrst motivations and give way to the third one. This goal could be achieved (a) by ensuring that proper candidates were elected as representatives, (b) by keeping representatives as separated as possible from the short-term and factional interests of their constituencies, and (c) by designing a system of checks and balances that would control potential abuses of any branch—and most notably, of the legislative branch—over other branches of government. As Hume—an author whim whom the Federalist were very familiar—had famously put it, “If separate interests be not checked, and not be directed to the public, we ought to look for nothing but faction, disorder, and tyranny from such government.” 60 Similarly, as Madison would state it in his Federalist no. 51, “you must ﬁrst enable the government to control the governed; and in the next place oblige it to control itself.” 61 Through which precise devices though? As for the discussion over the quality and independence of the representatives,
Wood, The Creation of the American Republic, p. 498. Quoted in Neal Riemer, “James Madison’s Theory of the Self-Destructive Features of Republican Government,” Ethics 65, no. 1 (Oct., 1954), 34-43, p. 35. 59 See Pocock, The Machiavellian Moment, chapter 15. 60 D. Hume, “Of Parties in General,” in Perspectives on Political Parties, S. E. Scarrow, ed. (New York: Palgrave MacMillan, 2002). 61 Madison, “The Federalist, no. 51”, in The Federalist, p. 269.
58 57

46

Chapter 1. From demophobia to demophilia

suﬀrage qualiﬁcations, the prohibition of binding instructions, and long-term oﬃce holding stand out. Regarding suﬀrage qualiﬁcations, it was argued that limitations on suﬀrage had to be provided, and that these had to aﬀect both passive and active suﬀrage. A. Pasive suﬀrage. Even though property qualiﬁcations for eligibility were proposed for the Congress, the quality and independence of the representatives was eventually achieved by enlarging the size of the electoral districts and by reducing the size of the two chambers of the Congress. By downsizing electoral districts, said Madison, “you render [representatives] unduly attached to [the local circumstances and lesser interests of their constituents], and too little ﬁt to comprehend and pursue great and national objects.” 62 Large electoral districts, on the contrary, separated representatives from of their electors, and their local and short-term interests, making it easier to focus on the long-term interests of the country. However, large districts also made much more diﬃcult for plain people to run for oﬃce, as the Antifederalists immediately protested. In small electoral districts no great resources were needed to run for oﬃce, since it was likely for the candidate to be already well known by the electorate. In large districts, by contrast, running for oﬃce was only available in practice for those who had the economic resources and spare time required to campaign. The same logic applied also to the size of the legislatures. Thus a maximum of one representative for every thirty thousand (article 1, section 2 of the Constitutions) was endorsed in the Constitution. B. Active suﬀrage. Property qualiﬁcations were also proposed—e.g. by Governeur Morris, who proposed that only land owners should be granted voting rights. Madison agreed with Morris in principle.63 However, he recalled that freeholders had precisely elected most of the state assemblies which democratic excesses they attempted to check now.64 Furthermore, he also suggested, in response to a similar proposal, that “landed possessions were no certain evidence of real wealth [since] many enjoyed them to a great extent who were more in debt than they were worth.” 65 Finally, the great diversity of suﬀrage qualiﬁcations among the states made it diﬃcult for the delegates to arrange an unitary suﬀrage disposition on which all the states could agree. Accordingly, delegates eventually decided that members of the House of Representatives ought to be elected by electors having “the qualiﬁcations requisite for electors of the most numerous branch of the state legislature” (art. 1, section 2 of the Constitution). Qualiﬁcations were limited thus to those already in force in each state not so much “due to reasons of principle, but of expediency,” as Manin points

Madison, “The Federalist no. 10,” in The Federalist, p. 47. “Viewing the subject in its merits alone, the freeholders of the Country would be the safest depositories of Republican liberty,” in Farrand, The Records of the Federal Convention of 1787, vol. 2, p. 203. 64 For further problems with limiting property qualiﬁcations to land ownership, see infra §1.3.3.1. 65 Farrand, The Records of the Federal Convention of 1787, vol. 2, p. 123.
63

62

1.2. Virtual representation

47

out.66 C. Binding instructions of the constituents. Even though the practice of instructing representatives had been a common practice in the colonies,67 it was categorically ruled out in the Philadelphia Convention. This practice, said Noah Webster, rested on the belief that “the constituents, on a view of their local interests ... or very imperfect information, are better judges of the property of a law, and of the general good, than the most judicious men are ... after attending to the best oﬃcial information from every quarter, and after a full discussion of the subject in an assembly, where clashing interests conspire to detect error and suggest improvement.” 68 In short, binding instructions were contrary to the very idea of a ruling elite of representatives in three senses. They denied (a) that they knew better than ordinary electors, (b) that they were independent from them and, (c) that they had the capacity to change their minds through public deliberation. D. Term in oﬃce. Proposals for long and, in some cases, life-long terms aimed at ensuring the independence of the representatives were made. As for the House of Representatives, the Antifederalists had appealed to the Whig motto that “when annual elections end, tyranny begins” to denounce two-years terms. Madison replied in the Federalist no. 53 justifying the length of the House of Representatives terms not only as a guarantee for the independence of the congressmen but also as necessary to enable them to acquire the “knowledge of the subjects on which [they are] to legislate.” 69 As for other chambers, though, Federalist did propose not only that Supreme Court justices had life-long appointments—which was ﬁnally included in the Constitution—, but also the presidency and the senate. “The second branch ought to be composed of men of great and established property—an aristocracy,” claimed Robert Morris. “And to make them completely independent they must be chosen for life, or they will be a useless body. Such an aristocratic body will keep down the turbulency of democracy.” 70 Regarding the system of checks and balances, the control and limitation of the democratic branch—the House of Representatives—pervaded the discussion. Although the threat of a tyranny of the few—i.e. of the British Crown—had dominated the debates during the revolution, now the threat of a tyranny of the many replaced it. As Robert Dahl pointed out, “Neither at the Constitutional Convention nor in the “Federalist Papers” is much anxiety displayed over the dangers arising from minority tyranny; by comparison, the danger of majority tyranny appears to be a source of acute fear” 71 A system of checks and balances—rather than a mere separation
Manin, The Principles of Representative Government, p. 107. See Reid, The Concept of Representation in the Age of the American Revolution, chapter 8. 68 N. Webster, “Government. The Practice of Instructing Representatives Absurd and Contrary to the True Principles of Liberty,” American Magazine 1 (1787-188), p. 206. 69 Madison, “The Federalist no. 53,” in The Federalist, p. 278. 70 Farrand, The Records of the Federal Convention of 1787, vol. 1, p. 517. 71 R. Dahl, A Preface to Democractic Theory, expanded ed. (Chicago and London: The University of Chicago Press, 2006), p. 9.
67 66

48

Chapter 1. From demophobia to demophilia

of powers, as Madison put it in the Federalist no. 48 criticizing Jeﬀerson and the Constitution of Virginia—was required to balance the democratic tendencies of the House of Representatives. A number of such checks were discussed but, for the sake of brevity, let us mention only three of them—the senate, the presidency, and judicial review of legislation. E. The senate. The upper chamber of the Congress was conceived as accomplishing two tasks. First, it was to represent proprietors’ interests in a speciﬁc chamber. As Abraham Baldwin of Georgia said, the task of the Upper House “ought to be the representation of property.” 72 Second, it was supposed to check the potential abuses of the ﬁrst chamber. “In tracing these evils to their origin every man [has] found it in the turbulence and follies of democracy,” said Madison, and he added that “some check therefore [is] to be sought for agst. this tendency of our Governments [and] a good Senate [seems] most likely to answer the purpose.73 Hence, the senate was not only considered a non-democratic institution but, actually, a check on democracy. Edmund J. Randolph also insisted that it was required “such a check as to keep up the balance, and to restrain, if possible, the fury of democracy.” 74 And Robert Morris concluded that the second branch ought to be a check on the ﬁrst—for without its having this eﬀect it is perfectly useless.—The ﬁrst branch, originating from the people, will ever be subject to precipitancy, changeability, and excess ... This can only be checked by ability and virtue in the second branch. On your present system, can you suppose that one branch will possess it more than the others? The second branch ought to be composed of men of great and established property—an aristocracy ... Such an aristocratic body will keep down the turbulency of democracy.75 F. The presidential oﬃce. The role of the presidency was partly modeled out of the British Constitution, which many Federalists admired as the most perfect institutional arrangement ever.76 True, some Federalists hesitated to see the president as an “elective king”—Madison denounced “the gross pretence of a similitude between a king of Great Britain, and ... the president of the United States.”
77

But for many

others the presidency was deﬁned as the monarchical branch of the Constitution, which was much more suitable to represent the organic interests of the country, rather than the particular interests of the existing factions. Parliaments were plural, and could hardly avoid disagreement and conﬂict. In clear contrast to parliaments, the presidency was unipersonal and, accordingly, could not disagree with itself. As Hobees had put it, “the legislator is he ... a Monarch cannot disagree with himselfe,
Farrand, The Records of the Federal Convention of 1787, vol. 1, pp. 469-470. Farrand, The Records of the Federal Convention of 1787, vol. 1, p. 51. 74 Farrand, The Records of the Federal Convention of 1787, vol. 1, p. 58. 75 Farrand, The Records of the Federal Convention of 1787, vol. 1, p. 517. 76 This was not as odd as it could ﬁrstly appear, since the revolution had not been carried out against the British constitutional system but on its very behalf—due to its pervasive violation by the British crown. 77 Madison, “The Federalist no. 67,” in The Federalist, p. 348.
73 72

1.2. Virtual representation

49

out of envy, or interest; but an Assembly may.” 78 Hence, the executive was not only deﬁned as unipersonal—in clear contrast with the executive of Pennsylvania, which was plural—and granted a powerful veto capacity, but it was also suggested by some Federalists that the president should should enjoy life-long tenure. The reason, according to Robert Yates of New York, was “that an executive is less dangerous to the liberties of the people when in oﬃce during life.” 79 Eventually, four-year terms of oﬃce and the possibility of being reelected indeﬁnitely were adopted. According to Madison, these were necessary conditions to “act his own opinion with vigour and decisions [to] contend ... the humours of the legislature.” 80 G. Judicial review. Even though judicial review was not included in the Constitution and would not begin to be used until 1803, after the Marbury v. Madison Supreme Court decision, it was widely discussed in the Philadelphia Convention, as well as in the years following the adoption of the Constitution. Only two thirds of the ﬁfty-ﬁve delegates took active part in the deliberations, and at least twenty-ﬁve among these stood up in favor, either directly or indirectly, of judicial review.81 During the ﬁrst Congress after the Constitution had been ratiﬁed, the discussions over the desirability of judicial review continued. For example, Fisher Ames claimed that it was the task of the judiciary to make ﬁnal decisions about legislation, so if courts found a clause unconstitutional, they ought to highlight it.82 On the other hand, the Antifederalists denounced the institution of judicial precisely because “the judicial under this system have a power which is above the legislative” 83 To be sure, there were a number of decisions before Marbury v. Madison in which the judiciary had already expressed the normative superiority of constitutional law, to which statutes had to conform, and in which the judiciary had expressed its own role in highlighting potential incompatibilities. However, Marbury v. Madison would establish the ﬁrst serious precedent. The case began with a petition to the Supreme Court made by William Marbury, one of the so-called “midnight judges” appointed by John Adams right before handing over the presidency to Jeﬀerson. Marbury claimed that he had been appointed by President Adams as Justice of the Peace in the District of Columbia but his commission had not been subsequently delivered by James Madison, Secretary of State under Jeﬀerson. Marbury petitioned the Supreme Court to force Madison to deliver the documents. However, the Court held that the provision of the Judiciary Act of 1789 that enabled Marbury to bring his claim to the Supreme Court was unconstitutional, because it purported to extend the Court’s original jurisdiction beyond that which article 3 established, and turned it down. With this clever move, Chief Justice John Adams ensured both the support of
T. Hobbes, Leviathan, R. Tuck (Cambridge: Cambridge University Press, 1988), pp. 185 and 132. Quoted in Waldron, The Dignity of Legislation, pp. 30-31. 79 Farrand, The Records of the Federal Convention of 1787, vol. 1, p. 300. 80 Madison, “The Federalist no. 71,” in The Federalist, p. 461. 81 See Gargarella, La justicia frente al gobierno, pp. 40-41; and C. Beard, The Supreme Court and the Constitution (Englewood Cliﬀs: Prentice-Hall, 1962). 82 Gargarella, La justicia frente al poder, pp. 44-45. 83 Brutus, “The Anti-Federalists no. 15,” in The Essential Federalist and anti-Federalist Papers, D. Wootton, ed. (Indianapolis, IN: Hackett, 2003), p. 92.
78

50

Chapter 1. From demophobia to demophilia

President Jeﬀerson to the decision and enormously increased the power of the Court by setting the precedent of judicial review of the constitutionality of statutes—which from that moment on could be turned down by the Court however democratic could they be—for the ﬁrst time. 1.2.3. Material independence The end of the revolutionary war by no means eliminated property qualiﬁcations for voting. Even though delegates in the Philadelphia Convention eventually decided to limit qualiﬁcations for active suﬀrage to those already enforced in the states (see supra, §1.2.2.2), the latter would long maintain previous qualiﬁcations and, in some cases, new ones would be added to the old ones—e.g. in Massachusetts. These property qualiﬁcations referred mainly to the ownership of real estate—a deﬁnition of property that made perfect sense in an eminently agrarian society as the late eighteenth-century America was. This deﬁnition of property and the very idea of material independence would go through major modiﬁcations in the early nineteenth century, as it will become clear below in section 1.3.3.1. And these modiﬁcations, in turn, would have an enormous inﬂuence over the deﬁnition of the employment relationship. For present purposes, these modiﬁcations are relevant for two reasons. First, because they help to explain the increasing demand for suﬀrage extensions and democratic participation. A demand that would be met in a number of state constitutional conventions called in the 1830s, ’40s and ’50s that were aimed, among other things, at removing some qualiﬁcations—notably, to wage laborers. And, second, because they help to explain the increasingly extended liberal view of the employment relationship—which had been traditionally seen as a form of “limited slavery,” as Aristotle put it—as a free and voluntary exchange among juridical equals.84 In turn, the reasons behind this view will be key to Part III of the dissertation, for they help to explain why the workplace tends nowadays to be seen as a sphere in which power relations are absent and that is, accordingly, beyond the scope of politics, including democratic politics, as we shall see in section 1.3.3. Now, before turning to the shift in the conceptions of property, material independence, and wage labor in nineteenth-century America, let us consider the three main reasons that were put forward to justify the requirement of material independence for being granted full citizenship in the late-eighteenth century America—(a) the lack of an independent will, (b) the lack of the leisure time required to get informed about and take part in politics and (c) not having a “stake in society.” 85 Let us see them in turn. A. Independent will. The Federalists were hardly original when they claimed that “such is the fragility of the human heart, that very few men who have no property, have any judgment of their own”, and can thus take part in politics autonomously.86
84 85 86

Aristotle, Politics, C. D. C. Reeve, trans. (Indianapolis: Hackett, 1948), 1260a-b. For the application of these ideas to the case of workplace democracy, see section 8.5 below. John adams to James Sullivan, May 26, 1776.

1.2. Virtual representation

51

This view was inherited from the classical republican tradition in which they had been intellectually raised. As current neorepublicans have emphasized and we shall see in some detail below in section 8.2, classic republicans had deﬁned freedom as non-domination or, in other words, as the absence of arbitrary interference.87 Relationships of domination included the relationship between masters and slaves. But they also included a large number of relationships that, during the American colonial period, had been referred to with the euphemism of relations of “friendship.” Fathers were “friends” with their sons, men were “friends” with their wives and patrons were “friends” with their journeymen.88 Unlike the former, the latter were in position to arbitrarily interfere—i.e., with no other restriction than their own will—in the courses of action of the former. As Sir Robert Filmer had declared in the seventeenth century regarding marriage and fatherhood, “The father of a family governs by no other law than by his own will.” 89 What made the former dominate the latter is that, unlike the latter, the former owned property, which was seen as the main source of domination. “Domination” comes from “dominium,” which in classical Latin means both “property” and “the ability to make free use of this property.” 90 It is hardly surprising thus that, within the republican tradition, it was believed that those who lacked property or dominium were (almost as a matter of deﬁnition) in all likelihood to be dominated by those on whom they depended to make a living. Property was seen as the main source of having an independent will, free from the inﬂuence and domination of third parties—it represented “the attributes of a man’s personality that gave him a political character.” 91 Slavery, parenthood, marriage and wage labor were thus relationships were those who lacked property—slaves, children, viwes and journeymen—were dominated by those on whom they materially depended—masters, fathers, husbands and patrons. Unlike the former, the latter owned property and were thus able to interfere the actions of the former with no other restriction than their own and arbitrary will. The main reason for not enfranchising those who lacked property was, thus, that their will could be easily inﬂuenced by those on whom they depended. Had the dependents been enfranchised, the only beneﬁciaries would have been those on whom they depended—the aristocracy. In a passage very often quoted by those supporting suﬀrage restrictions in the Early Republic, Blackstone had argued that

The true reason of requiring any qualiﬁcation, with regard to property, in voters, is to exclude such persons as are in so mean a situation that they are esteemed to have no will on their own. If these persons had votes, they would be tempted to dispose of them under some undue inﬂuence or other. This
87 See paradigmatically P. Pettit, Republicanism. A Theory of Freedom and Government (Oxford: Oxford University Press, 1997); and Q. Skinner, Liberty Before Liberalism (Cambridge: Cambridge University Press, 1999). 88 See Wood, The Radicalism of the American Revolution, p. 58. 89 Quoted in Wood, The radicalism of the American Revolution, p. 49. 90 See A. Domenèch and D. Raventós, “Property and Republican Freedom: An Institutional Approach to Basic Income,” Basic Income Studies 2 (2007): 1-8. 91 Quoted in Wood, The Creation of the American Republic, p. 219.

52

Chapter 1. From demophobia to demophilia

would give a great, an artful, or a wealthy man, a larger share in elections than is consistent with general liberty.92 And, as Gouverneur Morris put it, Give the votes to people who have no property, and they will sell them to the rich ... The time is not distant when this country will abound with mechanics and manufacturers, who will receive their bread from their employers. Will such men be the secure and faithful guardians of liberty. Will they be the impregnable barrier against aristocracy.93 This idea had been so deeply rooted that even some British Levellers had defended it during the Putney debates in 1647. For example, Petty, a Leveller spokesman had defended that “the reason why we exclude apprentices, servants and beggars is because they depend on the will of other men and they would be fearful of disgusting them.” 94 It is noteworthy that, even though the disenfranchisement of those under inﬂuence is justiﬁed above on behalf of the common interests of the Republic, justiﬁcations were provided also on the very behalf of the disenfranchised. Since they were not materially independent, granting them a say in public aﬀairs would serve as a way to strength the power of their masters, husbands and patrons and therefore would harm not only the interests of their communities but also their actual interests. According to Blackstone, As to qualiﬁcations of the electors. The true reason of requiring any qualiﬁcation ... in voters is to exclude such persons as are in so mean a situation that they are esteemed to have no will of their own. If these persons had votes, they would be tempted to dispose of them under some undue inﬂuence or other. This would give a great, an artful, or a wealthy man, a larger share in elections than is consistent with general liberty. B. Leisure time. In addition, it was argued that owning real estate provided the spare time required to get informed about and take part in politics in a meaningful way. James Harrington had expressed this idea quite clearly when he stated, regarding mechanics, that “till have ﬁrst feathered their nests, like the fowls of the air, whose whole employment is to seek their food, are so busied in their privates concernments that they have neither the leisure to study the public, nor are safely to be trusted with it.” 95 It is important to note that taking part in politics was not seen only as a privilege for those who owned property enough to be leisured, but also as a duty that they were expected to fulﬁll. Benjamin Franklin provides the most well-known
Quoted in Robert J. Steinfeld, “Property and Suﬀrage in the Early American republic,” Stanford Law Review 41, (1989), 335-376, p. 340. 93 Quoted in Keyssar, The Right to Vote, p. 18. 94 Quoted in A. de Francisco, Ciudadanía y democracia. Un enfoque repulicano (Madrid: Los Libros de la Catarata, 2007), p. 139. 95 J. Harrington, The Commonwealth of Oceana and A System of Politics, J. G. A. Pocock, ed. (Cambridge, Cambridge University Press, 1992), p. 138. For the inﬂuence of Harringtonian republicanism over the late eighteenth-century America, see. J. G. A. Pocock, The Machiavellian Moment. Florentine Political Thought and the Atlantic Republican Tradition (Princeton: Princeton Unievrsity Press, 1975), esp. ch. 14.
92

1.3. Actual representation

53

instance of this, for he stopped working as soon as he acquired enough wealth (at the age of forty-two) and devoted exclusively to public aﬀairs for the rest of his life. C. A stake in the society. Finally, it was claimed that the ownership of real estate served as a good indicator of having stable interests in the community—a “stake in the society”—and thus an incentive to use political rights responsibly. This was so, in turn, for two reasons. On the one hand, because real estate, unlike personal property, was not movable and therefore increased the likelihood of staying in the same state, providing owners with a long-term interest in their prosperity. (However, according to Jeﬀerson, this requirement could be satisﬁed by “either having resided a certain time, or having a family, or having property, any or all of them.” 96 ) On the other hand, because owners of real estate had to feel directly the consequences of the laws, since they had to pay taxes. Accordingly, they would have an incentive to acutely monitor the representatives, and prevent them from passing arbitrary statutes. James Iredell clearly synthesized these three reasons—freedom from inﬂuence, political competence and stable interests in the community—when he defended in 1774 that, if all men were independent, then everyone could have “an individual vote for a representative.” But since they were not, “there must be some restriction as to the right of voting: otherwise the lowest and most ignorant of mankind must associate in this important business with those who it is to be presumed, from their property and other circumstances, are free from inﬂuence, and have some knowledge of the great consequences of their trust.” 97 In short, the argument for the exclusion of the propertyless from full citizenship can be summarized as follows. (B1) Only those who are materially independent ought to be granted full citizenship.98 But since, as a matter of fact, in the late eighteenth century America (B2) Not everyone was independent,

It followed that (B3) Not everyone ought to be granted full citizenship.

1.3. Actual representation
This section presents the challenges to virtual representation by those who defended the alternative theory of actual representation, along with its relationship
Quoted in Wood, The Creation of the American Republic, p. 169. Quoted in Wood, The Creation of the American Republic, p. 168. 98 Material independence was considered a necessary condition for full citizenship. But, as Gundersen points out, it was not considered suﬃcient also, for many single women did own property and could meet property qualiﬁcations yet not enjoyed voting rights. See Joan R. Gundersen, “Independence, Citizenship, and the American Revolution,” Signs 13 (1987): 59-77, p. 65. Harrington had already expressed this non-biconditionality when he stated that “The man that cannot live upon his own must be a servant; but he that can live upon his own may be a freeman.” Harrington, The Commonwealth of Oceana and A System of Politics, p. 269.
97 96

54

Chapter 1. From demophobia to demophilia

with the increasing acceptance of “democracy” in American politics in the decades after the founding. It argues that the increasing acceptance of the democracy was in part—although only in part, as it will become clear in chapter 2—due to the increasing acceptance of the circumstance of disagreement and conﬂict on interests and values in the American society, and the crisis of the theory of virtual representation as a result. The more the fact of disagreement was accepted as a legitimate and unavoidable feature of the American society, the less credible virtual representation became, and the more intensely citizens demanded both suﬀrage extensions and the enactment of some of the institutional devices that had been traditionally seen as attached to the idea of democracy. In short, the increasing acceptance of “democracy” and democratic institutions can be partly explained as a consequence of the new and changing circumstances of the nineteenth century. Put simply, the theory of actual representation can be presented, in contrast to the theory of virtual representation, in the following steps: (C1) (C2) (C3) Pluralism tenet: A political community is a plural entity with conﬂicting interests and values. Anti-elitism tenet: Political decisions should be made by those familiar with the interests of ordinary people. Irrelevance tenet: Full citizenship is to be granted to every member of the political community despite of their being materially independent or not. Again, the section reconstructs these three tenets in turn.

1.3.1. Plurality of interests The pervasive use of the theory of virtual representation during the Philadelphia Convention and the ratiﬁcation process was closely related to the perception of an increasing factionalism in post-independence American politics. As we have seen above, this factionalism was denounced by many Federalists as part of the democratic tendencies of the 1770s and 1780s. Hence, the Federalists defended virtual representation as a way to overcome the plurality of interests between and within the states and to put forward a centralized government. And this central government, as it was eventually endorsed in the Federal Constitution, became a very successful alternative to more democratic institutional arrangements. But this perception of an increasing factionalism was also shared by many Antifederalists and ordinary people, who became increasingly aware of, and increasingly accepted, the existence of the fact of disagreement in the American society. As the nineteenth century progressed, this awareness would make turn the organicist conception of the society into less credible, and it would ﬁnally undermine the eﬀectiveness of virtual representation as a rationale for the sharp distinction between virtuous representatives and ordinary people the Federalists had argued for. As ordinary people became more aware of the idiosyncrasy of their interests and values, the demand

1.3. Actual representation

55

for actual representation became also irresistible and virtual representation could no longer be used as successfully as the Federalists had used it in the late 1780s. Of course, plural interests and values, along with the mobilizations aimed at advancing them, were not new for the American public. As soon as the Stamp Act riots broke out, many laborers had begun organizing themselves within their crafts and within their communities against the imports, including the younger and poorer artisans. True, these workers would still have to wait long to obtain the franchise. However, pamphlets were read to and by them, outdoor political rallies and demonstrations were open to them, and organizers of petitions collected signatures from both voters and non-voters alike, which made them more aware of their interests and more willing to organize themselves to advance them. As Nash, Smith and Hoerder have pointed out, [i]n the course of struggling for non-importation, mechanic consciousness, already awakened, was transformed. Most striking, artisans began to exert themselves as a distinct political entity. This is reﬂected in Philadelphia in the newspaper and pamphlet appeals, beginning in 1767, which were directed speciﬁcally at the mechanics as a separate interest.” 99 The evolution of the market economy, the creation of pressure groups and societies, and the split of the Federalists and the rise of the ﬁrst and second party systems would do nothing but make this change more acute. Let us brieﬂy describe these trends in turn. 1.3.1.1. Market economy Pluralism of values and interests became more pressing as the market economy evolved in American cities and ordinary people—mechanics and artisans mostly—increasingly saw their own interests as idiosyncratic. As Michael Merril has concluded in his summary of the current state of the art, the consensus among economic historians is that the transition to market economy in American occurred in the immediate aftermath of the revolutionary war. This view contrasts to the view of the revisionist historians of the 1950s, including Louis Hartz and Richard Hofstatder, who challenged the importance of the transition to capitalism in American history by insisting that the European settler societies were almost fully capitalist from their earliest days.100 This transition aﬀected ﬁnancial, productive and labor markets. As for capital markets, by the ﬁrst decades of the nineteenth century more people more often lent money to more distant debtors, and interest rates and the price of money began to get rid of their ancient customary restraints and gradually converged. As for productive markets, although the American society remained mainly rural compared to its EuroG. B. Nash, B. G. Smith and D. Hoerder, “Labor in the Era of the American Revolution: An Exchange,” Labor History 24 (1983): 414-439, p. 428. 100 See R. Hofstadter, The American Political Tradition and the Men Who Made It (New York: A. A. Knopf, 1948); L. Hartz, The Liberal Tradition in America (New York: Hartcourt, Brac & Co., 1955); and M. Merrill, “Putting “Capitalism” in Its Place: A Review of Recent Literature,” The William and Mary Quarterly 52 (1995): 315-326.
99

56

Chapter 1. From demophobia to demophilia

pean counterparts,101 most farmers were by 1800 much engaged in commerce and the exchange of goods; productivity increased fast, more goods were taken to wider and further markets, and economic inequality enlarged. Finally, as for labor markets, the ratio of persons engaged in agriculture to those engaged in commerce and manufacturers dramatically decreased and the rapid growth caused a greater division of labor and occupational specialization, mainly in the coastal cities.102 On the other hand, the master/servant relationship of hierarchical dependence went through a dramatic legal change and would eventually become a purely contractual engagement between juridical equals, as we shall see in detail below in section 1.3.3.103 All these trends would transform dramatically the structure of the American society, promoting a more self-interested and individualistic behavior, as Max Weber famously captured in Benjamin Franklin’s advices on investing and proﬁt-seeking.104 As market economy evolved, inequalities enlarged, workers increasingly specialized, and the awareness of the plurality of interests in America raised. As Wood points out, the interests of the post-revolutionary years were . . . more numerous . . . more popularly expressive of new widespread economic elements of the society than the colonial period had been. Americans virtually ceased talking about the people’s “interest” in the singular; the people’s “interests”—agricultural, commercial, manufacturing—were all plural now.105 Accordingly, more and more workers and ordinary people organized themselves to advance their particular interests by a number of means. For example, the printing of newspapers and pamphlets directed at each craft as a separate interest; the organization of craft societies and unions, such as the Patriotic Society of the Philadelphia mechanics; the arrangement of places to meet, such as the “Mechanics Hall” bought by the New York mechanics; or public demonstrations and strikes, such as the twelve major strikes carried out by various journeymen between 1786 and 1816 in Philadelphia—the ﬁrst major strikes by employees against their employers in American history.106
101 In 1800 nearly 90 percent of workers were still involved in farming, and only thirty-three towns had a population of twenty-ﬁve hundred or more. By contrast, in 1801 only 36 percent of English workers engaged in farming, and one-third of the people lived in cities. See G. S. Wood, The Empire of Liberty. A History of the Early Republic, 1789-1915 (New York: Oxford University Press, 2009), p. 318; and C. Clark, Social Change in America. From the Revolution to the Civil War (Chicago: Ivan R. Dee, 2006), p. 85. 102 See Keyssar, The Right to Vote, p. 328; Nash, Smith and Hoerder, “Labor in the Era of the American Revolution: An Exchange.” 103 See R. J. Steinfeld, The Invention of Free Labor. The Employment Relation in english and American Law and Culture (Chapel Hill: The University of North carolina Press, 1991); and M. J. Horwitz, The transformation of American law, 1780-1860 (Cambridge, Mass.: Harvard University Press, 1977). 104 M. Weber, The Protestant Ethic and the Spirit of Capitalism, trans. T. Parsons (New York: Scribner, 1958), pp. 48-50. 105 Wood, The Radicalism of the American Revolution, p. 247. 106 See among others Wood, The Empire of Liberty, p. 349; Nash, Smith and Hoerder, “Labor in the Era of the American Revolution: An Exchange;” Staughton Lynn, “The Mechanics in new York Politics, 1774-1785,” in Class Conﬂict, Slavery, and the United States Constitution (New York, Cambridge University Press, 1967), 79-108.

1.3. Actual representation

57

This organizational trend would become more pressing as the nineteenth century advanced. By the 1820s, it had become clear, as a craft wage-earners group put it, that “the interests of the journeymen are separate and in some respects opposite of those of their employers.” 107 In Philadelphia, still the leading manufacturing center in the nation, strikes become regular—no less than ten in 1820—, and journeymen formed the Mechanics’ Union of Trade Associations—the ﬁrst citywide union in America—, created the Mechanics’ Free Press, and in 1828 organized the Working Men’s Party with the goal of electing “men of our own nomination, whose interests are in unison with ours.” 108 1.3.1.2. Republican-democratic societies The French Revolution brought new uses of political terms to the American jargon. At ﬁrst, it fueled the anti-democratic rhetoric of many Federalists. “Look at France!,” claimed Noah Webster. “There you have a picture of real democracy.” “The French revolution has so discredited democracy,” wrote John Jay in a letter, “that I doubt its giving you much more trouble.” 109 And John Adams took advantage of the revolution to recall that “I was always for a republic, not a democracy, which is as arbitrary, tyrannical, bloody, cruel, and intolerable a government as that of Pharalais with his bull is represented to have been. Robespierre is a perfect exempliﬁcation of the character of the ﬁrst bellwether in a democracy.” 110 Yet the revolution also animated many discontents to organize themselves and show their opposition to the Federal government. Since the Federalists had so openly despised democracy and had used the term very much to discredit their enemies, some of these discontents were now willing to label themselves, very controversially, as “democrats.” And so they did. In the mid-1790s, at least forty-two political societies had appeared—and at least sixteen of them called themselves “democratic.” The ﬁrst one in doing so was the Democratic Society of Pennsylvania, and it was citizen Genêt, the French minister to America, who suggested the word “democratic” for their name. In contrast to the craftsmen organizations, though, these societies were not formed by rank and ﬁle workers but by members of the upper classes. Their aims were mainly to stimulate discussion, to support the French Revolution, and to criticize the Federal government. By no means could the Federalist cabinet claim to be representing the interests of the nation any more, they denounced. And, in addition to their criticisms to the Federalists, they also supported suﬀrage extensions, more frequent elections, and more power to the House of Representatives and to the state legislatures. By the end of the decade, most of these societies had disappeared, after being discredited for having supposedly been involved in the Whiskey Rebellion in Pennsylvania.111 As president Washington wrote in a letter, “I consider this insurrection
Quoted in Sean Wilentz, The Rise of American Democracy. Jeﬀerson to Lincoln (New York and London, W. W. Norton & Co., 2005), p. 282. 108 Wilentz, The Rise of American Democracy, p. 283. 109 Quoted in Morantz, “Democracy" and "Republic" in American Ideology, p. 149. 110 Baylin, The Ideological Origins of the American revolution, p. 282 n. 50. 111 The Rebellion broke out in 1794, and was carried out by Pennsylvania country farmers
107

58

Chapter 1. From demophobia to demophilia

as the ﬁrst formidable fruit of the Democratic Societies; brought forth I believe too prematurely for their own views, which may contribute to the annihilation of them.” 112 However, they were very innovative in giving democracy a better name. 1.3.1.3. The ﬁrst and second party systems113 In addition to pressure groups and societies organized to advance their economic and political interests, Jeﬀerson’s Republican party emerged as a ruling alternative to the Federalists. The split of the Federalist Party and the emergence of the ﬁrst party system provoked an increasing partisanship and political participation. Before turning to this phenomenon, it has to be recalled that by the end of eighteenth century parties and partisanship were still considered inimical to the organic interests of the society. As Hofstfader reports, eighteenth-century thinkers had “often postulated that society should be pervaded by concord and governed by a consensus that approached, if it did not attain, unanimity. Party, and the malicious and mendacious spirit it encouraged, were believed only to create social conﬂicts that would not otherwise occur.” 114 On the one hand, the new generations of Federalists surely failed to unseat Jeﬀersonian Republicans—even though they mimicked Jeﬀersonian strategies such as editing partisan newspapers, arranging social events such as barbecues or campaigning at local and state levels. However, their existence promoted political competition, which, in turn, promoted partisanship in politics and called into question the use of the organicist conception in American politics. That the word “logrolling” began to be used to name the daily practice of trading of votes among legislators should not surprise us.115 On the other hand, the increasing participation was not only due to party competition—actually, the Federalist party was still very active between 1796 and 1799, when participation increased slowly, and nonetheless moribund between 1804 and 1816, when it increased quickly—, but also due to the members of the Jeﬀersonian movement itself. The Jeﬀersonians were considered to represent the interests of
against a whiskey excise tax passed in 1791 to pay the war debt. Washington vigorously repressed for being “subversive of the national interest.” Thus, he ordered the states to call out some 15,000 militiament and twenty rebellious men were charged with high treason, though only two were convicted. See J. K. Martin, “The Whiskey Rebellion Redivivus,” in S. Boyd, ed., The Whiskey Rebellion. Past and Present Perspectives (Westport, Ct.: Greenwood Press, 1985), 3-7. 112 Washington to Henrry Lee, August 26, 1794. Quoted in Dupuis-Déri, The Political Power of Words, pp. 226-227. By contrast, Jeﬀerson would declare write that “The denunciation of the democratic societies, is one of the extraordinary acts of boldness of which we have seen so many from the faction of monocrats. It is wonderful indeed, that the President should have permitted himself to be the organ of such an attack on the freedom of discussion, the freedom of writing, printing and publishing.” Jeﬀerson to James Madison, December 28, 1794, in The Political Writings of Thomas Jeﬀerson, M. D. Peterson, ed. (Chapel Hill, NC: University of North Carolina Press, 1993), p. 119 113 The ﬁrst and second party systems are analyzed in more detail below. See infra, §2.2.2 and §2.2.3, respectively. 114 R. Hofstadter, The Idea of a Party System: The Rise of Legitimate Opposition in the United States, 1780–1840 (Berkeley: University of California Press, 1969), p. 21. See also N. L. Rosenblum, On the Side of the Angels. An Appreciation of Parties and Partisanship (Princeton: Princeton University Press, 2008), pp. 89-94. 115 See Wood, The Empire of Liberty, p. 330.

1.3. Actual representation

59

ordinary people and were commonly known as “democrats”—and the party itself as the “Democratic” party.116 As expected, this label was ﬁrstly used by the adversaries of the Jeﬀersonians to discredit them. Hence, during the presidential campaign of 1800, Olivier Ellsworth warned that, in case Jeﬀerson arrived to oﬃce, he would establish a “Jacobinal democracy.” And, for Hamilton, Jeﬀerson was “too much in earnest with his democracy.” 117 As Morantz points out, the “FederaIists attempted to “pin” the name “democratic” on their Jeﬀersonian opponents. With considerable lack of insight, they reasoned that if they could succeed in making the label stick, the Jeﬀersonians would be emphatically rejected at the polls.” 118 However, although the party would not be oﬃcially labeled as the Democratic Party—as we know it today—until the Party’s 1840 national convention, the Jeﬀersonians began using the term, among other reasons, to distinguish themselves from the Federalists—even though Jeﬀerson himself generally hesitated to do so. Thus, the split of the Federalist party and the new political strategies used by the Jeﬀersonians both promoted partisanship in American politics and mobilized the electorate. The organicist tenet lost credibility and ordinary people gave support to the Jeﬀersonian Republicans who, in addition to be seen as representing the interests of ordinary people, were increasingly known as “democrats.” Accordingly, democracy began to be seen much more favorably. However, competition among opposed parties with diﬀerent agendas would not be seen as a perfectly legitimate democratic practice until the emergence of the second party system. During the ﬁrst party system, both the Federalists and the Republicans still introduced themselves as the common interests of the nation and accused each other of betraying the republican principles on which they all agreed. In short, the acceptance of the fact of disagreement as a legitimate feature of the American society, along with partisanship and political struggle as the means to advance conﬂicting interests and values, had not fully arrived yet. And partisanship and political struggle would not turn out to be seen as a natural and even healthy feature of American politics until the harsh and very professionalized political competition emerged in the second party system. Political competition and factionalism would then replace the organicist tenet. As Morantz puts it,

[the ideal of the Founders had been] an organic society, in which right-thinking men dutifully united in the sober determination of the good of the whole. Parties destroyed this harmony.119

See D. H. Fischer, The Revolution of American Conservatism. The Federalist Party in the Era of Jeﬀersonian Democracy (New York: Harper & Row, 1965). Fischer argues that, though unable to win elections, the Federalists were in this period much more active and eﬀective than is generally believed 117 Quoted in Dupuis-Déri, The Political Power of Words, p. 227. 118 Morantz, “Democracy” and “Republic” in American Ideology, p. 151. 119 Morantz, “Democracy" and "Republic" in American Ideology, p. 189.

116

60 1.3.2. Ruling anti-elitism

Chapter 1. From demophobia to demophilia

Many of the pressure groups and democratic societies continued the Whig tradition of political thought and called into question the assumptions of the theory of virtual representation. As we have seen, by pointing out the plurality of interests within the American society against the organic conception put forward by the Federalists. But also by challenging the idea of disinterested representatives as the most desirable ones to run the government, and by supporting the opposing idea that ordinary people should be elected and that institutional devices should be implemented to make sure that representatives remained as accountable and as responsive as possible of the interests of their constituents. The latter challenge was put forward, at least, on three grounds. 1.3.2.1. Interests and self-interest First of all, the criticism of the impartial and disinterested ruling elite was in part a logical move once the organicist assumption had been called into question. Once the plurality of interests was accepted as an inevitable and legitimate feature of the American society, the call for virtuous representatives with the ability to overcome such factional interests turned out to be irrelevant if not absurd. Thus, it was assumed that any imaginable representative would have his own particular interests to promote, and that so they did the Federalists that had drafted the Constitution and had ruled the country afterward. In America, as Tocqueville would write somewhat exaggeratedly in 1840, “the period of disinterested patriotism is gone by forever.” 120 Similarly, as Fischer points out regarding the new generations of Federalists, “the pervasive holism and the honest undisguised elitism of the old school had become anachronisms.” 121 In fact, part of the old guard acknowledged this phenomenon. For example, in 1821 Madison called a constitutional convention in Virginia to extend the franchise to the propertyless because, as he claimed, “there are various ways in which the rich may oppress the poor ... It is necessary that the poor should have a defense against the danger.” 122 Thus, it is hardly surprising that oﬃcials began to be paid for the ﬁrst time and increasingly professionalized. In line with the traditional view, the Federalists had put forward a conception of politics as as a public spirited activity and a non remunerated one. Washington—one of the wealthiest man in the country at the time—had famously refused a salary as commander in chief. And Benjamin Franklin had outrageously opposed any attempt to remunerate public oﬃcials. (Of course, these attempts were often put forward by representatives who were not as well oﬀ as Franklin.) This was a conception of oﬃce-holding inherited from the elitist branch of classical republicanism.123 Thus, when Aristotle had discussed how to implement his politeia—a model on which many Federalists relied when trying to work out their
de Tocqueville, Democracy in America, chapter 14. Fischer, The Revolution of American Conservatism, p. 199. 122 Quoted in Clark, Social Change in America, p. 111. 123 For a historical distinction of the elitist and democratic branches of the republican tradition, see A. Domènech, El eclipse de la fraternidad (Barcelona: Crítica, 2004).
121 120

1.3. Actual representation

61

ideas of a mixed republic—, he famously called for the abolition of the misthós, the allowance paid to the members of the Court of the Five Hundred, the People’s Court, the Assembly and some magistrates of the Athenian democracy. The misthós had been introduced by Ephialtes in 461 B.C., and had been eliminated and reintroduced in diﬀerent versions afterward. If we have to believe Aristotle, it had increased the participation of the free poor citizens and make the upper classes drift away from political activity, thus turning Athens in a democracy—i.e. the “rule of the poor.” 124 It should be noted, however, that those who pointed out the plurality of interests against the organicism of the Federalists did not necessarily entail that everyone was by deﬁnition self-interested, but only that parochial interests indeed existed in America and should be taken into account. These interests included, most notably, the interests of the members of the ruling elite, who had presented themselves as disinterested and virtuous, but were now denounced for having particular interests too—and advancing them when possible—, as many other members of the society. To put it clearly, the claim was not so much that everyone was by deﬁnition self-interested, but rather that the ruling elite, however supposedly disinterested, had also interests as ordinary people did. According to Wood, the Federalists rather than the Antifederalists were actually the ones who were ruling out the possibility of ordinary people being disinterested (and thus able to take part in politics in a virtuous form) by drawing a clear line between “the rich and well-born,” on the one hand, and “the mass of the people [who] seldom judge or determine right,” on the other.125 Eventually, this distinction would undermine their capacity to attract the support of those who had despised for being purely self-interested. As Wood puts it, “by resting their government so completely on the assumption that most people were self-interested and absorbed in their private aﬀairs, the Federalists laid virtually no civic foundations for their scheme and weakened their ability to justify their peculiar disinterested leadership.” 126 1.3.2.2. Epistemic access The desirability of a sharp distinction between representatives and their constituencies was criticized not only on logical grounds—once the very possibility of disinterestedness had been criticized—, but also on epistemic ones. Even if there were disinterested politicians with the ability to overcome all factional interests—including their own ones—, they could not speak for ordinary people unless they were familiar with their “feelings, circumstances, and interests.” 127 The point had been already made by the Antifederalists during the ratiﬁcation
Aristotle, Politics, in Jonathan Barnes (ed.) The Complete Works of Aristotle (Princeton: Princeton University Press, 1984); and M. H. Hansen, The Athenian Democracy in the Age of Demosthenes, trans. J. A. Crook (Oxford and Cambridge, Mass.: Blackwell, 1991), p. 330. 125 Alxander Hamilton, in Farrand, The Records of the Federal Convention of 1787, vol. 1, p. 299. Thomas Govan has called into question, however, whether Hamilton actually said this, which contradicts Madison’s notes and Hamilton’s own outline of the speech. See T. P. Govan, “The Rich, the Well-born, and Alexander Hamilton,” The Mississippi Valley Historical Review 36 (1950): 675-680. 126 Wood, The Radicalism of the American Revolution, p. 264. 127 Quoted in Wood, The Radicalism of the American Revolution, p. 255.
124

62

Chapter 1. From demophobia to demophilia

process, who had claimed that “The well born, and highest orders in life, as they term themselves will be ignorant of the sentiments of the middling class of citizens, strangers to their ability, wants, and diﬃculties.” 128 And it gained a wider importance as the organicist tenet lost credibility. At a campaign rally in 1810, Jeﬀersonians advised the assembled mechanics, laborers and farmers to elect men of their own kind. The reason was an epistemic one: “does a nobleman,” they argued, “know the wants of the farmer and the mechanic?” Even more interestingly, during the very same campaign, they attacked Jonas Platt, the Federalist candidate, for promoting that elected representatives voted their conscience rather than merely reﬂected the interests of their constituents.129 Once the existence of a plurality of idiosyncratic interests was acknowledged, political representation turned out to be seen as a means to mirror factional interests in the legislative assemblies—“descriptive representation,” in Pitkin’s terminology. As Brutus had claimed in his rejoinder to The Federalist, “The very term, representative, implies, that the person or body chosen for this purpose, should resemble those who appoint them ... They are a sing—the people are the thing signiﬁed.” 130 As for actual devices to ensure actual representation, besides the size of the legislatures—which had to be large enough as to ensure that social diversity could be properly represented—, two means were proposed. First, to guarantee reserved seats for representatives of the main social groups—so they could actually represent their particular interests and, therefore, those of their constituents. As Brutus had put it, “each class ought to have an opportunity of choosing their best informed men for the purpose.” 131 And second, to provide citizens with the legally binding ability to instruct their representatives. These two means are obviously related and they were often demanded together indeed. But they can be nonetheless distinguished. On the one hand, binding instructions would make unnecessary for the representatives to be actual members of the group they represented. On the other hand, even if the former and the latter had equal interests, asymmetries in the level of knowledge about how to advance them, unforeseen contingencies, or mere trust could make the practice of binding instructions unnecessary or even counterproductive. 1.3.2.3. Common sense A third reason was put forward against the ruling elite tenet—namely, the priority of common sense over expertise as a requirement for taking part in politics. As part of the ideal of equality embraced in the Declaration of Independence, it was increasingly believed that all men were equal and with an equal capacity to reason about public aﬀairs. Two implications followed. Firstly, common sense was taken to be a capacity that, unlike expertise, everyone was in a position to have. Secondly, common sense
128 H. J. Storing, The Anti-Federalist (Chicago: University of Chicago Press, 1985), p. 123. See also Manin, The Principles of Representative Government, pp. 108 ss., and T. Ball, “A Republic—If You Can Keep It,” in Conceptual Change and the Constitution, T. Ball and J. G. A. Pocock, eds. (Lawrence: University Press of Kansas, 1988). 129 See H. Strum, “Property Qualiﬁcations and Voting Behavior in New York, 1807-1816,” The Journal of the Early Republic 1 (1981): 347-371, p. 367. 130 Storing, The Anti-Federalist, p. 123. 131 Storing, The Anti-Federalist, p. 125.

1.3. Actual representation

63

was increasingly believed to be at odds with the kind of very sophisticated education received by the gentry. The Charleston mechanics had already appealed to common sense during the revolutionary period. When William Henry Drayton despised their right to take part in political decisions because they lacked the education required to do so, they appealed to common—a common sense which lack “cannot be compensated by all the learning of the school.” 132 By the early nineteenth century, this view of ordinary people’s ability to take part in politics independently of their expertise would be so widely accepted that aristocratic gestures would even have a negative eﬀect over the candidates who exhibited them. Thus, when Simon Snyder was elected governor of Pennsylvania in 1808, he refused an honor guard at his inauguration because, as he claimed, “I hate and despise all ostentation—pomp and parade as anti-democratic.” And when his adversaries despised his humble origins by calling him and his followers “clodhoppers,” he and his supporters not only did not try to dismiss the label, but proudly began to use it to label themselves.133 Eventually, Andrew Jackson would acknowledge the importance of this trend and would exploit it masterfully during the 1928 presidential campaign, as it will become clear later. The revolutionary potential of this view only becomes clear once we recall the highly hierarchical character of colonial America, a society organized around a complex system of well established ranks, relations of patronage, and “friendships.” Deference to superior ranks had been always present, beginning by the gentry’s wearing of wigs or by ordinary people’s doﬃng of caps in presence of gentlemen. And dishonor was often paid with death through the frequent practice of dueling among gentlemen (Hamilton himself would die shot by Jeﬀerson’s Vice President Aaron Burr). Unsurprisingly, this social deference also applied to public aﬀairs, as it is shown by the fact that, until the 1770s, ordinary people tended to elect members of the gentry when given the opportunity to vote for public oﬃce, as we have shown above in section 1.2.2.134

1.3.3. Transformations of material independence The crisis of the ruling elite tenet aﬀected the rationale for property qualiﬁcations regarding both passive and active suﬀrage. More and more ordinary people demanded suﬀrage extensions so they could choose their own preferred candidates. Likewise, voting rights were eventually granted—notably, in the state constitutional conventions called in the 1830s, 1840s and 1850s—, though only gradually. This challenge aﬀected the three main basis on which material independence had been defended as a necessary condition for suﬀrage—as a means to prevent the likely inﬂuence by those on whom otherwise the propertyless were likely to depend, as a means for political competence and participation, and as a proxy for the attachment to the interests of the community.
132 133 134

Quoted in Wood, The Radicalism of the American Revolution, p. 277. Wood, The Empire of Liberty, p. 331. See Main, “Government by the People,” p. 392.

64

Chapter 1. From demophobia to demophilia

As Robert Steinfeld has pointed out, the challenge was developed along two very diﬀerent, and somehow contradictory, lines. One towards a change in the way material independence was conceived and another one towards the universal extension of rights and the abandonment of the requisite of material independence altogether—call them, for the sake of simplicity, the self-ownership position and the equality of rights position.135 Since these positions would dramatically shape the enfranchisement of parts of the population that had been excluded from political participation—namely, wage laborers—and the very idea of democracy, on the one hand, and the way the employment relationship was understood (an issue that will be central to Part III of the dissertation, as we shall see below), let us see them in turn. 1.3.3.1. Property as self-ownership Firstly consider the property as self-ownership position, according to which, Property as self-ownership.—Wage laborers are to be taken as materially independent, and thus as worth being granted full citizenship, because they are owners—owners of their labor force. In an age of commercial growth, property ceased to be identiﬁed with real estate and to be seen as a long-term attachment to the interests of the community. Rather, it began to be seen as a commodity to be exchanged in the marketplace. As Morton Horwitz has put it, “in the early years of the nineteenth century ... the idea of property underwent a fundamental transformation—from a static agrarian conception ... to a dynamic, instrumental, and more abstract view of property that emphasized the newly paramount virtues of productive use and development.” 136 This shift aﬀected wage laborers, who had been seen as propertyless and as being dependent and not worth being granted full citizenship hitherto. On the one hand, the number of wage laborers in contrast to farmers increased dramatically, which made the ideal of a society of yeomen embraced by some—notably, Jeﬀerson—appear to be less likely to ever occur. On the other hand, journeymen also found increasingly diﬃcult to carry out the three-step climb from apprentice to journeyman to craftsman, which for an artisan had been the only way to go from servitude to independence in the pre-revolutionary period.137 By contrast, rather than unpropertied and thus dependent, wage laborers were increasingly seen now as owners—owners of their own labor force, which could be rented in the market—and therefore as self-standing economic agents. Accordingly, employment ceased thus to be seen as a relationship of dependence and turned out to be seen as a purely contractual engagement between juridical equals. As a member of the Massachusetts State convention explained in 1853,
Steinfeld, “Property and Suﬀrage in the Early American republic.” See also Steinfeld, The Invention of Free Labor. 136 Horwitz, The transformation of American law, 1780-1860, p. 31. 137 Among Philadelphia cordwainers and tailors, for example, 38 percent of the journeymen managed in 1756 to become master craftsman within a decade. By contrast, only 3 percent of them did so during the ﬁve years after 1767. And the tendency persisted. Nash, Smith and Hoerder, “Labor in the Era of the American Revolution: An Exchange,” p. 426.
135

1.3. Actual representation

65

In a free government like ours, employment is simply a contract between parties having equal rights . . . The employed is under no greater obligation to the employer than the employer is to the employed . . . In the eye of the law, they are both freemen—citizens having equal rights.138

Despite this shift, it has to be noted that at that moment employment could hardly be deﬁned as a purely voluntary relationship between free and equal agents. At least, while legal sanctions against workers who quit before they had performed their contractual agreement were still into force. But non-pecuniary sanctions progressively disappeared as the nineteenth century progressed. As a result, workers gained the right to formally enter and exit their jobs and employment would be more easily deﬁned as a free and voluntary relationship.139 Hence, wage workers were not acknowledged as independent members of the community and thus as political equals because the link between property and independence dissolved. But rather because they were acknowledged as owners—owners of themselves and thus as legally entitled to rent their labor force in the labor market and to enter employment relationships as juridical equals. As Carole Pateman has pointed out, this kind of ownership rested (and rests) on a “political ﬁction.” Even though it is the case that parts of that property can be technically separated from its owner and transferred to the purchaser (e.g., nowadays sperm or some organs), there are other parts that cannot be technically separated from the owner (e.g., talents or skills); they can be separated and purchased only as if they were another asset.140 Nevertheless, thanks to this ﬁction wage workers turned out to qualify as owners and thus as independent members of the community while the link between ownership and independence remained intact. This shift had an important incidence over two additional groups—leisured landowners and paupers. Regarding the former, as the distinction between real and personal property dissolved and the increasingly extended wage labor gained respectability, landed property turned out be seen as “only one of the incidental rights of the person who possesses it,” as the New York Republicans claimed, rather than as “the attributes of a man’s personality that gave him a political character,”.141 Men who legally disposed of their own labor and were able to support their families were considered now as perfectly independent agents. As one reformer in the 1829 Massachusetts convention said of “laboring men”:

Quoted in Steinfeld, “Property and Suﬀrage in the Early American Republic,” p. 351. See Steinfeld, The Invention of Free Labor, and Horwitz, The transformation of American law, 1780-1860. 140 C. Pateman, “Self-Ownership and Property in the Person: Democratization and a Tale of Two Concepts,” The Journal of Political Philosophy 10 (2002): 20-53. See also Domènech, El eclipse de la fraternidad, pp. 94-95; C. B. MacPherson, The Political Theory of Possessive Individualism: Hobbes to Locke (Oxford: Oxford University Press, 1963); and K. Polany, The Great Transformation. The Political and Economic Origins of Our Time (Boston, Mass.: Beacon Press, 1957). 141 Quoted in Wood, The Creation of the American republic, p. 219, and in Wood, The Radicalism of the American Revolution, p. 264.
139

138

66

Chapter 1. From demophobia to demophilia

they have no freehold—no property to the amount of two hundred dollars, but they support their families reputably with their daily earnings.142 As we shall see below in Part III (§§ 7.3.5 and 8.5), this conception of wage labor contrasted with the classical republican one, which considered it as a humiliating and dependent form of work. Cicero, for instance, had famously argued that “vulgar are the means of livelihood of all hired workmen whom we pay for mere manual labor, not for artistic skill; for in their case the very wage they receive is a pledge of their slavery.” 143 Wage workers were increasingly seen now just as the contrary—as perfectly respectable and self-standing workers. And this shift brought leisured aristocratic landowners into contempt. Recall that, according to classic republicanism, ownership of property, and specially of property of land, had been seen as a precondition for having the leisure time required to get informed about and take part in politics—besides serving as a proxy for having stable interests and for having and autonomous will. The Federalist gentry had thus introduced themselves as owning such property and, therefore, as being leisured enough to devote their lives to the public realm. Now that conception had changed dramatically. Not only property of land had lost its aura of autonomy—let alone the technical problem of measuring who qualiﬁed actually as economically independent, due to the high degree of indebtedness of many landowners. But it was also the case that work was increasingly celebrated as an indicator of the commitment to the community. Thus an astonished Tocqueville would report not only how “honorable” work in America was, but also how honorable it was to “work to make money.” 144 And, as Wood points out, talented workmen and inventors were now receiving the civic wreaths that traditionally belonged to the virtuous and wise political elite. The latter, in turn, were now seen as men brought up “in idleness, dissipation and extravagance,” as the Jeﬀersonians referred to the Federalist gentry.145 Leisure was increasingly associated with idleness, rather than with independence and public spiritedness; and salaried work, in turn, with industriousness rather than with dependence. Furthermore, many salaried workers in the cities showed to be much better informed and more engaged in politics than the landowning gentry in the countryside, despite their not being propertied whatsoever in the classical sense. Let us turn now to the case of paupers. As wage labor gained respectability workers began to speak out against census suﬀrage and to claim for suﬀrage extensions, which were actually granted though only gradually. In numerous states, constitutional conventions aimed at amending suﬀrage qualiﬁcations and extending the vote were called in the 1820s, ’30s and ’40s. Eventually, a majority of states replaced their property owning qualiﬁcations either with taxpaying qualiﬁcations or with provisions for white manhood suﬀrage. However, except for the new Western states, most of them also included provisions for pauper exclusion. The rationale for these provisions was no other than the independence argument. In the 1831
142 143 144 145

Quoted in Steinfeld, “Property and Suﬀrage in the Early American Republic,” p. 364. Cicero, De Oﬁciis, trans. W. Miller (Cambridge, Harvard University Press, 1913), I. XLII. Alexis de Tocqueville, Democracy in America, . Quoted in Wood, The Radicalism of the American Revolution, p. 278.

1.3. Actual representation

67

Delaware convention, for example, the committee which added a pauper exclusion to the taxpaying qualiﬁcation—in eﬀect since 1792—explained to the convention that Paupers who live on the public funds and who were under the direction of others, who might control their wills, ought not to be permitted to vote.146 The same applied to women, native Americans and blacks, who with few exceptions would still have to wait long to enjoy voting rights. In addition to purely sexist and racist motivations, the independence argument was frequently used. At the 1821 New York convention, for example, a delegate opposed black suﬀrage because blacks were “incapable, in my judgment, of exercising that privilege with any sort of discretion, prudence, or independence.” 147 1.3.3.2. Equality of rights According to the ﬁrst line of thought presented above, wage workers demanded suﬀrage not because they claimed to have natural voting rights, but on their being materially independent. That is, although the notion of material independence was dramatically restated in the decades after the revolution, it kept its standing as a necessary condition for enjoying full citizenship. However, the very idea of material independence as a necessary condition for enjoying full citizenship did not go without challenge. The ideal of equality embraced in the Declaration of Independence had such a powerful normative force and contagious eﬀect that everyone found in it an intelligible language to express their demands.148 Hence, the unpropertied, who had been considered likely to be inﬂuenced and thus unable to have their own judgment, were now demanding the franchise not only qua owners of their labor force but just qua equals. That is, they demanded their rights not only on the basis that having a job made them independent, or that their paying taxes made them full members of the community. Rather, they argued that they should be granted political rights despite their having a job or paying taxes. Put simply, Equality of rights.—All members of the political community are worth being granted full citizenship despite their being materially independent or not. Suﬀrage extensions thus followed the path towards granting full citizenship despite owning property or not—rather than extending full citizenship by granting property to everyone, as Jeﬀerson had put forward in the 1770s. The link between property and democracy would thus dissolve as a consequence of the increasingly widespread equality of treatment. Likewise, material independence gradually ceased to be seen as a necessary condition for full citizenship. Even though this did not aﬀect slaves and native Americans—two issues that would have still to wait long—, once the suﬀrage had been extended, it seemed increasingly unjustiﬁed to exclude women and
Quoted in Steinfeld, “Property and Suﬀrage in the Early American Republic,” p. 362. Quoted in Keyssar, The Right to Vote, p. 45. 148 The beginning of the Declaration read as follows: “We hold these truths to be self-evident, that all men are created equal, that they are endowed by their Creator with certain unalienable Rights, that among these are Life, Liberty and the pursuit of Happiness.”
147 146

68

Chapter 1. From demophobia to demophilia

the poor, as John Adams had anticipated with fear. “The same reasoning which will induce you to admit all men who have no property, to vote ... will prove that you ought to admit women and children; for, generally speaking, women and children have as good judgments, and are as independent minds, as those men who are wholly destitute of property.” 149 Wood reports, for example, that employees increasingly hesitated to call their employers “master” or “mistress.” Rather, they called them “boss,” a derivation of the Dutch word for master, as an euphemistic substitute. And this shift also took place the other way around. For example, a minister of Maine used the euphemism “help” for a domestic maid because, unlike the word “servant,” it implied “a sense of independence and hope of rising in the world.” 150 Political rights would be claimed now on the basis of having equal rights rather than on being owner of property. As we have already seen (supra, §1.3.2.3), the common sense of the many ordinary and unpropertied people had been opposed to the political expertise of the propertied few. Nondemocratic republican political thought had seen the latter as a model for public spiritedness and virtuous citizenship. And denounced the former, on the contrary, for their being materially dependent and therefore self-interested and politically unable. Now, by contrast, property ceased to be seen as a “proof of superior virtue, discernment, or patriotism” because every man was considered to have not only a share in the common sense required to take part in public aﬀairs but also a right to do so, even though the formal acknowledgment of such equality would have still to wait long.151 1.3.3.3. Two liberal positions The two changes in the deﬁnition of material independence analyzed in this section—the self-ownership and the equality of rights positions—had political implications that departed from the nondemocratic republican position presented above (supra, §1.2.3). But they did so very diﬀerently. The proponents of the self-ownership position still accepted the formal structure of the argument put forward by nondemocratic republicans. Although they redeﬁned what being materially independent meant—thus including wage earners as materially independent—, they also accepted that, as a matter of fact, not everyone was independent—most notably, paupers, women, native Americans and blacks—and thus worth being granted full citizenship. By contrast, the proponents of the equality of rights position departed dramatically from both nondemocratic republicans and the proponents of the self-ownership position. Contrary to the conclusion reached by both of them, they did accept, as a matter of principle, that (B3*)
149 150 151

Everyone ought to be granted full citizenship.
Quoted in Keyssar, The Right to Vote, p. 1. Wood, The Creation of the American Republic, p. 346. Strum, “Property Qualiﬁcations and Voting Behavior in New York, 1807-1816,” p. 359.

1.3. Actual representation

69

But they did so by dropping premise (B1) altogether—i.e. by defending that full citizenship be granted to every member of the political community despite their being materially independent or not. It is noteworthy that, however separated in this period, proponents of the self-ownership and of the equality of rights positions, would eventually ﬁnd a path of convergence. The former had helped restate the notion of material independence so wage workers could qualify as being independent and thus as worth being granted full citizenship in contrast to paupers, women, Native Americans and blacks. However, as time went by, it became clear that their distinction was a weak one—which would be attacked on two basis. First, the distinction proved to be extremely vague in practical terms. As Steinfeld points out, the existence of cyclical and seasonal unemployment and of poor relief made technically diﬃcult to keep the two categories of the propertyless but independent wage laborers, on the one hand, and of the dependent paupers, on the other, sharply separated.152 Second, due to the increasing number of women and blacks entering the labor market as wage laborers, the deﬁnition of material independence as being actually salaried turned out to apply to virtually all workers. Thus all workers, including women and blacks, could be said to own their own labor force. Put diﬀerently, since virtually everyone turned out to be salaried, virtually everyone could be taken to exchange their labor force in the market. Likewise, since virtually everyone could be taken to own their labor force, virtually everyone could be taken as worth being granted voting rights. In short, the link between material independence and full citizenship gradually dissolved in practice, giving way to the following uniﬁed tenet: Equality of rights*.—All members of the political community are worth being granted full citizenship, either because all of them are materially independent due to their being owners of their labor force or because their being materially independent or not is not relevant at all. 1.3.4. Republican democracy In Gordon Wood’s inﬂuential account, the democratization process in America can be explained as a shift from a classic and nondemocratic republican society, on the one hand, toward a more modern and democratic liberal one, on the other. To be sure, this is an account perfectly consistent with the one developed here. As we have seen so far, as market economy evolved, ordinary people became more aware of the plurality of interests within the American society and the theory of virtual representation gradually lost its credibility. Once self-interestedness had been accepted as an inevitable and even desirable economic and political disposition in modern America, it became increasingly diﬃcult for the members of the ruling elite to present themselves as impartial and virtuous representatives and, in turn, ordinary people’s demands for being actually rather than virtually represented became irresistible. In a nutshell, the classical commitment to civic virtue gave way to the acceptance and
152

Steinfeld, “Property and Suﬀrage in the Early American republic,” pp. 370-374.

70

Chapter 1. From demophobia to demophilia

even celebration of self-interestedness and, as Wood puts it, “such a view marked and end to classical republicanism and the beginnings of liberal democracy.” 153 However, a third way between nondemocratic republicanism and democratic liberalism can be identiﬁed in the period studied—for the sake of simplicity, call it democratic republicanism. Although the Federalists had very sharply opposed republican government to democratic government, this distinction was rather artiﬁcial and contrary to the more traditional view that republics could be either aristocratic or democratic, as famously stated by Montesquieu.154 “Republics can be divided into democraticks and aristocraticks,” had a Massachusetts writer put it in the aftermath of the War.155 So, republican democrats of the period supported democracy without embracing liberalism and embraced republican political theory without giving up democracy. And they applied this thinking, at least, to the independence tenet embraced by the former and rejected by the latter. Consider it in some detail For democratic republicans the relationship between democracy and material independence was a two-way one. They believed that, as a necessary condition of democracy, “no citizen should be rich enough to buy another, and none so poor as to be forced to sell himself,” as Rousseau had famously claimed.156 Put it simply, democracy implied that two clauses had to be met, as it is clear from Rousseau’s statement. Call them the minimum property and the maximum property clauses. The former clause was deﬁned in absolute terms. For reasons already analyzed in detail (supra, §1.3.4), democratic republicans accepted that material independence served indeed as a necessary condition for freedom as non-domination—i.e. premise (B1) above—, just as nondemocratic republicans, such as the Federalists, did. (B1) Only those who are materially independent ought to be granted full citizenship.157 But since, contrary to their nondemocratic foes (but equally to the liberal democratic ones), they also considered that, as a matter of principle, it should be the case that (B3*) Everyone ought to be granted full citizenship,

They concluded that (B2*) Everyone ought to be materially independent.

Hence, the few republican democrats who defended this position stood up against both democratic liberals and nondemocratic republicans. Contrary to democratic
153 154

Wood, The Radicalism of the American revolution, p. 296. See C.-L. Montesquieu, De l’esprit des lois, Laurent Versini, ed. (Paris: Gallimard, 1995), Quoted in Wilentz, The Rise of American Democracy, p. 9. J.-J. Rousseau, The Social Contract, trans. Maurice Cranston (Baltimore: Penguin, 1968),

Book 3.
155 156

p. 96.

Material independence was considered a necessary condition for full citizenship. But, as Gundersen points out, it was not considered suﬃcient also, for many single women did own property and could meet property qualiﬁcations yet not enjoyed voting rights. See J. R. Gundersen, “Independence, Citizenship, and the American Revolution,” Signs 13 (1987): 59-77, p. 65. Harrington had already expressed this non-biconditionality when he stated that “The man that cannot live upon his own must be a servant; but he that can live upon his own may be a freeman.” Harrington, The Commonwealth of Oceana and A System of Politics, p. 269.

157

1.3. Actual representation

71

liberals, they did accept that material independence was a necessary condition for full citizenship. And, contrary to nondemocratic republicans, they also accepted that full citizenship—and thus material independence—should be guaranteed to as many citizens as possible. Even though it is not easy to apply the category to actual policies and proposals put forward in the period, redistributive policies put forward by radical republicans—including Thomas Paine, the Duane-Leib faction of the Pennsylvania Jeﬀersonians or workingmen intellectuals such as Stephen Simpson and Thomas Skidmore—could be seen as instances of this position158 Likewise, in her admonition to her husband John Adams to “Remember the Ladies,” Abigail Adams accepted the requirement of economic independence for political independence, yet argued for a revision in married women’s property rights. According to Joan Gundersen, “Because Abigail Adams accepted the belief that a voter needed economic independence, voting rights were irrelevant to her until the married had control of her own property.” 159 More radically, representatives of workingmen of the like of Stephen Simpson, Thomas Skidmore, and William Heighton responded to the novel phenomena of industrialization and wage labor by arguing that no property relations were legitimate if they allowed some to depend on others and that independence, in turn, required redistribution of property.160 Hence, the Report of the Committee of Fifty published in the the Working Man’s Advocate, the main organ of New York Workingmen’s Party, argued for ‘AN EQUAL AMOUNT OF PROPERTY ON ARRIVING AT THE AGE OF MATURITY, and previous thereto, EQUAL FOOD, CLOTHING, AND INSTRUCTION AT THE PUBLIC EXPENSE.” 161 But for democratic republicans the relationship between democracy and material independence also worked the other way around, implying also the maximum property clause. This latter clause was not deﬁned in absolute terms but in relative or comparative ones. It was believed that, in a democracy, inequalities should not be as large as to endanger political equality in decision making. This latter clause was acknowledged by William Findley, when he stated in 1786 that Pennsylvania was “too unequal in wealth to render a perfect democracy suitable to our circumstances ... If our wealth is less equal than our kind of government seems to require . . . Democracy must fall before it.” 162 As Andrew Shankman concludes commenting on Findley, for democratic republicans “Perfect democracy required perfect equality, not just before the law but also in wealth, in property.” 163 It was believed that those concentrating large amounts
See Thomas Paine, “Agrarian Justice,” in Political Writings, Bruce Kuklick, ed. (Cambridge: Cambridge University Press, 2000); Shankman, Crucible of American Democracy; and A. Gourevich, “The Unﬁnished Revolution: Freedom and Property in the Nineteenth Century Republicanism,” unpublished draft, Brown University. 159 Gundersen, “Independence, Citizenship, and the American Revolution,” p. 75. 160 See Gourevitch, “The Unﬁnished Revolution,” esp. section II. 161 The Working Man’s Advocate, Saturday, October 31, 1829, 1.1, page 1. Quoted in Gourevitch, “The Unﬁnished Revolution,” p. 34. 162 W. Findley, Address to the Assembly, March 31, 1786, in Mathew Carey, ed., Debates & Proceedings of the General Assembly of Pennsylvania (Philadelphia, 1786), p. 66. 163 Shankman, Crucible of American Democracy, p. 4.
158

72

Chapter 1. From demophobia to demophilia

of property would otherwise be in a bargaining asymmetric position and would be able thus to dominate the process of political decision-making. The debate between Robert Morris and William Findley over whether the Pennsylvania legislation should recharter the Bank of America can be seen, thus, as a struggle between nondemocratic and democratic republicans rather than, as Wood has suggested, as one between nondemocratic republicans and democratic liberals.164 Put brieﬂy, Findley believed that, even if Pennsylvania was not perfectly equal in 1786, inequality was not large enough yet to completely endanger democracy. No individual or group possessed the wealth required to control the political decision-making process and shape society to serve their private needs. The Bank of America, by contrast, was a huge concentration of wealth and would be able thus to threaten democracy. It is worth mentioning that the reasons put forward by democratic republicans to embrace the independence tenet were very similar to those put forward by nondemocratic republicans. As William Findley put it, when wealth was distributed in many hands, “every man in the disposal of his own wealth, will act upon his own principles. His virtue, his honour, his sympathy, and generosity, will inﬂuence his disposals and designs, and he is in a state of personal responsibility.” 165 Their argument was that, in case everyone owned property enough not to depend on others to make a living, they would have an incentive to monitor decisions made in the political sphere, behave responsibly toward the commonwealth and ensure the institutional conditions required for their material independence and equality—among others, agrarian laws and ﬁscal policies. The society of virtuous yeomen embraced by Jeﬀerson served as an obvious ideal for them. Note that, according to this conception, and contrary to what some contemporary commentators have written about the republican tradition, civic virtue was by no means intrinsically valued as a part of a broader conception of the good life (in contrast to what sometimes has been labeled as civic humanism or neo-Aristotelianism). Well on the contrary, civic virtue and political participation were valued merely as a means to secure citizens’ material independence, so they could behave as they pleased without being under others people’s arbitrary will. As Skinner points out when discussing this trend within republicanism—and most notably, Machiavelli’s Discorsi—, “the writers I am considering never suggest that there are certain speciﬁc goals we need to realise in order to count as being fully or truly in possession of our liberty.” 166 In short, if ownership of real estate was a necessary condition for being materially independent and this, in turn, for having the epistemic and dispositional capacity to take part in politics in a responsible way: Why not ensure that everyone owned
See A. Shankman, Crucible of American Democracy. The Struggle to Fuse Egalitarianism & Capitalism in Jeﬀersonian Pennsylvania (Lawrence: University of Kansas Press, 2004), pp. 3-9; Wood, The Radicalism of the American Revolution, pp. 256-259; and Wood, “Interest and desinterestedness in the Making of the Constitution,” in Beyond Confederation. Origins of the Constitution and American National Identity, R. Beeman, S. Botein and E. Carter II, eds. (Chapell Hill: University of North Carolina Press, 1987), 69-109. 165 Findley, “Address to the Assembly,” March 31, 1786, p. 66. 166 Quentin Skinner, “The Paradoxes of Political Liberty,” in The Tanner Lectures on Human Values, vol. 7 (Salt Lake City, The University of Utah Press, 1986), p. 233.
164

1.3. Actual representation

73

property instead of excluding a large majority of the population from full citizenship? This was the question democratic republicans posed to their nondemocratic republican foes. And it was the question they attempted to answer by embracing democracy without giving up the independence tenet, as liberal democrats did. Obviously, this answer would have to be deeply reconsidered when the economic and social circumstances of the American society dramatically changed as the nineteenth century drew on and beyond. However, the tradition of democratic republicanism has survived the test of time and is still well alive, as a number of recent political proposals have shown, and as I will try to show in Part III, chapter 8, of the dissertation as for the case of workplace democracy.167

Consider, for example, recent proposals to give a stake to every citizen as she reaches maturity, or to provide a basic income to every citizen, which have sometimes been conscientiously related to the tradition of democratic republicanism. As for the former, see B. Ackerman and A. Alstott, The Stakeholder Society (New Haven: Yale University Press, 1999), p. 11. As for the latter, see Domènech and Raventós, “Property and Republican Freedom;” and D. Raventós, Basic Income: The Material Conditions of Freedom (London, Pluto Press, 2007).

167

Chapter 2

Conceptual shift
2.1. Introduction
The previous chapter has attempted to show some of the social, political and ideological changes behind the increasing acceptance of the idea of democracy, along with the institutional devices traditionally attached to it, in the later decades of the ratiﬁcation of the Constitution. These changes can be summarized as a shift from the theory of virtual representation, still widely accepted in eighteenth-century America to the theory of actual representation, increasingly supported in the ﬁrst half of the nineteenth century. However, these changes do not fully explain why the Founders openly referred to the American constitutional system as an undemocratic system whereas in less than half a century virtually everyone—the heirs of the Federalists included—turned out to refer to very much the same system as a perfectly democratic one. Democracy could have gained increasing support, but the institutional arrangements included in the Constitution and enacted by the Federal government were still very much those of 1787. How could it be that very much the same devices turned out to be considered as perfectly democratic ones in less than half a century? This chapter shows that, besides its increasing acceptance, the idea of democracy also went through a semantic shift during the decades after the ratiﬁcation of the Constitution. Although some features of the concept of democracy—and some of the institutional devices associated to it—remained the same indeed, some others changed. This helps to explain why the Federalists believed that the representative government they put forward was not only diﬀerent from democracy but in fact a very check against it. And it also helps to explain why, some decades later, virtually everyone referred to very much the same system as an openly democratic one. In a nutshell, what the Federalists meant by “democracy” by 1787 is not exactly what the term turned out to mean half a century later. Likewise, the institutional devices attached to the former concept and the latter concept also changed.1 This is why John Adams, a harsh critique of democracy during the 1770s and 1780s, could write in 1814 that Democracy ... must not be disgraced; democracy must not be despised. Democracy must be respected; democracy must be honored; democracy must
For full accounts of this shift see Morantz, “Democracy" and "Republic" in American Ideology; B. Laniel, Le mot "democracy" aux Etats-Unis de 1780 a 1856 (Saint-Etienne: Publications d e l’Universite d e Saint-Etienne, 1995); and Dupuis-Deri, The Political Power of Words.
1

76

Chapter 2. Conceptual shift

be cherished; democracy must be an essential, an integral part of the sovereignty, and have a control over the whole government.2 The chapter is divided into three sections, including this introduction. Section 2.2 shows the inﬂuence of classical anti-democratic republican thought over the use of the term by the Federalists, its widespread derogatory sense during the 1780s and the shift in its meaning and in the institutional devices traditionally attached to it during the Jeﬀersonian and the Jacksonian periods. Next, section 2.3 compares the early and the later concepts as for ﬁve key issues—representation and popular sovereignty, feasibility, property rights, stability and the division of powers.

2.2. Brief history of the concept
This section brieﬂy traces the evolution of the concept from its rather derogatory sense in the 1780s to its celebratory one ﬁfty years later. It does so by analyzing three key steps. First, the inﬂuence of classic republicanism—on which the Founders had been intellectually raised—over their conception of democracy (§2.2.1). Second, Jeﬀerson’s presidency and the debate among Pennsylvania Jeﬀersonians over the meaning and scope of democracy (§2.2.2). And, ﬁnally, Jacksonian democracy and the second party system (§2.2.3). 2.2.1. The inﬂuence of the classics The inﬂuence of the classics in eighteen-century America was everywhere. As Wood has pointed out, classical inﬂuence was in the names of towns and streets, such as Syracuse and Troy, the label of political institutions, such as the senates and capitols, or the political pseudonyms, such as “Publius” and “Brutus.” 3 And so it did classical political thought—notably, classical republicanism—inﬂuence the revolutionary gentry. They had been educated in a system that gave the classics a central position and that had taught them how to read and write Latin, and often also Greek, ﬂuently.4 As Noah Webster put it, The minds of youth are perpetually led to the history of Greece and Rome or to Great Britain; boys are constantly repeating the declamations of Demosthenes and Ciciero or debates upon some political question in the British parliament.5
2 3

Adams to John Taylor. Quoted in Dupuis-Déri, The Political Power of Words, p. 242. G. S. Wood, “Comment on Galston Paper,” Chicago-Kent Law Review 66 (1990): 69-72, p.

72.

The medieval “trivium” (rhetoric, logic and grammar) and “quadrivium” (arithmetic, music, geometry, and astronomy) would dominate Western curricula until the nineteenth century. Far from abolishing the standard classical studies, the Protestant Reformation had only added programs to promote literacy in the vernacular languages. See Carl J. Richard, The Founders and the Classics: Greece, Rome, and the American Enlightenment (Cambridge, Ma., Harvard University Press, 1994), p. 20. 5 Quoted in Richard, The Founders and the Classics, p. 13.

4

2.2. Brief history of the concept

77

The evidence showing the use of the classics—and notably of the classic republican authors—by the revolutionary gentry is vast.6 Jeﬀerson shipped from Paris copies of Polybius and sets of ancient authors to James Madison. While chairing a committee aimed at suggesting books for the congressmen, Madison put Aristotle’s Politics in the top of his own list. John Adams insistently repeated that the best of Machiavelli, Harrington, Sidney, Locke and Montesquieu came from the classics, notably, Aristotle, Cicero and Polybius. James Wilson described the perils of the Shays Rebellion after Horace, by saying that “We walked on ashes concealing ﬁre beneath our feet.” 7 And so on and so forth. Unsurprisingly, the inﬂuence of the classics also inﬂuenced their concept of democracy. So they inherited the rather derogatory sense that the concept had had for classic republicans such as Aristotle, Polybius or Cicero. This is relevant for present purposes. Since their relationship with actual democracies was more theoretical than practical, the classical inﬂuence provides a key reason to explain why the Founders could be so hostile toward democracy. The term had been barely used in the colonies, at least in the daily language of ordinary people. Only the educated gentry used it, but even in these cases, as “a technical term of political theory.” 8 According to Roy Lokken, there was no controversy over the meaning of the term “democracy” in colonial America as there is today. The colonists gave little thought to it, and the word seldom appeared in their political writings, speeches, sermons, public papers, and private correspondence ... Their understanding of the term was derived from classical writers, specially Aristotle and Cicero.9 Now, when the term began to be more widely used in the 1770s and 1780s, the Federalists took advantage of its derogatory sense and insistently used it to label and thus discredit their adversaries. The latter, in turn, generally hesitated to use it to label themselves, and even when they did so—as Wordsworth in 1794 when he wrote that “I am of that odious class of men called democrats”—, it was used in a deﬁant manner, which shows how discredited the term was.10 Hansen has attributed this derogatory sense of the idea of democracy to a mischaracterization of the Athenian democracy inherited from the sources used by the Founders and other members of the educated gentry of the time. These included, most notably, Aristotle’s Politics, Plato’s dialogues (specially the Statesman and the Republic, books 8-9) and Polybius Histories, book 6. As Hansen concludes, “Aristotle’s misleading and and Plutarch’s distorted picture of Athenian democracy
See notably Richard, The Founders and the Classics; Pocock, The Machiavellian Moment; and Wood, The Creation of the American Republic. 7 R. M. Gummere, “The Classical Politics of John Adams,” Boston Public Library Quarterly 9 (1957): 167–182, and “The Classical Ancestry of the United States Constitution,” American Quarterly 14 (1962): 3-18. 8 Wood, The Radicalism of the American Revolution), p. 243. See also R. R. Palmer, “Notes on the Use of the Word ’Democracy’ 1789-1799,” Political Science Quarterly 68 (1953): 203-226, pp. 207 and 212. 9 Roy N. Lokken, “The Concept of Democracy in Colonial Political Thought,” The William and Mary Quarterly 16, no. 4 (Oct., 1959), 568-580, p. 570. 10 Quoted in Palmer, “Notes on the Use of the Word ’Democracy’ 1789-1799,” p. 207.
6

78

Chapter 2. Conceptual shift

obscured the Enlightenment’s understanding of classical Athens.” 11 Rome was taken as a model rather than Athens—and so the Founders set up a Senate that met on the Capitol rather than a Council of the Aeropagus. And when Athens was mentioned positively, the reference was Solon—who was considered to had established a mixed and balanced government—rather than Cleisthenes, Pericles or Demosthenes—who were held up, as Hansen points out, “as a bugbear to warn popular champions of popular rule against excesses of democracy.” 12 Not until the nineteenth century, with the rise of history as a separate scholarly discipline, would Athenian democracy be more accurately studied and its previous mischaracterization gradually abandoned. Then the principal sources referred to would be no longer Plato and Aristotle, but rather Herodotus, Thucydides, Demosthenes and, after its discovery in 1879, Aristotle’s Constitution of the Athenians. Accordingly, the previous negative view of Athenian democracy would be substituted by a more positive one. However, in eighteenth-century America, the concept of democracy was still dramatically inﬂuenced by “the anti-Athenian tradition which the Founders imbibed at an early age.” 13

2.2.2. Jeﬀersonian democracy A strong opposition to the Federalists and their policies had emerged within Washington’s government as early as his ﬁrst oﬃce term. An opposition that provoked a crisis within the Federal government, the creation of the Republican party and, eventually, Jeﬀerson’s presidential oﬃce-holding in 1801. The Federalists reacted by labeling the the Jeﬀersonians as “democrats” to discredit them. As an instance of this, Hamilton denounced Jeﬀerson for being “too much in earnest with his democracy.” 14 However, as we have already seen in section1.3.1.3), far from rejecting the label, the Jeﬀersonians began using it. They did so, among other things, to distinguish themselves from the Federalists. Since “federalism” had become a byword for repression and anti-republicanism, for elitism and privilege, the latter were now avoiding to refer to themselves as “federalists” and were stressing their republicanism. In 1801 the Federal-Republicans of Pennsylvania even dropped the term “Federal” from their name and they introduced themselves merely as “Republicans” from that moment on. As a consequence of this abuse, “republican” turned out to be increasingly emptied of an accurate meaning and Jeﬀersonians began labeling themselves as democrats to distinguish themselves from the Federalists. At that point the ideological battle was between the republican faction and the democratic faction. This shows that “democratic” was still a label for the movements associated to ordinary people’s interests—in contrast to those of the gentry—, rather than a label for the American political system as a whole.
M. H. Hansen, “The Tradition of the Athenian Democracy A. D. 1750-1990,” Greece & Rome 39 (1992): 14-30, p. 19. 12 Hansen, “The Tradition of the Athenian Democracy A. D. 1750-1990,” p. 18. 13 Richard, The Founders and the Classics, p. 124. 14 Quoted in Dupuis-Déri, The Political Power of Words, p. 227.
11

2.2. Brief history of the concept

79

But the Jeﬀersonians used also another and more subtle strategy. They began referring not only to themselves as democrats, but also to the American political system as a whole as a democracy. So, when a group of Federalists founded The Republican or Anti-Democrat in Baltimore in 1801, the editors of the Philadelphia Aurora—the most widely read Jeﬀersonian organ in the nation—immediately replied. Now, instead of defending Jeﬀersonians from the usual charges of being democrats, they argued that it was the very American political system, and not only the Jeﬀersonians, which was a democracy: Republic does not so precisely deﬁne the nature of our government, as democracy ... Democracy therefore especially designates our government than republic—as it literally declares ... that all power is from, and all sovereignty centers in, the people, the democracy.15 Three issues stand out here. First, by doing this, the Aurora editors were turning a dispute over the features of a political faction—the Jeﬀersonians—into a dispute over the features of the American political system as a whole This clever move left the Jeﬀersonians—who were the democrats—in the advantageous position of speaking not on behalf of a faction among others, but on the very behalf of America as a whole. As Dupuis-Déri points out, “Here the political-linguistic strategy of the Aurora was not only to stand as ’Democrats,’ but even to label the regime of the United States with the name ’Democracy.” ’16 The Jeﬀersonians were eventually very successful in moving the issue in this direction, as the subsequent replies of The Republican or Anti-Democrat show. “Those who attempt to convince the people of the United States,” read a later issue, “that they already live under a democracy, that curse of republics, that volcano of licentiousness, that gulph of liberty, that proliﬁc mother of faction, cruelty, injustice, sedition ... and tyranny ... do nothing but harm ... The Constitution of the United States is far removed from a democracy ... It is a republic.” Second, the use of the term in the editorial of the Aurora departed radically from ordinary references to the American constitutional system in the 1780s. To be sure, many had accepted that some parts of the system were democratic or at least more democratic than others—notably, the House of Representatives in contrast to the Senate or the presidential oﬃce. But virtually no one had dared to label the whole system as a democratic system before. Now, the whole system was to be labeled as democratic by the Aurora editors. And that this move was eventually successful is showed by the fact that many used the term in this sense in the years to follow. As Elias Smith would state in 1809, The government adopted here is a DEMOCRACY.17 Third, by identifying democracy with popular sovereignty the editors of Aurora were also departing from the distinction between the former and the latter widely assumed
15 16 17

Aurora, September 18, 1802. Quoted in Dupuis-Déri, The Political Power of Words, p. 227. Quoted in Dupuis-Deri, The Political Power of Words, p. 241.

80

Chapter 2. Conceptual shift

in the 1770s and 1780s, as it will become clear later (infra, §2.3.1). As Dupuis-Déri has points out, in the late eighteenth century “ ‘democracy’ and ‘popular sovereignty’ were by no means thought of as coextensive.” 18 Even further, as Shankman puts it, “in fact, the declaration of popular sovereignty had done much to undermine arguments for direct majority governance.” 19 The clever move of the Aurora succeeded in identifying the American system with a democracy and democracy, in turn, with the Jeﬀersonians themselves. This improved both the previously derogatory use of “democracy” and the position of the Jeﬀersonians in the political debate. But not everyone among the Jeﬀersonians was convinced by the idea of democracy the editors of the Aurora were putting forward. Jeﬀersonians had remained united while they shared a credible common enemy. But, as soon as they defeated the Federalists and Jeﬀerson got to oﬃce in 1801, a harsh internal debate triggered among them. Among other many issues, over the desirability of democracy and its implications not only regarding the devices for political decision-making and the enfranchisement of the citizenry, but also regarding highly controversial issues, such as property, employment contracts, and the organization of the economy. As Andrew Shankman has shown, this debate was especially vivid (and enlightening for present purposes) between the more radical Jeﬀersonians of Pennsylvania and the more moderate ones. The former were to be known as the Duane-Leib faction (after William Duane and Michael Leib, their two most outstanding leaders), the Aurora-men, the Jacobins, or the Malcontents. The latter, on the other hand, were known as the Rising Sun Party (after the tavern where they ﬁrst met in 1802), the Third Party, the Tertium Quids, or simply the Quids.20 The idea of democracy put forward by the Malcontents (as I will refer to them from now on) was hardly at odds with the one used by the Federalists to discredit their Antifederalist foes ﬁfteen years before. As for the system of checks and balances, they leaded the attack on the senate, on the governor’s patronage and veto power and, most notably, on the judiciary, demanding the impeachment of several judges in Pennsylvania—some of them being successful—and putting forward a system of arbitration as a substitute of courts.21 Put brieﬂy, their argument against an independent judiciary was that such an institution was unnecessary in a democracy—and even in contradiction with it. Judicial independence, Aurora argued, made perfect sense in Britain, in order to protect judges from the arbitrary interference of an obviously illegitimate power—the crown. However, in a democracy, judges could not be “independent of the control of free people” and, interestingly for present purposes,
Francis Dupuis-Déri, “The Political Power of Words: The Birth of Pro-democratic Discourse in the Nineteenth Century in the United States and France,” Political Studies 52 (2004): 118-134, p. 121. 19 A. Shankman, “Malcontents and Tertium Quids: The Battle to Deﬁne Democracy in Jeﬀersonian Philadelphia,” The Journal of the Early Republic 19 (1999): 43-71. 20 See Shankman, “Malcontents and Tertium Quids: The Battle to Deﬁne Democracy in Jeffersonian Philadelphia;” and Shankman, Crucible of American Democracy. 21 Between 1803 and 1806 the state legislatures passed some forms of arbitration legislation, though all of them were vetoed by the Jeﬀersonian governor Thomas McKean afterward. See Shankman, “Malcontents and Tertium Quids,” p. 56.
18

2.2. Brief history of the concept

81

much less so regarding the review of the constitutionality of the statutes passed by the legislature, which was seen as a blatantly undemocratic measure.22 In addition to this, the judges had been able to invoke English-inspired common law to shape the economic and social landscape and, more precisely, had interpreted it—including the common law of real estate—in ways protective of capitalist development.23 The Malcontents were not against commerce—and actually promoted it. But they believed that, unless properly constrained by the democratic legislatures, commerce would serve the interests of the few, enlarging their fortunes, reinforcing their position in the political decision-making process and eventually downsizing democracy.24 For them, democracy had a two-way relationship with the economy. On the one hand, it required economic equality to function properly and, on the other, it served as a means to keep economic inequalities under control. They were not concerned with economic inequality per se, but rather with the pernicious eﬀects of economic inequality in creating political inequality and, in turn, with the eﬀect of political inequality in enlarging economic inequality. In short, they did not demand economic equality by itself, but as a precondition of democracy. By subordinating private property to democracy—instead of the other way around—, democratic decision-making guaranteed a great degree of economic equality—what they used to refer to as the “happy mediocrity of condition.” And preserving and promoting this happy mediocrity of condition was, in turn, “our greatest security and our best preservative against the gradual approaches of arbitrary power.” 25 If large economic inequalities arose, a lack of democracy would be the cause; and more democracy, thus, the solution. The demands of the Aurora to undo the separation of powers, in addition to Michael Leib’s eﬀorts to present these demands as representative of the Pennsylvania Jeﬀersonians in the House of Representatives—of which he was a member—, made dissenting Jeﬀersonian speak out and organize themselves against the Malcontents. By doing this, the Quids—as they would be known by their foes—not only tried to inﬂuence the policies of Jeﬀerson’s party qua Jeﬀersonians but also intended to inﬂuence what democracy meant and implied qua democrats. They knew that appeals to democracy increasingly served to mobilize the electorate. So, instead of giving up the democratic label with which the Federalists had tried to discredit them, they claimed that they were—and not the Malcontents, who had monopolized the term hitherto—the true democrats. Accordingly, they denounced how the Malcontents had misunderstood and misused democracy in their newly edited Freeman’s Journal.26 If Malcontents’ “intemperate conduct [were] countenanced by the Repub-

22

Aurora, March 7, 1803, p. 31. Quoted in Shankman, “Malcontents and Tertium Quids.” p.

53.

Mainly, by promoting developmental use and free exchange of property, in contrast to the jurisprudence of the ﬁrst half or the eighteenth century, which had constrained the alienation and transfer of real property. See Shankman, “Malcontents and Tertium Quids,” pp. 51 ss. 24 On the relationship between democracy and economic inequality see infra §2.3.3. 25 Aurora, May 27, 1805. Quoted in Shankman, Crucible of American Democracy, p. 260. 26 The original name of the newspaper was the Philadelphia Evening Post, but they quickly changed it due to its shared name with the Hamiltonian New York Evening Post.

23

82

Chapter 2. Conceptual shift

licans as orthodox democracy,” the Quids argued, American politics would remain indeﬁnitely turbulent.27 The idea of democracy put forward by the Quids was in clear contrast to the idea of democracy the Malcontents had successfully put forward hitherto. The Quids emphasized the division of powers and individual and property rights in the place of majority rule. True, they were not the ﬁrst ones who defended the former against the latter. However, they were very innovative in that they did so in the name of democracy itself. According to them, the Malcontents had obliterated “the just and wise lines of distinction which have been marked out for diﬀerent branches of our government.” 28 And, by doing so, they had distorted the true meaning of democracy. Not so much because a properly functioning democracy had to be constrained (by property rights and a system of checks and balances), but rather because such constraints were part of the very conditions of a proper democracy. So, if the Malcontents attacked the judiciary as being potentially at odds with democratic decision-making, the Quids replied by emphasizing the importance of removing some issues from the control of the legislatures for the proper functioning of democracy. They maintained thus that the courts of justice are to be considered as the bulwarks of a limited constitution, against legislative encroachments. The exercise of power ... implies nothing more than that the people are superior to the legislature, and that the judiciary, as a coordinate branch of government ... is bound to regard the will of the people, as expressed in the constitution, in preference of the legislative body.29 In turn, by 1805 the Malcontents had begun to call into question the very nature of constitutions, arguing that they were aristocratic and undemocratic documents, resulting “from a disposition to controul future time.” 30 A position that was used by the Quids—who reacted by creating the Society of Constitutional Republicans—to point out the radicalism of the Malcontents’ deﬁnition of democracy, which would put “life, liberty, and property in danger of being cast aﬂoat on the boisterous sea of anarchy.” 31 And rather successfully argued that democracy had to respect, as a matter of deﬁnition, property rights, the normative precedence of the Constitution, and a clear division of powers—interestingly for present purposes, including the independence of the judiciary and judicial review by the Supreme Court. The contrary, they argued, was not democracy but “Jacobinism” and “turbulence”—which were the epithets that had been traditionally used to discredit democracy. Revolutionarily enough, the Quids were now using these epithets not to label what democracy was but to label what democracy was not—and ought not to be.
27 28

Quoted in Shankman, Crucible of American Democracy, p. 102. Evening Post, February 21, 1804. Quoted in Shankman, “Malcontents and Tertium Quids,”

p. 60.

Shankman, Crucible of American Democracy. p. 133. Aurora, March 21, 1805. Quoted in Shankman, “Malcontents and Tertium Quids,” p. 62. 31 Freeman’s Journal, August 31, 1805. Quoted in Shankman, “Malcontents and Tertium Quids,” p. 64.
30

29

2.2. Brief history of the concept

83

2.2.3. Jacksonian democracy The eventual dissolution of the Federalist party in 1815 brought a period of relative ideological calm regarding the use and meaning of “democracy.” A period that would not end until the emergence of the second party system in the late 1820s, when Andrew Jackson very successfully used democracy as a battle-cry. After having been defeated by John Quincy Adams in the presidential contest of 1824, Jackson decided to run again in 1828. But now he distinguished himself from other candidates by labeling himself as a democrat rather than as a republican—which was the label used by all the remaining candidates and the one Jackson himself had used hitherto. Certainly, Jackson’s use of “democracy” to label himself was not merely strategic. However, his success in the running for presidential oﬃce in 1828 had indirect consequences for the dramatic extension of the term “democracy” in the American political jargon and also for its very meaning. A number of events behind this extension stand out. First of all, it is worth pointing out that during the 1820s “democracy” had lost almost all its previous derogatory sense—even though it had still not gained the undisputed acceptance it would enjoy later and that some could still claim, as the Richmond Whig did in 1834, that under Jackson’s presidency “The Republic has degenerated into a democracy.” 32 In general, few believed democracy to be unfeasible or unstable, nor incompatible with individual rights, as it had been usually claimed for more than two thousand years. In clear contrast to the previous generation of the Founders, the political elites during the Jacksonian era had received little if any education in classical political theory and were free from its anti-democratic inﬂuence. As Mogens Hansen points out, “the leading Jacksonian democrats were middle-class business men with little education in general and none in classics.” 33 Furthermore, Many Jacksonians were even hostile to the classics and classical education, which had served hitherto as a proxy for distinguishing the gentleman from the common man, and was at odds with the new egalitarian-minded and often anti-intellectual Jacksonian era. It was usual, for example, to count Jackson’s lack of classical education among his virtues.34 Hence, their conception of democracy did not come from the classics, but rather from their own political experience—most notably, from Jeﬀersonian politics under which virtually everyone took shelter at that time. The emergence of the second party system would not only spread this positive sense of democracy, but would also legitimize the existence of opposed parties with diﬀerent agendas as a common democratic practice. During the ﬁrst party system both the Federalists and the Republicans had continued to present themselves as representatives of the shared interests of the nation as a whole. Thus, it was rather usual for them to denounce each other for having betrayed the republican principles on which virtually everyone relied at the time. By contrast, due to the harsh competition during the second party system Americans slowly came to accept the existence
Quoted in Wilentz, The Rise of American Democracy, p. 425. Hansen, “The Tradition of the Athenian Democracy A. D. 1750-1990,” p. 26. 34 See Edwin A. Miles, “The Young American Nation and the Classical World,” Journal of the History of Ideas 35, no. 2 (April - June, 1974), 259-274, pp. 264-265.
33 32

84

Chapter 2. Conceptual shift

of diﬀerent candidates with diﬀerent agendas as a legitimate fact of American politics. The fact of disagreement and political competition replaced the organic tenet that had been so powerful during the ﬁrst party system. In short, democracy not only gained wide support but was also associated with the institutional system of political competition between increasingly professionalized and large-scale campaigning parties. In an almost unprecedent commentary, Van Buren—the brains behind the Jacksonians and successor of Jackson as president—cherished party competition and deﬁned it as “a complete antidote for sectional prejudices by producing counteracting feelings.” 35 In this context of increasing competition Jackson not only labelled himself as a democrat, but actually labeled the party itself as the “Democratic Party” and eventually just as “the Democracy.” 36 As a matter of strategy, this served to distinguish himself from other candidates—namely, John Quincy Adams ﬁrst and Henry Clay later—and to show a higher responsiveness to the ordinary people than the other candidates. Certainly, the strategy was in line with the social and political changes of the period and Jackson’s own political agenda. Not only did many states called for constitutional conventions and extended the franchise, but also the participation rate increased dramatically. For example, turnout in presidential elections, had been only 27 percent in 1824. By contrast, it soared to 57 percent in 1828 and to 80 percent in 1834.37 Even though an ample majority of the population would remain disenfranchised—at least, until the adoption of the Fifteenth Amendment in 1870, the Nineteenth Amendment in 1920, the Indian Citizenship Act of 1924 and the Voting Rights Act of 1965—, it was widely believed that America was indisputably democratic.38 Right before passing away in 1826, Jeﬀerson had already stated the “[w]e of the United States ... are constitutionally and conscientiously democrats.” 39 And the Atlantic Monthly would write in 1872 that “the democratic principle ... reached its culmination about 1850.” 40 Thus, it is hardly surprising thus Jackson’s political agenda could gain the support he received. Explicitly supported by Jeﬀerson himself before his death, Jackson—a former general in the war of 1812—promoted short terms and rotation in oﬃce, supported (moderate) suﬀrage extensions, fought the Bank of America by vetoing its rechartering in 1834 and, more generally, represented the access of the ordinary and humble man to the higher levels of the decision-making process. “The reign of
Van Buren to Thomas Ritchie, 1827. Quoted in A. James Reichley, The Life of Parties. A History of American Political Parties (Maryland, Rowman & Littleﬁeld, 1992), p. 73. 36 “One of the ... parties has undergone various changes of name,” wrote the New York Evening Post rather apologetically in an editorial in 1834, “the other has continued steadfast alike to its appellation and to its principles is now, as it was at ﬁrst, the Democracy.” Quoted in Laniel, Le mot "democracy" aux Etats-Unis de 1780 a 1856, p. 226. 37 See Reichley, The Life of the Parties, p. 84. 38 In his account of the American democracy, Tocqueville famously acknowledged this inconsistency—advancing, for example, that the exclusion of the black population would be one of the most problematic issues in the future—, but resolved it stating that “once a people begins to interfere with the voting qualiﬁcation, one can be sure that sooner or later it will abolish it altogether.” De Tocqueville, Democracy in America, p. 52. 39 Quoted in Dupuis-Déri, The Political Power of Words, p. 227. 40 Keyssar, The Right to Vote, p. 63.
35

2.2. Brief history of the concept

85

KING MOB seemed triumphant,” famously stated Supreme Court Justice Joseph Story regarding Jackson’s multitudinous arrival to the White House in 1829.41 However, just as he promoted measures as the ones mentioned above, Jackson and the Jacksonians also strengthened central government, went back to colonial practices such as patronage and made extensive use of the executive veto. And, interestingly for present purposes, they did so in the name of democracy.42 As Wood puts it, Jacksonianism did not create democracy in America. But it legitimated it; it restrained and controlled it and reconciled Americans to it. It did so by infusing into American democracy more elements of monarchy than even the Federalists had dared to try. The Jacksonians did this, however, in the midst of the most enthusiastic democratic rhetoric that any modern country had ever experienced. Consequently, the centralizing and consolidating aspects of the Jacksoninan revolution have been obscured and generally lost to us.43 Only slowly did National Republicans began to react to the hegemony of the Jacksonians. It was after Jackson’s war on the Bank of America in 1834 when they reorganized themselves in a new party that brought together former National Republicans, conservative Democrats and a number of disperse state parties. They named themselves as the Whig Party, a label that had been previously adopted by Southern planters—who advocated the rights of the states and thus hesitated to call themselves National Republicans—and was commonly associated to both the revolutionary uprising against the British crown—in contrast to the Tories, the supporters of the Crown during the rebellion and later revolution—and the British Whigs of the seventeenth and eighteenth centuries. The British Whigs and the American Whig Party were far from having coincident ideological positions though. For example, the former had traditionally supported the House of the Commons against the House of the Lords„ while the latter defended a strong senate as “pre-eminently a conservative branch of the federal government.” 44 Thus, a Democrat could observe, not without sarcasm, that through “metempsychosis ... ancient TORIES now call themselves WHIGS.” 45 As their next step, the Whigs began mimicking Jacksonian campaign techniques and also calling themselves “democrats,” as Van Buren had perceptively predicted in 1834, when he ironically wrote that “I almost begin to pity the poor Whigs. Their next cognomen will be Democrats.” 46 The Whigs had adequately recognized that the Democrats had “in their own name, a tower of strength.” 47 And they believed that it would be diﬃcult to defeat them while they monopolized its use. As senator William
Quoted in Wilentz, The Rise of American Democracy, p. 312. See Marvin Meyers, The Jacksonian Persuasion. Politics & Belief (Stanford, Stanfor University Press, 1957), pp. 250-252. 43 Wood, The Radicalism of the American Revolution, p. 302. 44 Quoted in Reichley, The Life of Parties, p. 79. 45 Quoted in Reichley, The Life of Parties, p. 79. 46 Quoted in Morantz, “Democracy” and “Republic” in American Ideology, p. 243. A joke not without irony, since in 1840 Van Buren would be defeated in his running for a second oﬃce term, by one of those “poor Whigs” now turned democrats—namely, William Henry Harrison. 47 Quoted in Morantz, “Democracy” and “Republic” in American Ideology, p. 237.
42 41

86

Chapter 2. Conceptual shift

H. Seward wrote in a letter to a friend, “It is utterly impossible, I am convinced, to defeat Van Buren. The people are for him. Not so much for him as for the principle they suppose he represents. That principle is Democracy.” 48 Even though some Whigs still hesitated to embrace the democratic rhetoric,49 some of them like Seward and Thurlow Weed began publicly labeling themselves as “democrats.” Even if the policies they put forward were dubiously democratic in the classic sense, the Whigs had little more alternative if they wanted to present themselves as a credible and strong electoral alternative to the Jacksonians. So they denounced the despotic tendencies of Van Buren’s executive, organized “democratic clubs,” introduced their candidate—William Henry Harrison—as the “poor man candidate” and the “people’s candidate,” and claimed insistently that they were the “true democrats, the real friends of the people.” 50 As the Boston Quarterly Review reported in 1838, It is well known that for the last two years the Whigs have, to some extent, claimed to be democrats; and it is equally well known, because they themselves have acknowledged it, that they claim to be democrats only because they regard the people as so attached to the name, that they will not vote for a party which does not bear it.51 Given the increasing popularity of “democracy,” the best option for the Whigs would have been that the Jacksonians hesitated to adopt the term to label themselves. As it is shown in the matrix below, that would have kept the Whigs in a similar position to that of the Jacksonians, since the latter would not have beneﬁted from using the now very appealing label—which ﬁtted well their political agenda—and the former would not have been damaged by not using it—since the label did not ﬁt their political agenda so well.

Jacksonians not “democrats” Whigs not “democrats” Whigs “democrats” 0, 0 —

Jacksonians “democrats” 0, 1 1, 0

Table 1. Whigs and Jacksonians on self-labelling as “democrats”

However, once the Jacksonians had seen the beneﬁts in exploiting the term and had begun doing so, the initial tie situation was broken. Then, the Whigs could not fall behind and had the incentive to start using it as well. No doubt this was a hazardous strategy for them, since they would have to bear for a while the costs of the dissonance between their public image and their actual policies. But they had nothing to lose and, if successful, they would have achieved two very important
Quoted in Wilentz, The Rise of American Democracy, p. 482. For example, as late as 1837, the American Quarterly Review was still criticizing Jeﬀerson for turning America “from a republic into a democracy.” Quoted in Morantz, “Democracy” and “Republic” in American Ideology, p. 235. 50 See Morantz, “Democracy” and “Republic” in American Ideology, p. 256-159. 51 Quarterly Review, September, 1838. Quoted in Morantz, “Democracy” and “Republic” in American Ideology, p. 246.
49 48

2.3. Early and later concepts compared

87

goals and would be better oﬀ as a result. First, they would have reduced the electoral distance between them and the Jacksonians. Second, they would have changed the meaning of the term, and would have managed to deactivate one of the main ideological weapons of the Jacksonians. No sooner said and done. Not only did the Whigs conscientiously adopt the democratic rhetoric. They actually labeled themselves as “the true democrats.” For example, in an speech delivered in Nashville in 1840 Henry Clay could claim that I am a Democrat—was born a Democrat, have lived and shall die a Democrat, in the true and genuine sense of the term.52

2.3. Early and later concepts compared
Democracy is not only an essentially contested concept—and so it was in the late-eighteenth century America, when diverse political factions put forward their own understanding of democracy and attempted to make it succeed over other understandings.53 It is also a fuzzy one. What does it refer to—a political system, as it was for colonial political theorists; a condition of society, as it was for Tocqueville? Which institutional arrangements does it include—unicameral and unchecked legislatures, as the Duane-Leib faction claimed, a mixed system of check and balances, as the Quids defended? Which preconditions does it demand—economic independence, equality of outcomes, equality of rights? It is thus very complicated to deﬁne sharply and once and for all neither the concept of democracy nor the concepts of democracy as used in the period studied so far. Nevertheless, however contested and fuzzy, there are some features of the concept that can be easier to identify in the early period but not in the later one, and the other way around. Five of them stand out—representation and popular sovereignty, feasibility, property, stability and division of powers. Let us comparatively reconstruct them following Madison’s famous argument in The Federalist no. 10. 2.3.1. Representation and popular sovereignty It was a common belief in late eighteenth-century America that democracy involved direct participation and that it was not easily compatible with representative government.54 Government “of all over all,” had James Otis deﬁned it in 1766, in a formulation almost identical to Aristotle’s deﬁnition of democracy as a government were “the one party rule and the other parties are ruled in turn” 55 So, when some
Quoted in Morantz, “Democracy” and “Republic” in American Ideology, pp. 259 and 244. As Galie warned, the problem with this kind of concepts is not so much that they are essentially contested, but rather that we can sometimes forget that they are. Walter Bryce Gallie, “Essentially Contested Concepts,” Proceedings of the Aristotelian Society 56 (1956), 167-198. 54 However, suﬀrage by lottery, which Montesquieu had written to be “natural to democracy, as that by choice is to aristocracy,” was not seriously considered at that time. Charles Louis de Secondat Montesquieu, The Spirit of Laws, p. 11. On the history of suﬀrage by lot and its absence in eighteenth-century debates in America, see Manin, The Principles of Representative Government. 55 Quoted in Lokken, “The Concept of Democracy in Colonial Political Thought,” p. 577; and Aristotle, Politics, 1271b7.
53 52

88

Chapter 2. Conceptual shift

royal oﬃcers feared the rising power of colonial assemblies, they related them to the Puritan Revolution of 1641. However, as Lokken points out, few of them accused the assemblies of being democracies or promoting democratic policies, for they did not assemble the whole people. The minister of the First Church in Boston, John Cotton, replying any suspicion that the colonist in Massachusetts Bay could be implementing a democracy, wrote in a letter that “though it be status popularis, where people choose their owne governors; yet the government is not a democracy, if it be administered, not by people, but by governors, whether one (for then it is a monarchy, though elective) or by many, for then (as you know) it is aristocracy.” 56 Madison echoed this conception of direct participation as a necessary condition for democracy when he stated that [By] a pure democracy ... I mean, a society consisting of a small number of citizens, who assemble and administer the government in person.57 Many claimed thus that representation was an entirely new principle, somehow unknown by the classics, and that it was incompatible with democracy ex hypothesi. Madison was more careful, though, and acknowledged that “the principle of representation was neither unknown to the ancients, nor wholly overlooked in their political constitutions,” 58 since in “pure democracies of Greece” many of the executives functions had been performed by oﬃcials elected by the people and representing them. Hence, the diﬀerence between a democracy and a representative system of the kind the Federalists put forward—a republic, as Madison called it—was not so much “the total exclusion of the representatives of the people from the administration of the former,” but rather “the total exclusion of the people, in their collective capacity, from any share in the latter ” 59 As Bernard Manin has put it, “for Madison, representative government was not one kind of democracy; it was an essentially diﬀerent and furthermore preferable form of government.” 60 The sharp separation between representatives and their constituencies was not the only feature of the republican government put forward by the Federalists, though. This form of government also implied that property qualiﬁcations—as deﬁned in the state constitutions, though not in the Federal Constitution—ought to prevent a majority of Americans—notably, women, blacks, natives and wage workers—from enjoying political rights. Now, the obvious question is: how could the Federalists reconcile these apparently anti-popular implications of their position with the rhetoric of popular sovereignty they had persistently appealed to as the main rationale against the British crown ﬁrst and for the ratiﬁcation of the Federal Constitution next?61 The answer is that
Lokken, “The Concept of Democracy in Colonial Political Thought,” pp. 577 and 571. Madison, “The Federalist no. 10,” in The Federalist, p. 46. See similar deﬁnitions in “The Federalist no. 14” and “The Federalist no. 48,” in ibíd. 58 Madison, “The Federalist no. 63,” in The Federalist, p. 329. 59 Madison, “The Federalist no. 63,” in The Federalist, p. 329, italics in the original. 60 Manin, The Principles of Representative Government , p. 2. 61 In the latter case, the delegates conscientiously violated the instructions of the Continental Congress in writing a new constitution instead of ammending the Articles of the Confederation. But they justiﬁed their actions on the sovereignty of the people to alter or abolish the Articles.
57 56

2.3. Early and later concepts compared

89

they could do so because popular sovereignty did not imply neither actual inclusion nor actual participation of the citizenry at large. Even though blacks, women, native Americans, and wage workers were denied political rights, they were all said to be virtually represented by those on whom they depended—and who did have political rights. The same applied to the ruling elite of representatives. However separated and independent from ordinary citizens—and to some extent unaccountable to them—, the former were conceived as virtually representing the interests of ordinary people through their taking part in the shared set of interests of the whole community. As Wood points out, “what justiﬁed elite rule ... was the sense that all parts of the society were of a piece, that al ranks and degrees were organically connected through a great chain in such a way that those on the top were necessarily involved in the welfare of those below them.” 62 In sum, the legitimacy of government rested in popular sovereignty but not in democratic decision-making, which was not coextensive with popular sovereignty. The argument for this is as follows. Indeed, democracy implied popular sovereignty: (D1) D → PS

However, democracy also implied actual representation—not only in the sense of the formal inclusion of everyone but also through the presence of a number of social conditions and institutional devices aimed at making this participation actual and meaningful. It thus excluded virtual representation by deﬁnition: (D2) (D → AR) ∧ (D → ¬V R)

On the other hand, even though popular sovereignty was consistent with actual representation, it did not require it and was also perfectly consistent with virtual representation. As Lokken has pointed out, “in the political ethos of the seventeenth and eighteenth-century English world, representation did not involve direct responsibility to the electorate. The political thought of the time included the idea that the people are the source of all political power, but this idea did not extend to the continued exercise of political power by the people after that power had been contracted and delegated to the one or several persons chosen by the people to govern them.” 63 That is, (D3) ¬ (P S → ¬V R) ∧ ¬ (P S → AR)

It followed thus that (D4) ∴ ¬ (D ← P S)

James Wilson, for example, argued that “the very manner of introducing this constitution, by the recognition of the authority of the people, is said to change the principle of the present Confederation [but] The people therefore have a right ... to form either a general government or state governments ... This, I say, is the inherent and unalienable right of the people.” Jonathan Elliot, ed., The Debates in the Several State Conventions on the Adoption of the Federal Constitution (Philadelphia, J. B. Lippincott, 1836), vol. 3, pp. 455-457. 62 See Wood, The Creation of the American republic, pp. 499. 63 Lokken, “The Concept of Democracy in Colonial Political Thought,” p. 570.

90

Chapter 2. Conceptual shift

In conclusion, as Dupuis-Déri points out, in the late eighteenth century “ ‘democracy’ and ‘popular sovereignty’ were by no means thought of as coextensive.” 64 Even further, as Shankman puts it, “in fact, the declaration of popular sovereignty had done much to undermine arguments for direct majority governance.” 65 Let us turn now to the later concepts of democracy and popular sovereignty and the relationship between them, which would go through a remarkable shift during the ﬁrst half of the nineteenth century. By 1840 virtually everyone agreed that democracy did not only imply popular sovereignty, (D1) D → PS

But also the other way around. That is, few accepted that popular sovereignty did not imply democracy. Now they were believed to be coextensive precisely because the new deﬁnition of democracy was not at odds with virtual representation any more. To be sure, democracy still required actual representation and this was in fact enlarged—notably, through suﬀrage extensions occurred during the ﬁrst half of the nineteenth century. But, contrary to what was believed earlier, it did not exclude virtual representation in some branches of the political decision-making process. For example, institutions such as the Supreme Court, which had been originally seen as a check on the “excess of democracy,” gained now democratic respectability—even if their members were not accountable because they were not elected and they enjoyed life-long appointments. In sum, (D2*) (D → AR) ∧ ¬ (D → ¬V R)

On the other hand, as for popular sovereignty, it was believed that sovereignty resided in the people, but not only in the sense of citizens being virtually represented by a number of more or less independent and unaccountable representatives—as was the case in the earlier conception of popular sovereignty—but also in the sense of necessarily letting the people actually take part in the decision-making process (say, by extending suﬀrage, both active and passive). In sum, (D3*) (P S → AR) ∧ ¬ (P S → ¬V R)

It followed thus that democracy and popular sovereignty turned out to be eventually interchangeable. (D4*) ∴ D ↔ PS

2.3.2. Feasibility From the early idea that in a democracy the people necessarily met and exercised the government in person it followed, as a matter of logic, that no democracy could be
F. Dupuis-Déri, “The Political Power of Words: The Birth of Pro-democratic Discourse in the Nineteenth Century in the United States and France,” Political Studies 52 (2004): 118-134, p. 121. 65 Shankman, “Malcontents and Tertium Quids: The Battle to Deﬁne Democracy in Jeﬀersonian Philadelphia.”
64

2.3. Early and later concepts compared

91

realized in large communities. In large communities it was very complicated—if possible at all—to have all the people meet together in the same place. Even Rousseau had seen this limitation and had concluded that “if we take the term in the strict sense, there never has been a real democracy, and there never will be” 66 Thus for Madison A democracy, consequently, must be conﬁned to a small spot.67 This implication ruled out the feasibility of a democratic government in a territory as large as eighteenth-century America. An implication that disappeared once democracy left room for representation and turned out to be seen as viable for a large territory, such as America. The Federal government was then taken as a fait accompli. Accordingly, constitutional institutions—such as the Senate, the presidential oﬃce or the Supreme Court—and devices—such as large electoral districts and small legislative assemblies—that had been considered opposed to democratic decision-making hitherto (see supra, §1.2.2.2) turned out to be seen, precisely, as the solution for the feasibility of a democratic political system in a territory as large and complex as nineteenth-century America. 2.3.3. Individual and property rights Small-scale and democratic communities had also been traditionally seen to involve a negative externality other than its infeasibility—namely, to make the so-called tyranny of the majority more likely.68 As Madison put it, “The smaller the society, the fewer probably will be the distinct parties and interests composing it; the fewer the distinct parties and interests, the more frequently will a majority be found of the same party; and the smaller the number of individuals composing a majority, and the smaller the compass within which they are placed, the more easily will they concert and execute their plans of oppression.” On the other hand, democracies were taken to be feasible only in small territories. That is, precisely in communities were those conditions applied. This was the reason why, according to Madison, Democracies ... have ever been found incompatible with personal security, or the rights of property.69 The trade-oﬀ between democracy and personal liberty had been a permanent charge against democracy since antiquity. To be sure, Madison’s argument on the inverse relationship between the size of a community and the likelihood of majority tyranny was novel, and somehow at odds—as he was well aware of—with the classical assumption on the unsuitability of large territories for republican governments, as typically
J.-J. Rousseau, Du contrat social, in Oeuvres complètes, vol. 3, B. Gagnebin and M. Raymond, eds. (Paris: Gallimard, 1961). Quoted from J.-J. Rousseau, The Social Contract, in The Social Contract and Discourses, trans. G.D. H. Cole (London and Toronto, J.M. Dent and Sons, 1923), p. 32. See also Shanaysha M. Furlow Sauls, The Concept of Instability and the Theory of Democracy in the Federalist (Ph.D. dissertation, Duke University, 2008). 67 Madison, “The Federalist, no. 14,” in The Federalist, p. 63. 68 The expression comes from J. S. Mill, “On Liberty,” in Collected Works of John Stuart Mill, vol. 18, J. Robson, ed. (Toronto: University of Toronto Press). 69 Madison, “The Federalist no. 10,” in The Federalist, p. 46.
66

92

Chapter 2. Conceptual shift

expressed by Montesquieu. However, the incompatibility between democracy and individual rights—and most notably, with property rights—had been an ever-present tenet since the classics. It is worth recalling that Aristotle had deﬁned democracy as the rule of the poor rather than as the rule of the many.70 It was obvious for him, as it would be for the subsequent tradition of nondemocratic republicans, that democracy jeopardized wealthy citizens’ property and individual liberty more in general. The logic was apparently obvious: democracy would empower the poor and these, being a majority of the population, would use their political power to advance their own interests and trump the rights of the minority—notably, their property rights. To be sure, there was a great deal of rhetoric involved in this charge both when used by Aristotle and when used by the Federalists. Private property had hardly been endangered neither during the post-Ephialtic Athenian democracy nor during the American critical period—which threats, as Wood has recalled, were partly exaggerated to justify many of the constitutional reforms later introduced in the Philadelphia Convention. Nevertheless, however historically inaccurate or empirically unsound, the popularity of the tenet about how democracy would inevitably lead the poor to expropriate the rich is hardly questionable. George Mason and Elbridge Gerry, two prominent Antifederalists, had warned in 1787 about “the danger of the leveling spirit” ﬂowing from “the excesses of democracy.” 71 In addition to this, the relationship between democracy and property had also worked the other way around. As we have already seen in detail (supra, §1.3.4), it was believed that both a minimum property and a maximum property clauses were implied by the concept of democracy. The former, in order to ensure that everyone enjoyed the material independence required to take part in public aﬀairs. The second, in order to prevent the wealthy from controlling the process of political decision-making and making it serve their particular ends. As Andrew Shankman concludes in his analysis of the Pennsylvania Jeﬀersonians, “Perfect democracy required perfect equality, not just before the law but also in wealth, in property.” 72 Now, the shift in the concept of democracy during the ﬁrst half of the nineteenth century would aﬀect both ways of the relationship. Consider ﬁrstly the trade-oﬀ between democracy and property rights. To be sure, many during the nineteenth century would still believe that democracy had a tendency to equalize property, as Adam Przeworski has shown.73 However, the new deﬁnitions of democracy put forward by the Quids in Pennsylvania or the Whig Party at the Federal level included the protection of property rights as a very condition of democracy and thus resolved the trade-oﬀ by subsuming one of the elements to the other. This shift had been possible, in turn, because in almost all the states poor white men had secured the vote and, nevertheless, property had not been threatened. In addition, judges had become the arbitrators between public power and private rights of property. Accordingly, law
Aristotle, Politics, C. D. C. Reeve, trans. (Indianapolis: Hackett, 1948), Book 3, ch. 8. Wood, The Creation of the American Republic, p. 484. 72 Shankman, Crucible of American Democracy, p. 4. 73 A. Przeworksi, “Democracy, Equality, adn Redistribution,” in Political Judgement: Essays in Honour of John Dunn, R. Bourke and R. Geuss, eds. (Cambridge, Cambridge University Press, 2009), 281-313.
71 70

2.3. Early and later concepts compared

93

became more and more of a specialized and autonomous science removed from politics and carried out by only an educated few who needed to go through the specialized professional schools.74 Consider next the maximum and minimum property clauses, which were also dramatically aﬀected. As for the former clause, the inclusion of the inviolability of property rights as a condition of democracy made democratic decision-making to be to a great extent at odds with the maximum property clause. Once property rights were entrenched as a very condition of democracy, it was very reasonable to expect that the performance of unequally propertied and skilled individuals would result in the enlargement of economic inequalities. However, this side-eﬀect would be compensated, as the Quids had argued in Pennsylvania, by a widespread prosperity and social mobility. As for the minimum property clause, the two shifts in the deﬁnition and function of material independence described above (supra, §1.3.3) diminished the importance of material independence in the long run. On the one hand, even if many would continue to accept material independence as a necessary condition for full citizenship, the gradual redeﬁnition of the former as self-ownership would eventually make everyone to be considered independent and thus worth granting full citizenship. On the other hand, the powerful ideal of equality of rights endorsed in the Declaration of Independence would serve as a rationale for demanding the extension of full citizenship to all in spite of them being materially independent or not. These two trends would dramatically transform the conception of the employment relationship, which had been traditionally seen as a form of “limited slavery,” as Aristotle put it and would be increasingly seen as as a free and voluntary exchange among juridical equals.75 In turn, these two trends will be key to Part III of the dissertation, for they help to explain why the workplace tends nowadays to be seen as a sphere in which power relations are absent and that is, accordingly, beyond the scope of politics, including democratic politics. 2.3.4. Stability The incompatibility between democracy and individual rights, and the predictable conﬂicts between majorities, on the one hand, and holders of property rights, on the other, made Madison conclude that [Democracies] have ever been spectacles of turbulence and contention [and] have, in general, been as short in their lives, as they have been violent in their deaths.76 Even though the Athenian democracy, the ever-present reference for the alleged instability of democracy, had lasted more than a century and a half, this had been the traditional view among nondemocratic republicans—namely, that democracy was a very unstable form of government. For instance, in the Federalist no. 9 Hamilton had
74 75 76

See Wood, The Radicalism of the American revolution, p. 323. Aristotle, Politics, C. D. C. Reeve, trans. (Indianapolis: Hackett, 1948), 1260a-b. Madison, “The Federalist no. 10,” in The Federalist, p. 46.

94

Chapter 2. Conceptual shift

famously stated about Greece that “It is impossible to read the history of the petty republics of Greece and Italy, without feeling sensations of horror and disgust at the distractions with which they were continually agitated, and at the rapid succession of revolutions, by which they were kept perpetually vibrating between the extremes of tyranny and anarchy.” 77 However, as it was the case with feasibility, the suspicion about the stability of democracy was dramatically reduced once the American political system was relabeled as democratic and its stability was thus taken for granted. (This assumption would be seriously called into question, though, with the breaking out of the Civil War.) Of course, if the stability of the American democracy rested on controlling the causes of the potential clashes between factions—and, most notably, between “those who hold, and those who are without property,” which according to Madison had been “the most common and durable source of factions” 78 —the question was now through which precise devices could those clashes be avoided. If we take into account that a persistent charge against democracy had come from its being an “unchecked” form of government, the inclusion of the division of powers—and, more precisely, of the system of checks and balances—as part of the very deﬁnition of democracy would provide a solution for this problem.

2.3.5. Division of powers The revolutionary gentry had not referred to democracy in its own right but always used it comparatively, as one of the three basic forms of government: monarchy, oligarchy and democracy—i.e. rule by the one, by the few, or by the many. This distinction was as old as Plato, Aristotle, and Polybius.79 In addition, these three regimes were often considered as the three parts of a mixed government (or politeia). The idea of a mixed government inﬂuenced probably more than any other classic idea the political thinking of the Federalists. They saw the British constitution just as Polybius had seen the Roman Republic—as the most perfect instance of a mixed government. The uprising against the British was justiﬁed thus not as an uprising against the British Constitution but rather as an uprising against the degeneration of its mixture. Even though the form of government remained the same, King George III had corrupted both the House of Lords and the House of Commons and thus destroyed the balance between them. So the Philadelphia Convention was called in the name of the mixed government as deﬁned by the classics and occasionally implemented in several historical periods. As John Corbin wrote, “The theory of our Constitution derives from Aristotle, and was put into successful practice in Ancient Rome, in eighteenth-century England, and in our early state constitutions, before it was given perhaps its most perfect embodiment by the Convention of 1787.” 80
Hamilton, “The Federalist no. 9,” in The Federalist, p. 37 Madison, “The Federalist no. 10,” in The Federalist, p. 44. 79 See Plato, Laws, 756e-757a and Politicus, 291d-e, 303c; Aristotle, Politics, 3.7; and Polybius, Histories, 6.5-18. 80 Quoted in Gummere, “The Classical Ancestry of the United States Constitution,” p. 6.
78 77

2.3. Early and later concepts compared

95

Accordingly, when in the 1770s and 1780s “democracy” reappeared as part of the political jargon, it began to refer to particular devices or parts of the ongoing constitutions—notably, to the legislative assemblies—rather than to the constitutions as a whole. The other branches of the constitutions and many of their devices were seen, actually, as checks on the democratic branch. And this branch, in turn, was believed to have the tendency to trump individual rights and destabilize the other branches—as the Federalists believed it had actually happened to some state constitutions during the 1780s. That was how James McHenry of Maryland saw the relationship between democracy and constitutions when he denounced that, “Our chief danger arises from the democratic parts of our constitutions [i.e., the legislative branch]. It is a maxim which I hold incontrovertible, that the powers of government exercised by the people swallows up the other branches. None of the constitutions have provided suﬃcient checks against the democracy.” 81 By contrast, during the ﬁrst part of the nineteenth century, “democracy” ceased to be limited to the legislature and began to refer also to other branches of the system of checks and balances, including the Senate, the Presidential oﬃce, and, eventually, the system as a whole. As we have already seen (supra, §2.2.2), the so-called Quids of Philadelphia were very innovative when they denounced the Malcontents’ more classic conception of democracy. In their view, this conception obliterated “the just and wise lines of distinction which have been marked out for diﬀerent branches of our government.” 82 According to the Quids, a proper deﬁnition of democracy had to include the very idea of checks and balances. Otherwise, it would be hardly compatible with property rights and thus unstable. In order for a stable and rights-consistent deﬁnition of democracy to succeed, other branches of the government that had been previously considered as non democratic had to be drastically redeﬁned. The case of the Supreme Court is especially outstanding for present purposes. As we have seen, judicial review of legislation by the Supreme Court had been seen as an undemocratic mechanism and as a very check on democracy, by the Antifederalists ﬁrst and radical Jeﬀersonains next. The former had denounced the institution of judicial review because “the judicial under this system have a power which is above the legislative” 83 Similarly, the latter had considered that, in a democracy, judges could not be “independent of the control of free people” and, accordingly, that the review of the constitutionality of the statutes passed by the legislature was a blatantly undemocratic measure.84 By contrast, by mid-nineteenth century, judicial review had turned out to be considered not only as a democratic institution, but in fact as a necessary condition of a good-working and balanced democracy. As we have seen, this shift applie dto many other institutions and devices not be taken to be democratic traditionally. Hence, the senate, the presidential oﬃce and
81 82

Farrand, The Records of the Federal Convention of 1787, vol. 1, pp. 26-27. Evening Post, February 21, 1804. Quoted in Shankman, “Malcontents and Tertium Quids,”

83 Brutus, “The Anti-Federalists no. 15,” in The Essential Federalist and anti-Federalist Papers, D. Wootton, ed. (Indianapolis, IN: Hackett, 2003), p. 92. 84 Aurora, March 7, 1803, p. 31. Quoted in Shankman, “Malcontents and Tertium Quids.” p. 53.

p. 60.

96

Chapter 2. Conceptual shift

judicial review by the Supreme Court, as well as the American institutional system as a whole, would turn out to be considered as fully democratic. As Duﬀ Green would report in the Telegraph in 1828, If required to deﬁne their government, the people of this Union would say that all their institutions were purely democratic.85

85

Quoted in Morantz, “Democracy” and “Republic” in American Ideology, p. 164.

Concluding remarks to Part I
The inﬂuence of the American constitutional system can be hardly overestimated. Over the last two centuries not few of its core institutional features have been widely replicated by a large number of democracies around the globe. Notably, judicial review of legislation, which originated in the Early Republic, has had an astonishing worldwide expansion afterward—to the point that nowadays three-quarters of the world’s states have some form of judicial review.86 It is thus surprising that a political system that has served as a democratic model and has been replicated by so many democracies was nevertheless taken to be non-democratic by its founders. And not less striking is that, only a few decades after the ratiﬁcation of the Federal Constitution of 1787, virtually everyone—the heirs of the Federalists included—turned out to refer to very much the same system as a fully democratic one. The goal of Part I of the dissertation has been to explain this shift in the Early American Republic with the hope that it would shed light on the more general paradox presented in the Introduction—namely, that the worldwide expansion of democracy has gone hand-by-hand with a dramatic downsizing of its scope (expressed in the ever-growing presence of non-majoritarian institutions such as constitutional courts and central banks). Hence, chapter 1 has shown that, despite its rather derogatory sense in the 1780s, the term “democracy”—along with some of the institutional devices traditionally attached to it—gained increasing support as soon as ordinary people became increasingly aware of the plurality of interests and values across the American society, organized themselves to advance such interests and the theory of virtual representation—including the conception of the society as an organic entity with a common set of interests—used by the Federalists in the 1780s gradually lost its credibility and political eﬃcacy. Once we realize of the commonly derogatory sense of the term “democracy” in the 1780s, it should not come with surprise that the Federalists hesitated to use the term to label the political system they were putting forward. Likewise, we should not be surprised by the fact that virtually everyone referred to very much the same system as a democracy ﬁfty years later, when the term gained a widespread positive sense and was used as a battle-cry by Andrew Jackson ﬁrst and the Whig Party next. Chapter 2, in turn, has shown that, besides its increasing acceptance, the term “democracy” also went through a notable semantic shift in the aftermath of the ratiﬁcation of the Federal Constitution. Although some features of the concept of
86

See Horowitz, “Constitutional Courts: A Primer for Decision Makers,” p. 125.

98

Concluding remarks to Part I

democracy remained the same, some others shifted. Likewise, some of the institutional devices that had been traditionally taken to be at odds with democracy for their being constraints on democratic decision-making turned out to be seen not only as consistent with democracy but actually as necessary for a democratic system proper. This helps explain why the Federalists believed that the representative government they were putting forward was not only diﬀerent from democracy but in fact a very check against it. As Bernard Manin puts it, “for Madison, representative government was not one kind of democracy; it was an essentially diﬀerent and furthermore preferable form of government.” 87 And it also helps explain why, only some decades later, virtually everyone referred to very much the same system as an openly democratic one. In a nutshell, what the Federalists meant by “democracy” in the 1780s is not exactly what the term turned out to mean half a century later. It is reasonable to ask whether the shift in the acceptance of democratic institutions and in the concept of democracy goes beyond the period and country studied. To be sure, this is a question that exceeds by far the limited scope of the dissertation. However, a number of authors have shown that a similar shift took place in other countries, such as France or Poland, by the same time.88 Likewise, and going back to the American case, there is ample evidence that the shift continued well after the Jacksonian period. Not only regarding the acceptance of democracy and the extension of universal suﬀrage, but also regarding the shift in the very concept of democracy. For example, as late as 1928 a US Army training manual could still deﬁne “democracy” as follows: "Democracy, n.:- A government of the masses. - Authority derived through mass meeting or any other form of direct expression. - Results in mobocracy. - Attitude toward property is communistic ... negating property rights. Attitude toward law is that the will of the majority shall regulate, whether it is based upon deliberation or governed by passion, prejudice,and impulse, without restraint or regard to consequences. - Result is demagogism, license, agitation, discontent, [chaos]."89

Manin, The Principles of Representative Government, p. 2. See among others J. A. Christophersen, The Meaning of "Democracy:” As Used in European Ideologies from the French to the Russian Revolution (Oslo: Universitetsforlagets Trykningssentral, 1968); Dupuis-Deri, The Political Power of Words; Palmer, “Notes on the Use of the Word ’Democracy’ 1789-1799;” and P. Rosanvallon, "The History of the Word ’Democracy’ in France," Journal of Democracy 6 (1995): 140-154. 89 U. S. War Department, “U. S. Army Training Manual No. 2000-25,” Washington, D.C., November 30, 1928.
88

87

Part II

Constitutionalism

Chapter 3

Analytical and normative background
Part I of the dissertation has shown, among other things, how some of the devices that were originally designed to be checks on democracy turned out to be considered not only consistent with democracy but often also necessary for a political system to be considered democratic proper. Part II focuses on two of these devices—constitutional rigidity and judicial review—and on the very notion of constitutionalism. It is divided into four chapters. Chapter 3 provides the deﬁnition, both historical and comparative, of constitutionalism used throughout Part II, and analyzes the two main institutional devices of constitutionalism—constitutional rigidity and judicial review. The remainder of Part II analyzes in detail the most salient normative theories of constitutional democracy, along with the problems they present.1 According to these theories, constitutionalism need not be at odds with democracy, either because constitutional constraints are necessary for the democratic procedures to bring just outcomes about (chapter 4), or because they are the upshot of democratically made precommitments (chapter 5), or yet because they set up the procedural rules and preconditions required for the proper functioning of the democratic system (chapter 6). For example, as Eisgruber has recently put it regarding judicial review of legislation, “the institution of judicial review is a sensible way to promote non-majoritarian representative democracy; not surprisingly, it is becoming increasingly popular in democratic political systems throughout the world.” 2 However, as I will try to show, all these theories are ﬂawed from a democratic standpoint.

3.1. Introduction
Constitutionalism is often used as an umbrella-like term for very diﬀerent phenomena. Hence, it is not unusual to refer to Britain (which lacks a written constitution), Australia (which lacks a bill of rights), the Netherlands (which lacks a system of judicial review) or New Zealand (which lacks a rigid constitution) as constituAs Sunstein has put it, “on [the reasons for entrenching institutional arrangements and substantive rights] constitutional theory remains in a surprisingly primitive state.” C. Sunstein, Designing Democracy. What Constitutions Do (Oxford: Oxford University Press, 2001), p. 97. 2 C. L. Eisgruber, Constitutional Self-Government (Cambridge, Mass.: Harvard University Press, 2001), p. 210.
1

102

Chapter 3. Analytical and normative background

tional states.3 However, in this chapter constitutionalism will be deﬁned, in a more restricted way, as follows: Constitutionalism.—The limitation of the powers of the legislative body entrenched in a rigid document and enacted through judicial review.4 Before turning to the three sorts of normative theories that will be studied below, this chapter provides the analytical framework to be used throughout Part II. Section 3.2 brieﬂy presents the history of constitutionalism, its main deﬁnitional features and the main sorts of constraints usually attached to it. Section 3.3 analyzes the role of constitutional conventions vis-à-vis formal constitutional provisions. The two main institutional devices of constitutionalism are analyzed in some detail in remainder of the chapter—namely, constitutional rigidity in section 3.4 and judicial review in section 3.5.

3.2. History and deﬁnition
A. Brief history. The global expansion of constitutionalism during the last two centuries, and notably since World War II, can be hardly overestimated.5 In 1802 no constitutional system proper—i.e., including a written constitution with an entrenched bill of rights and judicial review—existed in the world. Nowadays, by contrast, at least 190 countries have written constitutions, with some exceptions, such as Bhutan, New Zealand, UK and Israel, which lack fully codiﬁed, entrenched constitutions. Even further, among these exceptions, Israel has a powerful Supreme Court that protects rights as part of the higher law and the UK and New Zealand have recently adopted charters of rights, even though without judicial review proper. Regarding entrenched bills of rights, at least 183 countries have adopted charters of rights during the last decades, often including not only political and civil rights but also social rights. And, as for judicial review, 101 out of the 106 constitutions written between 1985 and 2008 include it.6
3 See J. C. Bayón, “Democracia y derechos. Problemas de fundamentación del constitucionalismo,” Constitucionalismo y derechos fundamentales, J. Betegón, Francisco Laporta, J. R. de Páramo, Luís Prieto, eds. (Madrid: Centro de Estudios Políticos y Constitucionales, 2004), p. 67. 4 Even though constraints on the executive body are also included in the concept of constitutionalism, in this dissertation I will focus only on the limits on legislative action. 5 On the recent world-wide expansion of constitutionalism, see among others R. Hirschl, Towards Juristocracy. The Origins and Consequences of the New Constitutionalism (Cambridge, Mass.: Harvard University Press, 2004); M. Shapiro and A. Stone Sweet, “The New Constitutional Politics of Europe,” Comparative Political Studies 26 (1994): pp. 397-420; C. Neal Tate and Torbjörn Vallinder, The Global Expansion of Judicial Power (NYU Press, 1997); C. R. Epp, The Rights Revolution. (Chicago: The University of Chicago Press, 1998); N. Tate and T. Wallinder, eds. The Global Expansion of judicial Power (New York: New York University Press, 1995). J. Ferejohn, “Judicializing Politics, Politicizing Law,” Journal of Law and Contemporary Problems 65 (2003): 41-68; Sieder et al., The Judicialization of Politics in Latin America (London: Palgrave, 2005). For a reliable source of comparative data on constitutionalism, see www.concourts.net. 6 The data come from A. Stone Sweet, “Constitutionalism, Rights, and Judicial Power,” in Comparative Politics, D. Caramani, ed. (Oxford: Oxford University Press, 2008). The shift has gone hand-by-hand with the democratization processes in the last decades of the twentieth century. Hence, if at least 68 percent of countries were authoritarian in 1975, by the end of 1995 only about

3.2. History and deﬁnition

103

Of course, modern constitutionalism is older than 1803 (when Marbury v. Madison, the ﬁrst instance of judicial review, was rendered) and its origins can be traced back to the late Middle Ages and early modern times, when constitutions emerged as devices to constrain the power of absolute monarchies, such as those of the royal houses of Tudor, Orange, Habsburg, or Hohenzollern. The unconstrained power of the absolutist king was crystal-clearly expressed by Lous XV: In my person alone resides the sovereign power, and it is from me alone that the courts hold their existence and authority ... For it is to me exclusively that the legislative power belongs, without dependence and without partition. To me alone belongs the legislative power without dependence and without division.7 True, as Lutz has pointed out, seventeenth and eighteenth-century constitutions tended not to include a bill of rights establishing the limits of the power of the branches of government.8 Rather, they tended to include only a set of institutional rules deﬁning their functioning. This is why the American constitution did not include a bill of rights in the ﬁrst place and why England was often believed not to have a constitution proper—since, in spite of having a long tradition of bill of rights, including the Magna Carta of 1215 or the Petition of Rights of 1628, no set of rules specifying the division of powers was ever written down. Hence, the French Declaration of the Rights of the Man and of the Citizens was rather innovative when it included a bill of rights as a necessary—though not suﬃcient—condition of constitutionalism. Its art. 16 read as follows: A society where rights are not secured or the separation of powers established has no constitution at all.9 Now, as constitutional monarchies gradually gave way to constitutional democracies—even if often with a king rather than with an elected and revocable president as the head of the state—, constitutionalism was redeﬁned as the limitation of the power of the parliament rather than of the absolute monarch. When this happened, constitutionalism was ﬁrstly taken to be at odds with democracy due to its constraining nature. As we have seen in the Part I of the dissertation, constitutionalism was considered not only to be at odds with democracy but actually to be a check on democratic decision-making. Only later did constitutionalism turn out to be considered democratic, to the point that nowadays is usual to consider it as a necessary condition for a political system being fully democratic. It is not the case that the constraining nature of constitutions over ordinary legislative politics has turned out to be overlooked so as to stress their enabling nature. Rather, it is
26 percent were authoritarian. See D. Potter, “Explaining Democratization,” in Democratization, ed. D. Potter, D. Goldblatt, M. Kiloh and P. Lewis (Cambridge: Polity Press, 1997), p. 1. 7 Quoted in D. Held, Models of Democracy, 3rd ed. (Cambridge: Polity Press, 2006), p. 56. 8 See D. S. Lutz, “Introductory Essay,” in Colonial Origins of the American Constitution: A Documentary History (Indianapolis: Liberty Press, 1998). 9 S. E. Finer. V. Bogdanor and B. Rudden, Comparing Constitutions (Oxford: Clarendon, 1995). p. 210.

104

Chapter 3. Analytical and normative background

widely accepted that constitutions are constraining of the legislative process indeed, but that such constraints are necessary for the democratic system to work properly. Ronald Dworkin is an illustrative example of the mainstream usage. Hence, he deﬁnes constitutionalism as “a system that establishes legal rights that the dominant legislature does not have the power to override or compromise,” pointing out that “constitutionalism so understood is an increasingly popular political phenomenon.” 10 B. Comparative deﬁnition. Constitutionalism so deﬁned—what is sometimes termed as “new constitutionalism” or as “strong constitutionalism”—contrasts with both absolutist constitutions and the Westminster model of parliamentary sovereignty, but also with the new commonwealth model of constitutionalism.11 Consider these differences in turn. First, constitutionalism so deﬁned is at odds with the so-called absolutist constitutions, which reﬂect and entrench the absolute power of one-person or one-party rule rather than constraining it.12 The locus classicus is the French Charter of 1814, though a number of current constitutions (e.g. those of Sri Lanka, Togo, or Niger) could also be included under this umbrella. Second, constitutionalism as deﬁned here is also absent from the Westminster model of democracy, which concentrates all the power on the parliament. Three features stand out in the Westminster model. First, the constitution (if there is a written constitution at all) is not entrenched. That is, the constitutional provisions are amendable through the ordinary legislative procedure. Second, any act—including judicial acts—that contradict a statute turns out to be invalid. And third, no substantive constraints—i.e. rights—are constitutionally entrenched. However, it is to be noted that Britain, New Zealand, Canada and Israel—once the epitomes of the Westminster model of parliamentary sovereignty, along with the Third and Fourth French Republics—have recently embarked upon profound constitutional reforms so as to leave room for principles of constitutionalism proper in their political systems. Third, constitutionalism as deﬁned here also departs from the so-called new commonwealth models of constitutionalism that combine rigidity and judicial review with ensuring that the legislatures retain the last word on constitutional matters and can override courts decisions on the constitutionality of statutes.13 According to Gardbaum, the model—which is also referred to as “weak-form judicial review,” the
10 Ronald Dworkin, “Constitutionalism and democracy,” European Journal of Philosophy 3 (1995): p. 2. According to Dworkin, nowadays every member of the European Community as well as other “mature democracies” embrace the view that parliamentary sovereignty ought to be constrained through constitutionalization and judicial review. Ronald Dworkin, A Bill of Rights for Britain (Ann Arbor, MI: University of Michigan Press, 1990), pp. 13-14. 11 The term “new constitutionalism” can be found in R. Hirschl, “The New Constitutionalism and the Judicialization of Pure Politics Worldwide,” Fordham Law Review 75 (2006): 721-754. The term “strong constitutionalism” can be found in Bayón, “Democracia y derechos,” p. 401; and in J. Waldron “The Core of the Case Against Judicial Review,” Yale Law Journal 115 (2006): 1346-1406, pp. 1354-1357. 12 The term “absolutist constitution” comes from Stone Sweet, “Constitutionalism, Rights, and Judicial Power.” 13 See Stephen Gardbaum, “The New Commonwealth Model of Constitutionalism”, American Journal of Comparative Law 49 (2001): 707-760. See also Stepehn Gardbaum, “Reassesing the New Commonwalth Model of Constitutionalism,” International Journal of Constitutional Law 8 (2010): 167-206.

3.3. Constitutional conventions

105

“parliamentary bill of rights model,” or the “dialogue model” of constitutionalism—,14 is deﬁned by three main features. The former two distinguish it from the Westminster model—namely, it includes a legalized bill of rights and some form of judicial power to enact such rights by assessing legislation as well. The latter feature, by contrast, distinguishes it from constitutionalism as deﬁned in this chapter—namely, notwithstanding the role attributed to courts, the legislative branch retains the ﬁnal word to decide by ordinary majority vote on the potential trade-oﬀs between statues and such rights. According to Gardbaum, the Canadian Charter of Rights, the New Zealand Bill of Rights and the UK Human Rights Act all represent this new “commonwealth model of constitutionalism” which “decouples judicial review from judicial supremacy by empowering legislatures to have the ﬁnal word.” 15 C. Constitutional constraints. Consider now the sorts of constitutional constraints and the institutional devices used to enact them. Greatly simplifying, constitutions include two main sorts of constraining provisions of democratic decision-making. First, substantive provisions, which typically adopt the form of a bill of rights entrenched in the dogmatic part of the text. And, second, procedural provisions in the form of a set of rules constitutive of the decision-making institutions and included in the organic part of the constitution. These provisions can have a constraining eﬀect only if they are included in a constitution with legal normative precedence over ordinary legislation—that is, a constitution not only with logical precedence over ordinary legislation, which is the case when a norm is previous to another one in the sense of providing the conditions under which the latter can be produced, but also with normative precedence when trade-oﬀs between the former and the latter arise and a decision has to be made.16 Accordingly, some institutional constraints are to be enacted so as to ensure that this precedence is observed—typically, constitutional rigidity and judicial review of the constitutionality of executive and legislative actions. When taking this form, the checks and balances included in the constitution act as “restrictions embedded in restrictions” proper.17 In the remainder of the section, constitutional conventions, constitutional rigidity and judicial review are analyzed in turn.

3.3. Constitutional conventions
Before turning to the more formal institutional devices of constitutionalism, let us brieﬂy mention the diﬀerence between written constitutions and constitutional
See Mark Tushnet, Weak Courts, Strong Rights. Judicial Review and Social Welfare Rights in Comparative Constitutional Law (Princeton and Oxford: Princeton University Press, 2008) and Janet Hiebert, “Parliamentary Bills of Rights: An Alternative Model?,” Modern Law Review 69 (2006): 7-28. 15 Gardbaum, “The New Commonwealth Model of Constitutionalism,” p. 708. 16 According to Bayón, normative precedence of norm A over norm B implies both “active force” (norm A can modify norm B) and “passive force” (norm A cannot be modiﬁed by norm B ). See Bayón, “Democracia y derechos,” Appendix. 17 J. Elster, Ulysses Unbound (Cambridge: Cambridge University Press, 2000), p. 115.
14

106

Chapter 3. Analytical and normative background

conventions and the role of the latter in constitutionalism.18 Constitutions can be written or unwritten. The political systems of those countries that lack written constitutions—notably, Britain, New Zealand or Israel—are only regulated by unwritten constitutional conventions. Now, even if uncodiﬁed, such conventions can be as precise in their meaning as the written ones and do impose costs on political agents that may violate them, thus successfully constraining the decisions of the legislature and the executive.19 However, constitutional conventions are also present in countries that do have codiﬁed constitutions—which include most Western countries—, as a supplement to the written provisions. Consider, for example, the two-term limit on the tenure of the American presidency, which is not included in the Constitution but is strictly observed, at least, since Roosevelt violated it by occupying the presidential oﬃce four consecutive times.20 It seems reasonable, thus, not to make a clear-cut distinction between codiﬁed constitutions and constitutional conventions and to focus better on the sort of constraints they impose on political decisions. That said, even though some attention will be paid to constitutional conventions (and, notably, to informal constitutional rigidity in section 5.4.4), throughout the chapter I will refer mainly to written and formal constitutions. For two reasons. First, because the causal mechanisms behind constitutional conventions are more complex and remain poorly understood, as Elster points out.21 Second, because written provisions are normatively more relevant. Violations of conventions may bear political sanctions, such as electoral defeat or social revolt, whereas violations of written provisions may also trigger legal sanctions in addition to political ones. It might be replied that the distinction applies only as a general rule, for conventions are sometimes enforced by courts. For example, the Canadian Supreme Court ruled in 1981 that there was a constitutional convention requiring the government to seek the approval of the provinces before amending the constitution while at the same time ruling that the government was not legally required to do so.22 In turn, written constitutions are sometimes ﬂexible, as we shall see below (section 3.4). However, these
The term “constitutional convention” comes from Dicey, who deﬁnes them as “understandings, habits or practices which, though they may regulate the conduct of the several members of the sovereign power, of the Ministry, or of other oﬃcials, are not in reality laws at all since they are not enforced by the courts.” A. V. Dicey, Introduction to the Study of the Law of the Constitution, E. C. S. Wade, ed. (London: MacMillan Press, 1945), pp. 23-24. 19 Lijphart argues that formal provisions are more constraining then conventions because the latter tend to be “merely a more or less amorphous collection of basic laws and customs without even a clear agreement on what exactly is and what is not part of the unwritten constitution.” Arendt Lijphart, Patterns of Democracy. Government Forms and Performance in Thirty-Six Countries (New Haven, CT: Yale University Press, 1999), p. 218. However, this need not be the case. Take, for example, the case of the rules of etiquette, which are often implict though very precise in their prescriptions. 20 Diverse explanations are provided for the anomalies of Britain, New Zealand and Israel. For example, that Israel has never been able to reach an agreement on a common set of constitutional principles (mainly, as for the place of religion in the legal system), whilst Britain and New Zealand have not need to reach a formal agreement due to the existence of a strong implicit agreement on such principles. See Lijphart, Patterns of Democracy, p. 217. Note, however, that both New Zealand and Britain have incorporated charters of rights to their legal system in the last decades, thus coming closer to the more formal model of a written constitution. 21 Elster, Forces and Mechanisms in the Constitution-Making Process,” p. 365. 22 Elster, Ulysses Unbound, p. 97.
18

3.4. Constitutional rigidity

107

are marginal cases. As Lijphart points out, the judicial enactment of “entrenched clauses” and “basic laws” in New Zealand and Israel are exceptional since they can be easily removed or superseded. And the same happens with ﬂexible constitutions, which turn out to be very rarely used.

3.4. Constitutional rigidity
Constitutions regulate fundamental matters of political life as well as themselves. Hence, apart from ﬁrst-order provisions—which include the substantive and procedural provisions analyzed in section 3.2—, they also include second-order provisions regulating the ammendment (and the suspension, in case of emergency) of the ﬁrst-order provisions. Thus, they present diﬀerent degrees of rigidity depending on the complexity and stringency of the procedures by which their ﬁrst-order provisions can be amendment.

Figure 3. Cross-national pattern for amendment rate and diﬃculty, indicating

amendment strategy.23

Constitutional rigidity serves as a necessary condition for the normative supremacy of the constitution—i.e. for its normative precedence over ordinary legislation. And
D. S. Lutz, Principles of Constitutional Design (Cambridge: Cambridge University Press, 2006), p. 173.
23

108

Chapter 3. Analytical and normative background

normative supremacy, in turn, serves as a necessary condition for constitutionalism.24 If the legislative body has the ability to amend the constitution by means identical to the ordinary legislative procedure (i.e., if the constitution is ﬂexible) it is unclear how can that assembly be constrained by the constitution in any meaningful sense. Elster provides a number of instances of this. For example, during the Hungarian constitution-making process of 1989-1990, it often happened that, when the parliament noticed that a constitutional provision was at odds with legislation, it opted to amend the constitution rather than the statute. Also, when the constitutional court of India has overturned a piece of legislation as unconstitutional, the parliament has frequently replied by amending the constitution. And so on. In short, the greater the stringency of the amendment procedure, the higher the costs of amending the constitution, and the less likely that it be amended, as Figure 3 above shows. As Buchanan has put it, Constitutional rules have the eﬀect of increasing the costs of taking certain actions. It is more costly to take action in violation of a rule than it is to take the same action in the absence of the rule. If this much be acknowledged, then rules must matter if the basic law of economics is accepted. An increase in the cost of any choice alternative will reduce the resort to that alternative.25 Rigidity can be deﬁned, thus, as follows: Constitutional rigidity.—The greater stringency of the constitutional amendment procedure in contrast to the ordinary legislative procedure, being this stringency a necessary condition for the normative supremacy of the constitution and, thus, for constitutionalism. A number of institutional forms of constitutional rigidity (generally appearing in a combined form) stand out. Without pretense of completeness, consider the following ones:26 Parliamentary super-majorities; procedural delays; states ratiﬁcation (in federal systems); ratiﬁcation by referendum; and absolute entrenchment. Of course, absolute entrenchment or intangibility is a rather radical form of rigidity. However, it is not an unusual one. For example, article 79(3) of the German constitution of 1948 includes such a clause as for basic rights. Similarly, the Bulgarian constitution provides absolute immunity for fundamental rights, just as the German constitution does regarding its federalism or the Romanian one regarding its republicanism. However, it is unclear whether such provisions can be absolutely entrenched in practice. At least, for two reasons.
However, it does not serve as a suﬃcient condition, as Bayón shows. See his “Democracia y derechos,” Appendix. 25 J. Buchanan, “Why constitutions matter,” in Why constitutions matter, N. Berggren, N. Karlson, and J. Nergelius, eds. (Stockholm: City University Press, 2000), p. 14. 26 See Astrid Lorenz, “How to Measure Constitutional Rigidity. Four Concepts and Two Alternatives,” Journal of Theoretical Politics 17 (2005), pp. 339-361; Donald S. Lutz, “Toward a Theory of Constitutional Ammendment,” American Political Science Review 88 (1994), pp. 355-370; V. Ferreres, “Una defensa de la rigidez constitucional,” ed. F. Laporta, Constitución: Problemas ﬁlosóﬁcos (Madrid: Centro de Estudios Políticos y Constitucionales, 2000).
24

3.4. Constitutional rigidity

109

First, if the article setting up the amendment procedures of the constitution (including the speciﬁcation of which articles are intangible) is the highest-order article in the legal system, it is unclear whether that provision applies to itself or not—i.e. the much discussed “Ross’ paradox.” 27 For example, the article 290 of the Portuguese constitution of 1976 absolutely entrenched some principles, such as the collectivization of the means of production in its article 83. However, the article 290 itself was not entrenched absolutely and could be amended according to the procedure included in the constitution. So, in 1982 the article 83 was indirectly amended by means of eliminating the article 290 that prevented the article 83 from being amended.28 Second, both constitutional and extra-constitutional actions against articles allegedly entrenched once and for all are always possible. Even if the society attempts sometimes to move some issues out of its political control by entrenching them, there is no “outside of the society,” as Elster puts it.29 That is, all institutional rules, including the absolutely entrenched ones, turn out to be institutionally or otherwise revisable, even if often at high cost. For example, this can be done by not observing them or their amendment procedures. As an instance of the former, consider the annual budget deﬁcit ceiling of 3 percent of the GDP included in the European Stability and Growth Pact (as part of the Maastricht convergence criteria). The measure was adopted in 1997 by the seventeen state members of the European Union that take part in the Eurozone and unobserved without batting an eyelash by virtually all of them afterward. As an instance of the latter, consider the writing of a completely new constitution by the delegates of the Philadelphia Convention in 1787, who utterly ignored their mandate to limit themselves to drawing up and proposing amendments to the Articles of Confederation. This second argument raises the classic philosophical issue of whether the very notion of a self-binding sovereign makes sense at all for “he that can bind can release,” as Hobbes famously claimed. According to Hobbes, along with Bodin and Austin, if all law—including constitutional law—is the command of a sovereign agent, the idea that this agent could bind herself is incoherent except in a ﬁgurative sense. In Hobbes’ own words,

The sovereign of the Commonwealth, be it an assembly or one man, is not subject to the civil laws. For having power to make and repeal laws, he may, when he pleaseth, free himself from that subjection by repealing those laws that trouble him, and making of new; and consequently he was free before. For he is free that can be free when he will: nor is it possible for any person to be bound to himself, because he that can bind can release; and therefore he that is bound to himself only is not bound.30
Alf Ross, “On Self-reference and a Puzzle in Constitutional Law,” Mind 78 (1969): 1-24. See I. Sánchez-Cuenca, Más democracia, menos liberalismo (Madrid: Katz, 2010), p. 134. 29 Elster, Ulysses Unbound, p. 94. 30 Thomas Hobbes, Leviathan, Richard Tuck, ed. (Cambridge: Cambridge University Press, 1988 [1651]), p. 184. See also John L. Austin, The Province of Jurisprudence Determined (Cambridge: Cambridge University Press, 1995), lecture VI.
28 27

110

Chapter 3. Analytical and normative background

However, three problems arise with this view. First, it is not completely true that “he that can bind, can release,” for that would be true only if the sovereign could undo previous commitments without costs. However, in politics there is an asymmetry between the cost of breaking a previous commitment and the cost of not making it in the ﬁrst place. As Elster puts it, in politics a broken promise is worst than not having made it at all.31 Second, there is a diﬀerence between sovereignty and its exercise. In a democracy, sovereignty rests on the people and the branches of the state, at least in theory, only exercise it. The constitution constraints the branches of the state, not the people itself.32 Third, unlike the limited sovereign, the absolute sovereign has problems to credibly commit itself. The asymmetry between the former and the latter regarding credible signaling shows that constitutional constraints are not merely ceremonial and that they can have a constraining eﬀect indeed. As Elster points out, this diﬀerence was clear in China in the 1980s, after Deng Xiaoping liberalized the economy and private property was allowed in order to improve the economic growth.33 Since the economic agents could not know if those circumstances would last and if they would be able to keep their proﬁts, their time horizon was shortened and they often preferred to use their proﬁts for consumption rather than reinvesting them. Even if the Chinese leaders may have wanted to commit themselves to a hands-oﬀ policy, they could not do so credibly for they had all the power and thus the ever-present possibility to undo their reforms. Of course, an obstinate Hobbesian could always reply that these cases only present a sovereign who either lacks complete sovereignty or is not really constrained after all. In the ﬁrst and third cases, because an agent who is accountable for its broken promises or who depends on third parties’ investments is not an absolute sovereign proper. In the second case, because the constitution does not constrain the sovereign (the people) but its agent (the state). In short, Hobbes and Austin might have been right after all in claiming that a real sovereign cannot bind herself, as it is the case under dictatorships.34 However, real-world democratic politics consists of multiple agents whose sovereignty is limited and overlapping. These include not only the people and their institutional agent (the state) but also international institutions, transnational corporations, unions and more diﬀuse actors such as ﬁnancial markets. Further, the people themselves stand divided by widespread and persistent disagreements and conﬂicts of interests. Under these non-ideal circumstances, constitutional constraints do pose costs and constraints on ordinary decisions made by the people and the state. In short, it is naive to believe that constitutions are all-powerful

J. Elster, “Deliberation and Constitution Making,” in Deliberative Democracy, J. Elster, ed. (Cambridge: Cambridge University Press, 1998), p. 98. 32 See Wil Waluchow, “Constitutionalism,” The Stanford Encyclopedia of Philosophy (Fall 2008 Edition), Edward N. Zalta, ed., URL = <http://plato.stanford.edu/archives/fall2008/entries/constitutionalism/>. 33 See Elster, Ulysses Unbound, pp. 148-149. 34 Franz Neumann, for example, deﬁned dictatorship as “the rule of a person or a group of persons who arrogate to themselves and monopolize power in the state, exercising it without restraint.” F. Neumann, The Democratic and the Authoritarian State: Essays in Political and Legal Theory (Glencoe, IL: Free Press, 1957), p. 233, my italics.

31

3.5. Judicial review

111

institutional devices—just as it is to believe that they do not impose costs at all. Things are way more complex and context-dependent than that.

3.5. Judicial review
As we have seen, constitutional rigidity is a necessary condition of the normative precedence of the constitution and of its constraining nature, though not a suﬃcient one. Rigidity alone can hardly constrain the legislative and executive branches if nothing forces such branches to observe the constitutional provisions. As R. H. S. Crossman, the member of the British Labour cabinet responsible for the controversial 1968 immigration law denying entry into Britain to about a hundred thousand British subjects living in Kenya, would declare later on, this law “would have been declared unconstitutional in any country with a written constitution and a Supreme Court.” 35 This reasoning has lead some to claim that judicial review turns out to be a necessary condition of the normative precedence of the constitution. As we shall see immediately below, this is too fast a conclusion, for means other than judicial adjudication (such as popular mobilization, external pressure by international agents or expected electoral costs) can make such branches observe the constitution. However, their contingent relationship notwithstanding, during the last decades judicial review has turned into the most widespread institutional device for the enactment of the normative precedence of the constitution. Once an American anomaly (initiated after the Marbury v. Madison Supreme Court decision of 1803), nowadays it has become the norm rather than an exception—to the point that 101 out of the 106 constitutions written between 1985 and 2008 include it.36 This phenomenon is part of what has been termed as the “judicialization of politics,” which Ran Hirschl deﬁnes as “the ever-accelerating reliance on courts and judicial means for addressing core moral predicaments, public policy questions, and political controversies.” 37 Three issues regarding judicial review are addressed in this subsection—the diﬀerence between weak and strong judicial review, the diﬀerence between the American and the European forms of judicial review and the actual inﬂuence of judicial review over political decisions. A. Weak and strong judicial review. As we have seen in section 3.2, weak and strong forms of judicial review can be distinguished. According to the weak form typical of the new commonwealth model of constitutionalism, courts assess the constitutionality of legislation, but they lack the ﬁnal word vis-à-vis the legislatures and, therefore,
Lijphart, Patterns of Democracy, p. 223, my italics. See among others Stone Sweet, “Constitutionalism, Rights, and Judicial Power.” For the history of judicial review, see Larry Kramer, The People Themselves (Oxford: Oxford University Press, 2004); Roberto Gargarella, La justicia frente al gobierno. Sobre el carácter contramayoritario del poder judicial (Barcelona: Ariel, 1996); Mary Sarah Bilder, “Idea or Practice: A Brief Historiography of Judicial Review,” Journal of Policy History 20 (2008): 6-26; Ran Hirschl, Towards Juristocracy: The Origins and Consequences of the New Constitutionalism (Cambridge, Mass.: Harvard University Press, 2004). 37 R. Hirschl, “The New Constitutionalism and the Judicialization of Pure Politics Worldwide” Fordham Law Review 75 (2006): 721-754, p. 721.
36 35

112

Chapter 3. Analytical and normative background

the ability to turn down or modify statutes. In some versions of the weak form, courts can enforce their decisions, but the legislature may override the decision, restating the original statutes by some additional means. This is the case of Canadian Charter of Rights and Freedoms, according to which courts can decline to enact statutes if they violate any of the provisions included in the Charter whilst the legislature retain the power to insulate legislation from judicial scrutiny (even though it has to be noted that such power is rarely used). In another and weaker versions, courts enjoy the ability to declare the unconstitutionality of a given statute, but lack the ability to enforce such declaration so as to turn statutes down. This is the case of the UK under the Human Rights Act, according to which the courts can issue “declarations of incompatibility” that can trigger a fast-track legislative procedure carried out by a minister in order to remedy the incompatibility. However, unless a minister initiates the fast-track procedure, such declarations do not “aﬀect the validity, continuing operation or enforcement of the provision in respect of which it is given; and ... is not binding on the parties to the proceedings in which it is made.” 38 Now, throughout this chapter judicial review will refer only to its stronger form, which can be deﬁned as follows: Judicial review.—The authority of courts to assess the constitutionality of legislative actions, including the ability to decline to apply statutes, to modify the eﬀect of their application so as to make them compatible with the constitution and, in the limit, to strike them down altogether. B. American and European judicial review.39 Even though both the American and the European models of judicial review are part of the “post-war paradigm” of constitutionalism, in which courts have the last word over the constitutionality of statutes, salient diﬀerences between them stand out.40 The American model was very relevant in starting the practice of judicial review and promoting it world-wide (for example, by pressing post-WWII Germany and Italy to adopt written constitutions with a charter of rights and judicial review). Hence, the American model is found in Africa, Asia and the Caribeean, specially in countries that were part of the British Commonwealth or were under American occupation or inﬂuence, such as Japan and parts of Central and South America. However, in recent times the European model has been much more inﬂuential than the American one, which is believed to introduce confusion in the separation of powers. With the decolonization process new constitutionalism expanded, though in its European version rather than in the American one. Consider brieﬂy the history of the European model. By the time of Marbury v. Madison, the French legislature had banned judicial review of legislation and
Human Rights Act, 1998, c. 42, § 4(2), (6). Quoted in Waldron, “The Core of the Case Against Judicial Review,” p. 1355. 39 A. Ruiz Miguel, “Modelo americano y modelo europeo de justicia constitucional,” Doxa 23 (2000): pp. 145-160; A. S. Sweet, “Why Europe Rejected American Judicial Review—And Why It May Not Matter,” Michigan Law Review 101 (2003): pp. 2744-2780. 40 L. Eisenstat Weinrib, “The Postwar Paradigm and American Exceptionalism,” The Migration of Constitutional Ideas, Sujit Choudhry, ed. (New York: Cambridge University Press, 2006).
38

3.5. Judicial review

113

had set up the Napoleonic code system according to which courts must not take part in the making of legislation. The French system expanded through Europe during the nineteenth century. As time went by, speciﬁc organs were created to protect rights from administrative or legislative abuses, but such power was never conferred to courts. Finally, in 1970 a bill of rights was included in the French Constitution and a Constitutional Council was created in the Fifth Republic. The Council, however, exercises only pre-enforced and abstract review of statutes. Once a statute has come into force, no review is allowed. The French system is important for two reasons. First, because it spread across Europe ﬁrst and Africa and Latin America later, mainly through Portuguese and Spanish colonialism. Second, because it served as the stepping stone for the current European model. Hans Kelsen, the father of the European-style constitutional court, drafted the constitution of the Austrian First Republic and included a constitutional court of this sort among its most outstanding institutional features. During the inter-war period, constitutional review was stablished in Czechoslovakia (1920), Liechtenstein (1925), Greece (1927), Spain (1931) and Ireland (1937), even though only the latter survived the WWII. After the war, most European countries adopted the Kelsenian model in subsequent waves. In the American model, judicial review is decentralized, so any court has the ability to declare the unconstitutionality of statutes, with the Supreme Court serving as the last court of appeal for ordinary law as well as for constitutional law and thus addressing cases of constitutional law less often. By contrast, in the European model constitutional review of legislation is carried out by a specialized constitutional court, which is thus much more active than the American Supreme Court regarding constitutional decisions. As Dahl has pointed out recently, contrary to what is sometimes believed, it is the European and not the American model the one that has had more world-wide inﬂuence, as Stone Sweet’s more comprehensive analysis proves (see table 1 below).41

Region Europe Africa Middle East Asia and South Asia North America Central America South America Caribbean Total

American 5 12 2 18 2 3 3 8 53

European 31 29 5 13 0 3 4 0 85

Mixed 3 (1) 1 0 2 0 3 (1) 5 (3) 0 14

Other 1 6 3 11 0 0 0 1 22

None 2 3 1 0 0 0 0 0 6

Table 2. Regional distribution of models of judicial review.42
Robert A. Dahl, How Democratic is the American Constitution? (New Haven and London: Yale University Press, 2001), ch. 3. 42 Stone Sweet, “Constitutionalism, Rights, and Judicial Power,” p. 223.
41

114

Chapter 3. Analytical and normative background

C. Eﬀectiveness. A lot has been said about the “counter-majoritarian diﬃculty” that arises whenever a court—which members have not been elected and cannot be voted out—strikes down a decision made by democratically elected members of the parliament or the executive (see below, § 115). However, it has often been been replied that judicial review could be the “least dangerous branch” after all,43 since empirical evidence shows that, on average, constitutional courts tend not to make decisions against public opinion. According to Dahl’s classical study, for example, we should not worry too much about the counter-majoritarian diﬃculty for “policy views dominant on the Court are never for long out of line with the policy views dominant among the lawmaking majorities of the United States.” 44 Dahl’s study has been conﬁrmed in subsequent studies. For example, Stimpson, Mackuen and Erikson have shown that, in a scale of 1-100 of progressism-conservadurism, a shift in the public opinion of 1 point translates into a shift in Supreme Court Justices’ view of 0,3 points and of 0,9 in the House of Representatives or the Senate. Accordingly, the Supreme Court is only slightly slower in getting in line with public opinion (on average, 2,3 years) than the representative institutions (around 1 year).45 Diverse explanations have been provided for this phenomenon. Consider three of them. First, in Dahl’s classic account, the mechanism by which Supreme Court justices tend to conform to the constitutional views of the majority is the appointment process. Hence, as older justices die or retire (his analysis refers only to the American Supreme Court, in which justices have life-long tenure), they are replaced by new ones with shared views with the majority political coalition that appoints them. Second, more recently, Epstein, Knight and Martin have argued, contra Dahl, that justices tend not to oppose the view of the majority because of the institutional constraints imposed on the court by the system of separation of powers.46 The legislatures can always pass new legislation and institutional crisis can arise if the court systematically strikes down statutes—as it happened in the US in 1937, when Roosevelt threatened to “pack the [Supreme] Court.” For that reason, judges can anticipate this possibility and incorporate the preferences of the members of the other branches in their decisions, reducing the dissonance and improving the stability of the institutional system overall. Third, scholars who emphasize the inﬂuence of social movements over constitutional law have provided yet another explanation for why judges adapt their views to those of the citizenry—an explanation that is consistent, I believe, with both Dahl’s and Epstein et al.’s. Hence, according to Robert Post and Reva Siegel, social movements inﬂuence electoral politics, which in turn inﬂuences which judges get appointed to the court. This is consistent with Dahl’s account. In addition, social movements
The expression comes from Hamilton, The Federalist, no. 78. R. A. Dahl, “Decision-Making in a Democracy: The Supreme Court as a National Policy-Maker,” Journal of Public Law 6 (1957), pp. 279-295. 45 J. A Stimson, M. B. Mackuen and R. S. Erikson, “Dynamic Representation,” American Political Science Review 89 (1995): pp. 543-565. Cited in Sánchez-Cuenca, Más democracia, menos liberalismo, p. 177. 46 See L. Epstein, J. Night and A. D. Martin, “The Supreme Court as a Strategic National Policymaker,” Emory Law Journal 50 (2001): 583-612.
44 43

3.6. The counter-majoritarian diﬃculty

115

also put direct pressure on the court when properly organized and supported, making thus diﬃcult for judges not to take into account their views. This latter explanation is consistent with Epstein et al.’s strategic account of judges’ behavior, and helps explain why they often strike down legislation that is not supported by the large public but not the other way around. Conclusion: Despite the heated disputes they often provoke, constitutional courts are not all-powerful. Now, they are not mere paper tigers either. At least, for two reasons. First, in countries with a well established tradition of judicial review constitutional decisions made by courts have a pervasive impact on society and, in turn, on the decisions made by the legislature and the executive (i.e., the preferences of the citizenry and of the representatives are to great extent endogenous to the legal framework). Second, just as courts can anticipate and incorporate the preferences of the public and of the members of other branches, the members of the legislative and the executive branches can also anticipate the views of the constitutional court and strategically take them into account in order to rule out the possibility of their decisions being turned down. Therefore, their constraining nature—as well as the counter-majoritarian diﬃculty—should be taken seriously indeed.

3.6. The counter-majoritarian diﬃculty
So far in this chapter we have brieﬂy covered the history and forms of constitutionalism, along with its most characteristic institutional features—constitutional rigidity and judicial review of legislation. On the one hand, rigidity makes the constitutional amendment procedure more cumbersome than the ordinary legislative procedure carried out by legislatures. On the other hand, judicial review provides courts with the ﬁnal authority to assess the constitutionality of legislation, including the ability to overturn statutes that have been previously passed by democratic legislatures. Hence, constitutionalism thus deﬁned raises serious problems of justiﬁcation for everyone committed to democracy and majority rule. This is what Alexander Bickel famously termed as “the counter-majoritarian diﬃculty.” 47 Even though the expression tends to be used to refer only to judicial review, it is obvious that the ability of legislatures to counteract decisions made by the courts through constitutional amendment will depend on the rigidity of the constitution. Hence both constitutional rigidity and judicial review can be depicted as counter-majoritarian devices. Of course, the relationship between constitutionalism and democratic legislatures is not an all-or-nothing matter. A rigid constitution poses constraints on the
A. Bickel, The Least Dangerous Branch. The Supreme Court at the Bar of Politics (New Haven: Yale University Press, 1962), pp. 16-17. For the history of the “diﬃculty” from its origins in the Early Republic until present times, see B. Friedman, “The History of the Countermajoritarian Diﬃculty, Part One: The Road to Judicial Supremacy,” NYU Law Review 73 (1998): 333-433; "The History of the Countermajoritarian Diﬃculty, Part Two: Reconstruction’s Political Court," Georgetown Law Journal 91 (2002): 1-65; "The History of the Countermajoritarian Diﬃculty, Part Three: The Lesson of Lochner," NYU Law Review 73 (2001): 1383-1455; "The History of the Countermajoritarian Diﬃculty, Part Four: Law’s Politics,” University of Pennsylvania Law Review 148 (2000): 971-1064; "The Birth of an Academic Obsession: The History of the Countermajoritarian Diﬃculty, Part Five," 112 Yale Law Journal 153 (2002): 153-259.
47

116

Chapter 3. Analytical and normative background

decisions made by the elected representatives. However, the degree to which the counter-majoritarian diﬃculty is present depends on the stringency of the amendment procedure. And, as we have seen, this stringency can go from mere procedural delays to parliamentary majorities to the ratiﬁcation by non-elected bodies to absolute entrenchment in its more radical form. In short, the greater the stringency of the amendment procedure, the greater the contribution of constitutional rigidity as a counter-majoritarian device. As for strong judicial review, its contribution as a counter-majoritarian device depends on a number of factors.48 First, the larger the number of abstract and vague articles included in the constitution, the greater the need for constitutional interpretation and thus the discretion and power of constitutional judges. Second, the democratic pedigree of constitutional judges will depend on the how they are appointed, thus being higher if they are appointed by the parliament and for a limited term than if they receive life-long appointments. In short, constitutional rigidity and judicial review turn out to be more or less counter-majoritarian depending on these factors. However, to the extent that they are counter-majoritarian, they pose a serious problem of justiﬁcation from a democratic point of view. For anyone committed to democracy and majority rule as the central sources of political legitimacy, constitutional rigidity and judicial review are disturbing institutional devices that need to be justiﬁed. Consider the case of the Statute of Autonomy of Catalonia (an autonomous community in Spain), which provides Catalonia’s basic law. Even though the Statute was passed by the Catalan parliament and by both chambers of the Spanish parliament, and eventually ratiﬁed in referendum by a large majority of Catalans (73,9 percent of the votes) in 2006, the Spanish Constitutional Court ruled in 2010 that fourteen of its articles were unconstitutional and were thus turned down. Each time this occurs, the counter-majoritarian diﬃculty, and the need for sound reasons to address it, become more pressing. The remainder of Part II analyzes the main attempts to provide precisely such reasons. More precisely, it analyzes three types of theories that call into question the counter-majoritarian diﬃculty and justify constitutional rigidity and judicial review on democratic grounds—namely, pure instrumentalist theories, precommitment-based theories, and proceduralist theories. According to these theories, constitutionalism need not be at odds with democracy, either because constitutional constraints are necessary for democratic procedures to bring about just outcomes, such as the protection of human rights or the improvement of the economic output (chapter 4), or because they are the upshot of commitments that actual citizens have made democratically (chapter 5), or yet because they set up the procedural rules and preconditions required for the proper functioning of the democratic system (chapter 6). Nevertheless, as we shall see throughout these chapters, all of them turn out to fall short of justifying constitutionalism on democratic grounds.

See Bayón, “Democracia y derechos: Problemas de fundamentación del constitucionalismo,” pp. 72-74.

48

Chapter 4

Instrumentalist theories
4.1. Introduction
This chapter addresses pure instrumentalist theories of the legitimacy of democratic authority and their justiﬁcation of substantive constitutional constraints.1 According to these theories, democratic institutions and decision-making are legitimate if and only if they bring about a set of substantive and procedure-independent goals, such as the protection of fundamental rights, the improvement of the economic output, or the stability of the political system. Thus, a trade-oﬀ between constitutionalism and democratic legitimacy need not arise, for the legitimacy of democratic institutions rests exclusively on their ability to achieve some substantive goals that are entrenched in the constitutional text and thus advanced by constitutional means. If the only point of democratic institutions is to produce certain outcomes and it is proven that constitutional constraints on democratic decisions are necessary for such outcomes to be brought about, the apparent contradiction between the former and the latter immediately dissolves and the latter becomes necessary for the proper functioning of democracy. The chapter proceeds as follows. It opens by setting up the structure of instrumentalism (§ 4.2) and turns to two prominent instrumentalist accounts of the legitimacy of constitutionalism—namely, that it helps democratic decision-making protect fundamental rights (§ 4.2.1) and that it improves the economic performance of the system overall (§ 4.2.2). Three criticisms are raised next (§ 4.3).

4.2. The structure of instrumentalism
Democratic institutions can be normatively assessed with regard to their procedures, to their outcomes or to a combination of both. According to (pure) democratic proceduralism, democratic decisions are legitimate because they are made by
See, among others, R. Arneson, “Democratic rights at national and workplace levels,’’ The Idea of Democracy, D. Copp, J. Hampton and J. E. Roemer, eds. (Cambridge: Cambridge University Press, 1993), p. 119; “Defending the Purely Instrumental Account of Democratic Legitimacy,” Journal of Political Philosophy 11, no. 1 (2003), pp. 122–132; “Democracy Is Not Intrinsically Just,” Democracy and Justice, K. Dowding, R. E. Goodin and C. Pateman, eds. (Cambridge: Cambridge University Press, 2004), 40-58; R. Dworkin, “What Is Equality? Part IV: Political Equality,” University of San Francisco Law Review 22 (1989): 1-30; F. A. Hayek, The Constitution of Liberty (London: Routledge, [1960] 1976); P. van Parijs, “Justice and Democracy: Are They Incompatible,” Journal of Political Philosophy 4 (1996): 101-117;
1

118

Chapter 4. Instrumentalist theories

fair procedures.2 According to (pure) democratic instrumentalism, by contrast, they are legitimate because they bring about better outcomes than any feasible alternative. Both democratic proceduralism and democratic instrumentalism combine two tenets, one about what counts as legitimate authority and another one about why democratic institutions and decisions meet that criterion. In the case of democratic instrumentalism, they are the following two:3 Political instrumentalism.—Political institutions have legitimate authority iﬀ they produce outcomes C better than any feasible political alternative, being C a set of substantive goals independent from and prior to the procedures used to bring them about. and Democratic instrumentalism.—Under the present circumstances, constitutional democracies produce outcomes C better than any feasible political alternative. A number of accounts have been provided for what counts as good outcomes—i.e. what speciﬁc C is to be brought about in order for political decisions to be legitimate. Sections 4.2.1 and 4.2.2 will tackle two of the most prominent ones—namely, the protection of fundamental rights and the improvement of the economic output. Consider brieﬂy two additional accounts. First, Amartya Sen has famously claimed that no democratic country with a free press has ever suﬀered from substantive famine.4 Similarly, David Estlund has recently argued that democratic legitimacy is the upshot of the ability of democratic institutions to prevent “primary bads” (i.e. “war, famine, economic collapse, political collapse, epidemic, and genocide”) better than any other political regime. Even though he does not provide his own empirical evidence, he points at studies showing that democratic countries have a lower tendency of going to war or that they do better in preventing famines or dealing with
See Robert Dahl, “Procedural Democracy,” in Philosophy, Politics and Society, Peter Laslett and James Fishkin, eds. (Oxford: Blackwell, 1979); Waldron, Law and Disagreement; Richard Bellamy, Political Constitutionalism. A Republican Defence of the Constitutionality of Democracy (Cambridge: Cambridge University Press, 2007). 3 See R. Arneson, “The Supposed Right to a Democratic Say,” Contemporary Debates in Political Philosophy, T. Christiano and J. Christman, eds. (Oxford: Wiley-Blackwell, 2009), p. 197. The most well-known instrumental account of the justiﬁcation of authority is Raz’s so-called service conception of authority, according to which legitimate authority, while it does not require necessarily the explicit consent of the governed, has to exist for their shake, in their beneﬁt. According to his “normal justiﬁcation thesis,” “The normal way to establish that a person has authority over another person involves showing that the alleged subject is likely better to comply with reasons which apply to him (other than the alleged authoritative directives) if he accepts the directives of the alleged authority as authoritatively binding and tries to follow them, rather than by trying to follow the reasons which apply to him directly”. See S. Wall, “Democracy, Authority and Publicity,” Journal of Political Philosophy 14 (2006): 85-100. Now, according to Viehoﬀ it is a mistake to think, as both critics such as Waldron and Christiano and supporters such as Wall do, that the Razian account of authority leaves no room for justiﬁcations of authority grounded on the value of procedures. As the above-mentioned critics point out, since the NJT is concerned only with “the content of the directives themselves” rather than with “how directives are reached,” it has to be supplemented by a procedural account of legitimate authority. See D. Viehoﬀ, “Debate: Procedure and Outcome in the Justiﬁcation of Authority,” Journal of Political Philosophy 19 (2011): 248-259, section II. 4 See A. Sen, Development as Freedom (New York: Knopf, 1999).
2

4.2. The structure of instrumentalism

119

natural catastrophes.5 Indeed, the relationship between democratic political systems and the likelihood of famines taking place, torture levels, disappearances and mass killings has been widely demonstrated to be negative and linear and generally believed to result from the fact that in democracies political leaders who use violence against their citizens can be removed from oﬃce.6 Second, Adam Przeworski has argued that democracy makes possible alternation in oﬃce without bloodshed, since elections, when free and competitive, provide the losing parties with the strong incentive to wait until the next elections rather than turning to military coup or calling for revolution.7 In a similar vein, though aimed at explaining the raise of democracy, Acemoglu and Robinson have claimed that democracy improves the stability of the political system and that it reduces the likelihood of social revolutions by providing a way for the economic elites to commit credibly to the redistribution of wealth to the poor, thus reducing the likelihood of social revolutions aimed at redistributing wealth by means of violent expropriation.8 Thus far we have seen why political instrumentalists may have good reasons to embrace democracy. Consider now why they may also have good reasons to embrace constitutionalism. First of all. it is important to note that, since instrumentalist theories deﬁne outcomes as prior to and independent from the procedures that are expected to bring them about, the relationship between the former and the latter is taken to be totally contingent. Further, democratic procedures can also produce pathologies and undesirable side-eﬀects when inadequately constrained or unconstrained at all. Consider two instances of such pathologies. First, a parliamentary majority might well use its political power to trump the civil rights of a minority, or to expropriate private property and discourage foreign investment, or yet to exclude an opposing party from the political process jeopardizing the stability of the political system overall. This is the famous threat of a “tyranny of the majority,” as labeled by J. S. Mill.9 Second, social choice theorists have argued that democratic voting can result in intransitive and incoherent outcomes due to Condorcet’s paradox, thus enabling actors to control the agenda setting so as to advance their own interests.10 Consider
5 D. Estlund, Democratic Authority. A Philosophical framework (Princeton: Princeton University Press, 2008). For a critique of the weakness of the empirical evidence provided by Estlund, see E. Anderson, “An Epistemic Defense of Democracy: David Estlund’s Democratic Authority,” Episteme. A Journal of Social Epistemology 5 (2008): 129-139. 6 See C. Davenport and D. A. Armstrong II, “Democracy and the Violation of Human Rights: A Statistical Analysis from 1976 to 1996,” American Journal of Political Science 48 (2004): 538-554. For a recent instrumentalist argument for a human right to democracy based on this evidence, see T. Christiano, “An Instrumental Argument for a Human Right to Democracy,” Philosophy & Public Aﬀairs 39 (2011): 142-176. 7 A. Przeworski, “Minimalist Conception of Democracy: A Defense,” in Democracy’s Value. I. Shapiro, ed. (Cambridge: Cambridge University Press, 1999). 8 D. Acemoglu and J. A. Robinson, Economic Origins of Dictatorship and Democracy (Cambridge: cambridge University Press, 2005). I go back to their argument in section 5.4.4. 9 See J. S. Mill, “On Liberty,” in Collected Works of John Stuart Mill, vol. 18, J. Robson, ed. (Toronto: University of Toronto Press). 10 The locus classicus is W. Riker, Liberalism Against Populism: A Confrontation Between the Theory of Democracy and the Theory of Social Choice (San Francisco: W. H. Freeman, 1982). For a criticism based on a comprehensive study of the empirical likelihood of this problem, see G. Mackie, Democracy Defended (Cambridge: Cambridge University Press, 2004).

120

Chapter 4. Instrumentalist theories

it brieﬂy. Assume that there are three candidates—A, B, and C—and that there are three voters with the preferences shown in table 2 (the candidates are ordered in decreasing order of preference). Further assume that the elected candidate is to be chosen from the aggregation of all voters’ votes, with all the votes having the same weight, and using a majority rule. As it turns out, the use of majority rule produces an intransitive and cyclical outcome, for there is a majority of two voters that prefer candidate A over candidate B, a majority of two voters that prefer candidate B over candidate C and a majority of two voters that prefer candidate C over candidate A. Voter Voter 1 Voter 2 Voter 3 Preference 1 A B C Preference 2 B C A Preference 3 C A B

Table 3. Condorcet’s paradox: voters’ preferences Constitutional constraints are thus called upon to prevent these pathologies from happening and to ensure that the democratic process produces the expected outcomes. As for the ﬁrst problem, as Holmes has argued, the constitutionalization of civil rights creates “gag rules” that take away issues of vote-swapping and horse-trading, preventing them from being trumped by temporary majorities and ensuring the stability of the system overall. As for the second problem, it has been shown that the requirement of majorities larger than 64 percent—the sort of supermajoritarian requirement for a constitutional amendment—rules out the possibility of the Condorcetian paradox taking place.11 In short, the instrumentalist case for constitutional constraints on democracy can be divided into four steps. First, the legitimacy of political institutions depends only on the outcomes they bring about. Second, democratic institutions have legitimate authority over citizens due to their ability to bring about such outcomes better than any feasible political alternative. Third, democratic institutions can also produce pathological outcomes and undesirable side-eﬀects when unconstrained. Fourth, constitutional constraints are required in order to minimize those pathologies from happening and to improve the likelihood of the expected outcomes being brought about. Consider now two outstanding accounts that have been provided for what counts as good outcomes 4.2.1. Outcomes 1: fundamental rights A very inﬂuential justiﬁcation of constitutional democracy is that democratic decision-making ought to be constrained so as to protect citizens’ fundamental rights from the tyranny of parliamentary majorities. Even though the abuse of the expression has almost deprived it of its meaning, reasonable worries persist. This is so, not only because majorities can often be ill-intentioned toward minorities. But also due to the logrolling and horse-trading typical of parliamentary politics, on the one
11

See A. Caplin and B. Nalebuﬀ, “On 64%-Majority Rule,” Econometrica 56 (1988): 787-814.

4.2. The structure of instrumentalism

121

hand, and the possibility of majorities just erring, on the other. Hence the alleged necessity to insulate rights from daily politics and ephemeral majorities. The ﬁrst worry is expressed by Raz: “Legislatures because of their preoccupation with current problems, and their felt need to secure re-election by a public all too susceptible to the inﬂuences of the short term, are only too liable to violent swings and panic measures.” 12 The second worry, by Dworkin: “Any competent interpretation of the Constitution as a whole must therefore recognize ... that some constitutional rights are designed exactly to prevent majorities from following their own conviction about what justice requires.” 13 In short, one of the main reasons for posing constitutional constraints on ordinary political decisions is that such constraints provide a better protection for individual rights from temporary and fallible majorities.14 Richard Arneson has provided a very inﬂuential pure instrumentalist justiﬁcation of constitutional democracy based on the protection of fundamental rights. According to him, fundamental-rights instrumentalism provides “the most natural and compelling justiﬁcation of political regimes of substantive constitutional democracy under modern conditions.” 15 His argument, in a nutshell, is that democracy is a mere means for the securing of fundamental rights, which justiﬁes the inclusion of the latter in a rigid constitution enforced by non-elected judges holding ﬁnal powers of review. Now, for his argument to succeed he needs to make three steps—one conceptual, another one normative and yet another one empirical. First, he has to show the conceptual distinction between procedural rights and fundamental rights. Second, he has to show their normative distinction—i.e., the mere instrumental value of the former and the intrinsic value of the latter.16 And third, he has to show that constitutional democracy follows as the best institutional response to the second step. Let us set aside the latter one (which has been already addressed, even if only brieﬂy, in 4.2) and consider the former two in turn. A. The conceptual distinction. Unlike civil or economic rights, political rights are Hohfeldian power-rights. They provide their holders with the power to change the

J. Raz, Ethics in the Public Domain (Oxford: Clarendon Press, 1994), p. 260. R. Dworkin, Law’s Empire (Cambridge, Ma: Belknap Press, 1986), p. 376. 14 This view is often—even though not always—grounded on a mainly negative conception of fundamental rights, termed as “trumps” (Dworkin), as “side-constraints” (Nozick), as the “sphere of the undecidable” (Ferrajoli), or as “forbiden territory” (Garzón Valdés). According to Dennis Mueller’s, for example, “A right is an unconditional freedom of an individual to undertake a particular action or to refrain from such an action without interference or coercion from other individuals or institutions.” See D. Mueller, Constitutional Democracy (Oxford: Oxford University Press, 2000), p. 212. 15 R. Arneson, “Democratic rights at national and workplace levels,’’ p. 119. See also his “Defending the Purely Instrumental Account of Democratic Legitimacy;” “Democracy Is Not Intrinsically Just;” and “The Supposed Right to a Democratic Say,” Contemporary Debates in Political Philosophy, Thomas Christiano and John Christman, eds. (Oxford: Wiley-Blackwell, 2009), pp. 197-212. 16 It might be argued that someone who embraces an instrumentalism of rights across the board, and thus rejects the very idea of fundamental rights also embraces an instrumentalism of democratic rights. True. However, here I focus on a narrower position, as Arneson does as well—namely, the one which assumes that if there are fundamental rights, democratic rights are not among them. See Arneson, “The Supposed Right to a Democratic Say.”
13

12

122

Chapter 4. Instrumentalist theories

very framework of rights they and their fellow citizens have.17 As Arneson puts it, “The exercise of the vote is an exercise of power by the voter over the lives of other citizens.” 18 The former are ﬁrst-order rights, whereas the latter are second-order rights, for they refer to the former and can serve, provided that enough citizens exercise them in a certain, coordinated way, as eﬀective means, for example, to create same-sex marriage rights, to eliminate retirement rights or to modify the structure of property rights. In order to clarify the distinction, consider two possible objections. First, it might be objected that ﬁrst-order rights also give power to their holders. For example, a powerful foreign investor can be in a position to threaten the government with withdrawing her investment if workers’ rights are not curbed. Even if she does not hold formal political rights, she might well be said to have powerful political rights over workers’ rights. However, this example does not dissolve the distinction for she might be said to have de facto (and very powerful indeed) political rights that put her in a position to successfully change the framework of workers’ ﬁrst-order rights. Second, it might also be objected that ﬁrst-order rights give power to their holders not because they indirectly provide them with de facto political rights, but because they impose correlative duties on other people, thus aﬀecting the rights of the latter. For example, if I have a right to free speech, then you have a correlative duty not to interfere with my publishing of an op-ed piece in the newspaper. Likewise, if I have a right to be assisted in case of a car accident, then you have a correlative duty to stop driving and assist me. As Griﬃn has put it in his critique of Arneson, it actually happens that “rights of virtually any sort ... allow one to exercise some degree of power over others without ﬁrst getting their consent.” 19 However, this kind of power is very diﬀerent from the sort of power provided by political rights, which provide their holders with the power to decide over which rights other people have, creating, modifying and eliminating full sets of other citizens’ rights. For example, oﬃcials have this sort of power over their citizens, bosses over their workers and parents over their children not because they have rights that the latter have to honor (which they do), but rather because they are in a position to modify the very structure of the rights the latter have and to enforce them through the issue of binding commands. As Arneson puts it, “a right that constraints other people from engaging in a certain type of conduct toward the right-holder diﬀers from a right to set rules that might specify what others shall do across a broad range of important types of conduct.” 20 Now, assume that a distinction between ﬁrst-order and second-order rights holds. Which rights among the former count as fundamental? Arneson’s response to this question is ambiguous. In an early paper he included a set of Rawlsian-style rights among the fundamental ones, including the right to material resources, while in a later paper he leaves open the question of “what is the moral standard for assessing
W. N. Hohﬂed, Fundamental Legal Conceptions (New Haven: Yale University Press, 1919). Arneson, “Democratic rights at national and workplace levels,’’ p. 120. 19 C. G. Griﬃn, “Debate: Democracy as a Non-Instrumentally Just Procedure,” Journal of Political Philosophy 11, no 1 (2003): 111-121, p. 112. 20 Arneson, “Democracy Is Not Intrinsically Just,” p. 47.
18 17

4.2. The structure of instrumentalism

123

results that determines which ones are best.” 21 Neither response goes without problems, as it will become clear in section 4.3.2. However, let us assume for now that an adequate account of which rights count as fundamental can be provided. B. The normative distinction. The fact that democratic rights serve as an instrument to create, change and even suppress ﬁrst-order rights does not necessarily imply that they have mere instrumental value subordinated to the advancement of the latter. For that matter, a taxi driver might well serve me as an instrument to get home, but he obviously does not have mere instrumental value to me.22 Likewise, political rights do have instrumental value, for someone who lacks franchise cannot advance her ﬁrst-order rights properly. However, it might be argued that, when someone is disenfranchised, her condition as an equal member of the political community—and not only her political purchasing power, to put it in rather Schumpeterian terms—is also taken away from her to great extent. Democracy would thus qualify as a worthwhile moral goal in itself (say, due to the equality or the autonomy that it would acknowledge to each citizen). Now, according to Arneson, intrinsic accounts of democratic rights are ﬂawed, for political rights and democracy more generally lack intrinsic value and are justiﬁed only insofar as they help secure citizens’ fundamental rights.23 This is so for two reasons—namely, because procedural rights are by deﬁnition merely instrumental and because they provide power over other people’s fundamental rights and thus are to be assessed regarding whether they help advance or not such rights. Arneson provides two analogies to justify these two reasons. Firstly, he compares democratic procedures to searching for a genuine treasure, in which we asses our devices (e.g., maps) in purely instrumental terms.24 Maps are indeed useful and probably necessary for any attempt to gain the treasure, but it would be certainly awkward to say that their use has intrinsic value of any sort. Mutatis mutandis, democratic procedures, including democratic rights, turn out to be useful and probably necessary for any attempt to secure and advance fundamental rights, but it would be equally awkward to say that they have more than instrumental value. The second analogy compares democratic rights to parental rights, which also provide parents with great power over their children. Indeed, parents enjoy ample discretionary authority in the upbringing of their children—they can decide which
Arneson, “Democratic Rights at National and Workplace Levels,” p. 119. “Democracy Is Not Intrinsically Just,” p. 41 n. 3. 22 This example resembles Kant’s categorical imperative in its second formulation, which reads as follows: “Act in such a way that you treat humanity, whether in your own person or in the person of any other, always at the same time as an end and never merely as a means to an end.” However, it need not be justiﬁed on Kantian grounds. Immanuel Kant, Grounding for the Metaphysics of Morals, James W. Ellington, trans. (Hackett, [1785] 1993), p. 30. 23 Arneson goes beyond more moderate consequentialist theories—which might allow trade-oﬀs between intrinsic and procedural fairness, even if only to end up by defending that the latter trumps the former—by embracing a “best results” theory, according to which intrinsic fairness judgements should not be given any consideration in decision making and policy making at all. The label comes from C. Beitz, Political Equality. An Essay in Democratic Theory (Princeton, N.J.: Princeton University Press, 1989), p. 20. See Arneson, “Democratic rights at national and workplace levels,’’ p. 145 n. 15. 24 Arneson, “Democracy Is Not Intrinsically Just,” pp. 43-44.
21

124

Chapter 4. Instrumentalist theories

school to take them to, which religion to raise them in or which other kids can they hang out with. Now, according to Arneson, such discretionary rights are given to parents just because doing so proves to be a good way of ensuring each child’s right to a decent upbringing. As soon as parents undisputedly fail to ensure the fundamental rights of their children (say, because of malnutrition or sexual abuses), they also lose their legitimate discretionary power over them. Mutatis mutandis, political rights are given to citizens just because doing so proves to be a good way of ensuring that each citizens’ fundamental rights are protected. Hence, as soon as citizens fail to bring about this goal (say, because they use a parliamentary majority to suppress their right not to be tortured or the right from deprivation), they also lose their legitimate authority to exercise their political rights.25 4.2.2. Outcomes 2: economic performance Consider now a second instrumentalist approach to constitutional democracy, according to which constitutional constraints make democratic decision-making bring about greater stability, accountability and credibility and thus improve the economic performance of the system overall. As Berggren and Kurrild-Klitgaard point out in a survey of the recent literature on constitutional political economy, “This perspective entails a largely instrumental view of how a democracy should be designed. [It] implies that the choice of democratic building blocs, within a framework of ’minimal’ basic principles of democracy, is a rather open question being dependent on to what extent the alternative arrangements contribute to this satisfaction of the citizens’ preferences.” 26 To be sure, as Elster points out, the causal relationship between constitutionalism and economic performance goes in both directions. Constitutional constraints may improve economic performance and the enactment and eﬀective enforcement of such constraints, in turn, may be feasible only under a speciﬁc level of economic development.27 However, this section only focuses on the ﬁrst direction of the relationship—namely, the impact of constitutional constraints on economic performance and their justiﬁcation on that ground.28
25 Griﬃn and others have objected that Arneson does not capture adequately the relationship of paternity for he does not take into account the desire of parents to parent their children. According to Arneson, this reading rests on a misunderstanding, for his account leaves room for taking into account the deep and strong desires of parents to raise their own children as fundamental rights. As long as “natural parents’ natural inclination to be connected as primary guardians of their own children [proves to be] an important aspect of the aims and aspirations, satisfaction of which enters into their right to wellbeing, then that goes in the hopper of the fundamental moral rights.” Arneson, “Debate: Defending the Purely Instrumental Account of Democratic Legitimacy,” p. 128. 26 N. Berggren and P. Kurrild-Klitgaard, “Economic Eﬀects of Political Institutions, Whith Special Reference to Constitutions” in Why constitutions Matter, N. Berggren, N. Karlson, and J. Nergelius, eds. (Stockholm: City University Press, 2000). 27 According to Przeworski et al., for example, no democracy failed above a per capita GDP of $6055 (measured in 1985 purchasing power parity dollars), which was Argentina’s level in 1975. See A. Przeworski, J. A. Cheibub, M. E. Alvarez and F. Limongi, Democracy and Development: Political Institutions and Material Well-being in the World, 1950-1990 (Cambridge: Cambridge University Press, 2000), p. 106. 28 See J. Elster, “The Impact of Constitutions on Economic Performance,” Proceedings of the World Bank Conference on Development Economics 1994 (1995): 209-226, p. 209. The ﬁeld of constitutional political economy goes back to J. Buchanan and G. Tullock’s The Calculus of

4.2. The structure of instrumentalism

125

Before proceeding, three caveats are in order. First, the notion of economic performance is not an undisputed one. While some deﬁne it as the maximization of some aggregate utility or wealth, others such as Elster include economic security—deﬁned along prioritarian lines as the ensuring of the highest level of well-being for the worst-oﬀ—as one of its variables. Second, the causal relationship between constitutional constraints and the improvement of economic eﬃciency and security is too technical a question for my level of expertise. However, I think that it is worth sketching the main causal mechanisms through which, as it is often claimed, constitutionalism might bring about such outcomes. Third, contrary to what is sometimes assumed in constitutional political economy, the main focus of the section is not the economic impact of rights or democracy, for both rights and democratic institutions can be enacted by means other than speciﬁcally constitutional means—i.e., without being included in a rigid constitutional text and enforced through judicial review. Instead, the section is mainly focused on the economic eﬀects of rigidity and judicial review—i.e., constitutionalism proper.29 That said, let us mention three ways in which constitutionalism might help improve economic performance: A. Time consistency. Unless constrained to do otherwise, legislatures and administration can suﬀer from time inconsistency in two diﬀerent ways. First, by investing in short-term projects with smaller yields rather than in projects with larger yields in the long run. The main reason for this is that the short-termism of electoral politics. For example, in order to satisfy present voters, politicians in oﬃce might be tempted to invest in housing by providing buyers with tax beneﬁts so as to reduce the unemployment rate among construction workers rather than investing in education and infrastructures so as to change the economic structure of the country, even if in the long run this decision can create economic bubbles and worse economic outcomes overall. David Stockman, budget director during the Reagan administration, expressed this view with unusual sincerity. Stockman refused to address Social Security severe long-term problems in 1981 because, as he claimed, he had little interest
Consent (Ann Arbor: University of Michigan Press, 1962) and has developed in journals such as Constitutional Political Economy, founded by Buchanan in 1990. The locus classicus showing the relationship between constitutionalism and economic performance in seventeenth-century England is D. C. North and B. R. Weingast, “Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in Seventeenth-Century England,” Journal of Economic History 49 (1989): 803-832. 29 True, the causal relationship between constitutionalism and economic performance can be analyzed in terms of the economic impact of rights, federalism, the length of legislative terms, the electoral system and other institutional devices, regardless of whether these devices are included in a rigid constitution and enforced through judicial review or, by contrast, just in hands of the legislature, as it is the case in the Westminster model. Further, ordinary laws do limit the behavior of the persons who operate under them, but unless constrained they do not credibly limit the behavior of the legislator for she can change them at will—i.e., through a simple parliamentary majority, which is precisely what constitutionalism attemps to prevent. Actually, what diﬀerences constitutional laws from ordinary ones is that the former do constrain the behavior of the legislatures, for they are enforced by an independent body and changeable only through a procedure more stringent than the ordinary legislative procedure. For a survey of the former task, see P. Kurrild-Klitgaard and N. Berggren, "Economic Consequences of Constitutions: A Theory and Survey," Journal des Economistes et des Etudes Humaines 14 (2004): article 1. For the need to distinguish between rules and commitments to rules, see Sánchez-Cuenca, Más democracia, menos liberalismo, p. 137-140.

126

Chapter 4. Instrumentalist theories

in wasting “a lot of political capital on some other guy’s problem in [the year] 2010.” 30 Hence, the constitutionalization of certain burdens and duties can enlarge the time horizon of the government, thus avoiding short-termism and ensuring the provision of certain goods along time. Second, constitutional commitments can prevent the government from deviating from their plans, thus helping them to avoid the so-called hyperbolic discounting. Time discounting is the tendency of people to give greater value to rewards (and greater disvalue to punishments) as they come closer in time. At the personal level, such a discounting is closely related to the phenomenon of the weakness of the will, and it was captured in St. Agustine’s prayer: “Give me chastity and continence, only not yet.” 31 It can have the consequence of making people deviate from an unpleasant commitment right before it happens. An example at the institutional level might be the commitment of the UK Labour Party to put on the agenda the electoral reform, a commitment that had been in their electoral manifesto for a long time but was never carried out whenever they got to oﬃce. Another institutional example might be the tendency of the head of government to break her previous commitments not to run for oﬃce for a second term when the ﬁrst term is about to expire and no mandatory rotation is included in the constitution. In short, constitutional commitments might force governments to carry out their commitments by preventing them from deviating from their previous plans. B. Stability. Regardless of the speciﬁc content of the constitution, constitutional rigidity improves the stability of the institutional system overall by providing a set of higher-order laws that are very diﬃcult to change and impose strong constraints, in turn, on the changeability of ordinary laws through judicial review. This improved stability has a number of economic eﬀects. First, constitutional rigidity and judicial review reduce the likelihood of fast and continuous law-making due to short-termism. In turn, by improving institutional stability and predictability constitutionalism can expand the time horizon of the economic agents, providing them with the incentive to make long-term investments and contracts and thus improving the economic output. As Deng Xiaoping once said, “Without a stable environment, nothing can be achieved.” 32 Second, even though no constitutional article has ever prevented a coup d’état taking place, as Guillermo O’Donnell once commented, constitutional entrenchment and judicial review can reduce the likelihood of a cabinet taking advantage of its position to suspend democratic and civil rights, say, by ruling out opposing parties and media. By doing so, constitutionalism provides economic agents with the certainty that democratic institutions will last long (or at least longer than under a less constrained institutional framework). On the other hand, there is some evidence that
Quoted in P. Pierson, “Increasing Returns, Path Dependence, and the Study of Politics,” American Political Science Review 94 (2000): 251-267, p. 261. 31 Quoted in Elster, “The Impact of Constitutions on Economic Performance.” 32 D. Xiaoping, quoted in C. Wing-hung Lo, China’s Legal Awakening (Hong Kong: Hong Kong University Press, 1995), p. 298.
30

4.2. The structure of instrumentalism

127

democracies perform economically better than autocracies. The main reason for this is that authoritarian rulers are neither controlled by opposing parties nor can they lose elections and have to step down. Accordingly, they have no proper incentive to maximize the total economic output. Superior economic outputs are thus to be expected under stable and durable democratic institutions.33 Third, constitutionalism can reduce the large amount of resources devoted at manipulating law-making and lobbying MPs when no constitutional constraints exist. When certain issues are insulated from day-to-day politics by making them diﬃcult to change, time and money that would be otherwise devoted to trying to inﬂuence the reform can be spent diﬀerently. As Posner comments, “If the vote of a simple majority could change the basic form of government or expropriate the wealth of a minority, enormous resources might be devoted to seeking and resisting such legislation.” 34 C. Credible signaling. Even if truly committed, an unconstrained government might lack mechanisms to make its commitments seem credible to other economic agents—e.g., investors or workers. Hence, rigid and externally monitored and enforced constitutional constraints can help signal the willingness to uphold certain rules and therefore improve the credibility of the commitments made.35 As Mancur Olson has argued, an autocrat cannot credibly commit himself: “If he runs the society, there is no one who can force him to keep his commitments.” 36 As explained above, this was the case of China during the 1980s, when the economy was liberalized and private property allowed. Under such novel circumstances, the economic agents could not know if the commitment of Deng Xiaoping and the Politburo of the Communist Party was actual or pretended, if the system would last and, eventually, if they would be able to retain their returns. As a consequence of this, they often chose to use their proﬁts for housing or luxury consumption rather than reinvesting them. The Chinese leaders were truly committed to the reforms and the stability of the new system, as it turned. However, they could not signal credibly their commitment because they retained all the power. Had they self-bound through the constitutionalization of their commitments, they would have made it easier for them to signal them, thus improving the incentives of the economic agents to reinvest their proﬁts.37
A. Przeworski and F. Limongi, “Political Regimes and Economic Growth,” Journal of Economic Perspectives 7 (1993): 51-69. 34 Quoted in Elster, “The Impact of Constitutions on Economic Performance.” 35 Schwartzberg explains how the Athenians used entrenchment of some provisions to make their commitments more credible. M. Schwartzberg, Democracy and Legal Change (New York: Cambridge University Press, 2007), ch. 2. There seems to be a negative empirical relationship between the length of a constitution and economic wealth. A study of 38 countries in the late 1980s shows that the longer the constitution (measured by the number of articles as a proxy), the lower of GDP per capita. See A. A. Montenegro, “Constitutional Design and Economic Performance,” Constitutional Political Economy 6 (1995): 161-169. 36 M. Olson “Autocracy, Democracy and Prosperity,” in Strategy and Choice, R. J. Zeckhauser, ed. (Cambridge: MIT Press, 1991), p. 153. 37 See Elster, Ulysses Unbound, pp. 148-149.
33

128

Chapter 4. Instrumentalist theories

4.3. Criticisms
Three diﬀerent criticisms are raised in what follows, all of them aimed at calling into question the democratic legitimacy of instrumentalist accounts of constitutionalism. Section 4.3.1 argues that these accounts are unable to address Wollheim’s paradox, and turn out to be an incomplete normative account of political authority as a result. Section 4.3.2 shows that instrumentalist accounts turn out to be self-defeating because they do not address adequately the fact of disagreement that, as we have seen in Part I, turns out to be a necessary condition of democratic legitimacy. Finally, section 4.3.3 shows that pure instrumentalism turns out to be too radical an account of political legitimacy when we consider it from the point of view of actual citizens, who tend to take into account not only the outcomes of political decisions but also the fairness of the procedures employed to bring such outcomes about. Before proceeding, the following caveat is in order though. Even though the instrumentalist justiﬁcations of constitutionalism can be attacked on instrumentalist grounds, this is not the path followed here. Put diﬀerently, it may be accepted that the legitimacy of political institutions depends exclusively on the outcomes they bring about, yet criticize the ability of constitutional rigidity and constitutional courts with ﬁnal powers of review of legislation—vis-à-vis alternative institutional forms, such as Westminster-style parliamentary sovereignty or the so-called new Commonwealth models of constitutionalism—to better protect fundamental rights, or to produce better economic outcomes, or yet to generate more stability. To be sure, there are good grounds on which this sort of criticisms can be raised.38 However, this issue is mainly empirical and cannot be settled by mere philosophical analysis—let alone the methodological problems of analyzing the causal relationship between variables as complex as these, and the very problematic counterfactual comparisons that such analysis involve. This is the reason why the remainder of the chapter is limited to the analysis of instrumentalist theories of constitutionalism from the point of view of their democratic legitimacy. In short, I will leave aside the second tenet of democratic instrumentalism (supra, § 117), and focus on the ﬁrst one, i.e. Political instrumentalism.—Political institutions have legitimate authority iﬀ they produce outcomes C better than any feasible political alternative, being C a set of substantive goals independent from and prior to the procedures used to bring them about. 4.3.1. Wollheim’s paradox The ﬁrst problem with instrumentalist theories of democratic authority—at least, with pure instrumentalist ones, which do not attribute any normative weight to procedures at all—is that they are unable to address Wollheim’s paradox.39 Let us
38 For a discussion and summary, see Waldron, “The Core of the Case Against Judicial review,” pp. 1376-1386. 39 R. Wollehim, “A Paradox in the Theory of Democracy,” in Philosophy, Politics and Society, Peter Laslett and Runciman, eds. (Oxford: Basil Blackwell, 1962), pp. 71-87.

4.3. Criticisms

129

consider the paradox ﬁrst (epigraphs A-D) and turn to the problems of instrumentalism to address it satisfactorily next (epigraph E). A. The paradox. Consider ﬁrst Wollheim’s paradox. Take a citizen X who whole-heartedly believes that (A1) Policy A ought to be enacted.

However, since X is a democrat, she also believes that (A2) The policy chosen by the majority ought to be enacted.

Now, assume that on this occasion (A3) The policy chosen by the majority is policy B (which is incompatible with A). Therefore, from (A2) and (A3) it follows that our citizen also believes—at ﬁrst sight inconsistently with (A1)—that (A4) Policy B ought to be enacted.

A number of ways out of the paradox of X believing both in (A1) and (A4) have been provided ever since Wollheim’s paper ﬁrstly appeared in 1962. They can be divided into three main types—namely, those which attempt to collapse (A1) into (A4),40 those which attempt to collapse (A4) into (A1)41 and those which attempt to show that (A1) and (A4) need not be at odds.42 In this chapter I try to show why the two ﬁrst strategies fail and I try to provide an alternative solution to the paradox. According to this solution (A1) and (A4) can be consistently held because they are derived from two diﬀerent types of principles—a ﬁrst-order principle of substantive correctness and a second-order principle of procedural legitimacy. These principles provide an answer to two diﬀerent normative questions—namely, to “what ought to be done” and to “what ought to be done provided that we disagree on what ought to be done and that, nevertheless, we have to take a common course of action.” In short, X can observe policy B, which authority is granted as procedurally legitimate and respectfully obeyed, while at the same time being committed to policy A as the policy she takes as substantively correct through, say, her attending demonstrations, petitioning, or voting for A in referenda. Next, it will be shown that, once we accept the two-level solution, pure democratic instrumentalism turns out to be an incomplete normative account of political authority. According to pure instrumentalism, political institutions have legitimate
40 Among these, see T. Honderich “A Diﬃculty With Democracy,” Philosophy & Public Aﬀairs 3 (1973-4): 221-26. 41 Among these, see B. Barry, Political Argument (London: Routledge & Kegan Paul, 1965); D. Weiss, “Wollheim’s Paradox: Survey and Solution,” Political Theory 1 (1973): 154-70; and R. Pennock, “Democracy Is Not Paradoxical: Comment,” Political Theory 2 (1974): 88-93. 42 Among these, see Wollheim, “A Paradox in the Theory of Democracy;” R. Harrison, “No Paradox in Democracy,” Political Studies 18 (1970): 514-17; V. Haksar, “The Alleged Paradox of Democracy,” Analysis 37 (1976): 10-14; and J. Waldron, Law and Disagreement.

130

Chapter 4. Instrumentalist theories

authority if and only if they bring about outcomes better than any feasible political alternative. It thus collapses procedural legitimacy into substantive correctness and, accordingly, (A4) to (A1). However, pure instrumentalism cannot dismiss X’s commitment to (A4) quite so easily, for (A4) obtains from X’s acknowledgment of the fact of disagreement. And the fact of disagreement, in turn, is not to be taken as a contingent political incident, but as a necessary condition for political decision-making procedures being called for, for if all citizens agreed on what ought to be done no decision-making procedure would be required at all. As Przeworski has put it, democracy presupposes that “Interests or values are in conﬂict. If they were not, if interests were harmonious or values were unanimously shared, anyone’s decisions would be accepted by all, so that anyone could be a benevolent dictator.” 43 The section proceeds as follows. Epigraphs B and C discuss previous sttempts to solve Wollheim’s paradox. Epigraph D presents the two-level solution. Epigraph E shows why pure instrumentalism is unable to address the paradox. B. The Rousseaunian solution. A ﬁrst possible way out of the paradox consists in claiming, ﬁrst, that there is a contradiction between X being committed both to (A1) and (A4) and, second, that X can nevertheless change her mind once she realizes that B is the preferred solution by the majority, thus turning to believe that B rather than A ought to be enacted. In this case, it would be proven that the voter was not truly committed to believing that A ought to be enacted no matter what, for when she ﬁrstly expressed her preference this would have been hypothetical and dependent on what other people believed. In short, she would have rather believed that (A1*) A ought to be enacted provided that a majority is of the same opinion.

Three sorts of problems (conceptual, empirical, and normative ones) arise with this possible solution—call it the Rousseauian solution. Consider the conceptual diﬃculties ﬁrst. Assume that voters do address Wollheim’s paradox by committing themselves to the view that A ought to be enacted provided that a majority is of the same opinion. (In politics this can be often the case due to, say, strategic voting or bandwagon eﬀects.) If that is the case, then X’s preferred policy will eventually collapse into whatever she believes that it could cast the support of a majority of voters, be it B or A. The reason for this is that, if B is eventually chosen, X will admit, given that she is committed to (A2), that she was wrong in her calculus, thus changing her mind and turning to believe that B rather than A ought be enacted. And if, on the contrary, A is eventually chosen, X will keep believing that A rather than B ought to be enacted, but only because a majority has chosen A over B rather than because she whole-heartedly believes that A ought to be enacted, for if that was the case she would not have changed her mind in the former possibility.
Adam Przeworski, “Self-Enforcing Democracy,” in Oxford Handbook of Political Economy, Donald Wittman and Barry Weingast, eds. (New York: Oxford University Press, 2006), p. 312.
43

4.3. Criticisms

131

Of course, there is nothing inconsistent or irrational in someone behaving as X does. Now, the problem comes when we generalize this voting behavior and apply it to everyone else. Assume that (B1) Every voter believes that A ought to be enacted provided that a majority is of the same opinion. Then it follows that (B2) Every voter will cast their vote according to what they anticipate that a majority will vote for rather than to what they believe whole-heartedly. But then it also happens that (B3) Since every vote is a function of everyone else’s votes, the policy that the majority of voters vote for turns out to be indeterminate. As an analogy, consider the case of someone—call her Mary—who distrusts her own ability to wisely order in a restaurant and has decided to order whatever the majority of her friends orders or, in order to keep the analogy accurate, whatever she believes a majority of them would or will order. Nothing irrational or inconsistent thus far. However, her behavior makes sense only insofar as her friends do trust their own criteria. If all of them were as hesitant as Mary none of them would be in a position to delegate their decision for, being the decision of the majority a function of individual decisions that are, in turn, a function of the decision of the majority, the former would turn out to be indeterminate. Turning back to politics, as Graham puts it, “Since no one expresses a categorical opinion that B ... ought to be enacted, an aggregation of the votes will yield only the conclusion that the majority has chosen B -if-a-majority-chooses-it.” 44 Admittedly, the conceptual problem does not pose a problem to the solution insofar as enough citizens refrain from delegating their decision. This seems to be the usual case in democratic societies indeed. But when that happens, an empirical problem arises—namely, that the solution does not account precisely for such voters, i.e. for those voters who whole-heartedly believe that A ought to be enacted. This often happens both before and after the voting takes place. Thus, in democratic politics it is not infrequent for a voter to vote for A knowing how the voting will go and often, as Wollheim already anticipated, knowing quite well indeed that her vote will go against the majority will.45 Now, why would someone who is truly committed to (A2) vote for a policy that she knows in advance that won’t gain the support of the majority?46 Further, many of these voters still disagree with policy B after it has been backed by a majority of votes, often speaking out against it, yet without calling into question the legitimacy of its adoption procedure. As Waldron has pointed out, an adequate account of the law’s authority—including Wollheim’s paradox—must
44 K. Graham, “Democracy, Paradox, and the Real World,” Proceedings of the Aristotelian Society 76 (1976): 227-245, p. 229. 45 Wollheim, “A Paradox in the Theory of Democracy,” p. 80. 46 See Harrison, “No Paradox in Democracy,” p. 515.

132

Chapter 4. Instrumentalist theories

accommodate the fact that the law “makes its demand ex hypothesi on someone who thinks he has good ground for believing that the legislature is mistaken.” 47 The third problem does not aﬀect the voters who believe that A ought to be enacted even after realizing that B is the preferred option by the majority, but rather the voters who do conform their policy preferences to those of the majority. The problem is normative—namely, it is unclear why such votes should have any normative appeal or authority whatsoever.48 After all, democracy is, among other things, a procedure employed to reﬂect citizens’ preferences about policies that will aﬀect their lives, which will be funded with their taxes and which will be coercively enforced against them. Going back to the analogy, consider that Mary and her friends wanted to order a dish to share. Should Mary’s vote have the same weight in the ﬁnal decision as the votes casted by her less hesitant friends? In short, the three issues raised here pose a dilemma to the Rousseauian solution. The more citizens behave as Rousseauians, thus adapting their beliefs to the beliefs of the majority of their fellow citizens, the more indeterminate the upshot of the democratic procedure is and the less normative force it has. And, conversely, the more determinate and normatively forceful the decision is, the less relevant the Rousseauian solution turns out to be. C. The Hobbesian solution. Another possible way out of the paradox would be to collapse (A4) into (A1), i.e. to claim that X’s commitment to democracy is not strong enough to trump her belief in policy A and that her commitment to B is merely prudential. In that case, X would observe policy B (in case it was eventually enacted) in a rather Hobbesian fashion, i.e. not because she considered it to be just or legitimate but just because she would be worst-oﬀ otherwise. The most obvious reason for someone behaving this way is her understandable fear to be legally punished by the state or to suﬀer the social pressure if she behaved otherwise. But there are additional and perhaps more subtle reasons, such as rule-consequentialist ones. According to Weiss, for example, X can simultaneously believe that “A ought to be enacted” and that “I ought to do B ... because it is the law, and general law-keeping has desirable consequences.” Two problems with this solution—call it the Hobbesian solution—arise though. First, it implies that the only reason for X not enacting A is that she lacks the social and political resources required to do so. Hence, if X happened to enjoy the social and political power to enact A, she would enact it without batting an eyelash. Barry expressed this view with his usual clarity: If you adhere to any ideal or principle which does not include in it a reference to the opinion of others then it is logically possible that you might be the only person holding it. There could be a situation, therefore, in which you say ’So and so should be done’ and everyone except you says ’So and so
Waldron, Law and Disagreement, p. 101. See Wollheim, “A Paradox in the Theory of Democracy,” p. 81; Graham, “Democracy, Paradox, and the Real World,” p. 229.
48 47

4.3. Criticisms

133

should not be done.’ You are in eﬀect setting yourself up as a dictator. Of course, you probably won’t have the power to get what you want done against everyone’s opposition, nevertheless you are saying that if you had the power you would.49 Now, it is at least doubtful whether X’s wholehearted commitment to (A1) has to imply that X would enact A against her fellow citizens in case she enjoyed the power required to do so. This is a controversial issue for, even though it is sometimes the case that a person has such power without necessarily using it to enact A, a proponent of this solution can always reply that X is not truly committed to (A1), or that she does not really have power enough to enact A, or yet that she wrongly perceives that she does not have such power. However, it is plausible to think that X might be truly committed to (A1), have the (actual and perceived) power to enact A and yet hesitate to do so because she also believes that A ought not to be enacted by any possible means but only by legitimate ones. As Haksar has put it, “A person can be a genuine democrat, in the sense that he is morally committed to democracy, and yet not agree that the majority is right about what ought to be done. He can be a genuine democrat because he genuinely (and not just tactically or prudentially) accepts and believes that the majority has the (moral) right to authorize or refuse to authorize the use of certain means and certain compulsory measures.” 50 Second, assume that the Hobbesian solution is right and that the only reason that prevents X from enacting A, being X truly committed to (A1), is that she lacks the (actual or perceived) power to do so. Thus, if she had such power she would doubtlessly use it. The problem with this solution is twofold. First, it is highly questionable whether someone who is willing to enact her preferred policies regardless of whether her fellow citizens have consented to them or not could qualify as a democrat at all. After all, being a democrat implies, almost as a matter of deﬁnition, being morally committed to the decisions brought about by democratic means (accordingly, many Hobbesians openly assume that they are democrats only contingently.) Second, letting aside the deﬁnitional issue, rather than resolving the paradox, the Hobbesian solution only evades it, for the paradox consists in saying that two inconsistent pieces of legislation both ought to be enacted. As Barry points out, Hobbesian attempts to address the paradox, such as Weiss’s, do not have any obvious bearing on Wollheim’s paradox, which is supposed to involve a democrat in saying at the same time that A ought to be enacted and B ought to be enacted, where A and B are incompatible.51 D. The two-level solution. A number of attempts to address X’s commitments to both (A1) and (A4) without neither precluding the former to the latter nor vice versa have been provided. The solution proposed here—call it the two-level solution—draws upon the contributions of Wollheim, Harrison, Haksar and Waldron,
49 50 51

Barry, Political Argument, p. 59. Haksar, “The Alleged Paradox of Democracy,” p. 12. Barry, “Wollheim’s Paradox: Comment,” p. 318.

134

Chapter 4. Instrumentalist theories

taking them further when necessary. It attempts to show two things. First, that (A1) and (A4) can be consistently held when they are derived from diﬀerent types of principles—a ﬁrst-order principle and a second-order principle, respectively. Second, that these principles provide an answer to two diﬀerent normative questions—namely, to “what ought to be done” and to “what ought to be done provided that we disagree on what ought to be done and that, nevertheless, we have to take a common course of action,” respectively. Consider ﬁrst Harrison’s contribution, which attempts to show that, rather than a paradox of democracy, the paradox is just a particular case of the potential contradiction between general moral principles, just as principles like keeping promises and saving life can turn out to be incompatible under certain circumstances (e.g., when I have promised that I would go pick up a friend from the airport and on my way I ﬁnd someone whose life is endangered and needs my help). Further, according to Harrison, this proves that there is no paradox as such. In both cases two moral principles, not in themselves incompatible, become so in particular circumstances. If this were suﬃcient to create a paradox in democracy, it would be suﬃcient to create a paradox in promising. However, since it does not create a paradox in promising, it should not be taken to create one in democracy either.52 Harrison is correct in presenting the alleged paradox as a particular case of a more broad conﬂict between general principles. However, he does not capture the diﬀerence between his own case and Wollheim’s, since the commitments in conﬂict in his case and in Wollheim’s are of a very diﬀerent sort. In the former case, both commitments are to substantive moral principles, such as keeping promises and saving life. In the latter case, by contrast, the belief that “A ought to be enacted” expresses a substantive principle whereas the belief that “B ought to be enacted,” though also substantive, is derived from a procedural or second-order principle according to which the policy chosen by the majority ought to be enacted—i.e., premise (A2). In a nutshell, the principles from which (A1) and (A4) are derived pertain to two diﬀerent normative levels—a substantive level and a procedural one. (A1) follows from general commitment to what X takes as the correct policy to be enacted, being such commitment non-conditional on third parties’ commitments. By contrast, (A4) follows from a general commitment to what X takes as the legitimate source of authority, which can be diverse and not necessarily democratic, for example: (A2) (A2*) (A2**) (A2***) The policy chosen by the majority ought to be enacted. The policy chosen by the Pope ought to be enacted. The policy chosen by lottery ought to be enacted. The policy chosen by Glenn Beck ought to be enacted.

Another form to put the disanalogy—and to clarify the nature of these two-level principles—is to ask to which questions do Harrison’s and Wollheim’s commitments
52

Harrison, “No Paradox in Democracy,” p. 516.

4.3. Criticisms

135

provide an answer. In the former case, the two principles provide an answer to the question of what ought to be done. In the latter case, by contrast, “A ought to be enacted” provides an answer to the question of what ought to be done, while “B ought to be enacted” provides an answer to the question of what ought to be done provided that we disagree about what ought to be done and that, nevertheless, we have to make a collective decision on that issue. In short, while the former commitment provides an answer to the question of substantive correctness or justice, the latter commitment provides an answer to the question of legitimate authority in the face of existing disagreements that, for better or worse, pervade political decision-making. As Waldron has put it, A person who believes that A is the right decision but B the decision that should be implemented, is oﬀering answers to two diﬀerent, though complementary questions. That B should be implemented is his answer to the question, ’What are we to do, given that we disagree about whether A or B is just?’ That A is the right decision is his own contribution to the disagreement that called forth that question.53 The two-level solution is consistent with the one originally sketched by Wollheim. According to Wollheim, (A1) and (A4) need not be at odds when they result from diﬀerent types of principles, thus adopting diﬀerent meanings. Consider Wollheim’s example: “Jews ought to be given privileged treatment.” Even though this judgment seems to have a single meaning, it adopts very diﬀerent meanings depending on whether it follows, say, from a principle regarding the compensation for the persecuted or from an attitude of Jewish chauvinism. Mutatis mutandis, (A1) and (A4) need not be at odds when they follow, respectively, from what Wollheim labels as direct and oblique principles. Direct principles refer to the morality of actions, policies, etc. by means of some general moral expression (e.g. murder or envy), whilst the oblique principles refer to the morality of actions, policies, etc. by means of an artiﬁcial property bestowed upon them as the result of the action of some individual or institution. According to Wollheim, Two judgments of the form ’A ought to be the case’ and ’B ought to be the case’ are not incompatible even though A and B cannot be simultaneously realized if one of these judgements is asserted as a direct principle whereas the other is asserted as a derivation from an oblique principle.54 However, as it has been often pointed out, Wollheim does not elaborate on the alleged diﬀerent properties of the direct and oblique principles and leaves the issue open.55 What the two-level solution adds to Wollheim’s is that it deﬁnes such properties, in the speciﬁc case of the paradox, in terms of substantive correctness regarding the direct principle from which (A1) obtains and of procedural legitimacy regarding the
Waldron, Law and Disagreement, pp. 247-248. Wollheim, “A Paradox in the Theory of Democracy,” p. 85. 55 See, for example, Weiss, “Wollheim’s Paradox: Survey and Solution,” and Graham, “Democracy, Paradox, and the Real World.”
54 53

136

Chapter 4. Instrumentalist theories

oblique principle from which (A4) obtains, being procedural legitimacy a normative issue that only triggers in the context of pervasive political disagreement among citizens about what ought to be done when a common course of action has to be taken. Now, it might be asked whether there are any practical diﬀerences whatsoever between the two-level solution and the one provided by Harrison.56 Consider two possible replies. First, it can be replied that the two-level solution does resolve the paradox not only in theory but also in practical terms, for it presents X as observing policy A, which authority is granted as procedurally legitimate (i.e., for its being the upshot of a democratic procedure) and thus respectfully obeyed, and, at the same time, as being whole-heartedly committed to policy B through her attending demonstrations, petitioning or voting for B in referenda. Second, it might be further objected that the ﬁrst reply is question-begging for it pretends to resolve the paradox by displaying the commitments to (A1) and (A4) at two diﬀerent moments t1 and t2 rather than both at any given tn . In the latter case, as Harrison has pointed out, “’A ought to be enacted’ and ’B ought to be enacted’ [would be] surely incompatible whatever the origins of the pair.” 57 This could be the case when X tries to make up her mind about whether to obey or challenge (say, through civil disobedience) policy A. Admittedly, under these circumstances (in which X’s commitments to substantive and procedural principles come into conﬂict and a practical decision has to be made), X could only handle the case in the same way as she handles a conﬂict between two substantive principles or between two procedural principles to which she is morally committed. That is, she could only let one of the commitments trump the other one—often tragically indeed. The costs of not observing the law considered, it would be likely to expect X to bite the bullet and let her procedural commitment override the substantive one. However, in some cases she could let her substantive commitment trump the democratic one, thus deciding to disobey the law. But, then, what would be the practical diﬀerence between the two-level solution and the one provided by Harrison? Admittedly, probably none. Under these circumstances, the two-level solution would not provide X with practical guidance regarding how to overcome the dilemma without having to give up one of its horns. However, even though the solution would not make a practical diﬀerence, it would make a conceptual one, something that Harrison’s unfortunately does not, for he just presents it as a contradiction between principles that operate at the same level. A twofold distinction would be in order, thus, so as to keep the two-level solution meaningful—namely, a conceptual distinction between substantive and procedural or second-order principles, on the one hand, and between the conceptual distinction and its practical implications, on the other. Of course, then it would be possible to further object that, in the absence of a practical way out to X’s dilemma when having to decide between her commitments to (A1) and (A4) at any given tn , the alleged
56 57

Something along these lines can be found in Barry, “Wollheim’s Paradox: Comment.” Harrison, “No Paradox in Democracy,” p. 515.

4.3. Criticisms

137

solution would not have really provided a full solution to the paradox—if a solution at all. After all, Barry could have been right when he claimed that paradoxes—and, notably, Wollheim’s paradox—cannot be solved but only evaded.58 If that was the case, the two-level solution would have at least provided a better understanding of the paradox. E. Pure instrumentalism. Let us turn back to pure democratic instrumentalism, which, as we have seen in section 4.2, consists in the following two tenets. Political instrumentalism.—Political institutions have legitimate authority iﬀ they produce outcomes C better than any feasible political alternative, being C a set of substantive goals independent from and prior to the procedures used to bring them about, and Democratic instrumentalism.—Under present circumstances, constitutional democracies produce outcomes C better than any feasible political alternative. Now, assume that the two-level solution is right and that Wollheim’s paradox can be solved (or better explained if you wish) by making a distinction between (the belief in) substantively correct and procedurally legitimate policies. Then pure instrumentalism turns out to be unable to address the paradox because it reduces legitimate authority to the achievement of a set of procedure-independent goals C —no matter whether by democratic means or not, for democracy is valued only contingently according to its ability to bring about precisely such outcomes. In short, it collapses procedural legitimacy into substantive correctness. Consider it from the third-person standpoint. Assume that (C1) X embraces pure instrumentalism, that is, she values democratic procedures only insofar as they produce a set of outcomes better than any feasible political alternative And further assume that (C2) X is truly committed to (A1), that is, she is completely convinced that A is the best outcome that a decision-making procedure can bring about all things considered. It thus follows that (C3) X has no reason whatsoever to be committed to (A4).

In short, X falls under the Hobbesian solution due to her embracing pure instrumentalism. There is no need to repeat the reasons why that solution is ﬂawed, reasons that also apply to the current case. However, consider three possible objections and
“Indeed, one might make the observation that if an apparent contradiction can be shown not to be a contradiction at all it is not a paradox in the ﬁrst place. A paradox is by deﬁnition something to be evaded.” Barry, “Wollheim’s Paradox: Comment,” p. 318.
58

138

Chapter 4. Instrumentalist theories

how they can be replied by pointing out at the fact of disagreement not only as a more or less pervasive and resilient political feature but as a necessary condition for political decision-making procedures being called for. Firstly, it can be objected that proponents of democratic instrumentalism are not only instrumentalists—since they evaluate political institutions only according to the outcomes they bring about—, but also democrats—since they consider that, under present circumstances, democratic institutions tend to satisfy this normative criterion better than any other feasible alternative. True, this is the reason why a proponent of pure instrumentalism could relax her absolute commitment to (A1) and submit to the majority will, thus turning to believe that B ought to be enacted. A plausible reason for doing so is that X might admit her own moral fallibilism. As Arneson has claimed, “The instrumentalist as I conceive her is a realist about morality but can and should be a fallibilist about our present moral knowledge. There is moral truth, but our current epistemic access to it is uncertain, shaky.” 59 However, this case does not apply when X is completely convinced that A is the best possible outcome. In those cases, (C4) applies, that is, there are no reasons for X not to give up her democratic commitments and to impose the enactment of A if she happens to have the power to do it.60 Now, this is an implication that need not trouble pure instrumentalists at all, for they only embrace democracy conditionally and, therefore, they are happy to give it up under certain circumstances, such as the ones stipulated here. In other words, Wollheim’s paradox might be troubling for those who are whole-heartedly committed to (A2), but not for those who see their commitment to (A2) as merely contingent and are certain of (A1). However, the paradox cannot be dismissed quite so easily, for pure instrumentalism is dependent on the possibility of outcomes C being speciﬁed and it is a feature of political communities that their members disagree willy-nilly on which outcomes ought to be pursued and which policies are to be enacted in order to achieve them, as the proponents of actual representation denounced (see § 1.3.2). As we shall see below in section 4.3.2, pervasive political disagreements need not be unreasonable or ill-intentioned. On the contrary, they are often the natural consequence of the slippery and contestable nature of political issues, of citizens’ imperfect ability to get to know how to address such issues with certainty and of the opposing but legitimate interests that pervade day-to-day political deliberation and bargaining.61 Premise (A2) cannot be so easily dismissed precisely because it provides an answer to such disagreements in the form of a collective decision-making procedure, thus expressing citizens’ acknowledgment of the existence and pervasiveness of such disagreements.
Arneson, “Democracy Is Not Intrinsically Just,” p. 43. “If you happen to have the power to implement [the] correct assessment, you should do so, despite the fact that your assessment will not attract the unanimous assent of those aﬀected.” Arneson, “Democracy Is Not Intrinsically Just,” p. 54. 61 Of course, political disagreements are often unreasonable and ill-intentioned—say, I may want you to be kicked oﬀ my country by the government because America is for Americans and you are a damn immigrant, while you may want me to be expropriated by the government because I own a ﬁrm and surplus extraction equals exploitation. However, democracy is made for unreasonable citizens as much as for reasonable ones. Further, that citizens make unreasonable claims does not mean that they do not take some procedures to be more legitimate than others.
60 59

4.3. Criticisms

139

Secondly, it can be further objected that the fact of disagreement need not pervade all political decisions, for it is possible to specify a number of political goals that are beyond (or at least less aﬀected by) the fact of disagreement and which can serve as a rather easily computable set of proxies for the evaluation of the output of political institutions. Further, these goals need not be vague and general, as part of a sort of incompletely theorized agreement.62 Indeed, they are often quite precise. Consider, among probably many others, the rates of child mortality, life expectancy, literacy, security of personal property, unemployment, gender equality, inﬂation, corruption, violent deaths, imprisoned journalists or economic growth. However, as we shall see in detail below (§ 4.3.2), even if a consensus existed or could be reached on these and other political goals, disagreements are way more far-reaching and pervasive, thus persisting at three diﬀerent levels at least. First, when having to balance the weight of basic goals and rights so as to solve likely trade-oﬀs among them. Second, when having to decide the means by which those principles and rights are to be enacted. And, third, when having to enforce them to particular cases and unforeseen contingencies.63 Disagreements on outcomes C pervade these three levels, and citizens’ commitment to (A2) is precisely the way they acknowledge their pervasiveness and the need for collective decision-making procedures so as to address them. By dismissing X’s commitment to (A2), pure instrumentalism overlooks a central feature of politics—i.e., the fact of disagreement—and turns, therefore, into an incomplete normative account of political authority.64 Thirdly, it might be true that citizens disagree on these issues and that many of them are whole-heartedly committed to (A2) and therefore to (A4) as a consequence of their acknowledgment of the fact of disagreement. However, it can be further replied that democratic instrumentalism is not a psychological account of citizens’ current political behavior. Neither it is an empirical account of political authority. Rather, it is a normative theory of political authority and it is therefore independent both from the third-person standpoint and from the actual functioning of current political communities. After all, the charm of political candidates or the
62 See C. Sunstein, “Incompletely Theorized Agreements,” Harvard Law Review 108 (1995): 1733-1772. Even though this sort of agreements are not analyzed here, it has to be noted that they are not without problems. First, constitutional provisions that are very abstract increase the power of constitutional courts, for they provide them with more margin for constitutional interpretation. This is what Gargarella has termed as the “interpretive gap” and this is the reason why Kelsen always defended very detailed constitutions, in order to reduce as much as possible the discretion of the constitutional court. For example, is not the same to provide constitutional judges with the ability to strike down statutes that may may imply “cruel and unusual punishment,” as the US Constitution does, than to provide them with the power to turn down statutes enacting capital punishment. See J. J., Moreso, Constitución: Modelo para armar (Madrid: Marcial Pons, 2009), p. 143. Second, it is very questionable whether we can say that these agreements are actual agreements for, as Rescher has put it, “’agreement’ only exists where a question on the agenda of relevant questions is answered by both of us in the same way.” N. Rescher, Pluralism. Against the Demand for Consensus (Oxofrd: Oxford University Press, 1993), pp. 44-45. 63 See Bellamy, Political Constitutionalism, ch. 1. 64 Indeed, empirical evidence shows that citizens normatively assess political institutions not only according to their ability to produce their preferred outcomes but also according to the procedures used to bring shuch outcomes about. See section 4.3.3 for a cover of the social psychology literature on the topic.

140

Chapter 4. Instrumentalist theories

tunes of their campaigns may well aﬀect citizens’ political behavior, as Schumpeter cynically suggested. But it does not follow that such factors ought to be included in a normative theory of the legitimacy of political decisions. However, the fact of disagreement is not a mere psychological feature that can be left aside when building up a normative theory of political authority. Neither it is a contingent—though persistent and widespread—feature of political communities. Rather, it is a necessary condition of politics proper, for if no disagreement existed no political decision-making procedure—democratic or otherwise—would be required at all (this is why pure instrumentalist turns out to be self-defeating, as we shall see below in section 4.3.2, epigraph E). As Bobbio puts it, “What is democracy other than a set of rules (the so-called rules of the game) for the solution of conﬂicts without bloodshed?” 65 The relevance of Wollheim’s paradox for the debate between procedural and instrumental normative theories of democratic authority is that it makes crystal-clear the fact of disagreement that is the very bread and butter of politics. By dismissing X’s commitment to (A2), pure democratic instrumentalism overlooks the fact of disagreement, fails to address Wollheim’s paradox adequately and turns out to be an incomplete normative account of democratic authority. 4.3.2. Political disagreement This section elaborates on the criticism presented at the end of the previous section—namely, that pure instrumentalist theories of democratic legitimacy turn out to be self-defeating because they dismiss the existing disagreements on which outcomes are to be achieved through democratic procedures. It is divided into ﬁve epigraphs. The ﬁrst four ones deﬁne the circumstances or politics, along with the scope, causes and implications of the fact of disagreement. The last epigraph explains why pure instrumentalism turns out to be self-defeating under the circumstances of politics. A. The circumstances of politics. As we have seen above, according to pure democratic instrumentalism, democratic institutions and procedures have legitimate authority if and only if they bring about a set of substantive goals independent from and prior to the procedures used to produce them. Now, the problem with this account of legitimacy is that it dismisses the so-called circumstances of politics which, just as Rawlsian circumstances of justice, are two.66 First, the need for a binding collective agreement on a uniﬁed course of action due to the fact that all aﬀected parties would be better-oﬀ than if no agreement was reached. And, second, the existence of a persistent disagreement on which precise course of action ought to be taken (i.e. the fact of disagreement). To be sure, for a large number of matters disagreement need not be an issue, for unilateral decision is possible and no uniﬁed collective course of action is required. In these cases, each of us can do her own thing without having to reach an agreement
N. Bobbio, The Future of Democracy (Cambridge: Polity Press, 1987), p. 193. Waldron, Law and Disagreement, pp. 101-106; Andrew Weale, Democracy (Basingstoke: Macmillan, 1999), pp. 8-13; and Bellamy, Political Constitutionalism, pp. 20-26.
66 65

4.3. Criticisms

141

with other people on what ought to be collectively done. And the other way around, for a large number of matters coordinated action is to be reached, but disagreement need not arise. Now, we say that an issue is political when both conditions apply, that is, when a uniﬁed collective course of action is required so as to improve the situation of the parties and, at the same time, no agreement is likely to be reached on which precise course of action ought to be taken. Obviously, the fact of disagreement need not imply political paralysis, as political decisions taken everyday proves. But far from being the upshot of a wide consensus on what ought to be done, political decisions are taken, and have always been taken, under lively and often harsh disagreements among aﬀected parties. Under a well functioning democratic system such decisions are the upshot of the majority will expressed in the form of a parliamentary majority while, under a nondemocratic political system they are the upshot of the minority will. In neither case do they result from a consensus reached among the parties, as liberals do sometimes assume, but from a disagreement among them that is likely to persist after the decision has been taken. As Waldron puts it, The prohibition of child labour, the reform of the criminal process, the limitation of working hours, the dismantling of segregation, the institution of health and safety regulations in factories, the liberation of women—each of these achievements was secured in what I have called the circumstances of politics, rather than in anything remotely resembling the justice-consensus that Rawlsians regard as essential to a well-ordered society.67 B. The causes of disagreement. Consider now the causes behind the fact of disagreement as for two diﬀerent dimensions of political decision-making, namely, citizen’s imperfect ability to get to know which political course of action will produce better outcomes, the slippery and contestable nature of political decisions themselves and the interests that unavoidably pervade political decision-making. The ﬁrst cause is citizens’ imperfect ability to get to know what ought to be done when political decisions are to be made. This need not imply that good political decisions (or at least better ones) do not exist or that reliable knowledge about them ought to be ruled out, but only that stable agreement about them is unlikely to emerge. Citizens’ knowledge about political issues is way more slippery and imperfect than scientiﬁc knowledge, for it is pervaded by what Rawls labeled as the burdens of judgment, namely, “the many hazards involved in the correct (and conscientious) exercise of our powers of reason and judgment in the ordinary course of political life.” 68 Rawls famously distinguished between comprehensive (philosophical, moral and religious) doctrines on which consensus is unlikely to be reached, on the one hand, and principles of justice on which consensus can be expected to emerge, on the other hand, thus limiting the eﬀects of the burdens of judgment to the former. However, there is no reason why the burdens of judgment should not pervade principles of
67 68

Waldron, Law and Disagreement, p. 106. John Rawls, Political Liberalism (New York: Columbia University Press, 1993), p. 56.

142

Chapter 4. Instrumentalist theories

justice as they pervade moral, philosophical and religious doctrines. As Bellamy has argued, their existence raise doubts not only about a consensus on the good but also on the right. He takes the debate over whether free speech ought to be restricted in the case of pornography as an instance of this. It is worth quoting his analysis in extenso: First, it is notoriously diﬃcult to assess the evidence concerning the harm pornography may cause in inciting violence against women or lowering the esteem in which they are held and hold themselves. Second, even were there agreement on some of these facts, not everyone would weigh them in the same way because of the other four factors. Third, the concept of ‘speech’ has proved infamously slippery and vague, producing long discussions over the degree to which a performance or ﬁgurative art count as ‘speech’ and so on. This vagueness is not just superﬁcial nit-picking about the ordinary usage of words, but can extend to the core normative assumptions involved in certain key terms – for example, over whether ‘speech’ suggests some form of reasoned argumentation and what that entails. Disputes at this level can render terms ‘essentially contestable’, with a number of rival conceptions of a concept in play, each of which will highlight a diﬀerent (if often overlapping) sphere of applicability for any right. Fourth, diﬀerent life experiences are no doubt an important factor in explaining why most women – even those who do not agree with a ban – feel more strongly about, or have greater understanding of the feelings raised by, this issue than most men. Fifth, this is a matter where focusing on diﬀerent types of moral claim can produce diﬀerent assessments and make an ‘overall’, ‘on balance’ judgment hard to make. An assessment that looks at the consequences for social attitudes towards women is likely to be more sympathetic to restrictions than one focused narrowly on the deontological claims of particular parties. After all, speech is restricted in the case of tobacco advertising in part because widespread smoking is regarded as socially as well as personally harmful. Glamorising smoking is also seen as encouraging young people to take it up. Sexually explicit advertising on TV is limited for partly similar reasons related to public morality and its eﬀects on children. Finally, the issue potentially raises a clash of rights of either a logical or contingent character. In such cases, we are forced to compare incommensurable goods and whichever decision we take entails some loss. We may decide, say, that the risk to a woman’s right not to be discriminated against is less than the problems and possible consequences of restricting freedom of speech. But the comparison of the values involved is always going to be contentious because there is no metric for saying the one is more important than the other. After all, some would counter even a small risk of gender discrimination is worse than a greater than necessary curtailment of free speech.69
69

Bellamy, Political Constitutionalism, pp. 22-23.

4.3. Criticisms

143

It is to be noted that democratic instrumentalists do not dismiss the importance of moral fallibilism about political issues. According to Arneson, “The instrumentalist as I conceive her is a realist about morality but can and should be a fallibilist about our present moral knowledge. There is moral truth, but our current epistemic access to it is uncertain, shaky.” 70 Furthermore, Arneson makes clear that moral fallibilism need not trump democratic instrumentalism at all. True, it might be the case that our knowledge about which political goals are to be pursued is incomplete, thus making unlikely for an epistemic consensus to emerge. However, unless we deny the very idea of political correctness, this fallibilism need not trump the idea that political institutions should be designed so as to maximize the achievement of such goals. As Arneson puts it, the position of the instrumentalist resembles that of a treasure-seeker. “We are searching for genuine treasure, and our practices should be assessed instrumentally, by the degree to which they enable us to gain treasure. Our current maps guiding us to treasure are ﬂawed, and our current ideas about what ‘treasure’ is are somewhat crude, and we have reason to believe there are better maps to be located and better conceptions of ‘treasure’ to be elaborated.” 71 However, Arneson’s analogy fails to acknowledge the second cause behind the fact of disagreement—namely, the slippery and contestable nature of political decisions. It is not only the case that our political knowledge about what ought to be done is fuzzy, but also that political issues themselves are vague and contestable. The analogy between seeking a treasure and political decision-making does not hold because the former is mainly a technical matter of developing and applying a set of skills in order to reach a goal settled in advance (even if we accept that our idea of what “treasure” is might be somewhat crude), while politics is not only about making technical decisions about which are the best means to achieve certain goal settled at the outset, but also about the very goals that are to be achieved. And, even though good (or at least better) political goals existed and citizens’ approach to political issues could be undistorted by the burdens of judgment, the issue of which goals ought to be pursued is slippery and contestable by deﬁnition, notably, when we get down from general and abstract principles to actual policies and unforeseeable contingencies (as it will be analyzed in some detail in the next epigraph on the scope of the fact of disagreement). Finally, in addition to the slippery and contestable nature of the political issues and citizens’ imperfect ability to get to know how to address them with certainty, particular interests are to be taken into account as well when addressing the existence and pervasiveness of the fact of disagreement. Hence, opposing interests in the political arena, and not only the vagueness of political issues and the falibilism in getting to know them, make stable and uncontested consensus among citizens unlikely to emerge. It is certainly the case that political interests can turn disagreeing parties into unreasonable. But this need not be always the case. Consider, for example,
70 Arneson, “Democracy Is Not Intrinsically Just,” p. 43. This raises the question of whether the diﬀerence between them and pure proceduralists, such as Waldron and Bellamy, could be only theoretical. 71 Arneson, “Democracy Is Not Intrinsically Just,” pp. 43-44.

144

Chapter 4. Instrumentalist theories

the case of multinational countries such as Belgium, Canada or Spain in which it is usual to see MPs of the nationalist parties openly refer to the interests of their constituencies, which do often represent a large portion of the population. True, when the parties at stake move toward radicalism and their interests become harsh and purely selﬁsh, political paralysis can follow, as Belgians know well. However, under normal conditions we tend to assume that these MPs are just advancing policies that combine—in a diﬃcultly distinguishable way, not least because of the unavoidable opacity of their intentions—the interests of their constituencies with what they take as the general interest, defending them with what they take to be good reasons and being open to negotiation and deliberation in the parliament and elsewhere. The policies that they put forward might be completely mistaken, but we would be way to fast to condemn them as necessarily unreasonable for that reason. C. The scope of disagreement. It might be argued that the fact of disagreement does not (or at least need not) pervade all political decisions. This view is behind the persistent claim of many citizens who blame their representatives for not being able to overcome the horse-trading and buck-passing typical of day-to-day politics and to reach a much-needed consensus on fundamental political issues, such as civil liberties, education, health care reform or national defense. Since, on the other hand, virtually everyone seems to agree on a set of basic rights and political goals, a minimal agreement appears to be possible at some level of generality. As Rawls claimed, agreement on constitutional essentials is not only possible to emerge but should actually be entrenched so as to insulate it from the short-term dynamics of electoral politics. Bellamy has criticized this view by pointing out that it is very questionable whether libertarians, conservatives, social democrats and communists, among other ideologically segmented citizens, would reach an agreement (in Rawlsian terms, an overlapping consensus) even at the very level of principles and rights.72 However, I believe that it is possible to list a number of political goals which are beyond the fact of disagreement and which serve as a rather easily computable set of indicators for the evaluation of the output of political institutions, including democratic ones. As Rousseau put it, “some points on which all interests agree.” 73 Consider, among probably some other indicators, child mortality, life expectancy, illiteracy, security of personal property, unemployment, gender equality, inﬂation, corruption or the rates of violent deaths, imprisoned journalists or economic growth. Of course, I do not mean to say that these indicators are completely beyond disagreement or that they remain actually uncontested, but only that they are comparatively less pervaded by disagreement and actually less contested than other indicators, as the ever-present attempt to achieve them by governments of the most diﬀerent ideological sort proves. Now, even if a consensus could be reached on these and other political goals (which seems to actually happen, at least if we put them in comparative terms),
72 73

Bellamy, Political Constitutionalism, pp. 23-24. Rousseau, The Social Contract, p. 66.

4.3. Criticisms

145

disagreements are more far-reaching, persisting at three diﬀerent levels at least. First, when having to balance the weight of basic goals and rights so as to solve likely trade-oﬀs among them, second, when having to decide the means by which those principles and rights are to be enacted and, third, when having to enforce them to particular cases and unforeseen contingencies. Brieﬂy consider them in turn. First, moral principles and rights often give lead to incompatibilities and trade-oﬀs, such as the one between the right to privacy and the right to security when being searched before taking a plane. Further, the way such trade-oﬀs are to be addressed depends not only on the weight attached to each right but also on the way rights are conceived. When having to settle what counts as a right violation, for example, it is not the same to see rights as pre-political and absolute side constraints, as libertarians tend to do, or as mere social conventions, as utilitarians often do. As Waldron puts it, not only do people diﬀer “about what rights we have, they diﬀer also on what is being said when we are told that someone has a right to something.” 74 Second, even though there seems to be a wide consensus on the set of political goals listed above, it is much less clear by which means are such goals to be achieved. We know some rather uncontested facts provided by social scientists indeed—for example, that expansive monetary policies raise inﬂation, that investment in early education is more eﬃcient than investment in higher education, or that private provision of health care brings about a worse overall health outputs at higher cost than the public one. However, these and many other policies remain highly controversial not only for the reasons included among the causes of the fact of disagreement but also because of the fact that their achievement aﬀects other goals, raising new and often tougher trade-oﬀs. For example, libertarians and socialdemocrats will in all likelihood disagree on the appeal of a constrained monetary policy that may reduce inﬂation at the expense of raising the unemployment rate or on the desirability of investing in early education at the cost of enlarging the state intervention, let alone on the advisability of a public health care scheme. As Richard Nixon put it in his debate with Kennedy on TV, I know what it means to be poor. I know what it means to see people who are unemployed. I know Senator Kennedy feels as deeply about these problems as I do, but our disagreement is not about the goals for America but only about the means to reach those goals.75 Third, the interpretation of basic principles and rights turns out to be vague and incomplete when going down to particular cases and unforeseeable contingencies. As Sunstein has put it, even if we agree that freedom of speech is to be constitutionalized, it is by no means clear whether it is abridged by laws regulating disclosures by government employees, restrictions on pornography or laws prohibiting disclosure of confessions during a criminal trial. True, the need for interpretation varies according to the degree of detail and exhaustiveness of the constitutional text. However, since
J. Waldron, “Rights in Conﬂict,” Ethics 99 (1989): 503-519, p. 503. See http://watergate.info/nixon/60-09-26_ﬁrst-election-debate.shtml (Accessed on November 15 2011)
75 74

146

Chapter 4. Instrumentalist theories

no constitution can cover all cases—it would have to be as Borges’ map, which was as detailed as the territory that was supposed to map and, for that matter, as large as the territory itself—, interpretation is needed and discretion unavoidable. This is not a problem only due to the lack of exhaustivity and the need for the legal texts to be handable, but also due to the temporal impossibility to advance the unforeseen contingencies that might appear in the future. Further, even if such contingencies could be established ex ante, it is also desirable for our principles and rights to be ﬂexible enough to adapt to changing circumstances. In short, the fact of disagreement does not only pervade political issues at the level of fundamental principles and rights. It also pervades their application when having to balance the trade-oﬀs that are likely to emerge among them and to address particular cases and unforeseeable contingencies. D. A second-best rationale? It might be argued that the fact of disagreement only provides a second-best case against pure instrumentalism. Hence, according to this argument even though the rationale for political rights consists in their ability to bring about better outcomes, current contestability of political issues and citizens’ imperfect moral and political knowledge would trump the pure instrumentalist case. However, this involves a misunderstanding about the very idea of a second best, which implies that a ﬁrst best is possible but that, due to a constrained variable, under present circumstances due to a constrained variable.76 In this case, consensus on the precise outcomes against which political procedures are to be evaluated is the variable that turns out to be constrained due to the fact of disagreement. However, the fact of disagreement is not a mere psychological feature that can be left aside when building up a normative theory of political authority. Neither it is a contingent—though persistent and widespread—feature of political communities. Rather, it is a necessary condition of politics proper, for if no disagreement existed no political decision-making procedure—democratic or otherwise—would be required at all. As Przeworski has pointed out, democracy presupposes that “Interests or values are in conﬂict. If they were not, if interests were harmonious or values were unanimously shared, anyone’s decisions would be accepted by all, so that anyone could be a benevolent dictator.” 77 In short, rather than taking disagreement as a temporary and contingent—even if persistent—constraint of our thinking about politics, it has to be taken as a starting point. E. Why pure instrumentalism is self-defeating. By dismissing this necessary feature of politics—namely, the fact of disagreement—democratic instrumentalism turns into

Under such circumstances, trying to achieve the ﬁrst-best option can bring about worse outcomes than turning to the second-best option. Santiago Carrillo expressed this view clearly when he claimed that “one must have the courage to explain to the working class that is better to give the surplus to this sector of the bourgeoisie, rather than creating a situation that carries the risk of turning against us.” S. Carrillo, Demain l’Espagne (París: Seuil, 1974), p. 187. 77 A. Przeworski, “Self-Enforcing Democracy,” in Oxford Handbook of Political Economy, D. Wittman and B. Weingast, eds. (New York: Oxford University Press, 2006).

76

4.3. Criticisms

147

self-defeating, as Waldron has claimed. Here is why. According to pure democratic instrumentalism: (D1) Political decision-making procedures are legitimate if and only if they bring about procedure-independent outcomes C better than any feasible political alternative. From what it follows ex hypothesi that (D2) C can be speciﬁed.

However, it also happens that disagreements on which outcomes are to be reached are a necessary condition for procedures of collective decision-making being called for. Think about it a contrario. Assume that all citizens agreed on what ought to be done across the board. If that was the case, no decision-making procedure would be required, except for purely coordination matters.78 Put simply, (D3) Political decision-making procedures are required only if disagreements on which outcomes are to be reached exist. Now, if (D2) and (D3) apply, then it follows that (D4) No political decision-making procedure is required at all.

As Waldron has put it, “any theory that makes authority depend on the goodness of political outcomes is self-defeating, for it is precisely because people disagree about the goodness of outcomes that they need to set up and recognize an authority.” 79 Of course, the fact of disagreement can also pervade political decision-making procedures, thus preventing a consensus on the procedures to be used to handle substantive disagreements from emerging.80 However, the possibility of a regress of procedures should not worry us here. For present purposes is suﬃcient to show that the fact of disagreement is a necessary feature of politics proper and a necessary condition for political decision-making procedures—whether democratic or not—being called for, thus preventing such procedures from being assessed according to their ability to achieve a clear-cut set of non-relational outcomes C. As David Enoch has pointed out, we will have to deal about disagreements about outcomes—not less than about procedures indeed—“by doing politics: Non-ideal, messy, roughly democratic, not remotely egalitarian, somewhat reason-supported, somewhat forcerelated, every-day politics.” 81
Kelsen referred to “the ideal of solidarity of interests of all members of the collectivity” as a “metapolitical illusion,” expressing the idea that disagreement and conﬂict are necessary features of politics. H. Kelsen, Vom Wesen und Wert der Demokratie (Mohr: Tübingen 1920). H. Kelsen, La Démocratie. Sa Nature-Sa Valeur (Paris: Economica, 1988 [1929]), pp. 32-33. 79 Waldron, Law and Disagreement, p. 253. 80 The point is made, among others, by T. Christiano, “Waldron on Law and Disagreement,” Law & Philosophy 19 (2000): 513-543. 81 D. Enoch, “Taking Disagreement Seriously: On Jeremy Waldron’s Law and Disagreement,” Israel Law Review 39 (2006): 22-35.
78

148 4.3.3. Procedural minds

Chapter 4. Instrumentalist theories

As we have seen above, legitimacy is a central normative feature of political institutions. A feature that, according to political instrumentalism, depends exclusively on the outcomes that political institutions produce, regardless of the fairness of the procedures used to bring about such outcomes. To be sure, outcomes count. Hence, we observe and support political decisions because, apart from being afraid of being punished if we did otherwise, we do believe them to be adequate means for the provision of security, welfare and freedom in the form of goods and services. Likewise, we sometimes disobey such decisions, rebelling and speaking out against the government, when we perceive them to fail in the provisions of such goods and services (certainly along other factors, such as our belief that we are not alone in our claims or that they can be eﬀective in changing the status quo). As an instance of how widespread the outcome-oriented view of legitimacy is, consider the correlation between the public support for European integration and economic growth, as shown in the following graph.

Figure 4. Public support for European integration and economic growth.82

However, pure instrumentalism turns out to be too radical an account of political legitimacy when we consider it from the point of view of actual citizens, for they generally take into account not only the outcomes of political decisions but also the fairness of the procedures employed to bring them about. Despite the exclusive focus on outcomes of instrumentalism, empirical evidence shows that procedural fairness counts as well and that, actually, procedure evaluations often exceed outcome evaluations in both citizens’ conception of democracy, their trust in political institutions and their personal well-being.83 The social psychology literature on the topic is
82 S. Hix and B. Hoyland, The Political System of the European Union (London: Palgrave Macmillan, 2011). 83 See, respectively, R. J. Dalton, D. C. Shin and W. Jou, “Understanding Democracy: Data from Unlikely Places,” Journal of Democracy 18, no. 4 (2007): 142-156; C. Carman, “The Process

4.3. Criticisms

149

extensive but appears to converge on the centrality of procedural fairness for the perceived legitimacy of the decisions made.84 In a review of the research on the topic, Tom Tyler concludes that “studies of the legitimacy of authority suggest that people decide how legitimate authorities are, and how much to defer to those authorities and to their decisions, primarily by assessing the fairness of their decision-making procedures.” 85 Recent research conﬁrms the claim. For example, a recent study of the Scottish Parliament petitioning system from 1999 to 2006 ﬁnds “clear support to the hypothesis that it is individual-level assessments of procedural fairness that have the greatest inﬂuence on trust in the political system. When individuals saw the petitioning process as being fairer, they expressed greater trust in Scottish and UK political systems ... Likewise, outcome evaluations [i.e., whether the petitioners agreed with the statement Ì consider my petition to be a success’] fail to achieve statistical signiﬁcance in either of the two equations predicting system trust.” 86 Consider now two possible replies. It might be argued that procedural legitimacy is more extended in consolidated democracies rather than in new ones (if regarded at all), where citizens are more likely to value democracy because of its actual or expected outcomes. Indeed, it is frequently argued that support for democracy in developing countries merely signals a desire for a higher standard of living. Thus, as Przeworski has reported, in post-Pinochet Chile, 59 percent of respondents expected that democracy would attenuate social inequalities, and in post-communist Eastern Europe the percentage of citizens linking democracy and social equality ranged from 61 percent in Czechoslovakia to 88 percent in Bulgaria.87 If that were the case, the ﬁndings presented above would have to be qualiﬁed, for it could happen for citizens in consolidated democracies to regard procedures as valuable in themselves only as a reiﬁcation of what previously was taken as a mere means to achieve certain non-relational goals. However, in a review of the major cross-national surveys asking common people about their understanding of democracy (including 49 countries), Russell Dalton et al. have found that, even if outcome-related conceptions are slightly more extended in non-democratic countries and new democracies (specially immediately after transitions to democracy) than in established democracies, procedural considerations are still salient. As they conclude, “Relatively few people deﬁne democracy in terms of social beneﬁts ... These results undercut claims that supporters of democracy really mean they want higher living standards and similar beneﬁts”.88
is the Reality: Perceptions of Procedural Fairness and Participatory Democracy,” Political Studies 58 (2010): 731-751; and A. Stutzer and B. S. Frey, “Political Participation and Procedural Utility: An Empirical Study,” European Journal of Political Research 45 (2006): 391-418. 84 The locus classicus is W. Laurens Walker and J. Thibaut, Procedural Justice: A Psychological Analysis (Hilsdale, NJ: L. Erlbaum Associates, 1975). For a more recent review of the literature, see T. R. Tyler, “Social Justice: Outcome and Procedure,” International Journal of Psychology 35 (2000): 117-125. 85 Tyler, “Social Justice: Outcome and Procedure,” p. 120. 86 Carman, “The Process is the Reality,” p. 743, italics in the original. 87 A. Przewroski, “Democracy, Equality, and Redistribution,” in Political Judgement. Essays in Honour of John Dunn, R. Bourke and R. Geuss, eds. (Cambridge: Cambridge University Press, 2010). 88 Dalton, Shin and Jou, “Understanding Democracy: Data from Unlikely Places,” p. 147.

150

Chapter 4. Instrumentalist theories

Second, it may also be replied that political legitimacy need not depend on how citizens perceive it, for what they take to be legitimate might be relevant for a descriptive account of legitimacy but not for a normative one. If that were the case, legitimacy could still be deﬁned in purely instrumental terms, regardless of the fact that citizens systematically take into account procedures when assessing the legitimacy of political decisions. Accordingly, the instrumentalist case for constitutional constraints would remain justiﬁed. After all, past moral acceptance of slavery did not make it any more correct, nor does current moral acceptance of eating non-human animals turn it into any more correct either. However, there is a diﬀerence between the justiﬁcation of slavery or the consumption of non-human animals and the justiﬁcation of the legitimacy of democratic institutions and decisions, for democratic institutions and decisions are intended to be justiﬁed for the citizenry (while slavery and the consumption of non-human animals have been rarely intended to be justiﬁed for slaves and non-human animals). Hence, instrumentalist theories justify constitutionalism on the better outcomes they bring about for the citizenry but, on the other hand, they purport to justify it on grounds that are totally disconnected from them. As Robert Sudgen points out, “Arneson speaks of justifying constitutional constraints on democratic decision making. But to whom are these constraints being justiﬁed?” 89

R. Sudgen, “Justiﬁed to Whom?,’’ The Idea of Democracy, D. Copp, J. Hampton and J. E. Roemer, eds. (Cambridge: Cambridge University Press, 1993), p. 152.

89

Chapter 5

Precommitment-based theories
5.1. Introduction
Precommitment-based justiﬁcations of substantive and procedural constitutional constraints are addressed in this chapter.1 These theories do not justify such limits as protecting and advancing a set of non-relational goals—as instrumentalist theories do (see chapter 4 above)—, or as the procedural conditions and preconditions for the very functioning of the democractic institutions—as procedural theories do (see chapter 6 below)—, but as the upshot of citizens’ democratic decisions themselves. Further, they do not deﬁne such constrains as the upshot of idealized democratic procedures either—as contractualist theories do—, but of actual ones.2 Assume that citizens are the only source of sovereignty and the ones who ought to decide which rights (either substantive or procedural) they have, how should these rights be balanced when trade-oﬀs arise and by which means are they to be enacted, as Waldron has claimed.3 Then we should be happy, from a democratic point of view, when they decide to commit themselves to a set of constitutional principles and provide courts with the ability to review future legislative decisions that may jeopardize such principles. Hence, from a precommitment-based standpoint, constitutional constraints need not be at odds with democracy for they can be the upshot of an actual collective decision democratically made by the people themselves. As Hayek put it, “Only a demagogue can represent as ’antidemocratic’ the limitations which long-term decisions and the general principles held by the people impose upon the power of temporary majorities.” 4 The chapter is divided into three sections. First, the structure of democratic precommitments is outlined. Second, Ackerman’s account of dualist democracy is
Relevant contributions include J. Elster, Ulysses and the Sirens (Cambridge: Cambridge University Press, 1984); S. Holmes, “Precommitment and the Paradox of Democracy,” in Constitutionalism and Democracy, J. Elster and R. Slagstad, eds. (Cambridge: Cambridge University Press, 1988); S. Freeman, “Constitutional Democracy and the Legitimacy of Judicial Review,” Law & Philosophy 9 (1990): 327-370; R. Dworkin, A Bill of Rights for Britain (London: Chatto & Windus, 1990), pp. 36-37; B. Ackerman, We the People. Foundations (Cambridge, Ma.: Harvard University Press, 1991); J. Elster, Ulysses Unbound (Cambridge: Cambridge University Press, 2000); and J. Kis, “Constitutional Precommitment Revisited,” Journal of Social Philosophy 40 (2009): 570-594. 2 See typically J. Rawls, A Theory of Justice (Cambridge, Ma.: Cambridge University Press, 1971) or Jürgen Habermas, Between Facts and Norms (Cambridge, Ma.: MIT Press, 1996). 3 See J. Waldron, “Participation: The Right of Rights,” Proceedings of the Aristotelian Society 98 (1998): 307-337 and, with modiﬁcations, in Law and Disagreement, ch. 11. 4 F. Hayek, The Constitution of Liberty, pp. 106-107. Quoted in Ely, Democracy and Distrust, p. 63.
1

152

Chapter 5. Precommitment-based theories

presented as the most elaborated and inﬂuential instance of a precommitment-based approach. Finally, ﬁve diﬃculties of this approach are analyzed in turn.

5.2. The structure of democratic precommitment
A precommitment is a device through which an individual or collective agent binds her/itself in order to follow (or to increase the likelihood of following) certain course of action in the future. The agent can do so by manipulating the structure of future options open to her, either by increasing the costs of choosing alternative options to the preferred one or, in the limit, by ruling out the very possibility of taking such alternative options, the classic example being that of Ulysses tying himself so as to be unable to swing toward the sirens later on.5 In order to clarify this device regarding constitutionalism, let us analyze, ﬁrst, the distinction between personal commitments and institutional commitments, second, the way in which constitutional constraints might be viewed as democratic commitments of the second sort and, third, the reasons why citizens would decide to commit themselves to a set of constraining rules that constrain their future decisions. 5.2.1. Commitments Commitments can be institutional or personal depending on whether the agent commits herself to following certain rules or just certain course of action, respectively.6 As instances of personal commitment consider, besides Ulysses’ case, José María Aznar’s public statement that he would not run for a third term or Hernán Cortés’ burning up his ships as soon as he arrived to America. In the former case, Aznar did not exclude the possibility of running for a third term but only increased the cost of running for a third term, making more likely for him to do otherwise. In the latter, by contrast, Cortés ruled out the very possibility of his crew going back to Spain. However, both actions are instances of personal rather than institutional commitments for neither Aznar nor Cortes committed themselves to rules. Institutional commitments, by contrast, are those in which the agents commit themselves to a set of rules that constrain their future decisions.7 As a historical instance of a plausible institutional commitment, consider North and Weingast’s account of the Glorious Revolution. According to them, the key-factor of early modern England’s economic success rests on the constitutional reforms enacted during the Glorious Revolution. True, these reforms where forced, often violently, on the Crown rather than self-adopted. But, from the point of view of England as a whole, they
See Elster, Ulysses and the Sirens; and Ulysses Unbound. Elster also distinguishes between two types of commitments—intentional and functional—depending on whether they are intended or untinded yet act as if they were so. See Elster, Ulysses Unbound. 7 Another diﬀerence between personal and institutional commitments is that the former can be irreversible (as in Hernán Cortes’ case, in which there is no turning back), whereas the latter cannot for, as Elster points out, there is not an outside of the society—i.e., rules, including absolutely entrenched rules, are always changeable. See Sánchez-Cuenca, Más democracia, menos liberalismo, p. 147.
6 5

5.2. The structure of democratic precommitment

153

can be seen as commitments that successfully constrained the future power of both the Crown and the parliament. Actually, attempts to constrain the absolute power of the Crown were as old as the Magna Carta of 1215. However, those attempts had been largely unsuccessful. By contrast, after 1688, and with the regicide of 1649 contributing to the credibility of the threats posed by the opposition, the Crown was forced to observe its commitments and thus lose many of its legislative and judicial prerogatives, along with its ability to alter rules—notably, regarding taxation and budgeting—without the consent of the parliament. As North and Weingast famously put it, What established the government’s commitment to honoring its agreements—notably the promise not to appropriate wealth or repudiate debt—was that the wealth holders gained a say in each of these decisions through their representatives in Parliament. This meant that only if such changes were in their own interests would they be made. Increasing the number of veto players implied that a larger set of constituencies could protect themselves against political assault, thus markedly reducing the circumstances under which opportunistic behavior by the government could take place.8 5.2.2. Constitutions Constitutions can be deﬁned as democratic institutional precommitments if and only if two conditions are met. First, they can be deﬁned as institutional commitments as long as they take the form of a set of rules that constrain the structure of future options open to parliaments, that is, as long as they are enforceable against ordinary legislation through their being rigid and enacted by third parties such as constitutional courts. Constitutions can do this by making it harder for some options to be chosen (through the requirement of procedures more stringent than ordinary ones) or by removing them altogether (through absolute entrenchment). Second, constitutions can be deﬁned as democratic commitments if and only if the following two conditions are met: Demos condition.—A constitution can be deﬁned as a democratic commitment only if it is adopted by those citizens against whom the constitution is to be enforced later on and only by them (i.e., without the participation of external agents against whom the constitution is not to be enforced). and Procedure condition.—A constitution can be deﬁned as a democratic commitment only if it is adopted through the use of equal voting and (perhaps qualiﬁed) majority rule.9
North and Weingast, “Constitutions and Commitment,” p. 831. Social choice theorists have shown that majority rule is the only rule that satisﬁes basic conditions of fairness, equality and rationality. See A. Sen, Collective Choice and Social welfare (San Francisco: Holden-Day, 1970), 71-74; K. O. May, “A Set of Independent Necessary and Suﬃcient Conditions for Simple Majority Decision,” Econometrica 20 (1952): 680-684.
9 8

154

Chapter 5. Precommitment-based theories

Once these two conditions are met, constitutions can be counted as democratic institutional commitments made by the citizenry; and judicial review, as the obvious institutional device through which such commitments are enacted against the logrolling and horse-trading of temporary parliamentary majorities. As Freeman puts it, So conceived, judicial review is a kind of rational and shared precommitment among free and equal sovereign citizens at the level of constitutional choice. By the exercise of their rights of equal participation they agree to a safeguard that prevents them, in the future exercise of their equal political rights, from later changing their minds and deviating from their agreement and commitment to a just constitution ... By granting to a non-legislative body that is not electorally accountable the power to review democratically enacted legislation, citizens provide themselves with a means for protecting their sovereignty and independence from the unreasonable exercise of their political rights in legislative processes.10 5.2.3. Reasons Why would a people decide voluntarily to bind themselves, thus constraining their future ability to do as they please? Before turning to plausible reasons, an important caveat is in order. Even if all the below-presented reasons for constitutional precommitment were ﬂawed, a majority of citizens could still believe them to be right and therefore give support to self-binding themselves on such grounds. It is important to note that, from the point of view of what counts as a democratic precommitment, that would be a form of precommitment for they would be binding themselves, and it would be a democratic one for it would be backed by a majority. In short, even though there are good reasons for why the people would be interested in committing themselves to a rigid constitution, what has to be analyzed from the point of view of precommitment-based theories is not such reasons, but rather the structure of the commitment itself and its democratic pedigree. That said, consider now six plausible instances of such reasons.11 First, citizens could decide to withdraw some issues from ordinary politics because of a belief that such issues are extrapolitical and should not be aﬀected by what temporary and ever-changing majorities might think about them. Fundamental rights such as bodily integrity, protection against torture or freedom of conscience are plausible candidates. Second, though related to the ﬁrst reason, citizens may decide to constitutionally entrench a set of procedural or substantive provisions in the belief that doing so might improve the quality of the upshot of the decision-making process, as we have seen in chapter 4. At the individual level, this happens when people constrain their future behavior in order to overcome their own tendency to make unwise decisions
Freeman, “Constitutional Democracy and the Legitimacy of Judicial Review,” p. 353. On these and additional reasons for constitutional precommitment, see C. Sunstein, Designing Democracy. What Constitutions Do (New York: Oxford University Press, 2001), pp. 96-101; and Holmes, “Precommitment and the Paradox of Democracy.”
11 10

5.2. The structure of democratic precommitment

155

due to temporary passions, biases or interests—e.g., Ulysses tying himself in order to avoid being tempted to swing toward the sirens. Likewise, at the collective level, citizens could decide to tie themselves through a constitution to overcome their own tendency to make unwise decisions due to short-termism, weakness of the will or hyperbolic discounting. By doing so they can prevent primary bads from happening, ensure the rules of the democratic game as to make possible the alternation in oﬃce without bloodshed or reduce the likelihood of electoral cycles happening (at least, as for fundamental issues). Third, citizens may decide to entrench a set of procedural provisions in a constitutional text not because of the outcomes they might help bring about but rather because they believe them to be necessary for the very functioning of democratic institutions, as we shall see in chapter 6. Plausible candidates include the right to vote or freedom of speech, though they might included a longer list of rights, so as to prevent minority or marginal groups from being excluded from the political process. For example, the South African Constitution ensures minimum economic conditions on the belief that ordinary politics cannot be trusted to protect those who live on the margins and are likely to suﬀer social exclusion with the risk of not being able to take part in politics in a meaningful way. Fourth, the following reason could be added to the third one. Under conﬂicts of interests between citizens and representatives and imperfect information and limited civic engagement on the side of the former during ordinary politics, citizens may want to entrench the most fundamental issues (or for issues that might be easily manipulable and open to opportunistic behavior) in a rigid constitution during extraordinary moments of constitutional transition, when citizens are specially engaged in politics and deliberation is widespread and robust. This is one of the main arguments put forward by Ackerman in We the People, as we shall see immediately below. Fifth, in societies that are deeply divided on some particular issues, it may be perfectly reasonable for citizens to decide to take such issues oﬀ the political agenda so as to ensure civil peace and the stability of the democratic system proper. For example, in religiously divided society the constitution can have the function of banning the legislatures from legislating on religious matters, thus taking oﬀ the day-to-day political agenda a potentially explosive question like this, improving the stability of the system overall and allowing the society to reach agreements on less polarized issues. Sixth, a constitution can improve coordination and bargaining among diverse and often struggling parties by including provisions that are aimed at solving collective action problems, which can take the form of prisoners’ dilemmas, hold-up problems, unbearable transaction costs or the inability to make credible commitments. Consider only a prisoner’s dilemma sort of problem. According to Sunstein, “The leading example is the Full Faith and Credit Clause, which requires each state to enforce judgments rendered in other states. Without this provision, every state might have a strong incentive to refuse to enforce the judgments of other states; if Massachusetts chooses not to honor the judgment of a New York court against a Massachusetts cit-

156

Chapter 5. Precommitment-based theories

izen, then Massachusetts receives a short-term gain because the resources its citizen needs to satisfy any judgment remain within the borders of Massachusetts. But all states would be better oﬀ if the law bound each of them to respect the judgments of others. The Full Faith and Credit Clause ensures precisely this outcome, eﬀectively solving the problem.” 12

5.3. Ackerman’s dualist democracy
Ackerman has put forward a very inﬂuential precommitment-based account of constitutional democracy, which is now considered in some detail.13 In his account, a constitutional democracy—or, as he puts it, a “dualist democracy”—leaves room both for moments of constitutional politics, when citizens deeply engage in politics and commit to a set of constitutional provisions after considered reﬂection, and for ordinary politics, when engagement is lower but temporary majorities are unable to jeopardize citizens’ considered judgments because they have previously entrenched them in a rigid constitution against such possibility. A. Dualism. Dualist democracy is a third way between “monistic democracy” and “rights foundationalism.” In a monistic democracy the democratic process has normative precedence over any sort of constitutional check, including those checks that are necessary for democracy to operate. According to Ackerman, supporters of this system take the Westiminster parliamentarianism as a reference and include, among others, Woodrow Wilson, Charles Beard and Oliver W. Holmes. Rights foundationalism, in turn, equals rights instrumentalism as deﬁned in section 4. It depicts rights as conceptually diﬀerent from and normatively prior to the democratic process and is typically represented by the German constitutional system, in which fundamental rights are entrenched once and for all (to the extent that they are only amendable through the replacement of the entire constitutional text). It includes, among its supporters, legal scholars such as Ronald Dworkin, Richard Epstein and Owen Fiss. Dualist democracy contrasts with these two ends as follows.14 It contrasts with rights foundationalism because it does note deﬁne rights as pre-political and non-relational goals to which the institutional framework has to adapt. Rather, it deﬁnes them in a strictly relational way—namely, as the upshot of citizens’ considered democratic deliberation and consensus. Thus, in a dualist democracy rights are not deﬁned
Sunstein, Desingning Democracy, pp. 99-100. Ackerman, We the People: Foundations; and We the People: Transformations (Cambridge, Mass.: Harvard University Press, 1998). 14 Ackerman’s aim in We the People is not only descriptive but also normative. It is descriptive because he believes that the model of dualist democracy better describes the American constitutional system (in contrast to the above-mentioned British and German systems, which would be better characterized by monism and rights foundationalism, respectively). Hence, unlike British parliamentarianism (Ackerma’s book is previous to the enactment of the Human Rights Rights Act), the American system does have a written constitution with normative priority over the ordinary legislative process but, unlike the German Basic Law, the constitution can be amended when the people consider it to be necessary, which has happened, according to Ackerman, in three key moments, namely, the Founding , the Reconstruction and the New Deal. But Ackerman’s aim is also normative for he puts forward an idea of constitutional democracy that he believes to be superior to both monistic democracy and rights foundationalism.
13 12

5.3. Ackerman’s dualist democracy

157

as previous to the democratic process, but as the outcome of that process.15 On the other hand, it contrasts with monistic democracy because it does not deﬁne rights as the upshot of whatever democratic process, regardless of citizens’ degree of engagement, deliberation and support. By contrast, it deﬁnes them as the upshot of a mobilized deliberation in which citizens deeply engage in politics, discuss to each other at length and reach wide consensus on fundamental issues, including rights. A dualist democracy includes a two-track lawmaking system. First, a constitutional lawmaking track, in which citizens commit, after considered reﬂection, to a set of fundamental laws—including fundamental rights—by entrenching them in a rigid constitutional text. And, second, an ordinary lawmaking track, in which elected representatives draft and enact ordinary laws with little participation on the side of citizens. Constitutional lawmaking is the upshot not only of a deeper engagement and more considered reﬂection on the side of citizens, but also of a wider consensus and commitment among them. By contrast, during normal politics plurality of competing parties is the norm, with no faction being powerful enough to advance its agenda against that consensus. That is the reason why constitutional laws entrenched during constitutional moments, either aimed at amending the previous constitutional order or to create it for the ﬁrst time, have normative precedence over ordinary laws. They prevent fundamental issues—including fundamental rights—from being jeopardized by temporary parliamentary majorities that are poorly monitored by citizens during ordinary times. Accordingly, constitutionalism is not justiﬁed as an external constrain on democratic decision-making, but rather as the democratic upshot of citizens’ decisions under certain conditions of engagement and reﬂection that are absent during ordinary politics. It overcomes the apparent trade-oﬀ between democracy and constitutional rights assumed by both monists and rights foundationalists by showing that both elements can be accommodated in a consistent framework. Within this framework constitutional rights are deﬁned as the upshot of a democratic decision made by the people in constitutional moments, and the normative priority of the constitution over ordinary legislation is deﬁned as the priority of the considered judgment of the people (democratically expressed in the constitutional moment) over the ordinary legislative process. According to Ackerman, this framework dissolves the classical quarrel between monists and rights foundationalists: It allows an important place for the foundationalist’s view of ’rights as trumps’ without violating the monist’s deeper commitment to the priority of democracy. To grasp the logic of accommodation, suppose that a rights-oriented movement took the higher lawmaking track and successfully mobilized the People to endorse one or another Bill or Rights. Given this achievement, the dualist can readily endorse the judicial invalidation of later statutes that undermine these rights, even when they concern matters, like the protection of
15 “My understanding of the Constitution does not begin with inalienable rights; it starts with the eﬀort by the American people to govern themselves. Fundamental rights have constitutional status only if they have their source in a deliberate and mobilized aﬃrmation of principle by the American people.” B. Ackerman, “Rooted Cosmopolitanism,” Ethics 104 (1994): 516-535, p. 517.

158

Chapter 5. Precommitment-based theories

personal freedom or privacy, that have nothing much to do with the integrity of the electoral process so central to monistic democrats.16 B. Private citizenship. According to Ackerman, the main rationale for a dualist democracy is that it better suits “private citizenship,” which Ackerman deﬁnes as a third way between “public citizenship” and “perfect privatism.” The former accords to the idea of citizenship in the tradition of civic humanism as updated by Hannah Arendt and more recently by some communitarians. According to this idea, the public sphere is where freedom is realized, going thus ﬁrst for citizens than the private one. In Arendt’s words, the political sphere “is the realm where freedom is a worldly reality”.17 Now, even if true that millions of citizens in democracies read the papers, talk politics to their friends, relatives and fellow workers and volunteer in unions, NGOs and political parties, it is equally true that many citizens are closer to the latter ﬁgure—that of the perfect privatist, for whom the private sphere goes ﬁrst in her life to the point that she refuses to take part in politics altogether. Of course, the perfect privatist need not be self-interested. His understanding of what ought to be done may be richer, including the pursuit of spiritual and associational ideals along with more purely selﬁsh interests. What characterizes him is that, whatever the content of her private goals, they always trump her public ones. Private citizenship lies between these two ends and is, according to Ackerman, the general disposition among citizens during normal politics. Hence, as a general rule, citizens in modern democracies are far from being the zoon politikon typical of the public citizen, since they greatly value their private lives. But they are neither perfect privatists in the sense of not caring about politics at all. Private citizens balance these two ends by way of standing up against the government and/or the constitutional order when a pressing issue requires them to do so, while giving greater importance to their private life during more normal times at the same time. According to Ackerman, modern democracies have to accomodate this fact by “learning to economize on public-regarding virtue.” 18 Dualist democracy is precisely the institutional solution to that goal. C. Economy of virtue. The point of a dualist democracy is not only that it accomodates private citizens’ apathy during normal times and their engagement in constitutional moments. Neither it is only that it honors citizens’ considered constitutional commitments during more ordinary political times. Its point is also and foremost that it minimizes the potential negative eﬀects of private citizenship during normal politics. Here is way. In the absence of constitutional constraints, temporary parliamentary majorities and powerful political lobbies would be in a position to take advantage of citizens’ apathy so as to advance their own interests and trump citizens’ fundamental rights. Consider three instances of such abuses. First, unless constitutionally constrained, “taxes may be designed so that the social groups supporting the
Ackerman, We the People, pp. 12-13. Hannah Arendt, “What Is Freedom?,” in Between Past and Future (New York: Penguin Books, 1993), p. 154. 18 Ackerman, We the People, p. 239.
17 16

5.3. Ackerman’s dualist democracy

159

Opposition are forced to pay the bulk of the revenue; beneﬁts may be distributed so that the Government supporters appropriate the lion’s share.” 19 Second, a powerful political platform can take advantage of private citizens’ apathy by enacting policies that a more participatory citizenry would not tolerate. Third, material asymmetries can dramatically distort the political process due to the fact that some citizens have larger economic and informational resources and more time available to take part in politics, as the data in table 3 below suggest. Unless adequate constraints are constitutionally entrenched, political apathy and ignorance of ordinary citizens can put democracy into risk as much as the mobilized support of a malevolent majority. In short, a dualist democracy “seeks to protect [citizens’ fundamental rights] against erosion during more normal times, when the People are less involved in aﬀairs of state.” 20
Type of activity Voting Campaign work Campaign contributions Contact Protest Community activity Board membership Political aﬃliation Income group < $15,000 52% 4% 6% 25% 3% 13% 1% 29% Income group ≥ $75,000 86% 17% 56% 50% 7% 38% 6% 73%

Table 4. Percentage of Americans active in politics: High and low income groups.21

D. Constitutional amendability. Now, even though constitutionalism may prevent private citizens’ commitments from being abused by opportunistic politicians and powerful elites and inﬂuenced by material inequalities, such commitments cannot be said to be democratic unless citizens can successfully turn their considered demands into constitutional amendments or into a completely new constitution. According to Ackerman, the transition from ordinary to constitutional politics can be divided into four main steps—namely, signaling, proposal, mobilized deliberation and codiﬁcation. Let us see them brieﬂy in turn. During the signaling stage the movement earns the authority to place the issue on the constitutional agenda, above day-to-day competition among ideological factions. The issue is successfully signaled when it gains the suﬃcient depth (it involves a high degree of information exchange and deliberation among the citizenry), its breadth (it generates a large support) and its decisiveness (it defeats the competing alternatives). As Ackerman describes it, There will be times in the future, as in the past, when one or another group of private citizens wil begin to strike a responsive chord amongst a
Ackerman, We the People, p. 256. Ackerman, We the People, p. 19. 21 The data come from S. Verba, K. L. Schlozman and H. E. Brady, Voice and Equality. Civic Voluntarianism in American Politics (Cambridge, Mass.: Harvard University Press, 1995), p. 190.
20 19

160

Chapter 5. Precommitment-based theories

larger group of fellow Americans; the weak, fragmentary, disconnected political talk around dinner table and workplace will begin to focus on a particular movement’s constitutional agenda, and will give shape to the debate in more public forums; increasingly, the transformative initiative will dominate the country’s political life; elections will be fought on the movement’s agenda, in the process transforming the public’s understanding of the nature of the issues, as well as the character of possible solutions; then, after long struggles in many diﬀerent public places, a movement may at last gain the constitutional authority to speak, once again, for We the People.22 Regarding how to identify when a demand has reached suﬃcient depth, breadth and decisiveness, voting fails to be a useful tool, for it does not distinguish proposal that are deep, wide and decisive from proposals that are not. By contrast, constitutional rigidity provides for such a device, for it rules out proposal that are not deep, wide and decisive enough by making higher lawmaking much harder than normal lawmaking. In a second stage, more or less ambiguous claims give way to increasingly concrete proposals, formulated in a language that the majority of the population can support. Of course, the institutional means used to translate a more or less chaotic bunch of claims into an accurate and concrete set of proposals varies across countries with diverse political traditions. Hence, if hierarchically organized and all-powerful parties have been the rule in Europe, America has always had less powerful and disciplinary parties (even though this might have changed from mid-twentieth century onward). A third stage of mobilized popular deliberation follows, which lapse varies but goes from proposal to ratiﬁcation. At this stage, citizens’ engagement is deep and deliberation robust and often witty indeed. As Ackerman puts it, “Not that this period of mobilized deliberation is anything like a philosophy seminar. It is much more democratic; much more energetic; much more multivocal. There will be a great deal of passion and personality, action as well as argument, drama as well as debate—the stakes are much too high to imagine otherwise.” 23 A fourth stage of formal codiﬁcation follows when the previous stage does not fail due to the successful opposition of supporters of the traditional constitutional framework and the proposal eventually gains the support of a widespread majority. Codiﬁcation can take the form of a formal constitutional amendment procedure, such as the one speciﬁed in the art. 5 of the US Constitution. But it can also take a more informal path, as it happened under Roosevelt, when he claimed a mandate from the people and the Congress enacted transformative statutes that, even though they ﬁrstly challenged previous constitutional principles, were eventually taken as equivalent to a formal amendment.24
Ackerman, We the People, pp. 263-264. Ackerman, We the People, p. 287. 24 Of course, the Supreme Court of the so-called Lochner era overturned the statutes, to which the President and the Congress replied by issuing a second wave of statutes and threating to pack the Court, as part of a turbulent and mobilized public deliberation. Finally, the Court accepted the new statutes thorugh a number of opinions—beginning with “the switch in time that saved nine” of Justice Roberts in Adkins v. Children’s Hospital 1936 decision—that created a new doctrine compatible with the New Deal.
23 22

5.4. Criticisms

161

E. Judicial review. Thus far the stages of a constitutional moment. Consider now the Supreme Court from Ackerman’s standpoint, which can be seen under a completely new light. Namely, not as questioning democracy or “the People,” but rather as questioning the popular mandate of a few hundred politicians in Washington, D. C. (or in Berlin or Cape Town, it may be added). While these folks have all won election, they have normally done so on the basis of soft ballot s by a majority of private citizens who themselves recognize that they have not given the key issues the kind of sober consideration they deserve. President and Congress normally do not have the considered support of the American people in assaulting the principles established by past successes in cosntitutional politics. If the Court is right in ﬁnding that these politician/statesmen have moved beyond their mandate, it is furthering Democracy, not frustrating it, in revealing our representatives as mere “stand-ins” for the People, whose word is not to be confused with the collective judgment of the People themsleves.25 Note that Ackerman does not claim that the Supreme Court represents the people better than the Congress. On the contrary, “During normal politics nobody represents the People in an unproblematic way—not the Court nor the President nor the Congress nor the Gallup polls.” 26 The issue is not as easy as choosing the institutional arrangements that better represent citizens’ interests, for citizens’ interests during normal politics and during constitutional moments are not the same. “In the very process of mobilized debate and decision, many minds would change, many new directions would be explored before a new constitutional solution was reached.” 27 The aim of a dualist democracy is to make sure that the people can turn their more considered judgments into constitutional provisions during constitutional moments and that such judgments are not trumped by temporary parliamentary majorities during more normal times. Hence, “during more normal times, isn’t it better for the Court to represent the absent People by forcing our elected political/statesmen to measure their statutory conclusions against the principles reached by those who have most successfully represented the People in the past?” 28

5.4. Criticisms
A number of criticisims are raised in what follows. Before proceeding, a caveat is in order though. When the precommitment-based justiﬁcation is used by pure instrumentalists, their case turns out to be inconsistent, for instrumentalists should be indiﬀerent to how non-relational rights are constitutionalized as long as they are. If the achievement of outcomes C is the only (and suﬃcient) criteria for the legitimacy
25 26 27 28

Ackerman, Ackerman, Ackerman, Ackerman,

We We We We

the the the the

People, People, People, People,

p. p. p. p.

262. 263. 263. 264.

162

Chapter 5. Precommitment-based theories

of political institutions, it turns out to be irrelevant whether those institutions have the support of a majority of citizens or not, as long as they are eﬃcient in advancing C. 5.4.1. Categorial confusion As we have seen, according to precommitment-based theories, constitutional constraints on democracy need not be undemocratic insofar as they are the upshot of a democratic decision made by the people, say, in a constitutional moment. Hence, Dworkin once argued that this alone is enough to dispose the democratic objection against constitutionalism. Since polls revealed that more than 71 percent of people believed that the British democracy would be improved by the incorporation of a Bill of Rights to the British legal system, that alone would turn such an incorporation into democratic.29 Similarly, a recent opinion poll carried out in Spain has shown that the people trust the Spanish Constitutional Court much more than political parties and politicians who, interestingly enough, were the less trusted institutions by far.30 Now, Waldron has argued that this argument rests on a categorial confusion. Political arrangements—including constitutional constraints—do not become any more democratic just because they have been chosen by democratic means, for a majority of citizens could pass blatantly undemocratic constitutional provisions democratically without such provisions turning into democratic for that reason.31 What makes political arrangements be democratic is not the means by which they are adopted, but rather their intrinsic institutional features, such as whether they are inclusive of all citizens, whether they provide them with meaningful rights to take part in the decisions, whether majority rule is used as the decision-making procedure, and so on.32 Since Dworkin has modiﬁed his early view on the issue and has granted that constitutional arrangements do not become any more democratic just because they have been adopted by democratic means, Waldron’s reductio is generally taken to have settled the debate. However, I think that Waldron’s is too fast an argument, at least if we consider the meaning of “democratic” in daily usage when ascribed to a charter of substantive—rather than procedural—rights. Waldron may be right in pointing out that what makes a decision-making procedure democratic is not the means by which that procedure was arrived at, for the people may well adopt an autocratic procedure by democratic means without that procedure turning into democratic for that reason. Thus he is right in claiming that what makes a decision-making procedure be democratic depends on its intrinsic features. However, Waldron commits a non sequitur when he extends this argument
See Dworkin, A Bill of Rights for Britain, pp. 36-37. See http://blogs.elpais.com/metroscopia/2011/09/menos-respeto-a-los-que-mandan.html (Accessed January 4, 2012) 31 The typical example of undemocratic laws passed by democratic means is the fall of the Weimar Republic in the hands of the Nazis. This particular case is critically discussed below in section 6.2. 32 This point is made by Waldron, Law and Disagreement, pp. 255-257; and Bayón, “Democracia y derechos.”
30 29

5.4. Criticisms

163

to substantive constitutional provisions, such as those that tend to be included in a charter of rights, as well. In contrast to procedural provisions, what makes substantive provisions—such as the prohibition of torture, the respect for one’s private and family life, or reproductive rights—be democratic has little to do with their intrinsic features or, to put it more precisely, has less to do with their intrinsic features than it does in the case of procedural provisions. Rather, we tend to say that substantive provisions—at least those that are completely disconnected from the concept of democracy—are democratic because they have been adopted by democratic means and enjoy the support of a majority of the citizenry. To be sure, “democracy” is an essentially contested concept and the ascription of “democratic” to substantive provisions is contestable and vague. Indeed, we often claim that substantive provisions are democratic because they apply to all citizens or because they have inclusive eﬀects (similarly as we often say that Inditex or H&M have “democratized” fashion because they have made it available to ordinary people). However, these are derivative uses of the term, which clearly refers to the intrinsic features of procedural provisions but much less so to the intrinsic features of substantive provisions. As an instance of this, consider the following statement: (E1) The Belgian electoral law is very democratic,

in which “democratic” clearly refers to the intrinsic features of the electoral law (e.g. its proportionality, the compulsory character of its voting, its use of open lists, etc.) rather than to the means by which the law was adopted. And compare it to the following one: (E2) The Belgian stem cell policy is very democratic,

which makes little sense unless “democratic” refers to the means by which the policy was adopted—rather than to the intrinsic features of the policy. Hence, in the former case the subject of which “democratic” is predicated (the Belgian electoral law) is a procedural issue and, accordingly, “democratic” refers to its intrinsic features. By contrast, in the latter case the subject is a substantive issue that is disconnected from the concept of democracy (the Belgian stem cell policy) and, accordingly, “democratic” does not refer to its intrinsic features but to the means by which the issue was arrived at or adopted. In short, Waldron’s reductio may apply to procedural provisions, such as the electoral law, by showing that such provisions do not turn into any more democratic just because they have been chosen by democratic means. But, since substantive provisions are democratic precisely by virtue of the means by which they are adopted, Waldron’s reductio does not apply to substantive provisions, such as those that are usually included in a charter of rights. Notably, it does not apply to most of the provisions included in the European Convention on Human Rights that became part of the British domestic law through the Human Rights Act of 1998. And since, incidentally, Dworkin’s case was limited to the adoption of provisions that were mostly substantive, Waldron’s reductio does not apply to his argument.

164

Chapter 5. Precommitment-based theories

Let us turn back to Ackerman now. Ackerman does not evade the mistake pointed out by Waldron, for his argument does not apply only to substantive issues. In fact, his dualist democracy may well end up turning into all but democratic. In order to see why, take the signaling stage of a constitutional moment. According to Ackerman, there are three criteria that are to be met for a political signaling to be considered constitutional proper—its depth (it has to involve a high degree of information exchange and deliberation among citizens), its breadth (it has to generate a strong support) and its decisiveness (it has to defeat the competing alternatives). Now, consider voting rights.33 Unlike the so-called rights foundamentalists, Ackerman believes that these rights are to be taken as the upshot of the democratic process, rather than as being prior to it. That is what makes his account be democratic. Now assume that a proposal for a constitutional amendment canceling these rights is put forward and, after wide and open deliberation, it gains the depth, breadth and decisiveness (along with the normative criteria of the subsequent stages, which I am not taking into account for the sake of simplicity) to be turned into higher law. In contrast to the German constitutional system, in which fundamental rights including voting rights are entrenched against constitutional amendment, in a dualist democracy this possibility remains open to be taken by legal means. Commenting on whether judges should be able to turn down amendments that have successfully gone through all the stages of a constitutional moment but are obviously repellent, Ackerman claims that “I do not believe that judges would be justiﬁed in asserting a general authority to protect the fundamental principles of dualist democracy against repudiation by the People.” 34 However, as Herzog points out, this reponse raises a dilemma for Ackerman: “We want to know if ’the People’ can do anything they like and nonetheless remain ’the People’ or if some of their actions threaten or undercut their own status.” 35 If the former, Ackerman’s account of dualist democracy turns out to be potentially self-defeating, for it can end up being all but democratic. If the latter, his dualism turns into a form of rights foundationalism and, accordingly, it looses its democratic pedigree according to Ackerman’s own standards. Conclusion: what makes voting rights (either constitutionalized or not) be a constitutive condition of democracy does not depend on what the people decide, but on how the concept of democracy is deﬁned. A democratic system that passes a law cancelling such rights would turn into undemocratic, regardless of whether the law has gained the support of the people or not. 5.4.2. Constitutional disagreements The precommitment-based account of constitutionalism has two alleged virtues. First, it presents collective commitments to constitutional rules as analogous to individual commitments to personal rules (as when we save a percentage of our monthly
33 This case is adapted from D. Herzog, “Democratic Credentials,” Ethics 104 (1994): 467-479, pp. 468-472. 34 Ackerman, We the People, p. 16n. 35 Herzog, “Democratic Credentials,” p. 469.

5.4. Criticisms

165

income or we decide to smoke only after lunch). This is a very powerful analogy, for self-constraining decisions tend to be seen not only as autonomous decisions but actually as more autonomous than unconstrained ones. As Harry Frankfurt famously claimed, freedom of the will precisely consists in being able to have second-order desires that constrain our ﬁrst-order desires and judgments.36 Mutatis mutandis, citizens’ collective commitments to constitutional rules need not be undemocratic when they are depicted as institutional precommitments, for they can be seen as more autonomous than unconstrained actions by parliamentary majorities. Second, this account does not present such constitutional commitments as the upshot of an idealized or counterfactual collective decision, but of a collective decision taken by actual citizens. Thus, Ackerman appears to leave aside the contractualist idealized methodology and provide a rationale for constitutionalism that emerges from actual politics in extraordinary moments, i.e. when actual citizens let aside their factionalism, deliberate at length and end up speaking with one uniﬁed voice on matters of fundamental concern. These two features are related to each other and, as we shall see immediately, are equally mistaken. In a nutshell, ﬁrst, collective commitments are not personal commitments writ large for, unlike individuals, modern societies contain millions of people that are divided by political disagreements that prevent them from having or reaching a shared set of common goals that they can constitutionally impose upon themselves. As Hannah Arendt put it, there is politics because “not man, but men inhabit the world.” 37 And we all disagree with one another. Second, even though a set of shared common goals could be established, that could only be done only under some idealized assumptions, that is, at the expense of the realistic methodology that provides for the democratic pedigree of constitutional commitments as the upshot of actual citizens’ decisions. Consider the ﬁrst problem in some detail. The analogy between personal and collective commitments does not hold because it assumes that, under certain conditions (namely, in constitutional moments), political societies can let aside their day-to-day political diﬀerences, deliberate open-mindedly, achieve or ﬁnd out an homogeneous will and speak with a unique voice. Hence, as Olsen has put it, “Constitutional rules and change procedures achieve legitimacy from the consent of the governed. Agreement and allegiance are based on argumentation, persuasion and compromises more than on commands and coercion.” 38 However, as we shall immediately below, unlike individuals, political communities are pervaded by persistent and deep disagreements that prevent them from having such an uniﬁed will. Of course, the analogy between individuals and political communities would still apply if individuals were assumed to have multiple personalities, for example, in the
H. G. Frankfurt, “Freedom of the Will and the Concept of a Person,” Journal of Philosophy 68 (1971): 5-20. Frankfurt’s analysis is in line with Kant’s classical one in the Groundwork. See I. Kant, Grundlegung zur Metaphysik der Sitten, in Kants gesammelte Schriften, vol. 8, Königlichen Deutschen Akademie der Wissenschaften, ed. (Berlin: Walter De Gruyter, 1902-). 37 H. Arendt, Men in Dark Times (San Diego: Harcourt Brace & Co., 1968), p. viii. 38 J. P. Olsen, “Coping with Conﬂict at Constitutional Moments,” Industrial and Corporate Change 12 (2003): 815-842, p. 816.
36

166

Chapter 5. Precommitment-based theories

non-pathological sense of thinking, feeling and acting diﬀerently in diﬀerent contexts, as Jeﬀ Sebo has recently claimed.39 As Hume put it, I cannot compare the soul more properly to any thing than to a republic or commonwealth, in which the several members are united by the reciprocal ties of government and subordination, and give rise to other persons, which propagate the same republic in the incessant changes of its parts.40 Now, even if the analogy thus seen held, it would not be enough to make the precommitment argument work. The reason why it would not is that the precommitment-based case for constitutionalism does not depend on whether political communities are analogous to individuals or not, but rather on whether they share homogeneous interests under certain circumstances—namely, in constitutional moments. This is the crucial point. If that condition fails, then the precommitment-based case for constitutionalism fails as well, regardless of whether the analogy thus deﬁned holds or not.41 To be sure, Ackerman acknowledges the fact of disagreement not only during normal politics but also during the four stages of a constitutional moment. For example, during the deliberation stage: “Not that this period of mobilized deliberation is anything like a philosophy seminar. It is much more democratic; much more energetic; much more multivocal. There will be great deal of passion and personality, action as argument, drama as well as debate—the stakes are much too high to imagine otherwise.” 42 However, he believes that the success of a constitutional moment depends precisely on the possibility of overcoming this situation of sharp disagreement and confrontation so “the contending parties [are chanelled] into an energetic exchange of public views, inviting them to address each other’s critiques as they seek to mobilize deeper and broader support from the general public.” 43 Thus, what distinguishes normal from constitutional politics is that citizens’ political ignorance and apathy give way to a higher degree of information and mobilization and, more importantly for present purposes, that political disagreements give way to constitutional consensus (or at least to a higher one, for consensus is not an all-or-nothing matter) on a set of fundamental principles that citizens support not on prudential and strategic reasons but whole-heartedly after a long and admittedly diﬃcult process of signaling, proposal, mobilized deliberation and codiﬁcation. In short, for Ackerman’s argument to work, the following condition has to apply:
See J. Sebo, The Personal is Political (PhD dissertation, NYU, 2011). D. Hume, A Treatise of Human Nature, L. A. Selby-Bigge, ed. (Oxford: Oxford University Press, 1978 [1739-1740]), Book 1, part IV, section VI, p. 261. The analogy is also made, regarding precommitment-based accounts of constitutionalism, by José Juan Moreso. See J. J. Moreso, Constitucion: Modelo para armar (Madrid: Marcial Pons, 2009), p. 138. 41 It may be argued that this condition also applies to individuals. Hence, if they turn out to be constituted of multiple personalities, as Sebo claims, and this feture prevents them from having a uniﬁed will when they make precommitments, then it may be also questionable that we can refer to individual precommitments as actual precommitments. 42 Ackerman, We the People, p. 287. 43 Ackerman, We the People, p. 287, my italics.
40 39

5.4. Criticisms

167

Consensus condition.—Wider consensus can be reached in constitutional moments than during normal politics.44 However, this condition need not apply for at least three reasons. First, if it were true that consensus is wider in constitutional moments than during normal politics, qualiﬁed majorities could be required for the ratiﬁcation of constitutional texts. As Melisa Schwartzberg has pointed out, “if ratiﬁcation requires widespread support, framers may be induced to ensure that their provisions have broader justiﬁability.” 45 However, even though constitutions require qualiﬁed majorities for their amendment (often along with additional requisites), they only require, with few exceptions, simple majorities for their ratiﬁcation.46 Now, if the very point of constitutional precommitments is to prevent ordinary parliamentary majorities from trumping the considered judgments to which the people have previously committed during constitutional moments, why should such commitments be adopted by simple (rather than qualiﬁed) majority rule, which is the very same rule that parliaments use? A plausible reason for this is that constitutional commitments, at least at the ratiﬁcation stage, may not always enjoy a wider consensus than ordinary statutes during normal times. Second, constitutional moments tend to take place in very fragile periods of institutional crisis, transition from autocratic regimes and civil unrest or even civil wars.47 For example, the American Founding came right after the so-called critical period of the American history; the Reconstruction, after the Civil War; the New Deal, after the Great Depression.48 And these are not exceptions but the general rule. It goes without saying that the imprisonment of thousands of petty debtors and revolts across the country, a bloody civil war or the worst economic disaster in the history of capitalism are not the best conditions for considered judgment and open-minded deliberation. By contrast, normal politics better provides citizens with the calm and stability necessary for the considered reﬂection and good-faith deliberation required for a real consensus, rather than a mere bargaining, to emerge.49 Third, due to their constraining and rigid nature, the issues included in a constitution tend to last long and have an immeasurable impact on the political landscape and on citizens’ life. Of course, the parties involved in a constitutional moment are aware of this. If we add the conditions of confrontation and unrest described right above to the cocktail, it is reasonable to think that, rather than giving way to wide consensus, constitutional moments will deepen disagreements and polarize
Since Ackerman adopts a deliberativist view of constitutional moments (in contrast to normal politics), he is far from alone in embracing the consensus requirement. For example, in one of locus classicus about deliberative democracy, Joshua Cohen claims that “deliberation aims to arrive at a rationally motivated consensus.” J. Cohen, “Deliberation and Democratic Legitimacy,” in Democracy and Diﬀerence. Contesting the Boundaries of the Political, S. Benhabib, ed. (Princeton: Princeton University Press, 1996), p. 23. 45 Schwartzberg, Democracy and Legal Change, p. 24. 46 See Elster, Ulyses Unbound, p. 101. 47 See Acemoglu and Robinson, Economic Origins of Dictatorship and Democracy, p. 65; and Elster, “Forces and Mechanisms in the Constitution-Making Process,” p. 394. 48 It has to be noted that Ackerman acknowledges this. See We the People, p. 306. 49 Paradoxically enough, when calm and undisturbed circumstances obtain citizens tend not to back constitutional change.
44

168

Chapter 5. Precommitment-based theories

interests. Unlike ordinary statutes, which can be easily modiﬁed, the stringency of constitutional rules turns incentives to prioritize private interests into too strong to think otherwise. If the stakes during constitutional moments are “much too high,” as Ackerman claims,50 we should expect to see a lot of manipulation, strong pressure by powerful interest groups, violent backlash from radical groups, perhaps subtle (or even open) threats of a putsch by the military, large amounts of money invested by special interests, polarization and radicalization, and so on. Accordingly, during constitutional moments citizens might well be more engaged and information be more widely shared, as Ackerman claims. But it is unlikely for deliberation to be more open-minded, for the ﬁrst-order and hardly amendable status of the constitutional upshot provide citizens with the incentive to defend their interests as strongly as they can, rather than leaving such interests aside, and deliberation can push them toward extrems, rather than toward consensus.51 As Ginsburg has put it, “Groups can invest more energy in playing for rules at the constitutional level precisely because of the presumptively higher stakes in the selection of rules. This might lead to more rather than less rent-seeking behavior at the constitutional level.” 52 True, Ackerman acknowledges that mobilized deliberation always includes, in his own words, opportunism and strategic behavior, along with grievance, outrage, extremism and hate. But he concludes that, if predominant, these pathologies of democratic deliberation can only bring about the failure of the constitutional moment. However, all these phenomena need not make the constitutional moment fail. If the support for the proposed amendment is large enough, or is backed by groups with enough power, the deliberation can end up in an equilibrium among the opposed parties that might not endorse the outcome whole-heartedly but might have enough incentives to support it. Consider the Spanish transition to democracy as an instance of this sort of equilibrium.53 Even though it has been often pictured in terms of a great consensus on some basic principles which all or a large majority of Spaniards agreed to, it can also be pictured as a multilevel equilibrium among opposed parties in a very fragile and complicated environment (among many other destabilizing factors, the military putsch attempt known as “Operation Galaxia,” the tough economic crisis that provoked the Moncloa Pacts or the intense political violence by ETA and GRAPO). Thus it can be pictured as an equilibrium among democrats and representatives of the previous regime; among Federalists and supporters of government centralization; among the

Ackerman, We the People, p. 287. See C. Sunstein “The Law of Group Polarization,” Journal of Political Philosophy 10 (2002): 175-195. 52 T. Ginsburg, “Public Choice and Constitutional Design,” in Research Handbook in Public Choice and Public Law, D. Farber and A. J. O’Connell, ed. (Cheltenham: Elgar, 2010). 53 On the Spanish transition, see J. M. Maravall, The Transition to Democracy in Spain (New York: St. Martin’s Press, 1982); J. J. Linz, Obras escogidas. Vol. 4. Democracias: quiebras, transiciones y retoss, T. J. Milley and J. R. Montero, eds. (Madrid: Centro de Estudios Políticos y Constitucionales, 2009) and Problems of Democratic Transition and Consolidation: Southern Europe, South America, and post-Communist Europe (Baltimore: Johns Hopkins University Press, 1994); and G. O’Donnell, P. C. Schmitter, L. Whitehead, Transiciones desde un gobierno autoritario, vol. 4 (Buenos aires: Paidós, 1988).
51

50

5.4. Criticisms

169

moderate right-wing Popular Alliance, the centrist Union of the Democratic Centre, the Socialist Party, and the Communist Party; and so forth. To be sure, a large majority of Spaniards—who were very mobilized indeed—agreed that the regime was moribund, that a change toward some form of democracy was inevitable and that it was preferable to make that change without bloodshed (or with as little bloodshed as possible, for political violence accompanied the transition across the board).54 But this does not imply that the eventual constitutional upshot was grounded on a consensus rather than also, and probably predominantly, on a fragile bargaining where all parties (though some more than others) had to make concessions so as to make the transition possible. For example, the interests of the de facto powers of the army and the Catholic Church were widely represented in the text.55 In short, we can present the Constitution as the upshot of a precommitment made by the Spanish society taken as a whole. But we cannot present such a commitment as resulting from a wide consensus among the individual citizens rather than as a fragile though successful equilibrium among bargaining parties that made concessions but not necessarily changed their minds or constitued an uniﬁed will. The question, then, is that it turns out to be unclear why should this solution have a special normative weight (other than for prudential reasons) over ordinary legislative politics and be entrenched in a rigid constitution against the will of future generations. Time to sum up. The analogy between individual and collective commitments does not hold regarding constitutional moments because, unlike individuals, collectivities are pervaded by the fact of disagreement. This fact is likely to persist, ﬁrst, because otherwise qualiﬁed majorities could be used for the ratiﬁcation of constitutions, second, because the circumstances under which constitutions are adopted or amended are not the better ones for citizens to leave such disagreements aside and, third, because the stakes are too high due to the immeasurable impact and hardly amendable nature of constitutions. Rather than giving way to wider consensus in which the people speak with a uniﬁed voice, constitutional moments are better depicted as bargaining processes and constitutions as equilibria. As Schwartzberg concludes in her historical study of constitutional entrenchment and amendment, “Constitution-making assemblies, far from resembling agents in an original position or ideal deliberative forum, are ﬁrst and foremost legislatures, both for good and for ill.” 56 In turn, these doubts about the wider consensus of constitutional moments raise doubts about the precision of Ackerman’s methodology, which aims to leave aside the Rawlsian idealized methodology and provide a rationale for constitutionalism that emerges from actual politics in extraordinary moments, that is, when profound
In fact, the Constitution gained an almost unanimous vote in the parliament (excluding some votes against it from the members of the right-wing Popular Aliance and the abstention of the Basque Nationalist Party) and 87,7 percent of the votes in a referendum. 55 The Constitution entrenched the freedom of the Catholic Church to create its own schools and the right to receive subsidies from the state to make them work. As for the army, it was charged with the duty to guard the sovereignty, independence and the territorial integrity of Spain. 56 Schwartzberg, Democracy and Legal Change, p. 24.
54

170

Chapter 5. Precommitment-based theories

crisis make the people let aside their factionalism, deliberate honestly and eventually speak with one uniﬁed voice. However, as we have seen, constitutional moments are far from Ackerman’s picture of them. Further, they are probably in a worse position in terms of producing a consensus among citizens. By doing so, Ackerman treats constitutional moments as somehow more ideal than normal politics.57 This raises a dilemma for Ackerman—namely, either he maintains his non-idealized methodology and drops the consensus condition or the other way around, he maintains the consensus condition by dropping his non-idealized methodology.

5.4.3. Constitutions as other-binding devices As we have seen immediately above, constitutions need not be the upshot of a wide consensus after open-minded and informed deliberation by the people. Far from Daniel Webster’s vision of “one country, one Constitution, one destiny,” 58 constitutions are often fragile equilibria among disagreeing parties that lack the power to impose their particular will and thus have to bargain with each other under circumstances of crisis and unrest. As an instance of this, consider Acemoglu and Robinson’s inﬂuential account of the historical expansion of democracy, according to which the wealthy accepted to share political power with the poor—i.e., democracy—in order to avoid revolution, since otherwise revolution was the only way the latter had to achieve wealth redistribution in a nondemocratic regime.59 Historically, a twofold worry pervaded the debates on the likely consequences of this move. First, contrary to the interests of the wealthy, the poor could use their newly gained political power to expropriate the wealthy once enfranchised.60 Second, contrary to the interests of the poor, the wealthy could eventually refuse to redistribute once the revolutionary threat is over. Solution to the ﬁrst problem: constitutionalize property rights, so the poor cannot easily use their political rights to expropriate the rich. Solution to the second problem: constitutionalize democracy—along with a number of redistributive mechanisms (e.g., in the form of automatic stabilizers such as unemployment beneﬁts)—so the poor have a political device to ensure that wealth is in fact redistributed. Even if neither the the poor really wanted to expropriate the rich nor the rich refused to redistribute their wealth, they might have not been able to credibly communicate their commitments except through constitutional signaling.
See Bellamy, Political Constitutionalism, pp. 129 ss. D. Webster, Speech, March 15, 1837. 59 This is an oversimpliﬁed outline of their argument indeed. See Daron Acemoglu and James A. Robinson, Economic Origins of Deictatorship and Democracy (Cambridge: Cambridge University Press, 2006). For a rather comprehensive account of the reasons behind suﬀrage extensions, see Adam Przeworski, “Conquered or Granted? A History of Suﬀrage Extensions,” British Journal of Political Science 39, no. 2 (2009): 291-321. 60 As Thomas Macaulay put it in the 1842 speech on the Chartists, “The essence of the Charter is universal suﬀrage ... If you grant that, the country is lost ... My ﬁrm conviction is that, in our country, universal suﬀrage is incompatible, not only with this or that form of government, and with everything for the sake of which government exists; that it is incompatible with property.” Quoted in A. Przeworski, “Democracy, “quality, and Redistribution,” Political Judgement, R. Bourke and R. Geuss, eds. (Cambridge: Cambridge University Press, 2009), p. 303.
58 57

5.4. Criticisms

171

In short, as Elster has pointed out, constitutional provisions are best understood as devices for binding others rather than for binding oneself.61 Now, this need not trump the idea of constitutions as democratic institutional precommitments as long as we assume that the collective body who makes the commitment is pervaded by deep and persistent disagreements among its parts and that the rules to which the commitment is made are the upshot of a bargaining process among such parts. Further, it need not trump the idea of constitutions as democratic collective institutional precommitments as long as the resulting equilibrium is adopted by a majority of citizens; and, more precisely, as long as the demos condition and the procedure condition deﬁned in section 5.2.2, according to which Procedure condition.—A constitution can be deﬁned as a democratic commitment only if it is adopted through the use of equal voting and (perhaps qualiﬁed) majority rule. However, failure to meet these conditions happen along synchronic and diachronic lines, as we shall see. Let us analyze the synchronic dimension here and the diachronic one in the next section. As for the former dimension, at least three sorts of failures to meet the above-established conditions can occur. A. Tyranny of the minority. The ﬁrst sort of failure consists in a minority of citizens imposing their interests in the constitution against a majority of their fellow citizens. There are at least four senses in which this can happen. Before turning to them, consider a rather trivial point though—namely, that constitutions and constitutional amendments are generally drafted, and often ratiﬁed as well, by representatives rather than by the people themselves.62 This undermines a distinction made by Ackerman between constitutional lawmaking as made by the people and ordinary lawmaking as made by their representatives. In Ackerman’s words, “Above all else, a dualist Constitution seeks to distinguish between two diﬀerent decisions that may be made in a democracy. The ﬁrst is a decisions by the American people; the second, by their government.”
63

However, constitutions are not written down by

the people themselves. The people only ratify a text drafted by someone else—and only sometimes. To be sure, there is a distinction between the constitutive and the constituted powers. Unlike the latter power, which is exercised by the government proper, the former power cannot be exercised by the government either because no government has been properly constituted yet or because passing a constitutional amendment exceeds its authority. However, the constitutive power is not exercised by the people either—or not more than the constituted power, for in both cases it is
See Elster, Ulysses Unbound. Ely puts it similarly: “in fact there is no consensus to be discovered (and to the extent that one may seem to exist, that is likely to reﬂect only the domination of some groups by others).” 62 Of course, the point is not so trivial if representation is dismissed altogether, as some neo-Rousseauian critics of representation do. See, for example, B. Barber, Strong Democracy. Participatory Politics for a New Age (Berkeley and Los Angeles: University of California Press, 1984). For a criticism of Barber and a defence of representation as a sine qua non condition of democracy, see D. Plotke, “Representation Is Democracy,” Constellations 4 (1997): 19-34. 63 Ackerman, We the People, p. 6.
61

172

Chapter 5. Precommitment-based theories

exercised by someone on behalf of the people. In short, it is mistaken to oppose the people exercising the constitutive power to the government exercising the constituted power. This admittedly trivial point assumes that constitutional law-making is made by representatives on behalf of their constituencies. However, this need not be the case. Actually, it is not the case, at least, under four not uncommon conditions. Consider them in turn. First, even though we usually think of a constitutional assembly representing the diversity of social interests and deliberating about the principles that should be turned into higher law, this sort of assemblies is the exception rather than the rule. Thus, after analyzing ﬁfty-nine constitutional processes, Braulio Gómez found out that only twenty-ﬁve percent of them were conducted in an assembly, the remaining seventy-ﬁve percent corresponding either to negotiation tables between the outgoing regime and forces of the new emerging regime or to boards of experts.64 And when constitutional assemblies take place, they are often constituted by representatives who have not been elected by a majority of citizens or lack an interest to put forward democratic constitutional arrangements.65 In short, neither direct participation of citizens, nor indirect participation of them is the norm. Second, even when citizens are appropriately represented and the resulting text is submitted to referendum, the “upstream constraints” that are imposed on the assembly before it starts to deliberate may undermine its representativeness.66 These constraints are of two types. First, the agent who convenes the constitutional convention and, second, the mechanism used to select the delegates. None of them has to be necessarily democratic. For example, in the United States in 1787, the Philadelphia convention was called by the Continental Congress; in France in 1789, it was called by the King; in Germany in 1949, it was called by the Western occupying Allied Powers. To be sure, the resulting text can nevertheless be labeled democratic when having to be ratiﬁed by the legislatures or in referendum. Even when they are drafted by non-democratically called and appointed constitutional assemblies, and despite the fact that citizens only take part in the ratiﬁcation (typically, through referendum) at best, the ratiﬁcation process can force delegates to take into account the preferences of the electorate. Indeed, if the delegates know that the document they produce will have to be ratiﬁed by another body—namely, by citizens or their representatives—their perception of the preferences of that body will constrain their decision. However, this is not always the case and constitutions are often made unSee B. Gómez, “El control político de los procesos constituyentes” (PhD dissertation, Complutense University of Madrid, 2006). Quoted in Sánchez-Cuenca, Más democracia, menos liberalismo, p. 142. 65 For example, regarding the American Founding, Ackerman claims that “We must learn to see the Founders as they see themselves.” Ackerman, We the People, p. 165. However, most Americans were marginalized from the Founding, which was aimed at only white males and was ratiﬁed by less than two thousand people, as Ackerman keeps in mind (p. 307). Further, as we have seen in the ﬁrst chapter, the Founders were far from being democrats; actually, they drafted the Constitution with downsizing and checking democracy as one of their central aims. 66 See J. Elster, “Forces and Mechanisms in the Constitution-Making Process,” Duke Law Journal 45 (1995): 364-396, p. 373.
64

5.4. Criticisms

173

der non-democratic constraints, without mobilized deliberation at all and not being submitted to referendum for ratiﬁcation. As Ackerman himself acknowledges, the four-step process of signaling, proposing, deliberating and codifying that leads to a constitutional founding/amendment might end up in the congress or an assembly proposing a constitution/constitutional amendment without anything remotely similar to “the deep, broad and decisive support appropriate for a constitutional signal.” 67 Third, even when citizens are appropriately represented without signiﬁcant upstream constraints on the constitutional assembly and the ratiﬁcation of the ﬁnal text is submitted to referendum, powerful elites can be expected to attempt to inﬂuence—sometimes successfully indeed—both the drafting of the text and its ratiﬁcation. Consider a number of ways in which they can do so. First, by posing (implicit or explicit) threats on representatives and/or citizens if they deviate from their preferred outcome. Instances of such interferences are the “coups by memorandum” carried out by the Turkish military in 1971 and 1997, the coup attempt during the French Algiers crisis of 1958 that placed De Gaulle in charge of the Republic and gave way to the Fifth Republic (and, interestingly for present purposes, to the establishment of the Conseil Constitutionnel), or the coup threat posed by the military in Madagascar in 2010 in face of a potential constitutional reform. Second, powerful and organized interest groups (e.g. business organizations, religious groups, unions) can inﬂuence the constitution making at the expense of unorganized groups (e.g. the unemployed, homeless) either by directly lobbying the delegates of the constitutional convention or by trying to manipulate public opinion or by trying the inﬂuence delegates’ perception of citizens’ preferred constitutional choices.68 Fourth, even when citizens are appropriately represented without signiﬁcant constraints on the constitutional assembly and the will of the majority is adequately translated into the constitutional text, a majoritarian party can impose constraints upon the legislature and the executive on the believe that it will soon cease to be majoritarian (say, due to demographic reasons or to a predicted ideological shift). A good example is Giscard d’Estaing’s extension of the powers of the Conseil Constitutionneil at the end of his second term in order to reduce the power of the Communist Party in case it reached power in the next election.69 Likewise, a party (or a coalition) with a parliamentary majority may design the constitutional framework so as to increase its future power. For example, at the constitutional stage, large parties tend to prefer majority voting in single-member districts rather than a more proportional electoral system, which is preferred by small parties.70 B. Democratic indecisiveness. It can also happen just the contrary. A majority
Ackerman, We the People, p. 291. On the latter two options, even though regarding the ordinary legislative process, see S. Stokes, “Pathologies of Deliberation,” in Deliberative Democracy, J. Elster, ed. (Cambridge: Cambridge University Press, 1998), pp. 128-132. 69 See Elster, Ulysses Unbound, p. 171. 70 For speciﬁc examples, see Elster, “Forces and Mechanisms in the Constitution-making Process,” pp. 378-379.
68 67

174

Chapter 5. Precommitment-based theories

of citizens, even a qualiﬁed one, might want to amend the constitution yet not be able to do so because the formal constitutional amendment is never carried out or even blocked by the political elites. Taking into account the strong rigidity of many constitutions and the very complicated process of amendment of many constitutions, the latter possibility is more likely. It might be replied that it is adequate for a constitution to be very rigid so as to allow only the proposals for constitutional amendment that are supported without any doubt by a vast majority of well-informed, highly-engaged and publicly-spirited citizens. However, the reply is ﬂawed in two ways. First, such a device is not symmetric between past and current considered judgments” of the citizenry. Rather, it is biased toward the status quo, regardless of the preferences of the citizenry.71 Second, for the device to be justiﬁed, the constitution that the device is protecting from being amended needs to be free from the sort of pathologies depicted in the subsection immediately above. But that need not be the case, for such pathologies (e.g., critical circumstances preceding the constitutional moment, greater stakes and greater incentives to manipulate/advance factional interests, etc.) are as likely in the original constituent moment as in the amendment process. C. External pressures. Finally, it can also happen for a external agent to be able to impose their interests in the constitution against those of the citizenry as a whole. This was the case of the Japanese Constitution in the aftermath of the World War II, drafted by Douglass MacArthur, the Supreme Commander of the Allied Powers in Japan and the interim leader of the country from 1945 until 1948. But it is also the case when a powerful external agent imposes certain constitutional amendments on a country in exchange of something. Consider, for example, the “hidden letter” sent by the European Central Bank to a number of EU member states asking them to include a deﬁcit ceiling in their constitutions in exchange of buying their sovereign debt bonds in the secondary market. Among the latter countries, the Spanish case is specially outstanding for at least four reasons. First, the alleged “hidden letter” made the two main parties—the Socialist Party and the Popular Party—reach an agreement on the deﬁcit ceiling amendment in just a few days and pass the amendment in just a few weeks, even though the debate on amending the Constitution had been going on for years and had been blocked by precisely such parties. Second, the proposed amendments that had enjoyed citizens’ wide support for years were dismissed once again. Third, the debt ceiling amendment was in neither of the two parties’ manifestos and, accordingly, citizens had not voted for it. And fourth, the agreement was reached in August, when most citizens were on holidays; the amendment was passed in a few weeks, without suﬃcient time for citizens to get informed and deliberate about it; and the

71 In this sense, Ackerman acknowledges that Article 5 of the American Constitution is overly stringent, preventing constitutional moments from generating succesful constitutional amendments. Accordingly, he defends that it should be replaced by a “Popular Sovereignty Initiative.” Ackerman, We the People: Transformations, p. 415.

5.4. Criticisms

175

call for a referendum was despised, even though a ten percent of the members of either chambers of the Parliament could have made it mandatory. A ﬁnal caveat. Note that, even when externally imposed, constitutional constraints may well improve the performance of the country due to the reduction of temporal inconsistency, the protection of fundamental rights or the stability of the political system overall, as we have analyzed in chapter 4. All these points are obviously relevant from an instrumentalist point of view, and may justify such externally imposed constraints. Further, as long as they help to reach the mentioned outcomes, they may be said to be completely democratic as well from an instrumentalist point of view, for instrumentalist theories deﬁne democracy only according to the outcomes it brings about. However, from a precommitment-based standpoint, outcomes are irrelevant. What counts is whether the constitutional constraints are the upshot of a democratic commitment made by those citizens against whom such constraints are to be enforced. Hence, from a precommitment-based point of view, externally-imposed amendments unavoidably undermine the democratic status of the constitution. 5.4.4. Future generations and constitutional change Consider now the democratic status of constitutional precommitments across generations. It is a point as old as Jeﬀerson and Paine that fully democratic consitutions are not possible ex hypothesi, for constitutions are designed to be enforced against future generations that did not have a say in the making and ratiﬁcation of the constitution. “The earth belongs to the living, not to the dead,” wrote Jeﬀerson, from which it followed that, “by the law of nature, one generation is to another as one independent nation to another.” 72 And according to Paine, “Every age and generation must be free to act for itself, in all cases, as the ages and generations which preceded it.” 73 Of course, a founding generation can have long-term motivations and be concerned about future generations, but just as dictators can be benevolent and be concerned about their peoples, thus taking their preferences into account when making political decisions. (Actually, less than benevolent dictators have to do so as well to some extent, in order to ensure the stability of their regime and to minimize repression costs.) By contrast, democratic decision-making refers, as a matter of deﬁnition, to a sort of decision-making that makes compulsory, rather than only advisable, for those in power to take into account the preferences of their constituencies and provides citizens with the institutional technology necessary to ensure that their preferences are taken into account. This idea of democracy is compatible with ex ante and with ex post accountability, with direct and with indirect participation, with divided and with undivided government, and so on. But it does rule out decision-making that does not include formal institutional mechanisms to ensure that citizens have a say and that those in oﬃce have to take into account their preferences.
72

Jeﬀerson, Political Writings, p. 599; and “Letter to James Madison, September 6, 1789,” p.

116.

T. Paine, “The Rights of Man,” ed. B. Kuklick, Political Writings (Cambridge: Cambridge University Press, 2000), p. 63.

73

176

Chapter 5. Precommitment-based theories

Now, future generations cannot be said to be democratically represented by previous constitutional commitments because they are excluded from those mechanisms almost by deﬁnition. Put it in the jargon of the all-aﬀected principle, unlike the constituent generation, subsequent generations are aﬀected without having a say. Put it in the Aristotelian jargon, unlike the constituent generation, which members rule and are ruled in turn, subsequent generations are ruled without ruling at all. However, consider two possible replies.74 A. Amendability. A ﬁrst possible reply is that the people can always amend the constitution.75 As Justice John Marshall famously put it regarding the American constitution, “The people made the Constitution, and the people can unmake it. It is the creature of their own will, and lives only by their will.” 76 If this were the case, a core feature of democratic government would be satisﬁed—namely, the changeability of the laws in the light of new circumstances, beliefs and needs.77 However, even though the possibility of amending a constitution is certainly open to subsequent generations (with the exception of absolutely entrenched provisions), two problems arise with this reply. First, there is an asymmetry between adopting and amending a constitution, both formally and informally. Formally, because, as a general rule, constitutions only require simple majorities for their ratiﬁcation yet qualiﬁed majorities for their amendment, as we have seen above (§ 5.4.2). Informally, because there is an asymmetry between doing and undoing a constitution, just as there is an asymmetry between breaking a promise and not doing it in the ﬁrst place. Consider six mechanisms through which constitutions tend to be much more resilient to change than what is speciﬁed in the formal amendment procedure. 1. Material incentives. A constitution that has been in force for years and perhaps for decades creates an incentive structure that, even if ineﬃcient or unjust, makes it harder for citizens to be willing to modify it. For example, a constitution or a constitutional jurisprudence that promote ownership housing at the expense of rental housing may well produce a poorer economic output—for example, due to its tendency to create housing bubbles or to prevent labor mobility and thus increase unemployment. Yet, an amendment toward a diﬀerent model would face the likely opposition not only of current and would-be owners, but also of strong lobbies such as construction companies, mortgage banks and labor unions who may beneﬁt from the existing model.78 2. Learning eﬀects. The incentives created by the existing constitutional order need
The non-existence and non-identity problems are not addressed here. The locus classicus for such problems is D. Parﬁt, Reasons and Persons (Oxford: Oxford University Press, 1984). 75 This point is raised both by Ackerman, We the People; and Holmes, “Precommitment and the Paradox of Democracy,” p. 256. 76 Cohens v. Virginia, 6 Wheaton (19 US) 264, 389 (1821). 77 See J. Waluchow, A Common Law Theory of Judicial Review: The Living Tree (Cambridge: Cambridge University Press, 2007). 78 For a similar example, see A. Stone Sweet, “Path Dependence, Precedent, and Judicial Power,” in A. Stone Sweet, On Law, Politics, and Judicialization (Oxford: Oxford University Press, 2002), p. 118.
74

5.4. Criticisms

177

not be only material. They can be cognitive as well. Getting used to a new constitutional order is costly and time-consuming and leads to higher returns from continuing use. Therefore, the costs of amending that framework will also increase as time goes by and citizens get used to it. Consider, for example, the administrative structure of the state. Since it is very costly to get used to it and to know how it works, citizens will have a strong incentive not to modify it due to the high costs of getting used to the new structure. 3. Coordination eﬀects. These take place when the individual gains of performing an activity increase as other individuals join the same activity. It is also related to the ﬁrst mechanism, but it ampliﬁes its relevance as more people adapt their behavior to the existing order, thus increasing their beneﬁts and reducing the beneﬁts of those who hesitate to conform to it. Hence, the more people conform to the existing constitutional order, the more costly amending that order turns out to be and, in turn, the more people tends to conform to the existing order. As for this conformation, it can take place in the form of a genuine process of preference adaptation or as a mere falsiﬁcation of preferences resulting from the fear of social disapproval that expressing the view held in private would trigger.79 4. Status quo bias. Letting aside actual incentives to support the existing constitutional order, a constitutional amendment can also face the status quo bias of citizens that need not have an actual rational preference or incentive for the status quo per se but, nevertheless, are averse to change due to loss aversion or aversion to uncertain and unforeseen consequences of the proposed amendments. Tests of status quo bias in both experimental and ﬁeld settings have demonstrated that certain options tend to be preferred vis-à-vis alternative options only because the former pertain to (or are labeled as pertaining to) the status quo.80 I am not aware of evidence regarding constitutional change, but it is reasonable to expect many citizens to consider that a bird in the hand is worth two in the bush as for constitutional issues as well. 5. Expressive function of law. Citizens’ preferences are endogenous to the legal system. Law does not only reﬂect citizens’ will as channeled through the legislative process (assuming that this is roughly the case in a proper democratic system). As the literature on the “expressive function of law” has shown, law has a preference-shaping eﬀect as well.81 By making an statement about what ought to be done, law may inﬂuence dramatically citizens’ understanding of what ought to be done. Thus, there is evidence available that unenforced laws are observed
79 On preference adaptation, see J. Elster, Sour Grapes: Studies in the Subversion of Rationality (Cambridge: Cambridge University Press, 1983); on preference falsiﬁcation, see T. Kuran Private Truths, Public Lies: The Social Consequences of Preference Falsiﬁcation (Cambridge, Mass.: Harvard University Press, 1995). 80 For a survey of the evidence, see D. Kahneman, J. L. Knetsch and R. H. Thaler, “Anomalies: The Endowment Eﬀect, Loss Aversion, and Status Quo Bias,” Journal of Economic Perspectives, 5 (1991): 193-206. 81 See, among many others, C. Sunstein, “On the Expressive Function of Law,’’ University of Pennsylvania Law Review 144 (1987): 2021–53; K. Dau-Schmidt, “An Economic Analysis of the Criminal Law as a Preference-Shaping Policy,’’ Duke Law Journal 1 (1990): 1–38; E. B. Rasmusen and R. McAdams, “Norms in Law and Economics,” in Handbook of Law and Economics, vol. 2, A. Mitchell Polinsky and S. Shavell, eds. (Amsterdam: Elsevier, 2007).

178

Chapter 5. Precommitment-based theories

just by the fact of having been passed as part of the law. The preference-shaping eﬀect of law can take place at two diﬀerent levels. First, at an epistemic level, notably, when the law makes a prescription about a complex issue on which citizens have very imperfect information. Consider, for example, the eﬀect that German government’s decision to ban nuclear energy under its territory may have on citizens’ evaluation of the risks involved in this sort of energy. And, second, at a moral level, when the law modiﬁes not only citizens’ causal beliefs, but also their values. Patricia Funk has analyzed the recent legal abolition of compulsory voting in Switzerland. Even though the ﬁnes for not voting had only been symbolic, the legal abolition of the voting duty signiﬁcantly decreased average turnout.82 Mutatis mutandis, it is reasonable to expect constitutional law to have the same eﬀect on citizens evaluations—both epistemic and moral—of government’s authority and fundamental rights (if not greater, due to the relevance that citizens usually attribute to the constitution). Put diﬀerently, it is reasonable to expect a preference-shaping eﬀect from constitutions as well. In turn, this eﬀect can produce a constitutional circularity that can consolidate the status quo, thus making it even harder for an amendment to eventually take place.83 6. Power asymmetries. As Pierson has pointed out, power asymmetries can produce positive feedback over time.84 Hence, increasing returns processes can turn a situation of relatively balanced conﬂict in which one group has to impose its preferences on another openly, into a situation in which power relations become so unbalanced that anticipated reactions and ideological manipulation make open political conﬂict irrelevant. Consider a majority electoral system that distributes campaign resources according to the number of seats of each party. As a party obtains more seats in the parliament, it will also enjoy more campaign resources and will be able to increase the number of seats in the next election and, in turn, obtain even more resources, thus consolidating its ideological position in the eyes of the citizenry, becoming a natural beneﬁciary of strategic voting and being in a better position to exploit the electoral system. In short, positive feedback over time increases power asymmetries, enlarging the power of the parties that beneﬁt from the current constitutional order and providing them with an incentive to support the status quo. Most of this phenomena fall under the umbrella of what has been termed as path dependence, which according to Margaret Levi means that “once a country or a region has started down a track, the costs of reversal are very high. There will be other choice points, but the entrenchments of certain institutional arrangements obstruct an easy reversal of the initial choice.” 85 And regarding law-making, Stone Sweet has described it as follows: “At a beginning point, a range of choices, formats, or
See P. Funk, “Is There An Expressive Function of Law? An Empirical Analysis of Voting Laws with Symbolic Fines,” American Law and Economics Review 9 (2007): 135-159, p. 136. 83 See M. Abramowicz, “Constitutional Circularity,” UCLA Law Review 49 (2001-2002): 1-91. 84 P. Pierson, “Increasing Returns, Path Dependence, and the Study of Politics,” American Political Science Review 94 (2000): 251-267, p. 259. 85 M. Levi, “A Model, a Method, and a Map: Rational Choice in Comparative and Historical
82

5.4. Criticisms

179

templates for a particular form of behavior are available; at one or several ’critical junctures’ one of these choices gains an advantage, however slight, and this advantage is continuously reinforced through positive feedback. Ultimately, the choice becomes dominant, or ’locked-in’, as a relatively taken-for-granted state of aﬀairs.” 86 The relevance of this phenomena for the debate on the intertemporal democratic pedigree of constitutions should not be diminished. Put simply, here is why. For a constitution to be fully democratic across generations it ought to present a symmetry between the founding and subsequent generations, so the existing provisions could be amended by the latter provided that the amendment enjoyed the support of a majority of citizens. However, this symmetry does not exist. Consider now the second problem with the reply. The problem with arguing that constitutions are democratic accross generations because they can always be amendmened is that rigidity (at least, in its formal form) is a constitutive feature of constitutionalism proper, for the very point of constitutionalism is to limit the powers of the legislative and administrative bodies by entrenching a set of constraints in a rigid document, as it has been deﬁned at the outset of Part II of the dissertation (§§ 3.2 and 3.4). A constitution that does not constrain future decisions (that is, a completely ﬂexible constitution) falls outside the scope of constitutionalism. Hence, there is a contradiction between supporting constitutionalism, on the one hand, and appealing to the ﬂexibility and amendability of constitutions as a response to the counter-majoritarian diﬃculty, on the other, for the very idea of constitutionalism includes the rigidity of the constitution as a constitutive feature. B. Future beneﬁts and future duties. A second, more subtle, reply is the one provided by Stephen Holmes, who relies on an argument already put forward by Pufendorf in the seventeenth century to justify why the heirs of the crown were bound by the duties and commitments in which their ancestors had incurred.87 More precisely, Pufendorf analyzed whether a recently crowned king had the duty to pay the debts in which his predecessor had incurred, provided that sovereign debt was a crucial resource to fund the public expenditure and that few monarchs could go without using it. Pufendorf argued that, as a general rule, a commitment made by an individual or a generation cannot bind subsequent individuals or generations. However, this rule had a crucial exception, namely, “the act of a generation can only bind another one when a man [for example, a creditor] has acquired a right from him.” 88 More precisely, if someone inherits the property of someone else (for example, the throne), he also inherits the duties associated to that property (for example, the debts). Mutatis mutandis, if a subsequent generation enjoys the beneﬁts of a precedent generation, it also assumes its duties as well. Consider, for example, that the founding generation includes a
Analysis,” ed. M. I. Lichbach and A. Zuckerman, Comparative Politics: Rationality, Culture, and Structure (Cambridge: Cambridge University Press, 1997), p. 28. 86 Stone Sweet, “Path Dependence, Precedent, and Judicial Power,” p. 115. 87 Pufendorf, De jure naturae et gentium, Book 8, ch. 10. Quoted in Holmes, “Precommitment and the Paradox of Democracy,” p. 234. 88 Pufendorf, De jure naturae et gentium, Book 4, ch. 2. Quoted in Holmes, “Precommitment and the Paradox of Democracy,” p. 234.

180

Chapter 5. Precommitment-based theories

budget deﬁcit ceiling in the constitution so as to solve the problem of temporal ﬁscal inconsistency between governments, that is, to prevent governments (including the present government) from generating negative budget diﬁcits that current citizens enjoy but future citizens will have to pay.89 Further assume that, by doing so, the founding generation is assuming certain burdens from which the subsequent generations will beneﬁt. Would it be reasonable for these subsequent generations to beneﬁt from such burdens without assuming their costs as well (for example, by saying that the decisions of the founding generation do not bind them, for they did not have a say in such decisions)? Hence, Holmes appears to solve the intergenerational problem of constitutional constraints by presenting such constraints as inherited duties correlative to inherited beneﬁts. In short, if subsequent generations beneﬁt from the constitutional constraints adopted by the founding generation, then it makes complete sense to claim that they are also bound by them. Now, even if Holmes might be right as a matter of justice and constitutional constraints between generations might be just, the argument misses the point of whether they are democratic from a precommitment point of view, which is the issue in which we are interested in this section (rather than in whether they are just or not). From a precommitment point of view, a constitution counts as a democratic precommitment only if it is adopted (not implicitly, but through an actual institutional device such as voting) by the very same people against whom the constitution is to be enacted afterward. And the fact is that the asymmetry between generations persists, for the founding generation takes part in the adoption of the constitution, while the subsequent generations do not, regardless of whether they are beneﬁted from its existence or not. More precisely, the founding generation takes part in the adoption of the constitution through a simple majority procedure, while the subsequent generations can take part in its amendment through a more cumbersome rule, such as a qualiﬁed majority rule. Even more precisely, the founding generation takes part in a decision that breaks with the previous legal order (say, a dictatorship) by using a simple majority procedure, while the subsequent generation can amend the existing legal order only through a more cumbersome rule, such as a qualiﬁed majority rule. In short, the persistent problem is why a rigid constitution adopted by the founding generation should count as democratic when enforced against future generations (sometimes very distant in time indeed) when the latter did not have a say in its adoption nor can they amend it as easily as the founding generation adopted it. The problem is similar to the one described in subsection 5.4.3—namely, a minority imposing their will over a majority, only that across generations. Here the minority is the founding generation and the majority all future generations against whom the will of the former is constitutionally enacted. Robert Dahl has summarized both
89 Recall David Stockman’s declaration, who refused to address Social Scurity severe long-term problems in 1981, when he was budget director of the Reagan administration, because he had little interest in wasting “a lot of political capital on some other guy’s problem in [the year] 2010.” Quoted in Pierson, “Increasing Returns, Path Dependence, and the Study of Politics,” p. 261.

5.4. Criticisms

181

the synchronic and the diachronic problems regarding the American Constitution as follows: Why should we feel bound today by a document produced more than two centuries ago by a group of ﬁfty-ﬁve mortal men, actually signed by only thirty-nine, a fair number of whom were slaveholders, and adopted in only thirteen states by the votes of fewer than two thousand men, all of whom are long since dead and mainly forgotten?90 5.4.5. Private citizenship and participation An alleged virtue of a constitutional democracy is that its two-track lawmaking system provides both for citizens’ mobilized deliberation during constitutional moments and for a lower civic engagement during ordinary legislative politics, when citizens devote most of their time to their families, friends, jobs and hobbies while their fundamental rights are constitutionally protected against opportunistic behavior by temporary majorities and powerful lobbies. In a nutshell, constitutional democracy conforms well to private citizenship. However, to see constitutional democracy as the upshot of, and the solution for, private citizenship, rather than the other way around, implies too static a conception of citizenship. Not in the sense of taking political passivity and apathy as an exogenous disposition under any circumstances, for private citizens can stand up and speak out against the constitutional order during constitutional moments—just as they can turn to less engaged tasks during more ordinary times. But rather in the sense of taking citizens’ political passivity and apathy during normal politics as the norm and as something to which the institutional framework has to adapt, rather than as something to great extent endogenous to the constitutional system. Assuming that it is possible to overcome the problems raised in this section and that constitutionalism rather than feasible alternatives (such as Westminster parliamentarianism or the new commonwealth system) better protect citizens’ fundamental rights, constitutional entrenchment and judicial review have the side-eﬀect of making political change on fundamental issues very hard. As we have discussed in subsection 5.4.3 (epigraph B), a majority of citizens, even a qualiﬁed one, might want to amend the constitution yet not be able to do so due to the stringency of the constitutional amendment procedure. Of course, the capacity of citizens to raise their demands and modify the constitution varies from constitutional system to constitutional system. But, apart from the rigidity of the constitution raises the likelihood of phenomena of preference adaptation and status quo bias taking place. On the one hand, citizens might adapt their preferences to the existing constitutional order when they perceive that the possibility of advancing their current preferences through constitutional change is low and/or very costly. On the other hand, citizens can suﬀer from a status quo bias once the existing constitutional system has been consolidated due to the diﬃculty of changing it.
R. Dahl, How Democratic is the American Constitution? (New Haven: Yale University Press, 2003), p. 2.
90

182

Chapter 5. Precommitment-based theories

Accordingly, citizens can end up taking the constitutional system for granted and naturalizing it, rather than taking it as something that has been created and can be amended at will. In turn, this can spread the perception among them that their demands, even when widely discussed and supported, cannot be adequately channeled by the political system, thus providing them with the incentive to abandon useless political activism and turn to their private lives, where they perceive their actions as having a deeper and more visible eﬀect. In short, the apathy and passivity typical of ordinary politics is better understood as the upshot of the constitutional system and the stringency of its amendment procedures, rather than the other way around. In turn, two undesirable eﬀects follow from this apathy and passivity. First, continued apathy and passivity due to the unchangeability of the constitutional system can spread frustration among citizens, who can perceive that the most salient issues aﬀecting them are always out of their control and are decided by an elite of judges that are often as ideologically divided and politicized as they are. In turn, frustration can jeopardize the legitimacy of the political system in the long run. Legitimacy is a limited capital and a system can obviously run out of it, with dangerous consequences for its stability.Ackerman seems to be aware of this problem for he urges, in the last chapter of We the People, that a constitutional moment, including a new Bill of Rights, is needed in American politics in order to make citizens perceive that the Constitution is amendable and thus the upshot of a democratic precommitment. Otherwise, he warns, “the constitutional narrative I have told will lose credibility for the next generation.” 91 Second, even if we accept that private citizenship is not a ﬁxed feature of normal politics but partially a function of normal politics under stringent constitutional constraints, private citizenship can still be taken as something to be promoted and constitutional democracy as the best institutional technology to do so. This is Ackerman’s point, for he claims that a constitutional democracy is preferable to monistic democracy because it leaves room and better protects private citizenship, which under modern circumstances is preferable to public citizenship. However, for democracy to work properly, larger participation than the typical of the private citizen is required during ordinary legislative politics. Unless adequately monitorized by an informed and active citizenry, politicians in oﬃce will lack an adequate incentive to take into account ordinary citizens’ interests and the political process will be likely controlled by special interests. As Rawls thoughtfully explained,

Unless there is widespread participation in democratic politics by a vigorous and informed citizen body ... even the best-designed political institutions will eventually fall into the hands of those who hunger for power and military glory or pursue narrow class and economic interests ... If we are to remain free and equal, we cannot aﬀord a general retreat into private life.92
Ackerman, We the People, p. 296. J. Rawls, Justice as Fairness. A Restatement (Cambridge, Mass.: Harvard University Press, 2001), p. 141.
92 91

5.4. Criticisms

183

Ackerman is fully aware of this possibility for he devotes a large amount of time to refute the idea that a good-working democracy can go without civic virtue at all. According to him, “while a modern democracy must learn to economize on public-regarding virtue, there can be no hope of doing without it entirely—even in normal politics.” 93 Now, even if he does not specify the degree of public involvement needed for the democratic system to work properly (something that, nevertheless, could hardly be done in general terms, being so context-dependent an issue), he appears to believe that a low degree of such an involvement would suﬃce under proper institutional conditions. To be sure, the causal relationship between participation and accountability and responsiveness is an empirical and highly controversial matter. But it is not unreasonable to believe, ﬁrst, that a well informed and active citizenship is necessary for the political system to work adequately during normal times and, second, that the stringency of the amendment procedure and the large judicial power of constitutionalism promotes citizens’ apathy and passivity. Ackerman could reply that the informed and active citizenry required for a healthy democratic system could give way to what he labels as “coercive democracy”—that is, the imposition of “public citizenship” as compulsory for citizens that might not value political participation at all. However, this need not be the case for two reasons. First, because private citizenship could give way to a more engaged citizenship in case the constitutional constrains were relaxed, being such constraints one of the causes of citizens’ apathy and frustration. And, second, because alternatives to the civic humanistic model of public citizenship are available. Hence, according to classical republicanism, participation is not even remotely valued per se, but rather as a condition to prevent the corruption of the political elite. As Skinner recalls regarding classic republicanism, “the writers I am considering never suggest that there are certain speciﬁc goals we need to realise in order to count as being fully or truly in possession of our liberty.” “[They] merely argue that participation (at least by way of representation) constitutes a necessary precondition of maintaining individual liberty.” 94 Of course, the causal relationship between participation and the quality of the political performance is a mainly empirical and highly controversial issue. However, assume for the sake of the argument that informed and active participation—or at least a more informed and more active participation than the one assumed by Ackerman—was required for normal politics to function adequately. In that case, what would be the point in keeping the dualist distinction between normal and constitutional politics? As we have seen, Ackerman’s dualism is dependent on the existence of private citizens that only stand up and engage in politics during unsual constitutional moments, when they entrench their considered demands against the horse-trading of ordinary politics, when they can go back to their private aﬀairs and let their elected representatives do their job. But once we have accepted that private citizenship is not enough for the political process to remain accountable and that
Ackerman, We the People, p. 239. Q. Skinner, “The Paradoxes of Political Liberty,” in The Tanner Lectures on Human Values, VII (Salt Lake City: The University of Utah Press, 1986), 233; and Liberty Before Liberalism, 74-75.
94 93

184

Chapter 5. Precommitment-based theories

an alternative model to the Arendtian public citizenship is available, the rationale for constitutional entrenchment against ordinary politics decision-making seems less robust.

Chapter 6

Proceduralist theories
6.1. Introduction
Theories of procedural constitutional constraints are addressed in this chapter.1 According to these theories, democratic institutions cannot be said to be really democratic unless a set of constitutive conditions (e.g., “general elections, freedom of the press, freedom of assembly, and freedom of speech,” as Rosa Luxemburg put it criticizing the suspension of civil liberties by the Russian Bolsheviks) and preconditions (e.g., “decent wages and universal reading,” as John S. Mill put it) are met.2 Further, it is also argued that, unless entrenched in a rigid constitution and enacted by a constitutional court, such conditions and preconditions can be jeopardized by temporary parliamentary majorities, which can decide, for example, to disenfranchise a group of citizens, to ban a political party, or yet to suspend certain democratic rights altogether. In a nutshell, unless constitutionally constrained, ephemeral parliamentary majorities can downsize or even destroy democracy by democratic means. It is noteworthy that, unlike instrumentalist accounts of democratic legitimacy, procedural theories do not justify constitutional constraints independently from the democratic process, for democratic institutions can bring about blatantly unjust outcomes without such outcomes turning into democratic for that reason. Neither do they justify them as the upshot of democratic commitments made by the citizenry—as it is the case with precommitment-based accounts—, for citizens can commit themselves to blatantly undemocratic rules without such rules turning into democratic for that reason. On the contrary, procedural theories justify constitutional procedural constraints as the constitutive conditions and preconditions of the the democratic process proper, as the constitutional safeguards of its adequate functioning. The chapter is divided into four sections. First, the much cited instance of a democratic system destroyed by democratic means—namely, the fall of the Weimar Republic in the hands of the Nazis—is brieﬂy analyzed (§ 6.2). Second, preconditions-based
The locus classicus is J. H. Ely, Democracy and Distrust. A Theory of Judicial Review (Cambridge, Mass.: Harvard University Press, 1980). See also Holmes, “Precommitment and the Paradox of Democracy;” C. Sunstein, Legal Reasoning and Political Conﬂict (Oxford: Oxford University Press, 1996); and Dahl, Democracy and Its Critics (New Haven: Yale University Press, 1989). 2 Luxemburg is quoted in C. E. Schorske, German Social Democracy, 1905-1917: The Development of the Great Schism (Cambridge, Mass.: Harvard University Press, 1955), p. 324. Mill is quoted in Adam Przeworski, Democracy and the Limits of Self-Government (Cambridge: Cambridge University Press, 2010), p. xiii.
1

186

Chapter 6. Proceduralist theories

theories of constitutionalism are brieﬂy analyzed, notably, the rather standard account of Holmes and Sunstein and the more peculiar one of Habermas (§ 6.3). Third, Ely’s (and Holmes’, though more brieﬂy) “participation-oriented, representation-reinforcing” approach to judicial review is analyzed as the most inﬂuential instance of a purely procedural account of constitutionalism (§ 6.4). Finally, a number of criticisms are raised (§ 6.5).

6.2. Excursus: was Hitler democratically elected?
The case for the constitutional entrenchment of democratic procedures and preconditions is often made in purely theoretical or counterfactual terms. It may well be the case for a parliamentary majority to decide to disenfranchise a minority or to suspend some democratic rights altogether, as it is frequently argued. When made in reference to actual political systems, however, the Weimar Republic and the way the Nazis destroyed the democratic system from within is the ever-present reference. Indeed, it is virtually impossible to raise this issue in the classroom without having a student pointing out that Hitler was elected democratically and that, unless strong limits on democratic procedures are constitutionally entrenched, a new Hitler could get to oﬃce in the future and destroy the democratic system by democratic means again. (Certainly, a version of Godwin’s law for the classroom would not be completely useless.) Joseph Goebbles himself seemed to have agreed with this observation when he argued that “[i]t will always be one of the best jokes of democracy that it gives its deadly enemies the means to destroy it.” 3 In the aftermath of the World War II it became usual to believe that Goebbles had been basically correct. As it was widely assumed by scholars, citizens and politicians alike, the institutional system of the Weimar Republic had failed to provide enough constitutional guarantees for the protection of the democratic institutions against the Nazis. But post-war Germans knew better and the Basic Law of 1948 was in part the upshot of this shared perception. To be sure, the perceived democratic fragility of the Weimar constitution was by no means the only reason behind the absolute entrenchment of some of the provisions of the German Basic Law, which not only included entrenched procedural rights but also procedure-independent ones.4 But the story of the democratic fragility of the Weimar Constitution would be much cited in circumstances of actual or perceived hazard for the Federal Republic—such as the banning of the Socialist Reich Party by the German Constitutional Court in 1952—, providing a powerful rationale for the constitutionalization of the procedural conditions and preconditions of democracy. Now, the received wisdom notwithstanding, the rise to power of the Nazis is very far from being a clear-cut example of the destruction of a democratic system by perfectly democratic means.5 To be sure, Hitler had a plurality of votes when he was
Quoted in A. S. Kirshner, “Proceduralism and Popular Threats to Democracy,” Journal of Political Philosophy 18, no. 4 (2010): 405-424, p. 405. 4 See Schwartzberg, Democracy and Legal Change, pp. 153-192. 5 See R. J. Evans, The Coming of the Third Reich (New York: Penguin, 2005); H. A. Turner, Hitler’s Thirty Days To Power: January 1933 (Basic Books, 1997).
3

6.2. Excursus: was Hitler democratically elected?

187

appointed chancellor by president Hindenburg on January 30, 1933 (the NSDAP had gained 196 seats in the November election, vis-à-vis the 121 seats of the SPD and the 100 seats of the KPD). Further, the “Reichstag Fire Decree” that suspended many of the fundamental rights after the Reichstag building was set on ﬁre on February 27, 1933 was issued by Hindenburg following the emergency powers included in the art. 48 of the Constitution. Even further, after the election of March 1933, the NSDAP gained a 43,9 percent of the votes and, along with their Nationalist allies (the DNVP), not a plurality but a majority of 51,8 percent. Yet even further, the Enabling Act that provided Hitler with the power to enact laws without the participation of the Reichstag was passed by the Reichstag on March 23 1933 and signed by Hindenburg. However, the Nazis also took a large number of blatantly undemocratic measures in their road to absolute power. Consider just three of the most outstanding ones. First, most of the members of the KPD were unable to take part in the March election, either because they had been arrested after the Reichstag was burnt down (not less than four thousand members of the KPD were arrested during that night of February) or because they had been “invited” not to run for oﬃce by the SA under dead threat. Second, Hitler’s government had expropriated the KPD of their properties before the election and their election propaganda, as well as that of the SPD, had been banned. Third, The Enabling Act was passed by the Reichstag under two illegal circumstances. On the one hand, the minimum quorum requirement established by the constitution was modiﬁed without observing the constitutional amendment procedure so as to prevent the SPD deputies from blocking the law. On the other hand, members of the SA illegally entered the parliament during the voting, and threatened indecisive deputies. In short, it is at the very least controversial to use the Weimar Republic as an instance of the alleged threat of a parliamentary majority destroying the democratic system by democratic means. However, it is always possible to point at additional—though admittedly controversial—instances of this threat, such as the large number of left-wing parties, religious parties or Kurdish separatist parties that have been systematically banned in Turkey, the American Communist Party, which was outlawed by the Congress in the Communist Control Act of 1954 or the Basque separatist party Batasuna, which was banned in 2003 by the Spanish Supreme Court under the Law of Political Parties, which was amended by the parliament the year before so as to outlaw political organizations, including parties, associated with terrorism.6 Further, the possibility of undemocratic decisions being made by democratic
Article 9 of the Organic Law of Political Parties states that “a party will be declared illegal when its grave and continuous” activity “makes democratic principles vulnerable.” Prohibited activities include “promoting, justifying or excusing assaults against the life or the integrity of persons ... inciting, bringing about or legitimizing violence as a means for the achievement of political objectives ... politically complementing and helping the action of terrorist organizations for the achievement of their ends of subverting constitutional order or gravely altering public peace ... giving express or tacit political support, legitimizing terrorist actions or excusing and minimizing their signiﬁcance.” It has to be pointed out that the three cases are very diﬀerent. As for the ﬁrst one, most of the Turkish parties were banned by the Constitutional Court rather than by the National Assembly.
6

188

Chapter 6. Proceduralist theories

means remains always open as a counterfactual. Either way, there are good grounds to claim that, unless constitutionally entrenched, democratic conditions and preconditions can be unattended or even openly violated by parliamentary majorities unless they are constitutionally protected.

6.3. Preconditions-related constraints
A frequent criticism of pure procedural democracies—of which Westminster-style democracies are the most outstanding example—is that they do not protect adequately the preconditions that are to be met for the democratic process to work properly. By refusing to constitutionalize them, these democracies run the risk of letting such preconditions unattended at best and perhaps violated at worst. If a democratic system is to work adequately—or so the argument goes—, such preconditions have to be protected adequately and taken away from the logrolling and horse-trading typical of ordinary politics, being their constitutionalization the best way to do so. Before turning to pure proceduralist accounts, thus, it is worth considering brieﬂy two types of preconditions-based accounts of constitutional constraints. First, a more or less standard account, as exempliﬁed by Holmes and Sunstein, and, second, Habermas’ more peculiar account. Before turning to these accounts, consider two outstanding types of democratic preconditions. Firstly, there are those rights that, even if not strictly political, turn out to be necessary for the proper functioning of the democratic system. Consider, for example, civil rights such as free speech and religious freedom or economic rights such as universal instruction and freedom from starvation. Is democracy possible in countries in which citizens are prosecuted because of their religious beliefs, in which illiteracy is the norm, or in which citizens starve not to die out of hunger? In countries in which these conditions apply, it is diﬃcult not to see democracy and its policies as mere window dressing. As Nino points out, “democracy requires fulﬁllment of certain prerrequisites without which there is no reason to defer to the results of democracy.” 7 Secondly, there are structural (social and economic) conditions that, even if directly unrelated to the democratic process, can inﬂuence and distort the political process. Consider, for example, economic and social inequalities that may be large enough to distort the democratic decision-making through campaign ﬁnance, pork barreling, or threats posed by powerful investors (e.g. countries such as China or private hedge funds such as Paulson & Co or Morgan Chase) of massively selling sovereign debt so as to obtain political beneﬁts. As Larry Bartels has concluded from his recent study of the relationship between economic and political inequalities,
As for the second one, the act was largely ineﬀectual. In 1961 case case, Communist Party v. Catherwood, the US Supreme Court ruled that the act did not prevent the party from taking part in New York’s unemployment insurance system, and no administration has attempted to enforce it ever since. As for the third one, the law was passed by the parliament, which asked the Supreme Court (Spain’s highest court of appeal) to ban Batasuna under the amended law. Eventually, the Supreme Court unanimously approved the request, which was ratiﬁed by the Constitutional Court afterwards. 7 Nino, The Constitution of Deliberative Democracy, p. 138.

6.3. Preconditions-related constraints

189

“economic inequality clearly has pervasive, corrosive eﬀects on political representation and policy making ... In light of these eﬀects, the liberal hope for distinct ‘spheres of justice’ with ‘their boundaries intact’ seems naively fastidious and quite probably ineﬀectual.” 8 Unlike in Habermas’ account, as we shall see immediately below, these preconditions are generally acknowledged to be substantive and conceptually independent from—though related to—the democratic process. Among these are Stephen Holmes and Cass Sunstein.9 Even though they support, as a general rule, constitutional ﬂexibility and the collapse of the distinction between constitutional and ordinary legislation, they also consider that the rights that serve as democratic preconditions should not only be constitutionalized but actually entrenched against the possibility of being amended. As they put it, “a speciﬁed list of individual rights should be made immune to revision. This list should include, ﬁrst and foremost, rights that are indispensable to democratic legitimacy.” 10 Among them it is what Brettscheneider labels as “rights as procedural preconditions” to democracy, which may include the constitutional rights to free speech or the right to freely travel.11 This position does not go without problems, as we shall see below in section 6.5. Among other things, because insofar as rights are deﬁned as substantive, the problems with rights-instrumentalism analyzed in section 4.3 turn out to pervade preconditions-based approaches as well. Habermas’ account diﬀers from more standard accounts that acknowledge that democratic preconditions are substantive precisely in that his account is deﬁned as purely procedural. This is so because no pre-political or substantive rights can be worked out under what he calls the “conditions of post-metaphysical thinking.” 12 Under such conditions, basic rights can neither be grounded on religious arguments, nor in metaphysical or traditional ones. They can only be the upshot of the political process, which is the only source of legitimacy and, in turn, serves as a means for the protection of such rights.13 In short, according to Habermas, far from an oxymoron, constitutional democracy is the natural expression of public autonomy—as instantiated in democratic decision-making—and private autonomy—as instantiated in constitutionalized basic rights. Even though he is reluctant to absolute entrenchment,14 he does not consider
L. Bartels, Unequal Democracy. The Political Economy of the New Gilded Age (Princeton, NJ: Princeton University Press, 2008), p. 285. 9 See S. Holmes and C. Sunstein, “The Politics of Revision,” in Responding to Imperfection, Sanford Levinson, ed. (Princeton, NJ: Princeton University Press, 1995). 10 Holmes and Sunstein, “The Politics of Revision,” p. 279. 11 C. Brettschneider, Democratic Rights. The Substance of Self-government (Princeton and Oxford: Princeton University Press, 2007), p. 13. 12 See J. Habermas, Faktizität und Geltung. Beiträge zur Diskurstheorie des Rechtes und des demokratischen Rechtsstaats (Frankfurt am Main: Suhrkamp, 1992). 13 Habermas’ set of basic rights is ﬁve-fold: ﬁrst, classic liberty rights; second, communitarian rights; third, procedural rights; fourth, political rights; and ﬁfth, basic social rights and rights to protection against technological risks as well as basic ecological rights. 14 In his words, “The character of constitutional foundings, which often seal the success of political revolutions, deceptively suggests that norms out of time and resistant to historical change are simply ‘stated.’ The technical priority of the constitution to ordinary laws belongs to the systematic elucidation of the rule of law, but it only means that the content of constitutional norms is relatively ﬁxed. As we shall see, every constitution is a living project that can endure only as an ongoing interpretation continually carried forward at all levels of the production of law.” Habermas,
8

190

Chapter 6. Proceduralist theories

constitutionalism, including judicial review, to be at odds with democracy. “A constitutional court guided by a proceduralist understanding of the constitution does not have to overdraw on its legitimation credit.” 15 In turn, basic rights are also necessary preconditions for the proper functioning of the political process, which would otherwise be distorted. Hence, the need for both democracy and constitutional rights rests on the interdependence of both elements. As Habermas puts it, The demand to orient oneself to the common good, which is connected with political autonomy, is also a rational expectation insofar as only the democratic process guarantees that private individuals will achieve and equal enjoyment of their equal individual liberties. Conversely, only when the private autonomy of individuals is secure are citizens in a position to make correct use of their political autonomy. The interdependence of constitutionalism and democracy comes to light in the complementary relationship between private and civic autonomy: each side is fed by resources it has from the other.16 Now, it is at least doubtful whether Habermas’ account can live up to its purely procedural promise without becoming substantive. At least, for three reasons. First, Habermas presents his account as purely procedural because the political goals that are to be achieved remain open for the citizenry to decide. However, since the outcomes have to be rational to be legitimate—i.e., they have to be consistent, at least, with the basic rights required for the procedure to be legitimate—, eventually such goals are far from being really up for grabs. Second, Habermas’ set of rights is so expansive that it includes some rights that could be rarely said to be democratic preconditions. For example, it is diﬃcult to see how reproductive rights or ecological rights (included in the ﬁrst and ﬁfth categories of his set of rights, respectively) could be related to the proper functioning of the political process. Third, Habermas’ co-originality thesis implies that a decision suppressing some basic right ought to be considered necessarily undemocratic, even if the decision has been made by undisputable democratic means and regardless of whether that right is actually necessary for the proper functioning of the democratic process.17 However, this can lead to highly counterintuitive conclusions. For example, it would be very counter-intuitive to say that Swiss citizens’ decision to ban minarets in 2009—assuming that the decision did diminish the freedom of conscience of Swiss Muslims—made the referendum that lead to that decision less democratic.

6.4. Pure procedural constraints
Consider now pure procedural theories of constititutional constraints. According to these theories, while substantive issues should be left to legislatures, constitutional
Faktizität und Geltung. Quoted from J. Habermas, Between Facts and Norms (Cambridge, Mass.: MIT Press, 1996), p. 129. 15 Habermas, Between Facts and Norms, p. 279. 16 J. Habermas, “Constitutional Democracy: A Paradoxical Union of Contradictory Principles?,” Political Theory 29 (2001): 766-781, p. 781. 17 See Bayón, “Democracia y derechos.”

6.4. Pure procedural constraints

191

constitutional rigidity and judicial review are justiﬁed when they are exclusively aimed at entrenching the constitutive conditions of democracy and protecting them from potential threats in the form of, say, gerrymandering, the disenfranchisement of insulate minorities or the prohibition of political parties.18 Of course, pure proceduralist theories need not be at odds with precommitment-based accounts.19 However, in this section they will be analyzed separately. From a purely procedural point of view, even though removing substantive issues (such as abortion, national defense or ﬁscal policy) from the legislatures and placing them in the hands of constitutional courts is taken to be undemocratic, that need not be the case when the entrenched issues are the procedural conditions necessary for the very functioning and adequate working of the democratic institutions. As Robert Dahl puts it, “a court whose authority to declare laws unconstitutional was restricted to rights and interests integral to the democratic process would be fully compatible with the democratic process.” 20 Among the most commonly held procedural conditions, the following ones stand out:21

— Equal participation of all citizens, including the right to vote, organize, assemble, protest and lobby for their interests so as to inﬂuence the decision-making process. — Regular, free and fair electoral competition between parties and open to all citizens. — Vertical accountability, which includes information and justiﬁcation of the political actions of politicians and branches of government and the resulting punishment or compensation by citizens. — Horizontal accountability, which requires oﬃce-holders to be answerable to other institutional actors, such as the parliamentary opposition, the parliamentary investigative committees, the bureaucracies or the ombudsman.22

Ackerman refers to this variant as “sophisticated monism,” trying thus to make it clear that, despite its commitment to a set of procedural and sometimes substantive checks, its supporters see constitutional checks as anti-democratic and justify them only when strictly necessary to ensure the procedures to function properly, always subordinating them to the democratic process.” Ackerman, We the People, p. 8. The view is also consistent with what Estlund labels as “intrinsic democratic proceduralism,” according to which the value of political arrangements depends only on (a) their being democratic and (b) their not unduly undermining or threatening the possibility of democratic procedure in the future. See D. Estlund, Democratic Authority. A Philosophical Framework (Princeton and Oxford: Princeton University Press, 2008), p. 84. 19 This is actually the case of Ely, who argues that fundamental rights are a necessary condition for the exercise of self-government. 20 Dahl, Democracy and its Critics, pp. 191 and 359, n. 9. 21 See L. Diamond and L. Morlino, “Introduction,” in Assessing the Quality of Democracy, ed. L. Diamond and L. Morlino (Baltimore: The John Hopkins University Press, 2005). See also the indicators included in Freedom of the World Reports by Freedom House, the yearly Economist Democracy Index or the United Nations Development Program Human Development Report 2002, Deepening Democracy in a Fragmented World (Oxford and New York: Oxford University Press, 2002); and Polity IV reports, such as M, G. Marshall and B. R. Cole, Global Report 2009: Conﬂict, Governance, and State Fragility (Vienna: Center for Systemic Peace, 2009). 22 The term comes from G. O’Donnell, “Delegative Democracy?,” Journal of Democracy 5 (1994): 55-69.

18

192 6.4.1. Holmes’ constitutional grammar

Chapter 6. Proceduralist theories

Before turning to Ely’s more elaborated account, consider brieﬂy Holmes’ account of constitutions as constitutive—rather than regulative—rules. As Holmes famously argued, the above-mentioned procedural conditions need not be constraining of democratic decision-making because they contain constitutive rules, rather than regulative ones.23 Put brieﬂy, while regulative rules regulate already existing activities (e.g. “smoking is forbiden”), constitutive rules make an activity possible for the ﬁrst time (e.g. “the bishop moves diagonally”). Hence, procedural rules are constitutive because, rather than regulating already existing democratic decisions-making procedures, they create and enable them, for they establish how, when, and by whom are democratic decisions to be made. In short, while substantive constitutional rules are regulative and may constrain the exercise of democratic power, procedural constitutional rules are constitutive and do not constrain the exercise of such power—on the contrary, they create and organize it. As Holmes puts it, “constitutional rules are enabling, not disabling; and it is thus unsatisfactory to identify constitutionalism exclusively with limitations of power.” 24 As we shall see below in section 6.5 in more detail, Holmes’ account of procedural constraints as constitutive rules does not solve the counter-majoritarian diﬃculty. Put brieﬂy, even though he is obviously right in pointing out that democratic decision-making is not possible without democratic rules that enable it, it does not provide a reason for why the deﬁnition, interpretation, enactment and amendment of such rules should be taken away from the hands of ordinary legislatures by entrenching them in a rigid constitution and being interpreted and enacted by constitutional judges. However, before turning to this and other problems, let us turn consider John H. Ely’s inﬂuential account of judicial review of the procedural condition of democracy. 6.4.2. Ely’s procedural constitutionalism In Ely’s inﬂuential account of constitutional adjudication, judicial review is to be limited to the establishment and maintenance of the procedural conditions required to ensure that the legislative process is fair and that all citizens can take part in it on an equal footing. Hence, rather than aimed at ensuring that the democratic process brings about certain substantive outcomes—notably, the protection of fundamental rights—, as in the accounts analyzed so far, courts are to adopt a “participation-oriented, representation enforcing approach to judicial review,” 25 thus correcting potential distortions of the democratic process and ensuring that it remains open to everyone, especially to those who might have a more disadvantaged position in it. In this account, constitutionalism—and, more notably, judicial review—need not be at odds with democracy as far as only procedural constraints are included in the constitution and enforced by courts. In short,
23 See Holmes, “Precommitment and the Paradox of Democracy.” The distinction comes from J. Searle, “How to Derive ’Ought’ from ’Is’,? Philosophical Review 73 (1964): 195-240. 24 Holmes, “Precommitment and the Paradox of Democracy,” p. 227. 25 Ely, Democracy and Distrust, p. 87.

6.4. Pure procedural constraints

193

contrary to the standard characterization of the Constitution as ’an enduring but evolving statement of general values,’ ... in fact the selection and accommodation of substantive values is left almost entirely to the political process and instead the document is overwhelmingly concerned, on the one hand, with procedural fairness in the resolution of individual disputes (process writ small), and on the other, with what might capaciously be designated process writ large-with ensuring broad participation in the processes and distributions of government.26 Let us see Ely’s distinction between “fundamental values” and “the process of representation,” the reasons advanced for this process-enforcing approach to judicial review and, in the next section, the problems arising from such an account. Ely rejects what he labels as the “fundamental values” thesis—namely, the idea that constitutionalism and judicial review are to be aimed at ensuring the protection of substantive rights. This rejection is both descriptive and normative. It is descriptive for he believes that the American constitution is mainly procedural and that no substantive constraints can be enforced on its grounds. But the argument is normative as well, aﬀecting both the rights-based instrumentalism and the precommitment-based account of substantive constitutional constraints. The criticisms, though very similar in both cases, operate in three diﬀerent levels—meta-ethical, empirical and epistemological. Regarding rights-based instrumentalism, he believes that it is ﬂawed because of its dependence on supposedly objective substantive rights. “We shall sense in many cases that although the judge or commentator in question may be talking in terms of some “objective,” nonpersonal method of identiﬁcation, what he is really likely to be “discovering,” whether or not he is fully aware of it, are his own values.” 27 Regarding the precommitment-based justiﬁcation, he rules it out because of its dependence on a supposed consensus among the citizenry. According to Ely, this consensus does not exist in practice though: “in fact there is no consensus to be discovered (and to the extent that one may seem to exist, that is likely to reﬂect only the domination of some groups by others).” 28 And, even if did exist, it would be unlikely to discover it, at least, for judges when getting down to cases. Ely mentions the opinions of Justices Brennan and Marshall in Furman v. Georgia, arguing that the death penalty was unconstitutional because it was at odds with contemporary community values.29 According to Ely, it is unlikely for a consensus on those “community values” to exist but, even if it existed, “between courts and legislatures, it is clear that the latter are better situated to reﬂect consensus [even if] legislatures are imperfectly democratic.” 30 Note that Ely is here limiting himself to those decisions aimed at striking down legislation, letting aside the common law context, where the court ﬁlls the gaps left by the legislature and applies its legislation to speciﬁc cases. Ely’s target is thus
26 27 28 29 30

Ely, Democracy and Distrust, p. 87. Ely, Democracy and Distrust, p. 44. Ely, Democracy and Distrust, p. 63. Furman v. Georgia, 408 U.S. 238 (1972). Ely, Democracy and Distrust, p. 65. Ely, Democracy and Distrust, p. 67.

194

Chapter 6. Proceduralist theories

limited to the constitutional context, where the legislature has passed a law and the court is in a position to overturn it on the basis that the legislature does not reﬂect the consensus of the citizenry as reﬂected in the constitution. “A body that is not elected or otherwise politically responsible in any signiﬁcant way is telling the people’s elected representatives that they cannot govern as they’d like.” 31 According to Ely, judicial review turns out to be disturbing because the American constitutional system was originally designed to be mainly democratic, and because the subsequent two centuries have “strengthened the original commitment to control by a majority of the governed.” 32 To be sure, from a historical point of view, this is a rather controversial claim. Not so much because the Constitution may embrace substantive provisions on top of procedural ones, but rather because, as we have seen in chapter 1, the Founders did not intend the American constitution to be democratic at all. However, it is not totally incorrect if democracy is used in its current meaning as Ely appears to do. Also, for what it is worth here, the historical accurateness of the claim is irrelevant if what is to be analyzed is the democratic pedigree of judicial review from a point of view that assumes, as Ely does, that “whatever the explanation, and granting the qualiﬁcations, rule in accord with the consent of a majority of those governed is the core of the American governmental system.” 33 Assume that Ely’s case against the judicial review of parliamentary actions involving substantive rights is sound. Since judicial review is often attached to the protection of substantive rights against potential violation by the legislatures, does Ely’s position rule judicial review out altogether? Not that fast. According to Ely, judicial review “may be desirable or it may not, depending on the principles on the basis on which it is done.” 34 And, while it is not so when the protection of substantive principles is at stake, it turns out to be desirable concerning the very procedural conditions of democratic decision-making. Hence, according to Ely, judicial review is justiﬁed when it is limited to prevent those in power from using their position to block the channels of political change to those who remain outside (e.g., by limiting free speech or political association of opposing citizens) and from marginalizing discrete and insular minorities from taking part in the political process. Ely has in mind Justice Stone’s suggestion in the famous United States v. Carolene Products footnote four: “[P]rejudice against discrete and insular minorities may be a special condition, which tends seriously to curtail the operation of those political processes ordinarily to be relied upon to protect minorities.” 35 Ely provides a three-fold case for the participation-enforcing conception of constiEly, Democracy and Distrust, pp. 4-5. Ely, Democracy and Distrust, p. 7. 33 Ely, Democracy and Distrust, p. 7. 34 Ely, Democracy and Distrust, p. 5. 35 United States v. Carolene Products Co., 304 U.S. 144, 153 n. 4 (1938). Of course, as Tushnet and Waldron have pointed out, not every minority deserves this special treatment, for the few supporters of going back to the gold standard may lose systematically in every election without for that reason becoming a discrete and insular minority or turning to be marginalized. As Tushnet points out, “ “Every law overrides the views of the minority that loses ... We have to distinguish between mere losers and minorities that lose because they cannot protect themselves in politics.” Tushnet, Taking the Constitution away from the Courts, p. 359.
32 31

6.4. Pure procedural constraints

195

tutionalism. Only the third one is worth considering for present purposes. However, let us mention the former two. The ﬁrst one is that, as a matter of fact, the American constitution is procedural in two senses: “on the one hand, with procedural fairness in the resolution of individual disputes ... and on the other, with . . . broad participation in the processes and distributions of government.” 36 ensuring As for the

attempts to entrench substantive values in the text, “have been ill-fated, normally resulting in repeal, either oﬃcially or by interpretative pretense.” 37 Ely concludes from this fact that preserving fundamental values is not an appropriate constitutional task. However that is too hasty a conclusion, for constitutional systems other than the American one—notably, the German Basic Law—give more weight to substantive values, often entrenching them against the very possibility of being amended. The second case is that the procedural approach, unlike the substantive one, is not inconsistent with representative democracy, American or otherwise. But that reason turns out to be circular, for the inconsistency between democracy and substantive constitutional constraints is precisely what is to be proved. The third case for why procedural fairness should be guaranteed through judicial review rather than through the ordinary legislative process is not that judges have more expertise in procedural matters for, according to Ely, others—notably full-time participants and legislators (being many of them lawyers themselves—can also claim expertise on how the political process allocates voice and power. The diﬀerence relies on the institutional position of courts vis-à-vis legislatures and on the sort of incentives that the former and the latter have to reinforce or diminish the fairness of the political process.38 According to Ely, constitutional adjudication of the procedural type is to be viewed as analogous to an “antritust” rather than an interventionist orientation to economic aﬀairs. That is, rather than dictate substantive outcomes, constitutional adjudicatio should be limited to prevent that the “political market” malfunctions. In turn, malfunctioning happens when those in power use their powerful position, ﬁrst, to block the channels of political change to those who remain outside (e.g., by limiting free speech or political association of opposing citizens) or, second, to prevent a minority from being taken into account in the decision making. Bearing this in mind, the problem with elected representatives is that they are judges in their own cause when having to ensure that the political market remains fair and competitive. They have a clear incentive to use their position while they remain in oﬃce to advance their own interests and to increase their political power (for example, by making the electoral system more majoritarian, or by manipulating the campaign ﬁnance system or through gerrymandering).39 By contrast, appointed
Ely, Democracy and Distrust, p. 87. Ely, Democracy and Distrust, p. 88. For a criticism of this strictly procedural reading of the constitution, see Lawrence H. Tribe, “The Puzzling Persistence of Process-Based Constitutional Theories,” Yale Law Journal 89 (1980): 1063-1080. pp. 1065-1067. 38 Rawls also believed that this is how the issue of judicial review vis-à-vis parliamentary supremacy ought to be addressed—namely, by considering the incentives of the agents according to their institutional positions. See Rawls, A Theory of Justice, p. 198. 39 The point is also made by Elster: “A majority government will always be tempted to manipulate political rights to increase its chances of reelection. If it is free to change the timing of the election, it may choose a moment when economic conjuctures are favorable. If electoral district
37 36

196

Chapter 6. Proceduralist theories

judges are less prone to conﬂicts of interests of this sort because their permanence in oﬃce does not depend on the political cycle (even more when they enjoy life-long tenure). Moreover, unlike MPs, they do not control the political agenda. In a nutshell, judges are neither more skilled nor more able to grasp the procedural principles constitutive of the democratic process. However, they enjoy an institutional position that make them less prone to self-serving bias and more independent in their assessment of procedural principles.

6.5. Criticisms
Four problems with procedural theories are raised in what follows. Before proceeding, a caveat is in order though. Procedural theories cannot exhaust the set of rights usually linked to constitutionalism, for there are a number of rights that are usually included in a constitution yet are completely independent from the democratic process, such as reproductive rights. Of course, this cannot be a criticism of procedural theories, for this is precisely what they argue for—that the constitution should only include provisions related to the constitutive conditions and preconditions of the democratic process. However, it is important to note that, precisely for that reason, proceduralism only provides a rationale for a tiny version of constitutionalism—a much more tiny version than the one we see at work in most of the current legal systems around the world. Just to mention a notable example, proceduralism is unable to provide for what is usually taken to be the most important aim of constitutional adjudication—namely, the protection of process-independent yet fundamental rights from potential abuses by the legislatures and other branches of the state.40 6.5.1. Paradox of preconditions Consider ﬁrst the so-called “paradox of democratic preconditions” that follows from preconditions-based accounts.41 Assume that these accounts are correct and that democracy cannot work properly unless a number of substantive preconditions—such as the above-listed ones (§ 6.3)—are ensured via constitutional entrenchment. As noted above, even if substantive, these preconditions need not be justiﬁed as process-independent constraints. They can be justiﬁed because of their relationship to democracy, that is, as necessary for its proper functioning. Now, if such preboundaries have to redrawn because of population changes, the government may try to do so to its advantage. If the majority is free to change the electoral system—for example, proportional representation versus single-member districts—it may exploit this possibility for strategic purposes. if voters have to be registered before they can vote, the government may have an incentive to make registration more diﬃcult to disenfranchise de facto some of those who would have voted for the opposition ...” J. Elster, “Majority Rule and Individual Rights,” S. Shute and S. Hurley eds., On Human Rights: The Oxford Amnesty Lectures 1993 (New York: Basic Books, 1993), p. 182. 40 This is why Waldron refers to these theories as “non-core” cases. See Waldron, “The Core of the Case Against Judicial Rveiw,” pp. 1401-1406. 41 See C. Nino, The Constitution of Deliberative Democracy (New Haven: Yale University Press, 1997); J. Bohman, “Survey Article: The Coming of Age of Deliberative Democracy,” Journal of Political Philosophy 6 (1998): 400-425, pp. 403-404; and J. L. Martí, La república deliberativa. Una teoría de la democracia (Madrid: Marcial Pons, 2006), pp. 115-127.

6.5. Criticisms

197

conditions are constitutionally entrenched, then they are taken away from the hands of the legislatures, thus reducing the scope of what can be decided democratically. In short, if democracy is to work properly, a number of substantive preconditions are to be constitutionally enshrined, but then such preconditions are insulated from democratic decision-making and democracy turns out to be downsized as a result. As Nino puts it, “It seems that if we make provision of all these resources preconditions for democracy ... there will be very few questions for democracy itself to answer.” 42 Even further, it is reasonable to believe that the more ensured these preconditions are, the better the democratic process will work and, in turn, the less issues will be left for democracy to decide. Hence, as Bohman points out, “[D]emocracy could become a very robust method for deciding very little. Or, it could also easily become a very weak procedure for deciding too much.” 43 Even though both Nino and Bohman refer to theories of deliberative democracy, the same problem pervades preconditions-based theories of constitutionalism, which also fall into the paradox of preconditions. Hence, even if these accounts put democracy ﬁrst and deﬁne substantive constitutional constraints as subordinated to the democratic process, in their attempt to ensure the proper functioning of democracy, they end up insulating key political issues from what can be decided democratically, thus severely limiting the scope of democracy. The more issues are constitutionally entrenched, the better democracy functions, but the less issues are left for citizens to decide. And, conversely, the more issues are left for citizens to decide, the less issues are constitutionally entrenched, and the worse democracy functions.44 The result is paradoxical because, in order to ensure the adequate working of democracy, a fundamental principle of democracy is diminished—namely, that in a democracy, as Nadia Urbinati puts it, “all issues [are] an object of public evaluation and all values a matter of opinion and consent.” 45 An alleged solution to the paradox consists in trying to reach a balance between the requisites of democracy and its actual operation. Thus, according to Nino, “We must not try to make the democratic procedure so perfect, by strengthening its preconditions to the maximum, that its scope of operation shrinks so much that it embraces almost nothing but coordination problems as the direction of traﬃc.” 46 Rather, we should try to establish a baseline of preconditions and leave the remaining issues—including additional preconditions—open for citizens to decide. “There is a certain baseline below which the democratic process has lost all capacity to improve itself. Above the baseline, democracy replenishes itself, working for the fulﬁllment of its own preconditions.” 47 To be sure, Nino’s solution can be useful, since at the end of the day we may have to reach a balance between ensuring that some basic democratic preconditions are met—so the democratic process does not turn into a pantomime—and leaving enough issues open for decision—so democracy does not
42 43 44 45 46 47

Nino, The Constitution of Deliberative Democracy, p. 139. Bohman, “The Coming of Age of Deliberative Democracy,” p. 404. See Martí, La república deliberativa, p. 117. N. Urbinati, “Unpolitical Democracy,” Political Theory 38 (2010): 65-92, p. 65. Nino, The Constitution of Deliberative Democracy, p. 139. Nino, The Constitution of Deliberative Democracy, p. 140.

198

Chapter 6. Proceduralist theories

turn into mere window dressing by being emptied of content. This is what we do at the individual level when we reach a balance between ensuring the preconditions of our autonomy (e.g., by taking care of our health, so we can keep doing things autonomously) and exercising our autonomy (e.g. by doing risky things that may worsen our health).48 However, two problems arise with this balancing solution. First, as José Luís Martí has pointed out, Nino’s solution does not solve the paradox. It only provides a point of balance between the two extremes of democratic decision-making being emptied of content, on the one hand, and giving up the preconditions that ensure the fairness of the democratic process altogether, on the other.49 Second, Nino’s account assumes that a clear-cut set of democratic preconditions can be settled beyond dispute. However, disagreement is as pervasive regarding democratic preconditions as it is regarding non-relational goals, as we shall see immediately below in section 6.5.2. Nino’s account is correct in presenting the paradox—unlike Harrison’s alleged solution to Wollheim’s paradox, for example—as a trade-oﬀ between two issues of very diﬀerent sorts—procedural and substantive. However, his account does not quite rightly show that the procedural issue only triggers if disagreement on substantive issues exists, including disagreement on democratic preconditions such as civil rights and social and economic conditions. It seems very reasonable to claim that, unless a number of conditions are met, political decisions cannot be said to be truly democratic. However, as it will become clear in the next section, citizens disagree on such preconditions, and even on whether preconditions are required for the good working of democratic institutions at all. The procedural issue triggers precisely because (even more, only if) such disagreements exist—as a way to tackle them and reach a common course of action not only on the process-independent goals that are to be advanced by democratic means, but also on the preconditions that are to be met so as to improve the quality of the democratic system. Thus seen, the paradox dissolves.

6.5.2. Disagreements, yet once again It has been often argued that, even though disagreements are inherent to substantive political issues, they need not pervade the decision-making procedures used to deal with such disagreements. According to Anthony Giddens, Gaddaﬁ would have commented in a personal interview that citizens in Western countries disagreed on political issues as much Lybian citizens. To which Giddens would have replied that Gaddaﬁ was right, except regarding democracy, on which virtually everyone agreed as the best political system.50 In his review of Rawls’ Political Liberalism, Stuart Hampsire also defended this view. “Whithin diﬀerent moralities, liberal and conservative, the fairness of the actual outcome of a conﬂict will be evaluated diﬀerently, even though both sides recognize the fairness of the adversarial process. Outcomes
48 49 50

See Martí, La república deliberativa, pp. 125-126. See Martí, La república deliberativa, pp. 123 ﬀ. Unfortunately, I have not been able to ﬁnd out the source.

6.5. Criticisms

199

are by their nature open to dispute, but processes need not be.” 51 However, there are good grounds to believe that this need not be the case both regarding democratic preconditions and procedures. Consider possible disagreements on democratic preconditions ﬁrst and on procedures next. A. Disagreements on preconditions. To be sure, it is not the same to discuss which rights should be entrenched as a matter of justice, or which economic variables are to be maximized, rather than asking which constitutional rights are needed for democracy to work properly or below which threshold of GDP per capita and above which rate of inequality is democracy endangered. However, if we assume contra Habermas that, as we have seen above (§ 6.3), democratic preconditions are substantive, it is diﬃcult to see why the problems with instrumentalism raised in section 4.3.2 should not arise here as well. In the end, if citizens disagree on substantive issues—such as reproductive rights, immigration policies or health care schemes—they are likely to disagree as well on the preconditions needed for the good working of democracy—such as whether public schooling should be provided, whether hate speech should be guaranteed, whether borders should be closed so as to protect national identity, or whether hedge funds should be banned or regulated so as to prevent them from lobbying governments. Disagreements on procedures can take place at diﬀerent levels. Consider some of them. First of all, citizens may discuss whether democratic preconditions are needed at all. For example, for many democrats, democracy consists on a set of minimal rules—such as universal suﬀrage, fair elections and majority rule—that enable government formation and that are suﬃcient for a political system to qualify as democratic, while other citizens reasonably see this minimalist conception as aberrant and the resulting political system as a democratic pantomime.52 Second, even if citizens agreed that some preconditions are needed for the proper functioning of democracy, they would still disagree—sometimes reasonably, sometimes not so much—on which precise preconditions are implied by democracy. As we have seen in chapter 1, William Findley believed that “Perfect democracy required perfect equality, not just before the law but also in wealth, in property.” 53 However, many citizens would see this view as way too demanding, thus arguing for diﬀerent of combinations of rights and social conditions. Third, even if citizens agreed on the set of preconditions required by democracy, this counterfactual only applies if they also share a common conception of
Hampsire also claims elsewhere that “Universal agreement can be expected, in the name of rationality, only on the methods of fair argument which will arbitrate between the diﬀerent answers to these questions [of substantive justice], when an answer is needed for public purposes and social arrangements.” Both quotes come from J. Cohen, “Pluralism and Proceduralism,” Chicago-Kent Law Review 69 (1993-1994): 589-618, p. 589 and 590. 52 Actually, it could turn out for the relationship between democratic procedures and democratic preconditions to be empirically unsound. hence empirical studies could eventually show that participation was causally unaﬀected by the preconditions these theories have posited, as Brettschneider has pointed. However, I am not sure whether that would make disagreement among citizens disappear. See Brettschneider, Democratic Rights, p. 14. 53 Quoted in Shankman, Crucible of American Democracy, p. 4.
51

200

Chapter 6. Proceduralist theories

democracy, for diverse conceptions of democracy—direct or representative, federal or centralist, presidential or parliamentary, with a majoritarian electoral system or with PR—entail diverse sets of preconditions. However, as we shall see in the next epigraph, citizens are far from endorsing the same form of democracy. Accordingly, they also endorse very diﬀerent views on democratic preconditions. Fourth, even if citizens could overcome these disagreements and eventually reach a consensus on the set of preconditions that are to be constitutionally entrenched in order to ensure that democratic institutions work adequately, the above-mentioned problems of scope are likely to arise regarding preconditions as much as regarding process-independent goals. First, in the form of incompatibilities and trade-oﬀs not only between preconditions (for example, between free speech and the right to security), but also between preconditions and constitutive conditions of democracy (for example, between the right to association and the strict regulation or even ban on economic organizations such as the above-mentioned hedge-funds).54 B. Disagreements on procedures.55 Some critics of Ely’s account or procedural constitutionalism, such as Gargarella and Nino, have focused on his pluralist conception of democracy.56 Some others, such as Ackerman, have criticized Ely’s focus on discrete and isolated minorities rather than on unorganized and difuse majorities that might be more vulnerable.57 As Gaetano Mosca had put it in the 1930s, The domination of an organized minority ... over the unorganized majority is inevitable. The power of any minority is irresistible as against each single individual in the majority, who stands alone before the totality of the organized minority. At the same time, the minority is organized for the very reason that it is a minority.58 To be sure, these are sensible concerns. However, they all misﬁre, for Ely could easily accommodate them just by saying that constitutional adjudication should be limited to the protection of the constitutive features (though he would add preconditions as well) of the democratic process, whatever those features turn out to be. Now, these criticisms and Ely’s possible reply make a less easily-avoidable problem of Ely’s account emerge—namely, that scholars as much as citizens have divergent conceptions of democracy and, therefore, endorse divergent sorts of procedures as democratic ones. Citizens not only disagree on procedure-dependent or -independent substantive issues, they also disagree on procedural issues. That is, citizens disagree on whether abortion should be banned or not, on whether the inheritance tax should
The reasons behind these disagreement need not be very diﬀerent from the ones mentioned in section 4.3.2, epigraph B. 55 “Procedural and substantive values come ... as parts of a package; contrary to Hampshire’s claim, moral pluralism does not drive a wedge between them.” Cohen, “Pluralism and Proceduralism,” p. 591. 56 See Nino, The Constitution of Deliberative Democracy; Gargarella, La justicia frente al gobierno; For a comparison, see S. Linares, La (i)legitimidad democrática del control judicial de las leyes (Madrid: Marcial Pons, 2008), pp. 129-135. 57 B. Ackerman, “Beyond Carolene Products,” Harvard Law Review 98 (1985): 713-746. 58 G. Mosca, The Ruling Elite, A. Livingston, ed. (New York: McGraw-Hill, 1939), p. 53.
54

6.5. Criticisms

201

be abolished or not or on whether illegal immigrants should be regularized or prosecuted. But, even when they acknowledge that they disagree on these issues, that they have to reach a common course of action nevertheless and that they want this course of acton to be reached by democratic means, they also disagree on which decision-making procedures count as democratic and should therefore be used.59 Of course, these disagreements do not jeopardize the very idea of democracy. They only show that democracy is an essentially contested concept that leaves room for many possible institutional arrangements and decision-making procedures.60 Consider only a few of them: bicameralism vs. unicameralism, presidentialism vs. parliamentarianism, proportional representation vs. majoritarianism, participatory budgets vs. cabinet budgetary discretion, subsidiarity vs. centralization, division of powers vs. parliamentary sovereignty, and so on and so forth. Incidentally, these disagreements debunk Holmes’ characterization of constitutional rules as constitutive rather than regulative rules and, for that matter, his analogy between constitutional rules and grammar rules. The analogy does not hold because, unlike the rules of grammar, democratic procedures are not neutral.61 Diﬀerent procedures produce diﬀerent allocations of power, thus beneﬁting some citizens at the expense of others. In short, they are not a matter of mere coordination but a political and highly controversial one. Hence, it is perfectly reasonable to see them as a subject of harsh and lively disagreement and political debate for the reasons pointed out in section 4.3.2—brieﬂy restated, citizen’s imperfect ability to get to know which procedures will produce a fairer allocation of power, the slippery and contestable nature of democratic procedures themselves and the interests that unavoidably pervade debates on which procedures should be enforced and, thus, who will beneﬁt from them. In short, the alleged neutrality and constitutive nature of democratic procedures cannot justify their constitutional entrenchment, for they allocate power very diﬀerently depending on the speciﬁc procedures that are enforced and, accordingly, they are neither neutral nor merely constitutive.62 In short, as Waldron has put it, “Like all rights ... participation is also an appropriate target or subject matter for authority. In politics, the right to participate (its nature and limits) is one of the things on which we disagree and about which we have to decide. But, unlike other rights, the right to participate also presents itself as a possible answer (or part of answer) to the question of authority.” 63 This raises serious doubts on any attempt to entrench democratic procedures in a rigid constitution and enforce them via judicial review. To be sure, there is a clear distinction between procedural and substantive issues, as Ely protests against those who may reply that
See Herzog, “Democratic Credentials,” p. 468, n. 2; and Jane S. Schacter, “Ely and the Idea of Democracy,” Stanford Law Review 57 (2004): 737-760. 60 For the idea of an “essentially contested concept” see W. B. Gallie, “Essentially Contested Concepts,” Proceedings of the Aristotelian Society 56 (1956): 167-198. 61 It may be objected that the grammar rules are not neutral either, for they beneﬁts some speakers over others. Thus seen, the analogy would hold, but Homes’ point when using the analogy, which was show the constitutive and neutral character of constitutionally entrenched procedures would not hold. 62 See Sánchez-Cuenca, Más democracia, menos liberalismo, pp. 136-140. 63 Waldron, Law and Disagreement, p. 249.
59

202

Chapter 6. Proceduralist theories

participation is also a value and that, accordingly, judicial adjudication aimed at protecting it is not very diﬀerent from judicial adjudication aimed at protecting any other value.64 However, that participation is procedural rather than substantive does not turn it into any less immune to disagreement. These disagreements provide, in turn, a strong case for leaving its deﬁnition, instantiation and application on the hands of the legislatures rather than insulated from them.65 C. Disagreements on values. To close this section, consider a criticism of procedure-based constitutional constraints from the opposite end, namely, from among those who claim that Ely’s account falls short of an adequate and complete concept of constitutionalism for procedural constitutional provisions ought to be complemented by substantive ones. According to these critics, it is not possible to trace a clear-cut distinction between procedural and substantive issues because procedural issues imply a number of core substantive values that provide the rationale for why democratic procedures are worth being protected and constitutionally entrenched.66 For example, the values of mutual respect, equality or autonomy provide good normative grounds for democratic procedures and their constitutional protection, as Corey Brettschneider has claimed.67 Now, once we assume that democratic procedures are grounded on some values that serve as a litmus test against which to evaluate the fairness of the democratic procedures—or so the argument goes—, there is no reason why we should not use such values to evaluate the fairness of the outcomes and to justify the constitutionalization of some substantive constraints as well.68 For example, if we consider unfair a democratic decision that disenfranchise a minority of the population because it prevents some individuals from exercising their public autonomy, why should we consider fair a decision that prevents them from exercising their private autonomy (for example, a bill that prevents them from opting out of a compulsory organ donation program)?69 This argument diﬀers from the one put forward by Richard Bellamy, who claims
64 “Participation itself can obviously be regarded as a value, but that doesn’t collapse the two modes of review I am describing into one. As I am using the terms, value imposition refers to the designation of certain goods (rights or whatever) as so important that they must be insulated from whatever inhibition the political process might impose, whereas a participational orientation denotes a form of review that concerns itself with how decisions aﬀecting value choices and distributing the resultant costs and beneﬁts are made.” Ely, Democracy and Distrust, p. 75n. 65 This may obviously raise a regress on procedures, as it has been often argueed. See T. Christiano, “Waldron on Law and Disagreement,” Law & Philosophy 19 (2000): 513-543; A. Rieger, “Voting on Voting Systems, or the Limits of Democracy,” Analysis 71(2011): 641-642. For a response, see Waldron, Law and Disagreement, pp. 302-306. 66 See Tribe, “The Puzzling Persistence of Process-Based Constitutional Theories,” Yale Law Journal 89 (1980): 1063-1080; Bretschneider, Democratic Rights; Cohen, Pluralism and Proceduralism;” C. Fabre, Social Rights Under the Constitution. Government and the Decent Life (Oxford: Oxford University Press, 2000), p. 277; A. Kavanagh, “Participation and Judicial Review: A Repy to Jeremy Waldron,” Law & Philosophy 22 (2003): 451-486, pp. 468-469. 67 See Brettschneider, Democratic Rights. 68 For example, Tribe criticizes that “the constitutional theme of perfecting the process of governmental decision is radically indepeterminate and fundamentally incomplete. The process theme by itself determines almost nothing unless its presuppositions are speciﬁed, and its content supplemented, by a full theory of substantive rights and values—the very sort of theory the process perfecters are at such pains to avoid.” See Tribe, “The Puzzling Persistence of Process-Based Constitutional Theories,” p. 1064. 69 To be sure, these sorts of principles might fall short of addressing all the outcomes of the

6.5. Criticisms

203

that it is not possible to state what counts as a fair democratic process without staing what counts as a fair outcome. According to Bellamy, “you cannot judge whether the process is fair without a view of what counts as a fair outcome, and cannot judge a fair outcome without referring to some account of fundamental values.” 70 Rawls’ notion of “pure procedural justice” serves as a counterexample for Bellamy’s argument though.71 The usual example is that of a lottery, in which there is no criterion for what counts as a just outcome other than the procedure itself. But the evidence provided above in section 4.3.3 show that citizens clearly distinguish between outcomes and procedures, and that they often judge that political procedures are fair even if they consider that the outcome they bring about are not. Turning back to the previous argument, the attempt to settle a set of values behind democratic procedures and to infer a set of constitutional rights from them seems to be as subject to disagreements as the attempt to settle a set of disagreement-free procedures, and probably much more. Consider universal suﬀrage, which appears to be to great extent beyond controversy and disagreement (at least, in well-established democracies). Now, controversy and disagreement ﬁercely arise again when trying to settle the values behind it. Hence, while some would agree on the values pointed out by Brettschneider, others would embrace universal suﬀrage for completely diﬀerent reasons, such as its ability to bring about economic growth or political stability or to reduce the likelihood of social revolutions breaking out. As Jacques Maritain (1949, p. 9) liked to tell, at one of the meetings of a UNESCO National Commission where human rights were being discussed, someone expressed astonishment that certain champions of violently opposed ideologies had agreed on a list of rights. “Yes”, they said, “we agree about the rights but on condition that no one asks us why”. That “why” is where the argument begins. 6.5.3. Constitutive rules and rigidity Procedures-based theories provide a powerful case for constitutionalism—even if only for a tiny version of it. The reason is that, rather than grounding constitutional rigidity and judicial review on the protection of much disputed non-relational goals, these theories justify them only in order to protect the constitutive conditions (and perhaps preconditions) of democracy proper. Thus, as Holmes puts it, while constitutional entrenchment of non-relational goals can be easily seen as a constraint on democratic decision-making, constitutionally entrenched democratic procedures can be less easily seen as constraints, for they ensure the constitutive rules of democracy—i.e., conditions without which a political system cannot be democratic at all. Just as it is not possible to play chess unless the rules of the game are settled in advance, it is not possible to make democratic decisions unless the democratic rules
political decision-making, but would nevertheless be unreasonable to limit their scope to procedural issues. 70 Bellamy, Political Constitutionalism, p. 110. See also R. Dworkin, A Matter of Principle (Oxford: Clarendon Press, 1986), pp. 57-69. 71 See J. Rawls, A Theory of Justice (Cambridge: Harvard University Press, 1971), section 14.

204

Chapter 6. Proceduralist theories

and institutions are settled before. In short, constitutionalism so deﬁned, far from being at odds with democracy, it enables it and is required by it. Now, Holmes’ argument falls into a non sequitur. To be sure, Holmes is right (1) in distinguishing constitutive rules from regulative rules, (2) in depicting democratic procedures as the former, (3) in distinguishing constitutive rules, including democratic procedures, from the exercise of such rules in ordinary politics and (4) in depicting the former as conceptually prior to the latter. However, he does not explain why the former should be normatively prior to the latter in the sense of being constitutionally protected from it. That is, he does not explain why the deﬁnition, organization, enactment and amendment of democratic rules should be taken away from the hand of citizens and their representatives. Holmes is right when he claims that “Decisions are made on the basis of pre-decisions ... When entering the voting-booth, for instance, voters decide who shall be president, but not how many presidents there shall be.” 72 Surely, it is logically inconsistent for citizens to make decisions about D according to certain decision-making rules R and about those very rules R at the same time. But the inconsistency disappears as soon as decisions about D and decisions about R are made at diﬀerent times. And, once the inconsistency disappears, Holmes’ argument fails to explain why, unlike decisions decisions about D, decisions about R should be constitutionally entrenched and be left in the hands of courts rather than parliaments. If democratic rules were rules of pure coordination and were out of the discussion, Holmes’ conclusion would probably follow. However, three problems block the argument. First, far from being an all-or-nothing issue and requiring a clear-cut set of decision-making rules, democracy is consistent with a wide range of rules and institutions (even if obviously not with all; a country without elections or where protest is prosecuted can hardly count as democratic). Second, democratic rules are far from being merely of coordination for, as we have seen, diﬀerent procedures produce diﬀerent allocations of power, thus beneﬁting some citizens and groups at the expense of others. Third, the issue of which democratic rules are fair—and, therefore, of which allocations of power are fairer—remains open to reasonable (and sometimes not-so-reasonable) disagreement. Why should citizens and their representatives be prevented from being able to settle such disagreements democratically and thus from being able to decide by which precise rules do they want to govern themselves? In short, it is obviously the case that democratic decision-making requires a set of rules specifying who ought to make political decisions, how are decisions to be made and what are those decisions to be made about. Holmes is right in claiming that democratic rules are not constraining of democratic decision-making. But constitutionally entrenched democratic rules are, for there is no reason why they ought to be entrenched against deﬁnition, enactment and amendment through the ordinary political process.

72

Holmes, “Precommitment and the Paradox of Democracy.”

6.5. Criticisms

205

6.5.4. Nemo Iudex in Sua Causa Assume that the doubts presented thus far are sound, that is, that citizens disagree about democratic procedures and preconditions as much as they disagree about non-relational issues and that legislatures stand in a better institutional position vis-a-vis constitutional courts to address such disagreements. Now, even if legislatures may address disagreements better than courts because of their representative and openly confrontational and deliberative nature, they are also more prone to conﬂicts of interests regarding procedural issues than courts because such issues aﬀect them decisively. Thus, one of the main reasons for judicial review advanced by Ely is precisely that courts are in a better institutional position vis-à-vis legislatures regarding the impartial monitoring of the political process. Sure, it is perfectly reasonable for citizens to disagree on the electoral reform, the setup of the congress or the powers of the branches of government. Likewise, it is perfectly reasonable for them to think that if an agreement existed on these issues, constitutional entrenchment and judicial review would be good candidates to ensure them, but since disagreement persists accross the board it is better to leave them in the hands of their representatives. However, this is too naive an account of parliamentary representation, for it assumes that citizens and their elected representatives have common interests and that no information asymmetries exist among them. Now, once we relax this assumption and assume that politicians may have particular interests and do have asymmetric information, it follows that they may also have a strong interest to use their powerful position to strengthen their position and enlarge their power. For example, a large party will have a strong interest to reduce the number of seats in the parliament, to make the electoral system less proportional and to improve presidential powers. In short, according to Ely, judges and elected oﬃcials stand in a similar institutional position, have similar incentives and are equally likely to be biased regarding substantive issues. However, regarding the design of political procedures, judges are less prone to have conﬂicts of interests and to suﬀer self-served biases than elected oﬃcials. They are “comparative outsiders” because they have a weaker incentive to design political procedures so as to increase their power and advance their political interests than the latter.73 In a nutshell, elected oﬃcials are judges in their own cause while courts are not (or are less so). Now, as Bellamy and Shapiro have pointed out, how convincing is this argument in the light of the US Supreme Court decision in Bush v. Gore in 2000?74 This case is especially appropriate to analyze Ely’s argument for at least three reasons. First, it was limited to a procedural controversy regarding the electoral process in the presidential election and, more precisely, regarding the method for recounting ballots in the state of Florida. Second, the election in both Florida and in the country was too close and the parties too subject to conﬂicts of interests and self-serving bias to reach a decision that could be seen as remotely fair (for example, many of the
Ely, Democracy and Distrust, p. 103. See Bellamy, Political Constitutionalism, p. 118-120; “Introdution,” in Constitutionalism and Democracy, R. Bellamy, ed. (Aldershot: Ashgate, 2006), pp. xxix-xxx.
74 73

206

Chapter 6. Proceduralist theories

oﬃcials involved were political appointees and the state governor was a brother and a fellow member party of one of the candidates). Third, a tie-breaking third party was available so as to avoid conﬂicts of interests and ensure that a fair decision was reached—namely, the Supreme Court. However, as the public statement signed by over ﬁve hundred American las professors denounced, it was quickly shown that, rather than an impartial and uninvolved third party, the Supreme Court had acted as a party to the more general political disagreement, for justices were as aﬀected by the outcome of the election as as every other American citizen. As part of the political system, Supreme Court justices are dramatically aﬀected by the way the political system is organized and how power is allocated—let alone the fact that they tend to disagree with each other as much as citizens do, as their use of majority rule to reach decisions shows.75 As Bellamy puts it, “Because they are at the apex of the system, they are judges over their own constitutional competence on the matter too.” 76 Things being so, then there is no reason to give them priority over citizens that are equally aﬀected by the decision. As Waldron puts it, nemo iudex can be replied with another Latin adagio, namely, quod omnes tangit, ab omnibus approbetur.77 It may be replied that the US Supreme Court decision in Bush v. Gore does not ﬁt Ely’s argument because it does not deal with “insular and discrete minorities.” However, three sorts of additional cases can be mentioned. First, cases in which constitutional courts have turned down campaign ﬁnance legislation aimed at curbing the inﬂuence of powerful economic groups on political parties and, in turn, on the political process. Second, cases in which constitutional courts have banned political parties representing minorities, such as the Kurds in Turkey or separatist Basques in Spain. And third, in the limit, cases in which constitutional courts have often backed not only the banning of political parties but also of coups against the democratic system. To be sure, constitutional courts have sometimes protected the democratic system from procedural injustices. But for each constitutional decision aimed at entrenching the democratic system, it is possible to point at a constitutional decision backing and thus legitimizing military coups, as it often happened in Latin America during the twentieth century.78 To be sure, these concerns do not settle the issue on whether courts or parliaments are more suitable and have more adequate incentives to protect the procedural conditions and preconditions of democracy. This issue is mainly empirical and context-dependent. This is the reason why Waldron acknowledges that his case against judicial review applies only under certain circumstances, e.g. countries with good working democratic institutions.79 Now, this caveat also poses a burden on
75 On internal disputes within the Supreme Court, see E. Lazarus, Closed Chambers: The First Eyewitness Account of the Epic Struggles Inside the Supreme Court (New York: Times Books, 1998). 76 Bellamy, Political Constitutionalism, p. 119. For an analysis of the self-interested motivations of the US Supreme Court justices, see F. Schauer, “Incentives, Reputation, and the Inglorious Determinants of Judicial Behavior,” University of Cincinnati Law Review 68 (2000): 615-636. 77 Waldron, Law and Disagreement, pp. 297-298. 78 See Linares, La (i)legitimidad democrática del control judicial de las leyes, p. 92. 79 See Waldron, “The Core of the Case against Judicial Review,” p. 1360, n. 1.

6.5. Criticisms

207

supporters of constitutionalism and judicial review, for they also have to admit that judicial review is to be rejected in those cases in which the constitution contains obviously unjust provisions or the constitutional court is openly corrupt. Hence, when discussed at a theoretical level, legislatures and courts are to be compared either at their best light or at their worst one, so as to avoid fallacies of asymmetry.80 And, when discussed at the level of implementation, contextual analysis seems the most cautious option to follow.

See Stacey, “Democratic Jurisprudence and Judicial review”, Oxford Journal of Legal Studies 30 (2010): 749-773, pp. 768 ﬀ.

80

Concluding remarks to Part II
Over the last decades we have witnessed an astonishing expansion of constitutionalism worldwide, to the point that 101 out of the 106 constitutions written between 1985 and 2002 include some form of judicial review. This expansion has gone hand-in-hand with the conceptual shift in the concept of democracy presented in the Introduction and analyzed in Part I of the dissertation as for the Early American Republic. According to this shift, constitutional rigidity and judicial review are nowadays seen not only as consistent with democracy but also as fundamental parts of any democratic system proper. For example, in the 2004 Report on Democracy in Latin America Guillermo O’Donnell addresses the relationship between democracy and judicial review as a dilemma between the European model and the American model of judicial review, asking which of them is more democratic rather than whether judicial review is democratic at all or not.81 At a philosophical level, a number of theories have been elaborated so as to justify the democratic pedigree of constitutionalism. Part II of the dissertation has carefully analyzed the most prominent theories among these and has found that they all face serious problems to justify constitutionalism from a democratic standpoint. Three prominent types of theories have been analyzed—pure instrumentalist theories, precommitment-based theories, and proceduralist theories. According to pure instrumentalist theories, constitutionalism need not be at odds with democracy because constitutional constraints are necessary for the democratic procedures to bring just outcomes about (chapter 4). According to precommitment-based theories, constitutional constraints need not be at odds with democracy because they are the upshot of democratically made precommitments (chapter 5). Finally, according to proceduralist theories, constitutional constraints need not be at odds with democracy because they set up the procedural rules and preconditions required for the proper functioning of the democratic system (chapter 6). A number of problems of all these theories have been analyzed. However, a common problem to all of them has been pointed out—namely, that they all fail to address adequately the fact of disagreement that, as we have seen in Part I of the dissertation, made democracy gain legitimacy and be increasingly demanded as the nineteenth century progressed. This fact is not a contingent, though resilient and widespread, feature of our political communities that we can leave aside when elabG. O’Donnell, “Notas sobre la democracia en América Latina,” Programa de Naciones Unidas para el Desarrollo, ed. La democracia en América Latina. El debate conceptual sobre la democracia (New York: PNUD, 2004).
81

210

Concluding remarks to Part II

orating normative theories of political authority. Rather, it is a necessary condition of politics proper, for if no disagreement existed no political authority—democratic or otherwise—would be required in the ﬁrst place. Hence, any normative theory of political authority that depends on elements that are beyond dispute (either actually or counterfactually) turns out to be self-defeating. Now, it may be asked what is the alternative to strong constitutionalism provided that it is so problematic from a democratic standpoint and that we have a strong commitment to democracy nevertheless. It is beyond the scope of this dissertation to make a full case for an alternative to constitutionalism. To be sure, despite of the diﬀerences between them, it is obviously the case that parliaments are better suited vis-à-vis constitutional courts to address diversity and disagreement. The Spanish Cortes Generales, the Russian Duma, the British House of Commons, the French National Assembly, the Japanese Diet, the American House of Representatives, they all include hundreds of representatives and are organized for ideological confrontation. As Walter Bagehot described the House of Commons in 1867, “Here are 658 persons, collected from all parts of England [sic], diﬀerent in nature, diﬀerent in look and language.” 82 However, it is worth pointing out that the Westminster model of parliamentary sovereignty is not the only available alternative to strong constitutionalism. There are many intermediate models, including the new commonwealth forms of constitutionalism described in section 3.2 above. Further, constitutionalism is not an all-or-nothing matter. A constitutional system in which some constitutional provisions are absolutely entrenched is completely diﬀerent from a system that only requires submission to referendum for amendment. Likewise, a system of judicial review in which courts have the last word on all constitutional matters and can overturn any statute that may be rendered unconstitutional completely diﬀers from a system in which courts, by contrast, can only issue declarations of incompatibility and the parliament retains ﬁnal authority on constitutional matters.83

W. Bagehot, quoted in J. Waldron, The Dignity of Legislation (Cambridge: Cambridge University Press, 1999), p. 34. 83 For a gradualist approach to judicial review and an attempt to measure the degree of the democratic objection as for diverse forms of judicial review, see Linares, La (i)legitimidad democrática del control judicial de las leyes, ch. 6.

82

Part III

The workplace

Chapter 7

Analytical and normative background
Part III of the dissertation analyzes the possible extension of democratic decision-making to the workplace, a sphere that in our liberal democracies is usually taken to be beyond the scope of democracy. It is divided into two chapters. Chapter 7 introduces and deﬁnes the main forms of workplace democracy (WD), focusing on the German co-determination system and cooperativism. Next, it reviews in some detail the core arguments for and against WD regarding its feasibility and eﬃciency, and analyzes the so-called “epistemic” and “propertarian” arguments against WD. Finally, Bowles and Gintis’ theory of contested exchange—which they use to show that the workplace is a political sphere in which power is exercised and that ought to be democratized accordingly—is critically analyzed. Chapter 8, in turn, puts forward a novel, republican case for WD that partly draws upon the democratic branch of republicanism analyzed in Part I, section 1.3.4. First, two competing views are analyzed—ﬁrst, that the availability of an exit option alone ensures workers’ freedom from arbitrary interference and, second, that ensuring this sort of freedom requires workplace regulation but not WD. Next, the republican case for WD is presented as straightforward as possible. Finally, it is argued that WD requires, from a republican standpoint, not only workers’ control but also workers’ ownership of the ﬁrm due to bargaining power asymmetries and moral hazard problems arising from the formal separation of ownership and control rights. It is thus shown that cooperativism is ceteris paribus preferable to other forms of WD in which ownership and control rights are formally separated, such as co-determination.

7.1. Introduction
The current economic crisis has triggered the interest in workplace democracy once again, 2012 having been proclaimed International Year of Co-operatives by the United Nations. Even though workplace democracy was a usual battle-cry among workers and a central research topic in industrial relations, labor economics and political theory in the 1970s and 1980s,1 since the early 1990s the topic has raised less and less attention among scholars, politicians and workers alike. However, the greater resilience of cooperatives to the current crisis—partly due to their greater wage ﬂexibility and lesser need to make proﬁts for investors and bonuses for man1

See D. Christie, “Recent Calls for Economic Democracy,” Ethics 95 (1984): 112-128.

214

Chapter 7. Analytical and normative background

agers—has strengthened their presence in the world economy and triggered once again the interest in this and other forms of workplace democracy.2 Early experiments with workplace democracy (henceforth, WD) can be traced back at least to the Industrial Revolution and have developed into many diﬀerent forms, such as cooperatives, workers’ councils, the Israli kibbutzim, the German co-determination system or the US Employee Stock Options Plans, and has also served as an umbrella for dubiously democratic arrangements such as quality circles, just-in-time production, total quality management, ﬂexible practices and so forth. However, even though its very concept is inevitably fuzzy and contested, in its minimal account WD can be deﬁned as follows:3 Workplace democracy.—A form of managerial organization where, if not all, a majority of workers, though not necessarily only them, have equal control rights over the management of the ﬁrm. A number of arguments have been traditionally advanced for and against WD. The aim of this chapter is to introduce the main forms of WD and to analyze in some detail the main arguments for and against it. Section 7.2 distinguishes between control rights by capital or by labor and between public and private ownership of the ﬁrm. Two main forms of WD can be found depending on whether democratic ﬁrms are owned by external investors or by workers themselves—the German co-determination system and cooperativism, which are respectively analyzed in sections 7.2.1 and 7.2.2 respectively. Section 7.3, in turn, reviews the main arguments for and against WD. Section 7.3.1 addresses its feasibility, considering that the proposal is not on the agenda, that it does not seem to be supported by workers themselves, and that there may be more pressing issues right now. Section7.3.2addresses a number of eﬃciency-based arguments for WD (greater eﬀort and information-sharing, lower unemployment output, and better vertical and horizontal monitoring), as well as against WD (greater decision-making costs, diﬃculties to attract external capital and to diversify investment, and likelihood of free-riding due to common ownership in large democratic ﬁrms). Sections 7.3.3 and 7.3.4 address in some detail the so-called “epistemic” and “propertarian” arguments against WD, respectively. According to the former: Why should workers be granted a say in their running when they often lack the expertise required to make informed decisions about these topics? According to the latter: Why should workers be granted control rights over the ﬁrm when they are not its
See J. Birchall and L Hammond Ketilson, Resilience of the Cooperative Business Model in Times of Crisis (Geneva: International Labour Organization, 2009); R. D. Lansbury, “Workplace Democracy and the Global Financial Crisis,” Journal of Industrial Relations 51 (2009): 599-616. 3 For a sample of notably diﬀerent deﬁnitions see Christie, “Recent Calls for Economic Democracy”; R. Dahl, A Preface to Economic Democracy (Berkeley and Los Angeles: University of California Press, 1985); S. Bowles and H. Gintis, “A Political and Economic Case for the Democratic Enterprise,” Economics and Philosophy 9 (1993): 75-100; and N.-hê Hsieh, “Survey Article: Justice in production,” Journal of Political Philosophy 16 (2008): 72-100. For exhaustive overviews of different forms of WD, see H. Hansmann, The Ownership of Enterprise (Cambridge, Mass., Harvard University Press, 2000) and G. K. Dow, Governing the Firm. Workers’ Control in Theory and Practice (Cambridge: Cambridge University Press, 2003).
2

7.2. Deﬁnition and models

215

owners? Finally, section 7.3.5 analyzes in much more detail Bowles and Gintis’ theory of contested exchange, which they use to make a case for WD.

7.2. Deﬁnition and models
Democratic ﬁrms are often introduced as a third way between capitalist and socialist ﬁrms, which are often distinguished along the dimension of ﬁrms’ ownership. Thus, while ﬁrms are owned by private investors under capitalism, they are publicly owned under communism. However, as Dow points out, economic systems also diﬀer along another dimension namely, control over production activities.4 Here the distinction is not between public and private ownership of the ﬁrm, but rather between ﬁrms controlled by capital suppliers and labor suppliers. Hence, while the ownership of the ﬁrm refers to the ownership of non-human assets, such as ﬁnancial wealth or physical assets, control rights refer to the authority to make managerial decisions, which may be delegated to directors and managers who, nevertheles, are elected and remain accountable to those holding ultimate authority ﬁrms, shareholders; in democratic ﬁrms, workers. Even though it is a matter of discussion whether control rights and ownership of the ﬁrm can be sharply separated in practice, as we shall see in detail below in section 8.5, these are two very diﬀerent dimensions that, combined, provide the following matrix.
Private ownership Control by capital Control by labor Capitalist ﬁrm Private democratic ﬁrm Public ownership Socialist ﬁrm Public democratic ﬁrm

in conventional (and large)

Table 5. Ownership and control

As we can see, four alternatives follow from the combination of these two dimensions. On the ﬁrst row we have capitalist and socialist ﬁrms, which diﬀer in whether assets are privately or publicly owned but have in common control by capital either private or public. By contrast, ﬁrms pertaining to the second row have in common that they are controlled by workers, even if they can be privately or publicly owned. An actual instance of the latter case could be found in Yugoslavia under Tito, where the assets of ﬁrms were publicly owned by the state but workers within each ﬁrm had substantial control over their use,5 while a theoretical instance of it has been developed by David Schweickart idea of “economic democracy.” 6 Here I will focus on the ﬁrst column of the second row
4 5

namely, private demo-

cratic ﬁrms, which are also controlled by workers but are nevertheless privately
G. Dow, Governing the Firm, p. 3. It has been widely discussed whether workers had actual autonomy or were rather controlled by the Communist Party, not only because many board seats were appointed by the Party but also because the Party had de facto control over managerial decisions. See J. E. Roerner, A Future for Socialism (Cambridge, MA: Harvard University Press, 1994). 6 See D. Schweickart, “Economic Democracy: A Worthy Socialism That Would Really Work,” Science and Society 56 (1992): 9-38.

216

Chapter 7. Analytical and normative background

owned. Depending on whether ownership and control rights are separated or not, they can be classiﬁed in two main groups. On the one hand, democratic ﬁrms in which capital is supplied by external investors but workers have control rights over the ﬁrm’s decisions. On the other hand, democratic ﬁrms in which workers have control rights but capital is also supplied by workers themselves, who ﬁnance collective asset ownership through debt contracting or by drawing upon their own savings. Thus, while ownership and control rights over the ﬁrm’s decisions are separated in the former case, they are kept together in workers’ hands in the latter case. I will take the German co-determination system and cooperativism as the most outstanding instances of the former and the latter, respectively. Before considering these two sorts of ﬁrms in detail, two caveats are in order though. First, the separation of ownership and control rights to which I refer here is not to be confused with the “separation of ownership and control” debate in managerial theory.7 This debate refers to the separation between a principal (shareholders, who own the ﬁrm) and its agent (directors and top managers, who exercise the control of the ﬁrm) and the limited control shareholders often have over the board of directors (and these, in turn, over top managers). To be sure, this problem also applies to democratic ﬁrms. The only diﬀerence is that, while the agent remains the same, workers rather than shareholders are now the principal. Hence, (at least in large democratic ﬁrms) managers are hired and ﬁred by a board directors, but board members are not elected and held accountable by shareholders (or at least not only by them), but rather by the workers of the ﬁrm. Second, it can be called into question whether control and ownership of the ﬁrm can be separated in practice (see section 8.5 below). On the one hand, the conventional concept of “ﬁrm ownership” includes not only the ownership of non-human assets used by the ﬁrm and the claim on the net income, but also control over the ﬁrm’s decisions and the right to sell the preceding rights to a third party. On the other hand, even though they are often believed to be inextricably linked, control rights can be and are often de iure separated from ownership. Consider three examples of this separation. First, the Yugoslav system of economic democracy, in which workers were granted control rights over ﬁrms that remained property of the state. Second, ﬁrms with employee stock options plans in which workers often have residual claims and contribute capital but lack control rights. And, third, the German co-determination system, in which workers have control rights but assets are owned by shareholders, who retain exclusive claims over net income. 7.2.1. Co-determination The German co-determination system is the closest actual model to formal separation between ownership and control rights, for it gives workers control rights without requiring them to make any equity investment in the ﬁrm. The co-determination act, in force since 1976, makes compulsory for all limited liability ﬁrms with over
The locus classicus is A. Berle and G. C. Means, The Modern Corporation and Private Property (New York: MacMillan, 1932)
7

7.2. Deﬁnition and models

217

2,000 workers to have a two-tier board structure. On the one hand, the management board directly runs the ﬁrm, making day-to-day decisions. On the other hand, the supervisory board makes the more general decisions, such as the ratiﬁcation of important investments or the approval of the annual budget. In the management board only shareholders’ executives are represented, while in the supervisory board both shareholders and workers are represented on a “near-parity” basis. The size of the supervisory board varies proportionally to the number of workers in the ﬁrm, but the number of representatives of shareholders and employees remains always equal. (In turn, blue-collar and white-collar workers elect their delegates, which in turn elect board members, separately and in proportion to their numbers.) Thus co-determination can be deﬁned as follows: Co-determination. A system in which ﬁrms are owned by shareholders but

control rights are shared among shareholder and employees. Now, shareholders and workers are usually said to be represented only on a “near-parity” basis because the chairman of the board has a tie-breaking vote and tends to be elected from the group of shareholders. In theory, this person can be elected from either the shareholder or the employee group. But in case of deadlock, which given the composition of the board is likely, the chairman is elected by the shareholder group. In short, it is very stylized to consider co-determination as separating ownership and control of the ﬁrm, for shareholders retain rights along with their property ample control rights both by electing all the members of the management

board and half of the members of the supervisory board plus the chairman, who has a tie-breaking vote. Further, it might be called into question whether codetermination could be considered an instance of workplace democracy at all, for workers are not the only agents with control rights and the principle of “one worker, one vote” is thus violated. However, codetermination is still the most developed existing instance of a system in which workers are provided with control rights over decisions made in the ﬁrm without having ownership rights as well.8 7.2.2. Cooperativism Consider now democratic ﬁrms in which workers both have control rights and own the ﬁrm. Cooperatives are not the only case in which control rights and ownership are kept together in workers’ hands. For example, the employee stock options plans (ESOPs) developed in the United States also provide workers with property rights that range from minority stock ownership as part of a beneﬁt plan with little or no control rights at all to voting equity shares that provide workers with ample control rights over decisions made by the board. ESOPs have rapidly expanded throughout the United States during the last two decades due to the generous corporate tax
A theoretical model in which ownership and control rights are also separated with a more equal distribution of conrol rights among shareholders and employees is the one developed by Isabelle Ferreras, in which the management of the ﬁrm is divided into two houses, one with workers’ representatives and another with shareholders’ representatives. See I. Ferreras, On Economic Bicameralism (master diss., MIT, 2004).
8

218

Chapter 7. Analytical and normative background

deductions, reaching an amount of 11,000 ﬁrms with ESOPs in 1999, including 12 among the 400 largest private ﬁrms.9 However, it is very questionable whether ESOPs could be regarded as an instance of WD, for in most ﬁrms control over the ﬁrm remains in shareholders’ hands.10 Further, the system is strong in the US but not elsewhere. In the remainder I will focus on another, more democratic and more widely extended, form of WD in which control rights and ownership are held together in workers’ hands Cooperative. namely, cooperatives, which can be deﬁned as follows. A ﬁrm in which all workers and only workers have equal own-

ership and control rights over the ﬁrm. Of course, only rarely we ﬁnd cooperatives so-deﬁned in real-world economies. As Elster recalls, non-working owners, non-owning workers and unequal distribution of shares are common.11 When these deviations become suﬃciently large, ﬁrms cannot be considered cooperatives any longer. Consider these deviations in turn. First, even though cooperatives are meant to be owned only by workers, this is hardly the case. To be sure, cooperatives’ equity capital tends to come from its members, either in the form of membership fees or in the form of reserves accumulated out of retained earnings. But it often comes from private investors as well, who accordingly enjoy control rights accordingly. In Spain, for example, non-members contributions are allowed to up to 45 percent of the total equity capital.12 Second, even when only workers own the ﬁrm, shares are often unequally distributed. Now, even though in some forms of WD e.g. ESOPs unequal shares are translated into unequal votes, in cooperatives the norm is to follow the principle of “one worker, one vote,” regardless of the shares owned by each worker. Third, it is also quite often the case that some workers, and sometimes a majority of them indeed, are salaried employees rather than owners of the cooperative. For example, only one in three workers of Mondragón cooperatives are full members, the remainder being hired and paid a salary as workers in conventional ﬁrms do.13 Hiring salaried workers may suit cooperatives better for market changeability and ﬂuctuances in product demand, even though they often justify it due to the fact that the cooperative mentality is not present everywhere or that it can take time for new workers to develop it. All these deviations are quite common, a reason why it may be useful to provide a less stringent deﬁnition: Cooperative*. A ﬁrm in which a majority of workers have equal control

rights over the ﬁrms and a majority of shares are owned by them. The origins of cooperativism can be traced back at least to the Industrial Revolution, with outstanding examples such as Robert Owen’s cooperative experiments in New
Dow, Governing the Firm, pp. 78-79. See Dow, Governing the Firm, p. 81; Hansman, The Ownership of Enterprise, p. 69. 11 See J. Elster, “From Here to There; or, If Cooperative Ownership is so Desirable, why are there so Few Cooperatives?,” Social Philosophy and Policy 6 (1989): 93-111. 12 Cooperatives Act 26/1999 of July 16. 13 See www.mcc.es (Accessed October 28 2011)
10 9

7.2. Deﬁnition and models

219

Lanark, which included discounted retail shops where proﬁts were transferred to his employees, and the Rochdale Society of Equitable Pioneers, founded in 1844, in which 28 weavers and other artisans set up the society to open their own store selling food they could not otherwise aﬀord and that triggered the cooperative movement throughout the UK. Even though its early advocates were aimed at transforming the economic system, cooperativism has turned into a peripheral yet persistent form of economic organization. Thus, according to the International Co-operative Alliance (ICA), in 2004 100 million people worked in cooperative ﬁrms, and the top 300 cooperatives in the world had around the same output as the GDP of Canada. Now, the notable persistence of cooperatives in the world economy does not imply that their market presence is homogeneously distributed across all market sectors. In fact, cooperative ﬁrms are very rare in sectors in which capital-intensive investment is required and substantial diﬀerences among employees of the ﬁrm exist, such as manufacturing. Notable exceptions are Fagor, Europe’s ﬁfth largest domestic appliances manufacturer that pertains to Mondragón Cooperative Corporation, and the plywood manufacturing cooperatives in the American Paciﬁc Northwest, which in 1984 accounted for more than 10 percent of all plywood produced in the US.14 By contrast, cooperatives have a rather strong presence in the service sector, including law, accounting, investment banking, management consulting, advertising, architecture, engineering, and medicine. Companies in this sector are more labor-intensive and diﬀerences among their workers are less substantial. As Hansmann points out, even though the debate on WD usually focuses on industrial ﬁrms and rarely refer to partnerships and professional ﬁrms, the latter are precisely the most widespread forms of WD. A notable instance is that of transportation companies, which are among the sort of ﬁrms more often organized as cooperatives. In Sweden, for example, all taxicab services and 50 percent of truck transports are supplied by cooperatives. (A presence that sharply contrasts with manufacturing, where cooperatives are virtually non-existent.)15 As Dow’s review of the literature concludes, “workers’ cooperatives tend to favor labor-intensive industries that lack signiﬁcant scale economies, require few specialized physical assets, and have minimal barriers to entry.” 16 This is bad news for WD in developing countries, where the economy is heavily based on the industrial sector. But it is good news for Western countries, whose economies are heavily based on services. For example, according to the United States Department of Commerce, just approximately 22 percent of the total GDP of $25.8 trillion was based on industrial output in 2007.17

See, respectively, http://www.fagor.co.uk/about-fagor-i-77.html (accessed October 26 2011) and Hansman, The Ownership of Enterprise, p. 66. 15 See Hansman, The Ownership of Enterprise, pp. 67-68. 16 Dow, Governing the Firm, p. 49. 17 See B. F. Snowden, The Feasibility of Workplace Democracy in Large, Publicly Traded Corporations in the United States (PhD diss., Union Institute and University of Cincinnati, 2005), p. 11.

14

220

Chapter 7. Analytical and normative background

7.3. Normative arguments
Why is workplace democracy a topic worth being researched? After all, democratic ﬁrms are relatively rare and marginal institutions in our economic landscapes and there are many pressing issues worth being analyzed in political philosophy, including war, poverty, energy policies, exploitation, human rights, criminal justice, especiesism, social capital, inter-generational justice, wealth inequality, nationalism, climate change, global ﬁnancial markets or religious fundamentalism. As Philippe van Parijs often repeats, political philosophy is not (or ought not to be) “an idle game played for the pleasure of making subtle distinctions and smart points, but a crucial part of the urgent task of thinking up what needs to be done to make our societies and our world less unjust than they are, or even simply to avert disaster.” 18 There are at least four reasons why studying how the workplace is organized and whether workers should be given a say over managerial decisions may be important. First, because workers spend one third of their daytime in the workplace, probably more than anywhere else. Second, because decisions made by shareholders, directors and managers dramatically aﬀect workers’ lives, not only directly through day-to-day commands or the setting up of the working conditions, but also indirectly, as a side-eﬀect of strategic decisions such as production planning or relocation. Further, what happens in the workplace does not only have an instrumental value to the worker but also an intrinsic one, work being a central source of self-respect in modern economies.19 As Rawls put it, “The lack of ... the opportunity for meaningful work and occupation is destructive ... of citizens’ self-respect.” 20 Third, as we shall see below, workers’ participation can be way more eﬀective and meaningful in the workplace than in other spheres, such as national or even local politics, being the impact of each vote
1 n

(where n = the number of participants in the decision) and

being n much larger in the national and local political domains than in the workplace, even in corporations as large as Mondragón, which has around 80,000 workers.21 Fourth, apart from workers’ everyday life, raising income inequalities within ﬁrms and an increasing political inﬂuence of modern corporations, in what some have termed as the “new Gilded Age,” puts corporations and their management at the center of the public debate.22
P. van Parijs, “Contestatory Democracy versus Real Freedom,” Just Democracy. The Rawls-Machiavelli Program (Colchester: ECPR, 2011), p. 24. 19 This is shown by the reluctance to stop working of those who nevertheless could aﬀord doing it. The intrinsic value of work as a source of self-respect has been widely noticed at least since Max Weber. Marx and Peeters have recently studied the eﬀects of the Belgian Win for Life—a lottery game where winners are granted a periodically unconditional lifelong income of 1,000 euros—over its winners. Although acknowledging the obvious limitations of a study of this kind, they found out that no single winning person out of fourteen reduced the amount of working hours and most of the winning households did the same. See Axel Marx and Hansen Peeters, “An Unconditional Basic Income and Labor Supply: Results From a Pilot Study of Lottery Winners,” The Journal of Socio-Economics, 37 (2008), 1636-1659, at p. 1648. 20 J. Rawls, Political Liberalism (New York: Columbia University Press, 1996), p. lix. 21 See http://www.mcc.es/language/en-US/ENG/Economic-Data/Most-relevant-data.aspx (Accessed November 1, 2011) 22 For a recent analysis of these trends in the Americna context explicitly linking them to democracy in the workplace, see P. Frymer, “Labor and American Politics,” Perspectives on Politics 8 (2010): 609-616.
18

7.3. Normative arguments

221

7.3.1. Feasibility Even though WD was a usual battle-cry among workers and a pressing proposal for many policy-makers in the 1970s and 1980s, today there are reasonable doubts about its appropriateness and feasibility, at least, for three reasons. First of all, as Axel Gosseries has pointed out, “The issue is not at all on the political agenda.” 23 Second, regardless of whether it is on the ageda or not, the proposal does not seem to be clearly and explicitly supported by workers, who might be more interested in ensuring and improving their working conditions job and often in just getting a rather than in acquiring control rights over their companies decisions. Further,

even when they enjoy good working conditions and stable jobs, risk-averse workers may prefer not to invest their savings in their own ﬁrms, for if the ﬁrm goes bankrupt they would lose not only their savings, but also their jobs.24 Third, global trends toward working ﬂexibilization and precariousness, decreasing union aﬃliation and increasing unemployment has weakened workers’ bargaining position during the last decades and has made it more unlikely for a social dialogue on WD to ever take place.25 To be sure, these are sensible concerns that cannot be easily dismissed from a normative point of view, unless it is believed that feasibility does not pose any sort of constraints on normative theorizing. I do not want to raise the very complicated methodological controversy on ideal versus non-ideal theory in normative ethics and political philosophy. However, I assume that feasibility and real-world applicability do pose some constraints on normative philosophy. What can be good under certain circumstances can turn into catastrophic when such circumstances change. In short, if WD proved to be completely unfeasible, there would be little if any interest in analyzing its desirability. That said, it has to be recalled that, WD is not merely a theoretical and counterfactual proposal, but an ongoing reality with two centuries of history behind it and a persistent presence nowadays. Thus, in 2006 more than 100 million people worked in cooperative enterprises, according to recent data compiled by the International Cooperative Alliance. To be sure, WD is not in the political agenda at all, as Gosseries points out. However, the proclamation of 2012 as the International Year of Co-operatives by the United Nations may change this. Furthermore, there are good reasons to call into question the objections raised above. On the one hand, it is questionable whether workers really despise the democra23 A. Gosseries, “Taking Workplace Democracy Seriously.” www.politika.lv/en/topics/civil_society/15968/ (Accessed October 26, 2011) 24 “[W]orking men prefer to work where they can obtain the best wages and the most comfortable place, while at the same time they also prefer to invest their capital where they think it will bring the greatest interest or dividend, whether that place be the one in which they are employed or elsewhere.” Letter from William Nuttall to Auberon Herbert, quoted in Elster, “From Here to There,” p. 94. 25 Figures from the Bureau of Labor Statistics of the United States Department of Labor show that approximately 12.5 percent of jobs in the US are unionized, compared to a high of approximately 34 percent durunder Franklin Roosevelt’s administration. Quoted in Snowden, The Feasibility of Workplace Democracy in Large, Publicly Traded Corporations in the United States, p. 2.

222

Chapter 7. Analytical and normative background

tization of the companies in which they work. Of course, the perception varies a lot depending on the context. But in 1984, for example, 66 percent of workers supported it and, as recently shown by Isabelle Ferreras, in Belgium there is a widespread demand for increasing workplace participation among low-skilled employees, such as those working at supermarket cashiers, a sectors in which trade unions already have a strong presence. Further, preferences for WD may be endogenous. As Elster has pointed out, isolated democratic ﬁrms in an otherwise capitalist economy may make them appear as less attractive for the average worker than they would be in an economy were these ﬁrms were the norm26 . On the other hand, economic crisis does not need to reduce the presence of democratic ﬁrms. Rather the contrary. The presence of cooperatives tends to enlarge during economic crisis, such as the current one.27 This is so at least for two reasons. First, because, cooperatives are specially resilient to economic recessions, partly due to their greater wage ﬂexibility and the lesser need to make proﬁts for investors and bonuses for managers. Second, because failing conventional ﬁrms are often taken over by employees, who buy them out in a desperate attempt to avoid getting unemployed. For example, in 2009, the year following the beginning of the Great Recession, an increase of 15 percent in the creation of cooperatives was registered in Catalonia, in contrast to a reduction of 32 percent in the creation of conventional companies.28 Of course, the greater resilience of cooperatives to economic crisis is not new.29 Consider a number of examples. First, during the agrarian crisis of the 1870s cooperative rural banks dramatically expanded in quasi-industrialized countries, along with manufacturing, trading and processing agricultural cooperatives. Second, during the Great Depression in the 1930s in the United States, cooperative farm supply purchasing greatly expanded from $76 million in 1924 to $259 in 1934. And the same goes for other businesses, such as petroleum cooperatives supplying farmers and farmer cooperatives. As for the former, by 1935 there were around 2000 of them supplying products worth $40 million and, as for the latter, by the same time there were 10,500 of them with 3,66 million members. Third, during the economic crisis that lead to mass unemployment in Western Europe in the 1970s and 1980s, many employees responded with a wave of takeovers in the form of employee-owned cooperatives aimed at curbing the impact of massive job losses. Finally, similar cases can be found in Finland during the early 1990s recession, when over 1200 labor cooperatives were created under the support of the Finnish Ministry of Labor. In a nutshell, even though democratic ﬁrms do suﬀer during economic crisis, they demonstrate to be more resilient to them. If the current crisis lasts long, it clearly appears that we will see more rather than less democratic ﬁrms across the globe.

See Elster, “From Here to There,” p. 96. See Birchall and Ketilson, Resilience of the Cooperative Business Model in Times of Crisis; Lansbury, “Workplace Democracy and the Global Financial Crisis,” Journal of Industrial Relations. 28 The data come from the Catalan Federation of Labor Cooperatives. See http://www.cooperativestreball.coop/v2/inici/comunicacio/mostra.php?id_pagina=9&idPare= (accessed March 15 2010). 29 See Birchall and Ketilson, Resilience of the Cooperative Business Model in Times of Crisis.
27

26

7.3. Normative arguments

223

7.3.2. Eﬃciency Let us now turn to another key normative aspect of WD its productive eﬃ-

ciency, understood in terms of level of inputs used per unit of output. Consider ﬁrst the main comparative advantages of democratic ﬁrms and their main disadvantages next. Of course, I do not pretend to cover all the potential and existing advantages and disadvantages from the standpoint of eﬃciency, which would make the section unnecessarily long, but only the main ones. A. Eﬃciency. It has been argued that democratic companies present three major eﬃciencies compared to non-democratic ones. — Eﬀort and information-sharing. It has been often argued that workers in democratic ﬁrms tend to ﬁnd their jobs more satisfying and meaningful, for they have a say in their companies’ decisions.30 Accordingly, it is very likely for productivity to increase, since workers’ supply of eﬀort increases and turn-over, absenteeism and strikes rates decrease. In addition, it is likely for product quality to improve, since workers are more willing to share their expertise and private information with managers and co-workers. And it is likely for job rotation to decrease, since workers are less likely to quit their jobs. In addition, all these eﬀects are likely to be greater in cooperatives, in which, in addition to control rights, workers also participate in proﬁts and wage inequalities are lower.31 — Unemployment. It has also been shown that, in times of economic crisis, democratic ﬁrms ten to produce a lower unemployment rate than non-democratic ones due to greater wage ﬂexibility (workers are more willing to moderate their salaries) and relocation of workers across diﬀerent departments of the ﬁrm. Again, this eﬀect is likely to be greater when workers are the owners of the ﬁrm.32 — Vertical and horizontal monitoring. Firms are structures of delegation between principals and their agents. Put simply, in conventional ﬁrms shareholders delegate to the board of directors, who in turn delegate to managers, who ﬁnally delegate to workers at the shop-ﬂoor level. Under more than likely circumstances of heterogeneous interests and asymmetric information among principals and agents, agency loss is served and agents have an incentive to shrink at best and behave opportunistically against the shareholders’ interests of maximizing proﬁts at worst. A number of means can be put in practice in order to address agency loss, including continued supervision of performance or salaries adjusted to productivity. However, all these means are costly and unavoidably imperfect, let alone speciﬁc problems such as the reduction of product quality. WD addresses the problem of agency loss by providing workers and managers alike with
30 See R. Sennett and J. Cobb, The Hideen Injuries of Class (New York: Knopf, 1972), pp. 235-236. 31 See S. Bowles and H. Gintis, “A Political and Economic Case for the Democratic Enterprise,” Economics and Philosophy 9 (1993): 75-100; H. M. Levin, “Worker Democracy and Worker Productivity,” Social Justice Research 19 (2006): 109-121. 32 R., Parks, D. Kruse and J. Sesil, “Does Employee ownership Enhance Firm Survival?,” Employee Participation, ﬁrm Performance and Survival, V. Perotin and A. Robinson, eds. (Amsterdam: JAI Press, 2004).

224

Chapter 7. Analytical and normative background

an incentive not only not to shrink but also to monitor each other. Even though perverse incentives may arise as well due to the democratic organization of the workplace, as we shall see below in section 8.5, workers in democratic ﬁrms have an incentive to monitor both managers’ performance and how their co-workers perform, with a reduction in the loss of agency and the costs of supervision and discipline as a consequence of it.33 B. Ineﬃciency. Brieﬂy consider now some of the main productive ineﬃciencies that have been pointed out. (Some frequent criticisms are not addressed though, such as rigidities of democratic ﬁrms in hiring and ﬁring or their potential tendency toward wage maximizations.)34 — Decision-making costs. Democracy takes time hierarchical decision-making.35 at least more than vertical and This is as true in the political sphere as it is in

the workplace. These costs go in three steps. First, the costs involved in electing representatives (MPs and the president, in politics; delegates and board members, in ﬁrms) and in making them accountable and responsive to their constituencies, including the need for some familiarity with what is going on in the ﬁrm on the side of electors (citizens in the former case; workers in the latter). Second, decisions are to be made collectively, which implies that coalitions will have to be formed, bargaining processes will have to be carried out and decision-making will be slow. Now, it might be argued that these are features shared both by democratic and non-democratic ﬁrms, for in the latter there is also a chain of delegation, there are costs associated to ensure accountability and responsiveness and decisions are often made collectively (say, by the assembly of shareholders, the board of directors, an so on). Yet there is a crucial diﬀerence. Namely, that in conventional ﬁrms there is a single well-deﬁned objective net present value of the ﬁrm’s proﬁts to maximize the , while in democratic ﬁrms there is a

greater plurality of objectives because there is a greater plurality of interests. As Hansmann points out, this is not to say that there is no room for conﬂict among shareholders in a conventional ﬁrm, for they can have diﬀerences in tax status, risk preference, or liquidity. But these diﬀerences are modest compared to those in a democratic ﬁrm.36 — External capital and risk aversion. It is a point often repeated that democratic ﬁrms tend to have problems to attract foreign capital and tend to be more risk averse.37 This is not a problem for codeterminated ﬁrms, in which capital is supplied by shareholders. But it is a problem for cooperatives, in which capital
See Bowles and Gintis, “A Political and Economic Case for the Democratic Enterprise.” See, respectively, J. Meade, “The Theory of Labour-managed Firms and Proﬁt Sharing,” Economic Journal 82 (1972): 402-428; and J. Vanek, The General Theory of Labour Managed Market Economies (Ithaca, NY: Cornell University Press, 1970). 35 See J. Dreze, “Some Theory of Labour Management and Participation,” Essays on Economic Decisions under Uncertainty (Cambridge: Cambridge University Press, 1987), pp. 375 ﬀ. Cited in Elster, “For Here to There,” p. 94. 36 See Hansmann, The Ownership of Enterprise, p. 62. 37 See M. C. Jensen, and W. H. Meckling, “Rights and Production Functions: An Application to Labor-managed Firm and Codetermination,” Journal of Business 52 (1979): 469-506.
34 33

7.3. Normative arguments

225

is supplied by its members, either in the form of membership fees or in the form of reserves accumulated out of retained earnings or yet by investing their pension savings in the ﬁrm. This feature has two negative consequences. On the one hand, it prevents workers from diversifying risk, for they reduce not only the diversiﬁcation of their investment portfolio but also the diversiﬁcation between their portfolio and their human capital. Put simply, it makes workers put all their eggs in the same basket, for if the ﬁrm goes bankrupt they not only lose their savings but also their jobs. Reasonably enough, this makes them more risk-averse and trumps eﬃciency-enhancing decisions that would otherwise be made. On the other hand, it makes them dependent on loans, which are diﬃcult to get when they are aimed at purchasing ﬁrm-speciﬁc assets, i.e. assets that are not easily fungible. This may explain why democratic ﬁrms are rare in capital-intensive sectors, and notably in sectors in which investment in ﬁrm-speciﬁc assets is required, but nevertheless have a strong presence in labor-intensive sector and in sectors in which capital investment is required but the assets are easily fungible, such as transportation or taxicab services.38 — Common ownership and free-riding. It is often argued that when workers use an asset they do not own, they may have no incentives to use it properly and take care of it, mainly when they are paid for the production output rather than for the production process and the way they treat the means of production (more on moral hazard problems arising from the separation of ownership and control on section 8.5). This is why some workers, such as truck drivers or carpenters, own their own tools, even if they work for a salary. And it is a reason sometimes used to argue for cooperativism, in which workers collectively own the means of production and thus have an incentive to take care of them. However, as Alchian and Demsetz have argued, joint ownership in large ﬁrms can produce free-rider problems. Assets that are owned by everyone may in practice be owned by no one if the number of owners is large enough.39 Now, it may be argued that this problem can be avoided in small cooperatives for, as Elinor Ostrom has famously shown, small-scale communities often resolve this collective-action problem.40 However, this solution is very problematic from the point of view of eﬃciency, for it rules out the exploitation of scale economies by means of enlarging the size of the ﬁrm, which is crucial for improving its eﬃciency. In short, an eﬃciency-based paradox follows from the joint ownership typical of cooperatives. The smaller the number of members of the cooperative, the more eﬃcient is the use of the assets but, in turn, the less eﬃcient is the exploitation of scale economies. And the other way around: The larger the number of members, the more eﬃcient the exploitation of scale economies is but the less eﬃcient the use of the assets turns

38

See Dow, Governing the Firm, pp. 47-50; and Hansmann, The Ownership of Enterprise, pp.

75-77.

39 A. Alchian and H. Demsetz, “Production, Information Costs, and Economic Organization,” American Economic Review 62 (1972): 777–795. See also Dow, Governing the Firm, p. 167. 40 See E. Ostrom, Governing the Commons. The Evolution of Institutions for Collective Action (New York: Cambridge Univeristy Press, 1990).

226 out to be.

Chapter 7. Analytical and normative background

However, at this point it is preferable to be cautious. First, because the issue of economic eﬃciency is too technical and complex for my level of expertise. Second, because, as pointed out by Dow and Hodgson, we lack data enough to draw conclusive and robust conclusions.41 Third, as noted by Hansmann, because the strong presence of democratic ﬁrms in certain economic sectors but not in other ones shows the context-dependency of eﬃciency issues, to the point that, if we leave aside considerations other than eﬃciency, we may well end up concluding that WD is a good idea in some sectors (say, services) but not in other ones (say, industries requiring large ﬁrm-speciﬁc investment). C. Additional eﬀects. Apart from their strict economic output, it may be worth mentioning three additional eﬀects that democratic ﬁrms may bring about. First, their likely impact on job security and workers’ health in terms of reducing the likelihood of occupational hazards taking place. Provided that workers are given a say over the setting up of working conditions, it can be expected that they will improve those conditions not only by ensuring that existing security standards are respected but also by providing their own knowledge so as to amend and improve such standards.42 Second, given the widespread and growing political apathy in Western democracies, a number of authors, such as Pateman and Mason, have defended WD not because of its potential economic or health beneﬁts, but due to its potential side-eﬀects over citizens’ engagement in the political sphere. According to the so-called “spillover thesis,” participation in the workplace provides an eﬃcient and eﬀective means of increasing participation in politics beyond the workplace.43 Given the amount of time workers spend in the workplace, participation in the workplace could promote the development of attitudes and psychological qualities necessary for a wide and genuine participation in the political sphere. However, as recently noted by Neil Carter, the existing evidence on the relationship between WD and political participation is far from conclusive and often at odds with the spillover thesis.44 For example, a study carried out by Greenberg et al. using a sample of 1,247 workers of democratic and non democratic companies shows that the relationship is much weaker than anticipated by Pateman and Mason.45
Dow, Governing the Firm, pp. 261-262; G. L. Hodgson, “Organizacional Form and Economic Evolution: A Critique of the Williamsonian Hypothesis,” Democracy and Eﬃciency in the Economic Enterprise, U. Pagano and R. Rowthorn, eds. (London: Routledge, 1996). 42 T. Theorell, “Democracy at Work and its Relationship to Health,” Emotional and Physiological Processes and Positive Intervention Strategies, en P. L. Perrewé and D. C. Ganster (Amsterdam and London: Elsevier, 2004). 43 See C. Pateman, Participation and Democratic Theory (Cambridge: Cambridge University Press, 1970) and R. Mason, Participatory and Workplace Democracy (Carbondale, IL: Southern Illinois University Press, 1982). 44 See N. Carter, “Political Participation and the Workplace: The Spillover Thesis Revisited,” The British Journal of Politics & International Relations 8 (2006): 410-426. 45 E. Greenberg, L. Grunberg and K. Daniel, “Industrial work and political participation: Beyond ’simple spillover’,” Political Research Quarterly 49 (1996): 287–304.
41

7.3. Normative arguments

227

Finally, given the global trend toward growing inequalities of size and weight among ﬁrms and of income and wealth within ﬁrms, WD has been sometimes advocated as a way of curbing such inequalities and of controlling, from within ﬁrms themselves, their inﬂuence over decisions made beyond the economic sphere notably, those made by democratic governments. During the last decades we have witnessed a rising income inequality in general and within ﬁrms in particular not observed since the Roaring Twenties in what some have termed as the “new Gilded Age.” 46 For example, according to a recent report by the International Labor Organization, the average wage diﬀerences of CEOs and average workers in the US grew from 370 to 1 in 2003 to 521 to 1 in 2007. For the same period, in the Netherlands the grew from 50 to 1 in 2003 to 100 to 1 in 2007.47 In addition, a similar trend has been observed regarding the size and weight of global corporations. For example, by 1996 51 out of the 100 largest world economies were already private companies, while only 49 were countries. For example, Wal Mart was larger than Poland, General Motors was larger than Denmark and Toyota was larger than Norway.48 It is questionable why democracy within ﬁrms would reduce their political inﬂuence or their rent-seeking behavior. There is no reason why workers ought to be less interested in manipulating the political environment so as to receive transfer payments or tariﬀ protections from the government. By contrast, wage inequalities within ﬁrms is a diﬀerent issue that WD may help to address.49 Indeed, in his study of American factories owned and/or operated by workers, Carnoy and Searer concluded that “in each of the cases studied, if the companies owned by the workers’ wages did not completely uniform, at least matches were signiﬁcantly in comparison with capitalist-owned enterprises and even the public bureaucracy.” 50 7.3.3. Expertise Firms are complex institutions that operate in constantly changing economic environments. Their management involves decisions about investment policies, production engineering, contracting and compensation and on budgetary planning, among many other technical and complicated issues. Why should workers be granted a say in their running when they often lack the expertise required to make informed decisions about these topics? As an executive said when asked about WD, “What? And let the monkeys run the zoo!” 51 Call this the epistemic argument against WD, which can be summarized as follows:
L. M. Bartels, Unequal Democracy: The Political Economy of the New Guilded Age (Princeton: Princeton University Press, 2008). 47 International Labour Organization, World of Work Report: Income Inequalities in the Age of Financial Globalization (Geneva: International Institute for Labour Studies, 2008), pp. 18-20. For an analysis of the eﬀects of economic inequalities over the political sphere (at least in the US), see L. Bartels, Unequal Democracy. The Political economy of the New Gilded Age (Princeton: Princeton University Press, 2008). 48 See S. Anderson i J. Cavanagh, “Corporate Empires,” Multinational Monitor 17 (1996). 49 For arguments for WD based on wage equalitity, see e.g. R. Plant “Socialsim, Markets and End States,” Market Socialism, J. Le Grand and S. Strin, eds. (Oxford: Clarendon Press, 1989). 50 M. Carnoy and D. Searer, Economic Democracy. The Challenge of the 1980s (New York: Sharpe/Pantheon Books, 1980), p. 175 51 Christie, “Recent Calls for Economic Democracy,” p. 115)
46

228 (A1)

Chapter 7. Analytical and normative background

In order to function properly and be eﬃcient, complex institutions, including ﬁrms, are to be run by those who have the expertise required to do so.

(A2) (A3)

Workers, at least a majority of them, lack this expertise. Firms are not to be run by their workers.

Two replies can be raised against the epistemic argument. First, consistently extended, it allows for an analogous criticism in the political sphere.52 Since ordinary citizens often lack the political knowledge required to take part in the government of an institution as complex as the state, why should they be able to do so? If WD is unjustiﬁed for epistemic reasons, then the government of those who know best—epistocracy, in Estlund’s insightful terms—follows also in the political arena.53 Those who consider that this argument fails to undermine the principle of political equality in the government of the state should also reject it (by modus tolens) in the government of the ﬁrm. And, on the contrary, those who consider that the argument succeeds they should consistently apply it to democracy across the board.54 Put simply, (A4) (A5) (A6) States are also complex institutions and not all citizens have the political expertise required to control the state. Not all citizens should control the state (from A1 and A4). Since (A5) is nowadays unacceptable, (A1) should be rejected—i.e. expertise should not be a necessary condition for having control rights. Second, even if we accepted premise (A1), it seems unlikely for shareholders to actually have more expertise than workers. Moreover if we consider that, unlike shareholders, workers spend a third of their daytime working in their respective frms. To be sure, this is the reason why shareholders delegate control to a board of directors, these in turn to top managers, and so forth. However, there is no reason why WD ought to be incompatible with delegation of decision-making about speciﬁc tasks to speciﬁc agents who can be democratically elected by workers and revoked by them. As Elster puts it, “Competence and expertise can be hired.” 55 In fact, division of labor is the norm in democratic ﬁrms as much as in non-democratic ones, for it is true that blue-collar workers, for example, may not have suﬃcient knowledge of ﬁnance or management to run the ﬁrm. As Hansmann puts it, “An individual employee need not herself have the expertise to make managerial decisions in order to exercise her voice eﬀectively ... She need only be able to vote intelligently in electing the ﬁrm’s directors” 56 To be sure, the institutional forms and means through which democracy can be realized within ﬁrms are as plural and open for debate as those in the political
I use here the so-called “parallel-case argument,” which will be discussed below. D. Estlund, Democratic Authority. A Philosophical Framework (Princeton: Princeton University Press, 2008). 54 A current example of this position is J. Brennan, “The Right to a Competent Electorate,” The Philosophical Quarterly 61 (2011): 700-724. 55 Elster, “From Here to There,” p. 95. 56 Hansmann, The Ownership of Enterprise, p. 115.
53 52

7.3. Normative arguments

229

sphere. But they need not exclude extensive use of division of labor and delegation of speciﬁc tasks to speciﬁc agents. As Walzer points out, “in a developed economy, as in a developed polity, diﬀerent decisions are made by diﬀerent groups of people at diﬀerent levels of society. The division of power in both cases is only in part a matter of principle; it is also a matter of circumstance and convenience.” 57 These two replies cast doubts about premises (A1) and (A2)—the former by trying to undermine the link between expertise and control and the latter by trying to refute the lack of workers’ expertise and by trying to show how this lack of expertise, even if true, could be consistently combined with WD through division of labor and delegation of control over ﬁrm’s decisions. However, it may be replied that they both neglect the relevance of agency problems in democratic ﬁrms and the epistemic preconditions required for a meaningful exercise of the democratic rights, including those of workers within a democratic ﬁrm. If we call into question the quality of a democratic system with, say, widespread illiteracy among its citizens and we do support public funding of education as consequence of it rather than assuming that democracy is not suitable for such countries under such circumstances, we ought to do the same regarding large epistemic asymmetries within democratic ﬁrms. In this regard, Kasmir has documented the eﬀects of the lack of expertise and spare time available over workers’ representatives in the Social Council of Fagor (the instance through which shop-ﬂoor workers are represented to management at one of the Mondragón Cooperative Corporation’s cooperatives) and over their ability to properly evaluate the proposals raised by management and to amend them if required.58 She has also shown the general apathy and refusal to participate produced by the absence of the epistemic conditions required for the meaningful exercise of control rights on the side of workers. According to Kasmir “normatively the Social Council extends considerably more rights to cooperative workers, while these rights to participate and plan are protected managerial prerogative in private frms. In practice, however ... structural factors (not having enough time, receiving too much complex information, not being able to seek the help of outside professionals ...) can limit its eﬀectiveness.” 59 Hence, it seems reasonable to conclude that a set of epistemic preconditions are to be met in order to hold directors and managers accountable and for control rights to be meaningfully exercised by workers. These may include, among other measures, permanent professional training and time assigned in workers’ schedule for the evaluation, amendment and elaboration of board proposals. Just to be clear, this does not need to mean that division of labor and delegation of complex tasks to speciﬁc agents should be excluded from democratic ﬁrms. Neither does it mean that eﬀective and meaningful exercise of control has an intrinsic value further beyond ensuring that directors and managers are held accountable, that agency loss is minimized, and that workers’ interests are taken into account in managerial decisions. But, even if we assume that division of labor and delegation
57 M. Walzer, Spheres of Justice (New York: Basic books, 1983), p. 302. See also Elster, “From Here to There,” p. 95. 58 S. Kasmir, The Myth of Mondragon (Albany: SUNY Press, 1996). 59 Kasmir, The Myth of Mondragon, p. 143.

230

Chapter 7. Analytical and normative background

of control are necessary and desirable in large and complex companies, a minimum competence on workers’ side—mainly, on less qualiﬁed, and therefore more vulnerable, workers’ side—is required if monitoring is to remain accurate and elected agents are to be held accountable. 7.3.4. Ownership Consider now WD from the point of view of the ownership of the ﬁrm. Why should workers be granted control rights over the ﬁrm when they are not its owners? Workers can always choose to work for democratic ﬁrms, take over their own ﬁrm in case it goes bankrupt or buy a majority of its voting shares—through ESOP or in the stock market—and thus gain control rights. As Nozick put it, “persons may form their own democratically-run cooperative ﬁrms. It is open to any wealthy radical or group of workers to buy an existing factory or establish a new one, and to institute their favorite microindustrial scheme; for example, worker-controlled, democratically-run ﬁrms.” 60 But, while they do not do it, shareholders remain the only legitimate owners of the ﬁrm and, accordingly, retain the rights derived from that ownership—among others, the right to possess the ﬁrm and to exclude others from using it, the right to decide how and by whom it may be used, the right to the residual income from it and the right to sell, modify and destroy it.61 Call this the propertarian argument, which runs as follows: (B1) (B2) (B3) Only shareholders are the legitimate owners of the ﬁrm. Ownership implies having control rights over the ﬁrm. Only shareholders have the right to control the ﬁrm.

Accordingly, a compulsory WD would violate the right of shareholders to use and run the ﬁrm they own as they wish (let aside, for the sake of the argument, prerogatives voluntarily awarded by shareholders to workers). As we shall see, proponents of WD do not stand empty-handed when facing the propertarian argument. But before turning to possible replies to premises (B1) and (B2), the following caveat is required. Even if we accepted that shareholders are the only owners of the ﬁrm and that ownership does imply full control rights, it may be replied that there is no reason why the implementation of WD ought to imply a violation of such rights, for it need not be compulsory. To be sure, some have argued for an inalienable right to WD.62 However, instead of through, say, expropriation or prohibition of non-democratic ﬁrms, the implementation of WD could be done gradually and voluntarily, by providing legal advice or through tax beneﬁts and direct subsidies such as those that, for the record, numerous non-democratic ﬁrms receive.63
R. Nozick, Anarchy, State and Utopia (New York: Basick Books, 1974), p. 250. L. Becker, Property Rights. Philosophic Foundations (London: Routledge and Kegan Paul), pp. 18-20. 62 See Dahl, A Preface to Economic Democracy; Walzer, Spheres of Justice; and D. Ellerman, Property and Contract in Economics. The Case for Economic Democracy (Oxford: Blackwell, 1992). 63 Some have argued that this shows that democratic ﬁrms are not eﬃcient, for if they were they would be created voluntarily, without requiring these advantages. See Jensen and Meckling,
61 60

7.3. Normative arguments

231

As Bowles and Gintis point out, “to argue against mandatory workplace democracy is to critique a straw man and to elide the fundamental issue, which concerns whether policies promoting workplace democracies are justifed in the interest of giving workers the opportunity to participate in these forms of governance.” 64 However, even if not compulsory, a rationale is still required for the state promotion of democratic ﬁrms at the expense of non-democratic ones. To be sure, there is a notable diﬀerence between using the coercive force of the state to ban non-democratic ﬁrms and using its public resources to discriminate them in favor of democratic ones. However, in both cases public means are used to beneﬁt one managerial option at the expense of the other one. In this sense, the propertarian argument still poses a serious threat to the justiﬁcation of WD. Let us then turn to its premises. Firstly consider premise (B2), according to which shareholders’ ownership of the ﬁrm implies their having exclusive control rights over it. In order to call this idea into question, it is possible to use again the analogy between WD and political democracy used above to reduce ad absurdum the epistemic objection. If we accept that property rights imply exclusive control rights, census suﬀrage appears to follow in the political arena. Why should all citizens enjoy equal political rights when they contribute such disparately—if at all—in the ﬁnancing of the state revenue through their taxes? Since, as Dow points out, “nothing about the ownership of a physical asset logically entails the owner to run the ﬁrm,” those who ﬁnd the objection unsound in the political sphere have to explain why they ﬁnd it sound—in case they do—within ﬁrms.65 And, on the contrary, those who ﬁnd the objection sound ought to admit that its scope goes beyond the ﬁrm and aﬀects the government of the state as well. In short, (B4) (B5) (B6) Not all citizens own property (neither do they pay taxes). Not all citizens should have control rights over the state decisions (from B2 and B4). Since (B5) is nowadays unacceptable, (B2) should be rejected—i.e. ownership should not be a necessary condition for having control rights. Consider now premise (B1)—namely, that only shareholders are the legitimate owners of the ﬁrm—and two possible replies. First, it may be argued that a ﬁrm is a set of agents supplying inputs such as ﬁnancial wealth, physical assets, raw materials, labor and so on. And that, since one of such inputs—namely, labor—cannot be legally owned, it is simply incorrect to say that ﬁrms are or can be owned by their shareholders. For example, Wal-Mart includes the workers of Wal-Mart, but Wal-Mart’s shareholders cannot legally own such workers without them technically becoming slaves. As Dow puts it, “in a society where slavery and indentured servitude
“Rights and Production Functions.” For a reply regarding why this need not be the case due to endogenous preference formation, adverse selection, discrimination and externalities, see Elster, “From Here to There,” esp. section III. 64 S. Bowles and H. Gintis, “Is Workplace Democracy Redundant in a Liberal Economy?,” Democracy and Eﬃciency in the Economic Enterprise, U. Pagano and R. Rowthorn, eds. (London: Routledge, 1996), p. 66. 65 Dow, Governing the Firm, p. 34.

232

Chapter 7. Analytical and normative background

are not legally enforceable, no one can own a frm because a ﬁrm is a set of human agents.” 66 However, it can be replied that talking about the “ownership of the ﬁrm” is just a way of talking about the ownership of non-human assets, such as ﬁnancial wealth or physical assets, and that workers are not owned but hired —just as some non-human assets are rented rather then owned. Further, shareholders and their agents (directors, managers) have limited authority over their employees. Unlike non-human assets, workers enjoy a number of labor rights, such as safety conditions and protection from discrimination. Furthermore, unlike under nineteenth-century Master and Servants Acts, or in current black markets of forced labor, or yet under circumstances of monopsony in the labor market, workers in free markets have the ever-present right to quit whenever they please. Whether these rights are enough to protect them from abuses is a completely diﬀerent issue that will be discussed below. Consider now a second criticism of premise (B1), which will be analyzed in some detail in the remainder of the section, along with some possible replies. According to a commonly held idea, shareholders’ property rights over the ﬁrm are negative rights—that is, they are rights from interference by third parties, including state intervention and, notably, state intervention aimed at advancing democracy within the ﬁrm. In the standard deﬁnition, Negative and positive rights.—The holder of a negative right is said to be entitled not to be interfered by third parties—including the state—, whereas the holder of a positive right is said to be entitled to the provision of a good or a service, again, from third parties—usually being the state. This conception is sometimes referred to as the “full liberal ownership” conception.67 Under it, shareholders’ property rights over the ﬁrm (actually, property rights in general) are said to be purely negative because they only require that third parties, including the state, refrain from interfering (e.g. through taxation, regulation and promotion of WD) in the owner’s possession, use, transfer, etc. of the assets of their property. Two reasons have been advanced against deﬁning property rights as negative ones. The ﬁrst one is that, as a matter of fact, enforcing property rights has public costs that are mainly funded out of taxing property (be it through real estate tax, income tax or VAT, among other). Since these costs are large—often larger than those posed by positive rights—and can only be funded out of taxing property, it is inconsistent to deﬁne property rights (including shareholders’ property rights over the ﬁrm) as restrictions against the very state intervention required to fund their enforcement costs. For example, Holmes and Sunstein have shown that just in 1996 the protection of property rights posed a cost of approximately $6,622 million on the US Federal budget. In order to put it in perspective, the education budget for the
Dow, Governing the Firm, p. 107. The expression comes from A. M. Honoré’s classic paper, “Ownership,” Oxford Essays in Jurisprudence, A. G. Guest, ed. (Oxford, Clarendon Press, 1961).
67 66

7.3. Normative arguments

233

same year was of approximately $2,132 million.68 As Elizabeth Warren has recently put it, There is nobody in this country who got rich on his own. Nobody. You built a factory out there—good for you! But I want to be clear. You moved your goods to market on the roads the rest of us paid for. You hired workers the rest of us paid to educate. You were safe in your factory because of police forces and ﬁre forces that the rest of us paid for. You didn’t have to worry that marauding bands would come and seize everything at your factory, and hire someone to protect against this, because of the work the rest of us did. Now look, you built a factory and it turned into something terriﬁc, or a great idea—God bless. Keep a big hunk of it. But part of the underlying social contract is you take a hunk of that and pay forward for the next kid who comes along.69 The second reason is related to the ﬁrst one—namely, that the very existence of property rights, and not only their enforcement, depends on state interference.70 As an instance of this, Murphy and Nagel take pre-tax income as an instance of this and argue that it is not possible to take it as independent from government interference. This is so because, despite the strong feeling that we are absolutely entitled to our pre-tax income, its very existence depends on the previous existence of money, banks, ﬁrms, stock exchanges, patents, and so forth. And these, in turn, depend on the existence of a previous system of criminal, contract, corporate, property and tort law. It is thus inconsistent to deﬁne pre-tax income independently from the institutional framework of which is partially a function. And the same goes for shareholders’ property rights over the ﬁrm. Since they are equally dependent on the institutional and legal framework that enable them, it is inconsistent to deﬁne them as negative rights. As Murphy and Nagel put it, “property [is] what it is created by the tax system, rather than what is disturbed or encroached on by the tax system. Property rights are the rights people have in the resources they are entitled to control after taxes, not before.” 71 Consider now three replies to these criticisms. First, it can be argued that Holmes and Sunstein overestimate the public costs of property rights and underestimate the (raising) role of private agencies—such as private prisons and private security companies—in their enforcement.72 Unlike prisons and security companies, some state services may not be privately provided ex hypothesi (e.g. adjudication, which cannot be totally supplemented by private arbitration). However, previous, current and counterfactual situations in which property rights arise and are enforced in the
See S. Holmes and C. Sunstein, The Cost of Rights. Why Liberty Depends on Taxes (New York and London: Norton. 1999), pp. 233 ﬀ. 69 The statement got viral in the blogosphere. It can be found, among many other places, at http://blogs.chicagotribune.com/news_columnists_ezorn/2011/10/warren.html (Accessed October 31 2011). 70 See Holmes and Sunstein, The Cost of Rights; and L. Murphy and T. Nagel, The Myth of Ownership: Taxes and Justice (Oxford, Oxford University Press, 2002). 71 Murphy and Nagel, The Myth of Ownership, p. 175. 72 See D. Schmidtz, “Property and Justice,” Social Philosophy and Policy 27 (2010): 79-100.
68

234

Chapter 7. Analytical and normative background

absence of a state are thinkable.73 Taking into account the unavoidable limits of private supply of goods and services, property rights may be hardly depicted as purely negative under present circumstances. However, the degree and forms of state interference required for the existence and enforcement of property rights—and thus the degree to which such rights can be said to be negative—vary dramatically from context to context. For example, property rights over ﬁrms can be said way more negative under the First Industrial Revolution, when the state interference was virtually nonexistent, than under the Second one, when the state played a crucial role in enabling monopolies at home and opening new markets abroad. Second, it is possible to accept that both the existence and enforcement of property rights pose costs on the public budget and yet deﬁne them as partially negative. To be sure, the state apparatus required for the existence and enforcement of property rights against third parties is costly. Since such costs are mainly funded out of taxing property, it is inconsistent to deﬁne property rights as restrictions against the very state intervention required to enable and enforce such rights. However, that does not turn such rights into purely positive rights, at least, for two reasons. First, since such costs are still aimed at enforcing a right from interference (both from private individuals and from the state, including monitoring public oﬃcials so as to prevent them from abusing their powers), they could still be considered as partially negative, rather than as completely positive.74 Second, even though shareholders beneﬁt from the institutional framework that enables and enacts their property rights, they are not the only ones in taking advantage of that framework—notably, workers also beneﬁt from its existence. They both beneﬁt from the existence of roads, schools and police, and not only the former, as Elizabeth Warren’s statement seems to imply. Likewise, they both pay for such services, either in the form of capital gains and property taxes or in the form of income tax and VAT. Shareholders may be wrong when they argue to have absolute property claims against state interference. But Holmes and Sunstein and Murphy and Nagel are equally wrong when they argue that they do not have any such claims at all. Third, it is possible to assume that state interference is required for property rights to exist yet reject that this interference has to be at odds with the negative deﬁnition of property rights when it follows the beneﬁt principle. That is, when such an interference is aimed at enabling and enforcing property rights—or, more broadly, at allowing the market to operate—and is funded out of taxes levied upon the taxpayers according to the beneﬁt they obtain from the existence of that institutional framework. Hence, assume that the institutional framework that enables and enacts property rights does pose serious burdens on the public budget and, accordingly, on the taxpayer. And further assume that some of the elements of the framework cannot be supplied privately, either because they cannot be privatized ex hypothesi
See F. Maultzsch, “Morals and Markets: The Signiﬁcance of Pre-Tax Ownership,” The Modern Law Review 67 (2004): 508-523; and J. P. Sterba, “Review of The Myth of Property,” Ethics 114 (2004): 628-631. 74 See A. Gewirth, “Are All Rights Positive?,” Philosophy & Public Aﬀairs 30 (2002): 321-333; G. Brennan, “’The Myth of Ownership.’ A Review Essay,” Constitutional Political Economy 16 (2005): 207-219.
73

7.3. Normative arguments

235

(e.g. criminal justice) or because they are public goods that would otherwise be underprovided by the market (e.g. national defense or clean air).75 Now, even though it follows that state interference via tax imposition is required if a regime of private property is to exist at all, this does not need to trump the idea of shareholders’ property rights as negative rights as long as that interference follows the beneﬁt principle. Once shareholders have paid for what they get—namely, the legal and institutional framework required for the existence and enforcement of their property rights over the ﬁrm—there is no reason why they should not be left alone to do as they please with their shares. However, the problem with the beneﬁt principle is one of measurement. Since not everyone beneﬁts equally from the provision of public goods, we need to ﬁgure out whom, what and to what extent to tax. And the problem is that the market price system does not work here, because regarding the provision of public goods the state operates as a price-discriminating monopoly. In short, the attack launched by Holmes and Sunstein and Murphy and Nagel on property rights as negative rights proves to be way too radical, for there are good grounds on which shareholders can claim to have negative rights over the shares they own and from being interfered by the state in such rights. However, it opens the door for a less absolute and negative conception of ownership in which state interference is justiﬁed as a matter of logic rather than on perfectionist grounds. 7.3.5. Power and contested exchange Part of the previous replies to the epistemic argument and the propertarian argument—namely, the reductiones in (A4)-(A6) and (B4)-(B6)—are based on an analogy between ﬁrms and the state that has been referred to as the “parallel-case argument.” 76 According to this argument, once democracy is accepted in the state, then there is no reason why it should be rejected in the workplace, in which power relations between employers and employees analogous to those between oﬃcials and citizens take place. As Robert Dahl famously put it, “If democracy is justiﬁed in governing the state, then it must also be justiﬁed in governing economic enterprises; and to say that it is not justiﬁed in governing economic enterprises is to imply that it is not justiﬁed in governing the state.” 77 To be sure, it may be replied that the analogy does not hold because ﬁrms and states are very diﬀerent. For example, while ﬁrms are for-proﬁt, clearly states are not.78 However, here I wish to focus on another potential objection—namely, that
As it is well known, public goods are nonrival and non-excludable. Since their consumption by one individual does not reduce availability of the good for consumption by others, and since no one can be eﬀectively excluded from using the good, markets tend to be unable to provide these goods optimally, if at all. 76 The label comes from J. Cohen, “The Economic Basis of Deliberative Democracy,” Social Philosophy and Policy 6 (1989): 25-50. For diﬀerent versions of the argument, see Hsieh, “Justice in Production,” pp. 86-91. For a careful criticism, see C. López-Guerra, “Against the Parallel Case for Workplace Democracy,” Revue de Philosophie économique 8 (2008): 11-28. 77 R. A. Dahl, A Preface to Economic Democracy (Berkeley: University of California Press, 1985), P. 111, italics in the original. 78 See López-Guerra, “Against the Parallel Case for Workplace Democracy.”
75

236

Chapter 7. Analytical and normative background

even if we accept that relations of power, such as those between the state and its citizens, ought to be managed democratically, it happens that, unlike relations between the state and its citizens, employment relations are not relations of power, for they are contractual relations in which the parties have entered voluntarily. Compulsory labor could have been the norm in nineteenth-century England, when Master and Servant Acts were still in force and employees were criminally prosecuted for quitting their jobs. And it is so, nowadays, in black markets of prostitution and forced labor, which according to the International Labor Organization include at least 12.3 million of forced workers.79 However, in free and competitive labor markets workers can exit their jobs as freely as they enter them. Accordingly, any attempt to democratize employment relations would turn out to be redundant at best, for they are already free and voluntary, and an illegitimate interference in a voluntary agreement at worst. As Bowles and Gintis put it, “if the capitalist economy is a sphere of voluntary private interactions, what is there to democratize?” 80 In this section I will present Bowles and Gintis’ challenge to this view. As we shall see, according to them, employment relations are power relations indeed, for they imply a contested exchange between the employer and the employee. This, in turn, would give way to their democratization.81 Put diﬀerently, the existence of power relations within the ﬁrm would not only justify the use of the analogy between ﬁrms and the state in (A4)-(A6) and (B4)-(B6), it would also provide a straightforward argument for WD. As Bowles and Gintis put it, “Because the employment relationship involves the exercise of power, its governance should on democratic grounds be accountable to those most directly aﬀected.” 82 The section analyzes the theory of contested exchange along with the problems it presents. First, the liberal conception of employment as absent of power relations is brieﬂy presented. Second, Bowles and Gintis’ theory of contested exchage is reconstructed in detail. Finally, a number of problems are raised. 7.3.5.1. The liberal conception of wage labor Unlike in much of classical political and legal thought, which considered employment relations as a partial form of slavery in which the employee temporarily waived her freedom by working for a salary, in neoclassical economics and liberal political
79 The ILO estimates that of these 12.3 million, 20% are coerced by the state (largely into military service), 14% are forced sex workers, and the remaining 66% are coerced by private agents in other industries, such as agriculture, mining, ranching, and domestic service, B. and Belser, P. (eds.) (2009), Forced Labor: Coercion and Exploitation in the Private Economy, Boulder: Lynne Rienner Publishers. 80 S. Bowles and H. Gintis, “A Political and Economic Case for the Democratic Enterprise,” Economics & Philosophy (1993): 75-100, p. 97. 81 See S. Bowles, and H. Gintis, “Contested Exchange: Political Economy and Modern Economic Theory,” American Economic Review 78 (1988): 145-150; “Contested Exchange: New Microfoundations of the Political Economy of Capitalism,” Politics and Society 18 (1990): 165-222; “Power and Wealth in a competitive Capitalist Economy,” Philosophy and Public Aﬀairs 21 (1992): 324-53; “A Political and Economic Case for the Democratic Enterprise;” “The Revenge of Homo Economicus: Contested Exchange and the Revival of Political Economy,” Journal of Economic Perspectives 7 (1993): 83-102; “Is Workplace Democracy Redundant in a Liberal Economy?,” Democracy and Eﬃciency in the Economic Enterprise, U. Pagano, et al., eds. (London: Routledge, 1996). 82 Bowles and Gintis, “A Political and Economic Case for the Democratic Enterprise,” p. 75.

7.3. Normative arguments

237

philosophy employment relations tend to be seen as voluntary relations between free individuals and competitive labor markets, in turn, as non-political. As we have seen above (§1.3.3.1), Cicero had perfectly illustrated classical thought by deﬁning the wages of “of all hired workmen whom we pay for mere manual labor, not for artistic skill [as] a pledge of their slavery.” 83 Taking into account that the classics deﬁned freedom precisely in opposition to slavery, it is unsurprising that employment relations were naturally considered to be power relations. By contrast, in neoclassical economics and libertarian political philosophy power is absent from employment relations, at least, under certain circumstances—namely, as long as labor markets remain competitive and workers can freely enter and exit their jobs. As Abba Lerner famously pointed out, in neoclassical economics “an economic transaction is a solved political problem.” 84 Since in free and competitive markets transactions are made freely and voluntarily, the only existing power is, in fact, purchasing power. Of course, it has been widely debated whether this feature applies to ﬁrms as well for, unlike in market exchanges, in the employment relationship employers continuously exercise their authority within ﬁrms in the form of commands and supervision. According to the conception of the ﬁrm as a nexus of contracts, it is mistaken to draw a clear line between ﬁrms and markets and employment relations are just another kind of market contracting.85 Even though employment relations may diﬀer from other market contracts in the continuity of association among employers and employees, the relationship among them is far from being a power relation. In Alchian and Demsetz’s formulation, for example, an employer has no more authority over an employee than a customer over a grocer. Of course, both the employer and the customer can punish their counterpart by no longer dealing with her. But an employer ﬁring or threatening to ﬁre an employee is not exercising her power any diﬀerently from a customer who stops or threatens to stop buying from her grocer. In Alchian and Demsetz’s words, It is common to see the ﬁrm characterized by the power to settle issues by ﬁat, by authority, or by disciplinary action superior to that available in the conventional market. This is delusion ... The ﬁrm ... has no power of ﬁat, no authority, no disciplinary action any diﬀerent in the slightest degree from ordinary market contracting between any two people ... Telling an employee to type this letter rather than to ﬁle that document is like my telling a grocer to sell me this brand of tuna rather than that brand of bread.86
83 Cicero, De Oﬁciis, trans. W. Miller (Cambridge, Harvard University Press, 1913), I. XLII. The word in the Latin original version is “servus” rather than “sclavus.” However, it has to be noted that in the Roman Republic “servitude” was synonymous with “slavery.” This meaning would remain stable until the High Middle Ages (c. 1000-1300), when the term would acquire its more usual medieval meaning, i.e. as the status of those peasants who were bound to a manor and subject to a levy of their labor, production and/or income by the lord of the manor. 84 A. Lerner, “The Economics and Politics of Consumer Sovereignty,” American Economic Review Proceedings 62 (1972): 258-66, p. 259. 85 See A. A. Alchian and H. Demsetz, “Production, Information Costs, and Economic Organization,” American Economic Review, 62 (1972), 777–795; and Jensen and Meckling, “Rights and Production Functions.” 86 Alchian and Demsetz, “Production, Information Costs, and Economic Organization,” p. 777.

238

Chapter 7. Analytical and normative background

The transition from the conception of employment as a form of slavery, in which the worker temporarily gives away her freedom to the liberal conception of employment as a voluntary transaction between free and self-standing individuals, had an important role in the development of industrial capitalism, at least in two ways. On the one hand, the conception of wage labor as free labor made it possible for American abolitionists to distinguish between slavery and wage labor. This use contrasted to that of the slaveholders in the South, who used the term “wage slavery” to discredit the supposed freedom of industrial workers in the North and, therefore, to justify slavery. Paradoxically enough, socialists were not the only ones to denounce wage labor in factories as wage slavery. According to Carsel, “By 1860 [the term “wage slavery”] had become a standard part of the proslavery rationale, and some publicists ... made it a central thesis in their arguments.” 87 On the other hand, as we have seen above (§1.3.3.1), the extension of wage labor to millions of workers during the Industrial Revolution turned out to be possible only because the status of wage laborers was assimilated to that of capitalists. Namely, as both being owners—the former of her labor force, the latter of her capital—and thus as being able to freely exchange their properties in the market. As Antoni Domenech has pointed out, this shift from partial slavery to freedom was possible thanks to “the legal ﬁction according to which the male have-nothings were also ’free owners:’ Owners of their labor force, and with an equal legal capacity—with an equal freedom—to make legal actions and business (civil contracts) from their ’property.” ’88 The ownership of the worker over her own person is a legal ﬁction—or a “political ﬁction,” as Pateman has put it—because, even though there are parts that can be separated from the owner and exchanged in the market—e.g. esperm or some organs—, other parts—e.g. abilities and talents—cannot be technically separated. They can be exchanged only as if they were another productive asset.89 As Karl Polanyi put it, “Labor [cannot] be detached from the rest of life, be stored or mobilized ... The commodity description of labor ... is entirely ﬁctitious.” 90 If workers are deﬁned as owners of themselves, it seems obvious that them being hired needs not to reduce their freedom, as long that they do it voluntarily. Further, if they are deﬁned as owners of themselves, why should they be prevented from selling themselves as slaves, as long as they do it voluntarily? Taken to its logical consequences, the liberal conception of wage labor as free labor calls into question the inalienable rights against slavery. If workers can be hired, there is no reason why they should not be bought, provided that they are sold voluntarily. Hence, Nozick could famously argue on Anarchy, State and Utopia that “The comparable

W. Carsel, “The Slaveholders’ Indictment of Northern Wage Slavery,” Journal of Southern History 6 (1940): 504-520, p. 504. 88 A. Domènech, El Eclipse de la fraternidad (Barcelona: Crítica, 2004), pp. 94-95. 89 See C. Pateman, “Self-Ownership and Property in the Person: Democratization and a Tale of two Concepts,” Journal of Political Philosophy 10 (2002): 20-53, pp. 26-27. 90 K. Polanyi, The Great Transformation. The Political and Economic Origins of our Time (Boston: Boston: Beacon Press, 1944), p. 71

87

7.3. Normative arguments

239

question about an individual is whether a free system will allow him to sell himself into slavery. I believe that it would.” 91 In short, the liberal conception of employment contracts as free contracts in which power is absent makes it possible to draw a clear line between states and ﬁrms. In the former domain public oﬃcials do exercise coercion over citizens, while in the latter the employer is unable to exercise coercion of any sort over the employee, for the employee has voluntarily agreed to the terms of the employment contract and can quit the job at will and unilaterally.92 Accordingly, the analogy between ﬁrms and states does not hold and the replies to the epistemic and the propertarian arguments fail. 7.3.5.2. Contested exchange Against the liberal conception of employment relations, which are regarded as non-political relations, Bowles and Gintis attempt to demonstrate “what is transparent to the unschooled—that the capitalist economy confers power upon those who occupy leading positions in the business world—is denied by the sophisticated.” 93 According to Bowles and Gintis, employment relations are relations of power because they involve a “contested exchange” between the employer and the employee, which opens the door to their democratization. “Because the employment relationship involves the exercise of power, its governance should on democratic grounds be accountable to those most directly aﬀected.” 94 This section reconstructs the theory of contested exchange regarding wage labor,95 and a number of criticisms are raised in the following one. Firstly consider the deﬁnition of “power” and “contested exchange” employed by Bowles and Gintis. Power relationship.—Agent A has power over agent B if, by imposing or threatening to impose sanctions on B, A is capable of aﬀecting B’s actions in ways that further A’s interests, while B lacks this capacity with respect to A. According to this deﬁnition, the advantageous and asymmetric imposition of sanctions—or of threats thereof—is a suﬃcient condition for the exercise of power. It is not a necessary one though, for it does not exclude additional instances under which A can be said to have power over B (say, under more stringent conceptions of power). Hence, the deﬁnition of Bowles and Gintis is rather minimal and hardly disputable. Even though there may be additional conditions under which power relations can
R. Nozick, Anarchy, State and Utopia (New York: Basic Books, 1974), p. 331. Interestingly, in the early twentieth century, in the heyday of the monopoly capitalism that characterized the second Industrial Revolution, Friedrich Nauman, founder of the German Liberal Party, had distinguished liberalism from socialism, precisely because of their willingness to accept or reject this distinction. Nauman said: "If regarding the state there are only diﬀerences of degree between socialism and liberalism, their relationship are more complicated when taking into account the major economic enterprises that have appeared in recent years. These are private organizations (...) That is how liberalism considers them. But, on the other hand, they are also major capitalist organizations such as the state (...) That is how they appear in the eyes of Socialists.” Quoted in Domènech, El eclipse de la fraternidad, p. 171; my italics. 93 Bowles, and Gintis, “Contested Exchange,” p. 145. 94 Bowles and Gintis, “A Political and Economic Case for the Democratic Enterprise,” p. 75. 95 Bowles and Gintis also apply the theory to ﬁnancial markets, which are of course beyond the scope of this dissertation.
92 91

240

Chapter 7. Analytical and normative background

indeed take place, it could hardly be denied that, when their conditions apply, power is exercised. Consider the example of a dictatorship, in which the dictator has the ability to aﬀect the actions of her citizens for her own interests, and citizens lack a similar capacity in relation to the dictator. On the other hand, an exchange is contested under the following conditions: Contested exchange.—An exchange between A and B is contested when B possesses an attribute that is valuable to A, is costly for B to provide, yet is not adequately speciﬁed in an exogenously enforceable contract. When the terms of a contractual relationship can be fully speciﬁed ex ante and are costlessly enforceable by a third party—e.g. a court—the exchange can be said to be uncontested. Thus many of the market exchanges are uncontested because the terms of the contract are well speciﬁed at the outset of the exchange and these terms can be enforced by a third party. However, there are also many cases in which the terms of the relationship cannot be fully speciﬁed ex ante and/or cannot be exogenously enforced. These exchanges are said to be contested. These conditions obtain under a number of common circumstances. For example, when the terms of the relationship cannot be fully speciﬁed in the contract—or it would be too costly to do so—because unforeseeable contingencies are likely to appear (say, change of the economic cycle); when there is no third party to which the contracting parties can turn to (say, when A and B are two sovereign states); or yet when there is a third party but the relevant evidence cannot be adequately measured (say, eﬀort). In all such cases, the contested attribute that A wants to get from B has to be obtained ex post, through the use of an endogenous enforcement mechanism, such as monitoring, sanctioning or designing an incentive scheme aimed at obtaining that desired attribute from B, given that it is costly for B to provide it. Now, according to Bowles and Gintis, the employment relationship follows this scheme. For the following reasons. The employment relationship begins when, in exchange for a salary, the worker agrees to submit to the authority of the employer. An employee diﬀers from an independent contractor—e.g. a free-lance worker—because the employee is subject to the authority of the manager of the ﬁrm, while an independent contractor acts autonomously. However, the employer does not contract the labor of the worker, but his ability to work, that is, her labor force. The employer hires the worker’s ability to work, from which the employer has yet to obtain actual work. The relationship between the employer and the employee can be characterized then as an agency relationship. Agency relationship.—There is a relationship of agency when a principal A authorizes an agent B to act on her behalf but cannot costlessly control the behavior of B and would like B to take some action that B would otherwise not undertake. In the employment relationship all these clauses apply. First, the employer hires a worker to act on her behalf, say, to perform certain task in the ﬁrm. Second, the employer tries to obtain the greatest quantity and quality of labor from the

7.3. Normative arguments

241

worker, while the worker ﬁnds this costly and has an incentive to shrink. That is, the employer desires the employee “to take some action that B (the employee) would otherwise not undertake.” Third, the employer “cannot costlessly control the behavior of employee B.” That is, the employer cannot costlessly ensure that the employee will perform the assigned task at the desired degree of eﬀort. In the absence of a third party, the employer has to devise a set of mechanisms to ensure that the worker performs the task and works at the desired level of eﬀort. In order to do this, the employer can attempt to specify a set of quantitative properties of the job in the labor contract, including speciﬁc tasks, salary, breaks, etc. Generally, these properties can be exogenously enforced. Thus, a “no call, no show” form of absence, for example, can lead to exogenously enforceable dismissal. However, there are a number of qualitative properties—such as the employee’s eﬀort, intensity and initiative—that are very diﬃcult to specify ex ante and that are equally diﬃcult to enforce exogenously. As De Francisco has put it, “you cannot go to the court and tell the judge that you have ﬁred the worker because she did not work hard enough.” 96 These properties are contested willy-nilly and turn the employment relationship into a contested exchange. According to Bowles and Gintis, An employer oﬀers a wage in exchange for which the employee oﬀers not some fully speciﬁable pro quo, but rather at best an unenforceable promise to perform at an adequate level of intensity, care, and initiative. The employer must develop a monitoring system to determine whether this level has been achieved in any particular case and must have means to discipline an employee whose performance is deemed unsatisfactory.97 To the extent that the employment relationship contains contested properties that can not be speciﬁed ex ante and/or enforced by means of exogenous mechanisms, the capitalist has to devise endogenous enforcement mechanisms in order to extract such properties. Which are these mechanisms? The mechanization, sequencing and standardization typical of Taylorism, as well as piecework, productivity bonuses, and internal promotion are attempts to address the problem of agency loss in the employment relationship and ensure that employees work hard. In the case of Taylorist practices and piecework, by reducing the weight of the contested elements of the relationship via specifying and standardizing as much as possible the tasks to be performed. In the case of productivity bonuses and internal promotions, by providing incentives for workers to supply endogenously the contested elements of the relationship. Without denying the importance of these and other mechanisms, Bowles and Gintis focus on the endogenous enforcement mechanism that they consider to be central to the employment relationship—namely, contingent renewal or, in other words, the threat of dismissal. According to them, the threat of dismissal is the main
96 A. de Francisco, Ciudadanía y democracia. Un enfoque republicano (Madrid: Los Libros de la Catarata, 2007), p. 73. 97 Bowles and Gintis, “Contested Exchange: Political Economy and Modern Economic Theory,” p. 146-147.

242

Chapter 7. Analytical and normative background

disciplinary mechanism used by the employer to obtain the contested properties from the employee. A mechanism that, on the other hand, the employee lacks. Is it thus possible to conclude that the employment relationship is a relationship in which the employer has power over the employee? Not that fast. The threat of dismissal is only credible if the dismissal implies a cost for the worker, that is, if the worker cannot easily ﬁnd a similar job in the market. However, in competitive equilibrium labor supply equals demand and, when the market clears, agents are indiﬀerent between their current transactions and pursuing the next best alternative. In such circumstances the threat of dismissal is not credible because the employee can ﬁnd a similar job elsewhere. In short, as Bowles and Gintis put it, “to the extent that such sanctions are the basis of power, neither owners, managers, nor creditors have power in Walrasian competitive equilibrium.” 98 In imperfect labor markets the employee may lack a real exit option, for the labor market does not clear. That is, there is a group of workers involuntarily unemployed—an “industrial reserve army,” to use Marx’s famous expression—and willing to take B’s job in case she is ﬁred, which turns the threat of dismissal into credible. Put diﬀerently, the existence of involuntary unemployment turns the exit option into costly, which in turn provides the employer with an advantageous position over the employee. By contrast, under competitive equilibrium the labor market clears and, accordingly, the employee does enjoy a costless exit option that balances the asymmetric bargaining position between the employer and the employee. Under such conditions, employer A cannot credibly threaten to ﬁre employee B, or at least not more than B can threaten A with quitting and, accordingly, no power can be said to be exercised according to Bowles and Gintis’ deﬁnition. In Samuelson’s famous words, “in a perfectly competitive model, it really doesn’t matter who hires whom; so let labor hire capital.” 99 And, as expressed by Gauthier, “the perfect market, were it realized, would constitute a morally free zone, a zone within which the constraints of morality would have no place.” 100 That is, a zone free of power relations, as he makes it clear: the presupposition of free activity ensures that no one is subject to any form of compulsion ... Each chooses for herself what to produce, what to exchange, and what to consume ... given the freedom of market interaction, the introduction of moral constraints would be the introduction of the very compulsion that we suppose morality is intended to counteract.101 Now, as shown by Shapiro and Stiglitz, when the employer can only imperfectly control the actions of the employee, the labor market does not clear, for the existence of involuntary unemployment serves as a necessary disciplinary mechanism
Bowles and Gintis, “Power and Wealth in a competitive Capitalist Economy,” p. 326. P. Samuelson, “Wages and Interests: A Modern Dissection of Marxian Economics.” American Economic Review 47 (1957): 884-912, p. 894. 100 D. Gauthier, Morals by Agreement (Oxford: Clarendon Press, 1986), p. 121. 101 Gauthier, Morals by Agreement, p. 96. According to Daniel Hausman, if this argument were correct, "morality [or, for what concerns us here, a possible democratization] would have applicability to markets only in virtue of the imperfection of these markets.” D. M. Hausman, “Are Markets Morally Free Zones?,” Philosophy & Public Aﬀairs 18 (1989): 317-333, p. 317.
99 98

7.3. Normative arguments

243

to ensure that employees work at the desired level of eﬀort.102 If the labor market cleared, workers would lack adequate incentives not to shirk, for they would be able to ﬁnd a similar job easily in case they were caught shirking. In short, involuntary unemployment turns out to be eﬃcient, for it serves as a disciplinary mechanism for those who do have a job, thus increasing their productivity. According to Shapiro and Stiglitz: Under the conventional competitive paradigm, in which all workers receive the market wage and there is no unemployment, the worst that can happen to a worker who shirks on the job is that he is ﬁred. Since he can immediately be rehired, however, he pays no penalty for his misdemeanor. With imperfect monitoring and full employment, therefore, workers will choose to shirk ... With unemployment, even if all ﬁrms pay the same wages, a worker has an incentive not to shirk. For, if he is ﬁred, an individual will not immediately obtain another job. The equilibrium unemployment rate must be suﬃciently large that it pays workers to work rather than to take the risk of being caught shirking.103 Given that in competitive equilibrium—let alone imperfect markets—involuntary unemployment exists, workers are not indiﬀerent to losing their jobs. Accordingly, the threat of dismissal turns out to be credible and the employer is able to impose sanctions that the employee, by contrast, cannot impose. Thus, Bowles and Gintis are in a position to say that the employment relationship between employer A and employee B is a relation of power even in competitive equilibrium. And that, for three reasons: (C1) (C2) In competitive equilibrium there are unemployed workers who would like to take B’s job, so A’s threats of dismissal are credible. Accordingly A can impose sanctions on B and, in particular, can threaten B with applying them to obtain from B the desired level of eﬀort and promote her (A’s) interests. (C3) (C4) B lacks the ability to impose similar sanctions on A. Employer A has power over employee B.

7.3.5.3. Objections Bowles and Gintis theory of contested exchange has been criticized, among other things, for the narrowness of their motivational assumptions; for not taking into account contextual factors—say, the unemployment rate, the economic cycle or the role of unemployment beneﬁts—that dramatically inﬂuence bargaining positions in the employment relationship, for not distinguishing between large and small and medium businesses; for not considering additional endogenous enforcement mechanisms; or for failing to identify power relations when contested elements are absent or marginal to the relationship. The remainder of the section addresses all these
C. Shapiro and J. Stiglitz, “Unemployment as a Worker Discipline Device,” American Economic Review 74 (1982): 433-444. 103 Shapiro and Stiglitz, “Unemployment as a Worker Discipline Device,” p. 433.
102

244

Chapter 7. Analytical and normative background

criticisms and argues that the theory of contested exchange remains incomplete and probably mistaken. On the one hand, it falls sort of capturing a number of very relevant instances of power relations in the workplace and, on the other hand, it includes a number of them that need not be power relations. A. Motivational assumptions. It can be objected that the theory of contested exchange is grounded on over-simplistic motivational assumptions—namely, that both employees and employers are self-interested utility-maximizers, so that the former are only interested in maximizing proﬁt and the latter would shirk unless properly constrained. Of course, it may be replied that Bowles and Gintis are exclusively aimed at internally refuting the neoclassical view of the employment relationship as non-political. That is, that even if we assume the neoclassical framework—including the assumption of economic agents as self-interested—it is possible to show that the employment relationship is a power relationship. However, Bowles and Gintis not only aim at internally refuting that corollary. They also want to develop a theory that can account for the way in which actual power relations evolve and surplus is obtained in a capitalist economy. Now, a central assumption of the theory of contested exchange is that employees will only work at the desired level of eﬀort when they have incentives to do so—namely, not to be dismissed. However, by doing so the theory leaves uncovered a number of forms of endogenous enforcement, such as when the employee perceives the contract as legitimate or just. Bowles and Gitis assume that the parties behave strategically, that they have conﬂicting interests and, accordingly, that employees will work hard only under commands, supervision, threats of dismissal and eﬀective sanctions. However, employees often identify with the company because they fear that it may go bankrupt or consider that they are well treated and paid. Under these circumstances, the threat of dismissal is still present, it plays a secondary role. Hence, even if we accept the assumption of self-interest, Bowles and Gintis do not take into account those cases in which workers—even if they may consider that the terms of the relationship are unjust—are interested in the survival of the ﬁrm in which they work. In these cases, as Wright and Burawoy put it, “given these shared interests, it is possible some degree of real reciprocity in the relationship between workers and capitalists.” 104 On the other hand, once we relax the assumption of self-interest, they do not explain those cases in which workers perceive the employment relationship as just and in which the threat of dismissal plays a minor role in the process of endogenous enforcement. In these two cases, the employment exchange can still be said to be contested, for B possesses an attribute that is valuable to A, costly for B to provide, yet is not adequately speciﬁed in an exogenously enforceable contract. However, it is at least questionable whether it can be said to be a power relationship (at least, in all cases), for B supplies the contested attribute not because she fears being ﬁred but because she considers that the terms of the relationship are
E. O. Wright and M. Burawoy, “Coercion and Consent in Contested Exchange,” Politics and Society 18 (1990): 251-266, p. 79.
104

7.3. Normative arguments

245

just or at least in her beneﬁt.105 B. Context-dependency. It may be also objected that, even if we accept that employees are self-interested utility maximizers and that the threat of dismissal is the main endogenous enforcement mechanism, the costs that the employee assumes in case of being ﬁred—and therefore the employer’s capacity to credibly use it as a disciplinary device—depends on a number of contextual factors that Bowles and Gintis do not take suﬃciently into account. These factors are both objective and subjective. On the one hand, even if the labor market cleared, personal investment in ﬁrm-speciﬁc human capital may lock-in the employee. And there are further subjective costs, such as those linked to loosing the current job and seeking a new one. As pointed out by Elster and Moene Ove, “A job change may involve a change of residence, with all the attendant psychological costs. On the other hand, lead to the breakdown of speciﬁc business friends, imposes similar costs of divorce.” 106 On the other hand, when assessing the inﬂuence of involuntary unemployment on the costs assumed by the worker in case of dismissal, Bowles and Gintis do not take into account two outstanding contextual factors. First, the economic cycle, that can dramatically increase or reduce the unemployment rate and thus the costs assumed by the employee in case of dismissal. And, second, the sum and duration of unemployment beneﬁts, along with additional public services, such as public schooling or health care, that dramatically aﬀect the bargaining position of the worker. The existence of these factors open the door for an scenario in which there is involuntary unemployment, the employment relationship is very contested (say, because the tasks to be performed by the employee are diﬃcult to measure and monitor) and, nevertheless, the existence of a strong welfare system—including generous unemployment beneﬁts and public housing, health care and schooling—balance the bargaining asymmetries between employers and employees. Put diﬀerently, one can think of circumstances in which the conditions stipulated by Bowles and Gintis obtain and in which, nevertheless, we can hardly say that the employment relationship is a power relationship, or at least not to the same extent as when the worker lacks a strong welfare system. As noted by Wright and Burawoy, “more important than the shortage of labor demand is the issue of public welfare provision. [It is] in the absence of real public welfare provision [when] the workers face a situation of extreme necessity if they lose their jobs.” 107

Of course, it can be argued that these sorts of relationships still involve power, only that more subtle. Hence, according to Wright and Burawoy “This is precisely what the concept of hegemony is meant to convey: a system in which the conﬂicting interests of workers and capitalists are coordinated in such a way that workers spontaneously consent to their own exploitation.” Wright and Burawoy, “Coercion and Consent in Contested Exchange,” p. 79. However, this is too fast a conclusion, for it rules out the possibility of the relationship being actually just, a condition that may often obtain. 106 Elster and Ove Moene, Alternatives to Capitalism (Cambridge: Cambridge University Press, 1989), p. 21). 107 Wright and Burawoy, “Coercion and Consent in Contested Exchange,” p. 82.

105

246

Chapter 7. Analytical and normative background

C. Uncontested exchange. Finally, it may be called into question the actual importance of the contested elements in the employment relationship and the extent to which it can be said that, when these elements are absent or they turn out to be marginal, the relationship between the employer and the worker cannot be said to be a power relationship. As noted above, the scientiﬁc organization of work characteristic of Taylorism was, among other things, an attempt to eliminate the presence of contested elements from the employment relationship by trying to specify ex ante the terms of the relationship through the standardization of the tasks to be performed. It may be the case that, during the last decades, the post-Fordist organization has in many ways replaced classical Taylorism, giving more weight to contested elements, such as the initiative of the worker or his “productive intelligence.” Now, on the one hand, it is not clear whether Taylorism has been actually replaced or whether it has lost its predominant position in the management of ﬁrms. As noted by Boltanski and Chiapello, Taylorism may have been reduced in certain sectors, especially in the continuous process industries, such as petrochemistry or steel. But it does coexists in other sectors, such as clothing industry or construction, in which Taylorism would have not only not disappeared but would have been strengthened, along with additional pre-Taylorist cases such as small businesses. In short, even though post-Fordism may have expanded during the last decades due to the introduction of quality circles, schedule ﬂexibility or just-in-time production, “the dominant tendency is the reproduction of previous Taylorist forms of organization.” 108 Moreover, the introduction of ISO 9000 family of management quality standards and procedures developed to meet such standards, has led some to talk about “participatory Taylorism.” 109 To be sure, these trends do not call into question the soundness of the theory of contested exchange, but only its relevance. However, this is not irrelevant, for Bowles and Gintis theory may leave important instances of employment relationships involving power uncovered and cover some instances that may not involve the exercise of power at all.

108

L. Boltanski and È. Chiapello, The New Spirit of Capitalism (London: Verso, 2005), p.

218.

See A. Lahera, La participación de los trabajadores en la democracia industrial (Los Libros de la Catarata: Madrid, 2004), ch. 4.

109

Chapter 8

Republicanism at work
8.1. Introduction
All the arguments analyzed in the previous chapter, along with some others, need to be taken into account if a conclusive case for WD is to be made. The aim of this chapter is more humble though. It draws upon republican political philosophy, as presented in chapter 1 and revitalized during the last three decades, and analyzes whether WD turns out to be necessary to ensure workers’ freedom from arbitrary interference within the ﬁrm.1 A two-step argument is defended. First, it is argued that a republican case can be made not only for the availability of an exit option from the ﬁrm and for workplace constitutionalism (i.e. the regulation of the workplace by setting clear and speciﬁc standards to which managers have to conform in the exercise of their authority), but also for WD. Second, it is further claimed that WD requires, from a republican standpoint, not only workers’ participation but also workers’ ownership of the ﬁrm. The chapter is in ﬁve sections. Section 8.2 provides the republican framework used throughout the chapter. Section 8.3 shows the forms of arbitrary interference that are likely to appear in the ﬁrm and the problems of two competing views to WD in addressing them—namely, (a) that the availability of an exit option from the ﬁrm suﬃces to prevent workers from arbitrary interference and (b) that ensuring republican freedom within the ﬁrm requires workplace regulation but not workplace democracy. Sections 8.3.1 and 8.3.2 show, respectively, that costless exit from ﬁrms is neither possible nor desirable both in perfect and imperfect labor markets and that exhaustive workplace regulation is neither possible nor desirable due to the incompleteness of labor contracts. Section 8.4, in turn, presents the republican case for WD as straightforward as possible, and addresses a number of relevant objections. Finally, section 8.5 discusses whether WD requires worker ownership or not from a republican standpoint. Bargaining power asymmetries and moral hazard problems arising from the formal separation between ownership and control rights are analyzed and the ceteris paribus desirability of cooperativism over other forms of WD in which control rights and ownership are separated, such as co-determination, is defended.
1 For previous discussions of WD from a republican standpoint, see N. Hsieh, “Rawlsian Justice and Workplace Republicanism,” Social Theory and Practice 31 (2005): 115-142; and B. H. McDonnell, “Employee Primacy, Or Economics Meets Civic Republicanism at Work,” Stanford Journal of Business and Finance 13 (2008): 334-393.

248

Chapter 8. Republicanism at work

8.2. Republican freedom
As we have seen above in chapter 1, the republican tradition of political thought is concerned with freedom as non-domination as a central political value.2 According to this conception, a free person is a person without a master, i.e. a person who does not live under the arbitrary will or dominium of others. In addition, as we shall see in sections 3 and 4, both political participation and material independence are taken to be necessary, though not suﬃcient, conditions of freedom as non-domination. As for now, let us focus on the deﬁnition of republican freedom. Its historical priority notwithstanding, the republican conception of freedom is usually analyzed in comparison to its liberal counterpart. Consider the latter, according to which an individual or collective agent is free as long as she is not interfered with in her choices and actions—be it through actions, coercitive threats or manipulation. In Berlin’s (1958: 7), famous words, “I am normally said to be free to the degree to which no human being interferes with my activity.” 3 Two counterintuitive implications follow from the liberal conception. First, we ought to accept that a slave with a benign master is to some extent free as long as her master forbears from interfering in her choices and leaves her do her own thing. Second, we ought to accept, as Hobbes and Bentham did, that any form of legal interference, however apparently justiﬁed, reduces by deﬁnition the freedom of the agent on whom the interference is visited. In Bentham’s words, “As against the coercion applicable by individual to individual, no liberty can be given to one man but in proportion as it is taken from another. All coercive laws, therefore ... are, as far as they go, abrogative of liberty.” 4 By contrast, republican freedom is deﬁned as non-domination rather than as non-interference. And domination, in turn, is deﬁned as the capacity to interfere arbitrarily in the choices and actions of another agent. Thus republicanism is a stricter principle in one dimension—there must not even be any potential interference—but a weaker principle in the other dimension—actual or potential interference is allowed as long as it is not arbitrary. The classic instance of someone who lacks freedom as non-domination is a slave, for she lives under the ever-present threat of being interfered with by her master arbitrarily, that is, without having to take into account the slaves’ interests whatsoever. From a republican standpoint, the capacity of the master to interfere arbitrarily diminishes the ability of the slave to deﬁne and pursue her ends with self-conﬁdence and raises the likelihood of the slave’s choices conforming to the preferences of the master. This is so because the slave is aware that her master enjoys an ever-present position to interfere whenever the slave deviates from her master’s preferences. According to Pettit, the impact of such a position
2 For careful and exhaustive analysis of the republican tradition, see among many others P. Pettit, Republicanism (Oxford: Oxford University Press, 1997); Q. Skinner, Liberty Before Liberalism (Cambridge: Cambridge University Press, 1999); A. Domènech, El eclipse de la fraternidad (Barcelona: Crítica, 2004); and F. Lowett, A General Theory of Domination and Justice (Oxford: Oxford University Press, 2010). 3 I. Berlin, Two Concepts of Liberty (Oxford: Oxford University Press, 1958), p. 7. 4 J. Bentham, “Anarchical Fallacies,” in Works of Jeremy Bentham, vol. 2 (Edinburgh: William Tait), p. 503.

8.2. Republican freedom

249

over the slave’s choices can occur in three diﬀerent ways. First, it may impact on the latter’s deliberative choices themselves, so she decides not to make them altogether. Second, it may impact on the set of options that the latter sees as available to her by removing some of the options from that set. And third, it may replace a subset of her options by another subset of options that she considers to conform better with her master’s tastes.
5

Just to be clear, from a republican standpoint, the freedom of the slave is diminished both by the capacity of the master to interfere arbitrarily in her choices and actions and by the extent to which such an ability is exercised. Here I take the stand of Skinner (1999) rather than of Pettit (2002). For the former, republican (or neo-Roman) freedom depends both on actual and possible interference, while for the latter it depends predominantly on the possibility of being interfered with. I do not discuss which account is historically more accurate. The intuitive idea is that the slave’s freedom is indeed diminished by the very possibility of her being interfered with, as both Skinner and Pettit believe. But, pace Pettit, her freedom is also and dramatically diminished when she is actually interfered with arbitrarily, or when such interference is more likely to happen. In turn, the extent and likelihood of actual arbitrary interference, and not only its possibility, will raise the likelihood of the slave conforming to the preferences of her master.6 To illustrate the diﬀerence between the liberal and the republican accounts of freedom, consider the two cases mentioned above. In the ﬁrst case, the republican tradition argues that, even if the slave is not actually interfered with, she is still unfree because her master enjoys the capacity to arbitrarily interfere in her choices and actions, which, in turn, will raise the likelihood of the slave conforming to the preferences of her master. Consider the case of a woman who depends on her husband as her only source of income. Further assume that the husband agrees to share his income with her and to make all the household decisions on a parity basis. Now, if the husband used his economic position to impose his interests, by doing so he would diminish his wife’s freedom indeed. But even if he does not use that position to advance his interests, his economic position will in all likelihood aﬀect the wife’s behavior as for those decisions involving spending—such as whether to buy a house or rent one, whether to take kids to a public or a private school or even whether to get divorced. From a republican standpoint, the wife’s liberty is aﬀected even when she is not actually interfered with by her husband.7
P. Pettit, “Republican Liberty: Three Axioms, Four Theorems,” in Republicanism and Political Theory, ed. Cecile Laborde and John Maynor (Oxford: Blackwell, 2008), pp. 106-107. 6 See Skinner, Liberty Before Liberalism; and P. Pettit, “Keeping Republican Freedom Simple. On A Diﬀerence with Quentin Skinner,” Political Theory 30 (2002): 339-356. 7 Here it is worth mentioning the social experiments with the negative income tax (NIT) in ﬁves cities in America and Canada between 1968 and 1980. Since in practice the NIT provided a source of income for many women who had lacked it hitherto, one of its side-eﬀects was that the divorce rate raised (for the Canadian case see D. Hum and S. Choudry, “Income, Work and Marital Dissolution: Canadian Experimental Evidence,” Journal of Comparative Family Studies 23 [1992], 249-265). Now, it is reasonable to think that most of the women who decided to get divorced used their newly gained economic autonomy to satisfy a preference they already had. But it is not unreasonable to think that the NIT could have made some women change their minds about their own marriage and open a new option to them or replace other options they could have seen as the
5

250

Chapter 8. Republicanism at work

As for the second case, in contrast to the liberal view, from a republican standpoint laws need not diminish the freedom of those against whom they are enforced as long as they are not arbitrary.8 Consider the following example:9 Tanya lives in a small, newly created country in Eastern Europe. Perhaps the most important issue in the region is the treatment of a disenfranchised minority that lives throughout the country. Tanya truly dislikes the minority and wants to further damage them if she can. While public opinion concerning the minority varies greatly, the government has taken the side of the minority. Consequently, a ban has been placed on any action or public speech that is intended to hurt the disenfranchised minority. In other words, the government has made laws against hurting the minority, but Tanya wishes she could hurt them. Jonathan Phillips presented subjects with this case and asked them whether they considered that the ban diminished Tanya’s freedom or not. Interestingly, rather than saying that Tanya’s freedom was completely diminished by the ban, the mean response was more close to “Not at all” than to “Completely.” Phillips’ ﬁnding is in line with the republican conception of freedom (though he analyzes it in diﬀerent terms). According to this conception, legal interference can occur without substantial loss of liberty when it prevents agent A (say, Tanya) from interfering the choices and actions of agent B (say, the disenfranchised minority) in an arbitrary manner, i.e. without being forced to track the interests of B.10 In that case, the law may diminish to some extent A’s freedom, for it actually interferes in her choices and actions. But it does not do so completely, for it makes compulsory for A to track the interests of B and thus interferes in A’s choices and actions in a non-arbitrary and non-dominating way. Of course, law can turn into arbitrary itself, as it will become clear in section 4. However, it need not be so as long as it does track the interests of those against whom it is enforced. Moreover, from a republican standpoint law turns out to be the guarantee of freedom from arbitrary interference itself because—and as long as—it ensures that the interests of the interferee are adequately tracked. Law ensures that such interests are adequately tracked not only when they are taken into account, for if someone is looking for a way to hurt someone else (as in the case of Tanya), she might keep track of what her victim prefers and systematically attempt to thwart each of her goals. It does ensure that such interests are adequately tracked when forces the interferer (say, Tanya) to take the interests of the interferee as having
only ones available to them. Just to be clear, what is important from a republican standpoint is that, whether they changed their minds or not and whether they eventually divorced or not, the women who had been dependent on the income of their husbands hitherto had also been (only in case they lacked other meaningful alternatives) under the latter’s ability to interfere them at will, and thus dominated. I thank B for pressing me to clarify this point. 8 Pettit, “Republican Liberty: Three Axioms, Four Theorems”; and “Law and Liberty,” in Law and Republicanism, ed. S. Besson and J. L. Martí (Oxford: Oxford University Press, 2009). 9 J. Phillips, “Freedom: Morality and Folk Intuitions,” unpublished manuscript, Yale University, 2010, p. 16. 10 “Republican Liberty: Three Axioms, Four Theorems,” p. 117.

8.3. Republicanism in the workplace

251

the same normative weight as her own interests. In Harrington’s words, republican freedom is not deﬁned as ‘immunity from the law,’ but rather as “immunity by the laws.” 11 Now, even if we accept this conception of the law, it is still an open question which laws would better ensure freedom from arbitrary interference in the workplace. In the next two sections I turn to the forms of arbitrary interference that are likely to appear in the workplace and the possible ways of addressing them.

8.3. Republicanism in the workplace
Put brieﬂy, in the liberal conception a person is free if and only if she is not actually interfered with in her choices. By contrast, according to the republican conception, Freedom as non-domination. X is free iﬀ she enjoys immunity against the

possibility of being interfered arbitrarily. Now, what forms of arbitrary interference are to appear within the ﬁrm and how are they to be addressed from a republican standpoint? In contrast to self-employment, the very point of the employment relationship is the voluntary subordination of the worker to the command and supervision of the employer regarding the material details of the job. It is thus inevitable for the employer to interfere continuously in the employee’s performance in the form of commands and supervision. Now, such interference need not be arbitrary as long as it is adequately checked, that is, as long as the she is forced to track the interests of the employee on which the interference is visited. As we shall discuss immediately below, there are number of ways in which the employer may be forced to track the interests of the employee, such as the employee’s ever-present ability to quit in case her interests are not taken into account, the legal regulation of the employment relationship, or workers’ control rights over the decision-making of the ﬁrm. To be sure, none of these means is intended to prevent the employer from issuing commands and supervising the employee’s performance. But they may prevent such commands and supervision from becoming arbitrary. The most obvious instances of workers subject to arbitrary interference are those of forced labor and child work, which are involuntary by deﬁnition.12 But arbitrary interference, either actual or possible, may also take place in non-forced but inadequately checked working environments. Nien-hê Hsieh has listed three levels of decision-making in which arbitrary interference can occur: (a) decisions directing a worker to perform certain task or to limit her performance, such as favoritism in allocating work and overtime or verbal and physical abuse in the supervision (including sexual harassment); (b) decisions aﬀecting not the particular actions of the worker but the conditions under which such
J. Harrington, The Commonwealth of Oceana and A System of Politics, ed. J. G. A. Pocock (Cambridge: Cambridge University Press, 1992), p. 20. 12 They include violations of two of the fourfold set of fundamental labor rights included in the ILO 1998 “Declaration on Fundamental Principles and Rights at Work,” the major international consensus document on the issue.
11

252

Chapter 8. Republicanism at work

actions are performed, such as capricious rescheduling of working hours, discrimination in promotion and compensation or assignment of tasks involving hazardous performance;13 and (c) decisions that are neither made directly about the worker nor about her working conditions, but nevertheless aﬀect her as a side-eﬀect, such as investment policies, production planning or relocations.14 Arbitrary decisions can be and are often made by shareholders, directors and managers at these levels. Since they occur in the setting where people spend one third of their lives and can dramatically aﬀect them and their families, it is reasonable to take them seriously indeed. Now, what is it to be done about them? As we have seen, from a republican standpoint adequate protection from arbitrary interference ought to be aimed at eliminating not only actual decisions of this sort, but also the very possibility of their being made.15 And adequate protection, in turn, will be ensured as long as it forces the actual or potential interferer to track the interests of the interferee. However, this aim can be achieved in a number of possible ways. Let us consider two of them in this section and turn to a third one—namely, WD—in section 4.16 8.3.1. Exit option A ﬁrst response is that, in free labor markets, workers can always quit their job. As a consequence of this, employers are forced to track the interests of their employees and arbitrary interference turns out to be impossible. As Arneson puts it, the freedom of the individual on a modern labor market willy-nilly confers on each person a considerable degree of control in the form of exit rights ... One can generally escape the reach of ... unwanted policies by quitting one’s job and taking another.17 Consider two versions of this response—a strong version and a weaker one. According to the former, in free and competitive labor markets little is to be done apart from ensuring that they remain competitive and workers can freely enter and exit their jobs. Arbitrary interference could have been the norm in nineteenth-century England, when Master and Servant Acts were still in force and employees were prosecuted
As Hodson points out, arbitrariness can happen at this level not only due to the employer’s actions but also due to her forbearance in the maintenance of the technical viability of production, including job safety and security conditions. See R. Hodson, Dignity at Work (Cambridge: Cambridge University Press, 2004), p. 93. 14 Hsieh, “Rawlsian Justice and Workplace Republicanism,” p. 122, and “Justice in Production,” p. 91. 15 To illustrate how the ability to interfere arbitrarily can aﬀect workers decisions, Stuart White gives the example of a worker who says to himself, “I had better not go to those gay clubs any more because if my boss ﬁnds out he might sack me, and I will then be destitute. Instead, I had better go to the Young Conservative’s Association to impress him”. See S. White, The Civic Minimum (Oxford: Oxford University Press, 2003), p. 47. 16 Just to be clear, even if we accept that the interests of workers are to be tracked in order to prevent their being interferred arbitrarily, this does not carry us all the way to an argument for WD. Sure, the paper will attempt to show that WD is in a better position to do this. But such a rationale for WD is far from being obvious and has still to be provided. I thank D for pressing me to clarify this point. 17 “Democratic Rights at National and Workplace Levels,” p. 139.
13

8.3. Republicanism in the workplace

253

for quitting their jobs. And it is so, nowadays, in black markets of prostitution and forced labor, in monopsonistic labor markets or in cases in which illegal immigrants decide not to quit under the threat of being reported and repatriated. These are all instances of compulsory or quasi-compulsory labor, in which workers are in a permanent position to be interfered by their bosses with not much more restraint than their capricious will. However, this need not be the case in free, competitive and full-clearing labor markets as long as they are kept free and competitive. In such markets, or so the argument goes, workers freely enter their jobs and can use their right to quit as an ever-present threat against any ex post managerial decision—i.e. not contractually speciﬁed at the outset of the employment relationship—that could lead to arbitrary interference and abuse, ruling out the possibility of arbitrary interference taking place.18 A weaker version of the exit option argument assumes that power asymmetries between employees and employers can arise in free labor markets and argues for making the exit option as costless as possible through welfare provision of unemployment beneﬁts, health care or a basic income. According to Brian Barry, for example, a basic income “is the most practicable (perhaps the only practicable) way of counteracting the excessive power of employers over workers.” 19 Now, even though the availability of an exit option is a necessary and vital condition to prevent forced or quasi-forced labor and to ensure that employers and managers have to track workers’ interests (for they can otherwise decide to quit), it turns out to be insuﬃcient, at least, for three reasons. First, imperfect labor markets do have involuntary unemployment, which has disciplinary eﬀects over employed workers by making it costly for them to quit (since they would not be in a position to ﬁnd another job easily). It is thus mistaken to say that employees are in a symmetric bargaining position to contest managerial decisions or that they can use the right to quit as an ever-present credible threat against any attempt of arbitrary interference. As long as this is the case, employers remain unforced to track the interests of their workers and are eo ipso able to make arbitrary decisions aﬀecting their employee’s choices and actions. Second, even if labor markets cleared, there are additional exit costs that workers have to bear when they quit their job and can lock-in employees.20 Three stand out: First, workers’ investment in developing ﬁrm-speciﬁc human capital (the so-called too-much-invested-to-quit eﬀect); second, the searching and transition costs from one
As Alchian and Demsetz’s famously put it, “What ... is the content of the presumed power to manage and assign workers to various tasks? Exactly the same as one little consumer’s power to manage and assign his grocer to various tasks ... Telling an employee to type this letter rather than to ﬁle that document is like my telling a grocer to sell me this brand of tuna rather than that brand of bread”. A. A. Alchian and H. Demsetz, “Production, Information Costs, and Economic Organization,” American Economic Review 62 (1972): 777–795, at p. 777. This is not Arneson’s position, though, for whom “the right of the manager [to command] involves power or dominion over the lives of others, whereas the right of the subordinate to quit does not”. R. J. Arneson, “Defending the Purely Instrumental Account of Democratic Legitimacy,” p. 125. 19 B. Barry, Why Social Justice Matters (Cambridge: Polity Press, 2005), p. 212. 20 See J. Elster and K Ove Moene, “Introduction,” in Alternatives to Capitalism (Cambridge: Cambridge University Press, 1989), p. 6; Hansmann, The Ownership of Enterprise, p. 71; and Hsieh, “Rawlsian Justice and Workplace Republicanism,” pp. 129 ﬀ.
18

254

Chapter 8. Republicanism at work

job to another; and, third, the psychological costs in quitting work altogether—work being a relevant source of self-respect in itself in our capitalist economies.21 Of course, all these costs vary enormously both within and across ﬁrms, as well as across time and place. For example, while an experienced surgeon in Rotterdam may be able to ﬁnd a similar job just by walking across the street, an unskilled worker in Detroit may ﬁnd it very diﬃcult to do the same. Accordingly, employers may have an uneven capacity to interfere arbitrarily with the various subsets of employees under diﬀerent circumstances. Now, even though employees who possess scarce and valued skills, such as a surgeon in Rotterdam, may be de facto immune to arbitrary interference, employees who lack such skills are not, as Dahl has pointed out.22 Hence, additional means to exit rights may not be necessary for the former—just as Warren Buﬀet may not need voting rights to de facto inﬂuence the political process—, yet they are necessary to ensure the immunity from arbitrary interference of the latter. Third, even if exit were completely costless, it could happen that job alternatives for workers were as despotic, unregulated and arbitrary as their current job. In that case, exit option would be of little value. Consider as an analogy an archipelago of countries that enforced open borders policies despite their being all dictatorships. Would the fact that citizens could costlessly exit their countries turn the power of their dictators into any less arbitrary?23 Adam Michnick, the Polish democratic dissident, was once presented with a similar counterfactual, to which he replied the following: “If forced to choose between General Jaruzelski and General Pinochet, I would choose Marlene Dietrich. The alternative is absurd and irrational. It oﬀers me the choice, as I ﬁght for democracy in a dictatorial system, of sitting in prison either as a Communist or as an anti-Communist.” 24 In a nutshell, though probably necessary, it appears that the availability of an exit option alone is not a suﬃcient condition to prevent arbitrary interference.25
The intrinsic value of work as a source of self-respect has been widely noticed at least since Max Weber. Marx and Peeters have recently studied the eﬀects of the Belgian Win for Life—a lottery game where winners are granted a periodically unconditional lifelong income of 1,000 euros—over its winners. Although acknowledging the obvious limitations of a study of this kind, they found out that no single winning person out of fourteen reduced the amount of working hours and most of the winning households did the same. See A. Marx and H. Peeters, “An Unconditional Basic Income and Labor Supply: Results From a Pilot Study of Lottery Winners,” The Journal of Socio-Economics, 37 (2008), 1636-1659, at p. 1648. 22 R. Dahl, “A Right to Workplace Democracy? Response to Robert Mayer,” The Review of Politics 63 (2001): 249-253, p. 252. 23 Pettit makes a similar point in “Law and Liberty,” at p. 52. As we have seen above, the analogy between ﬁrms and states I use here—the so-called parallel case argument (Cohen, “The Economic Basis of Deliberative Democracy”)—has not gone without controversy. See Claudio López-Guerra, “Against the Parallel Case for Workplace Democracy,” Revue de philosophie économique, 8 (2008), 11-28. The main criticism is that ﬁrms and states are diﬀerent enough for any analogy between them to work out. (For example, unlike states, ﬁrms are primarily proﬁt-seeking.) However, even if it is true that the analogy hardly applies all things considered, it does apply once the exercise of domination in each sphere is isolated from other variables, since forms of arbitrary interference are likely to appear in both states and ﬁrms unless adequately checked. And that is enough for the present case as long as we assume that any conclusion reached from the standpoint of republican freedom is inconclusive unless properly balanced with the rest of the relevant variables. 24 Quoted in A. S. Kirshner, “Proceduralism and Popular Threats to Democracy,” Journal of Political Philosophy, 18 (2010), 405-424, at p. 411. 25 To be fair to republican authors arguing for a basic income, it has to be noted that none
21

8.3. Republicanism in the workplace

255

8.3.2. Workplace constitutionalism An additional response to the ﬁrst one is that workplace regulation, but not democracy, is required in order to protect workers from abuses and arbitrary decisions of their employers—call it workplace constitutionalism.26 This is the sort of regulation we currently ﬁnd, somehow or other, in our advanced economies in the forms of international labor standards, such as those settled by the International Labor Organization; workers’ rights, such as those included in domestic constitutional documents and labor legislation; or professional and craft standards, such as those established at the sectorial level. From a republican standpoint such regulations are properly settled as long as they reduce managerial discretion and, accordingly, managers’ ability to make arbitrary decisions, including sexual harassment, racial discrimination in compensation and promotion, capricious distribution of overtime, assignment of hazardous tasks and so forth. Like the exit option argument, workplace constitutionalism (henceforth, WC) is concerned with the forms of arbitrary interference that can appear at the three levels of managerial decision-making mentioned above. Yet, unlike the exit option argument, it does not attempt to grant workers the right to costlessly quit, so they can use it as an ever-present threat in their day-to-day individual relation with the management. Rather, it deals with arbitrary interference by trying to set clear and speciﬁc standards to which managers have to conform in the exercise of their authority, reducing thus their discretion. This appears to be a much better means if arbitrary interference is to be prevented since, as Hodson has reported, the likelihood of arbitrary interference is much lower when command and/or supervision is indirect and standardized (e.g., through bureaucratic rules or craft and professional criteria) rather than direct and personal.27 In short, the aim of WC is not to balance bargaining power asymmetries between employers and employees, but to reduce managers’ discretion as much as possible and thus rule out their ability to make arbitrary decisions. Just to be clear, under WC employees’ and managers’ authority is not ruled out, for they still have ample margin to run the ﬁrm in one way or another, say, when there are changes in demand or new product strategies have to be implemented. However, they have to conform to a number of standards that reduce their discretion in the exercise of their authority. For example, under WC employers and managers can of course decide whether employee A will transport some merchandise to Philadelphia
of them ever said otherwise. Raventós, for instance, argues for a basic income just as part of a set of six necessary but not independently suﬃcient conditions for republican freedom. See Daniel Raventós, Basic Income: The Material Conditions of Freedom (London: Pluto Press, 2007), pp. 62-63. Even Barry’s statement is clearly more grounded in realpolitik than principled. 26 Domènech, El ecplise de la fraternidad; and Hsieh, “Rawlsian Justice and Workplace Republicanism.” Bottomley uses the term “corporate constitutionalism” in a diﬀerent sense, just to point out the political nature of ﬁrms and to “suggests that there are values and ideas in our public political life that can provide useful insights when considering the legal regulation of corporate governance and decision-making.” S. Bottomley, The Constitutional Corporation (Aldershot: Ashgate, 2007), p. 12. 27 R. Hodson, Dignity at Work (Cambridge: Cambridge University Press, 2004), p. 88, table 4.1.

256

Chapter 8. Republicanism at work

or rather to Pittsburgh and whether employee B rather than employee C will be promoted. But they cannot require employee A to drive above the number of hours permitted by the health and safety regulation statute, just as they cannot violate the anti-discrimination law in promoting employee B rather than C. Since republicanism is concerned with workers’ immunity from arbitrary interference as a necessary and suﬃcient condition of their freedom and WC appears to provide such an immunity through workplace regulation, it might seem that WC turns out to be suﬃcient to ensure republican freedom within the ﬁrm. A number of problems regarding WC arise though (which are analyzed in the next section in relation to WD). For now, they can be summarized in two main points: 1. WC does not prevent workplace regulation from being arbitrary itself, for it does not guarantee that the interests of workers are taken into account in its drafting and passing. 2. WC does not prevent workplace regulation from being arbitrarily enforced to particular cases and unforeseeable contingencies typical of the economic activity in a complex ﬁrm, since it does not guarantee that the interests of workers are taken into account in the process of enforcement. Before turning to analyze these two reasons, consider the following counterintuitive implication of WC. Let us assume that properly designed legal protections alone could ensure immunity from arbitrary interference within the ﬁrm. And let us turn next from the workplace to the political sphere. Then, why should we have a quarrel with non-parliamentary constitutional monarchy à la Bismarck—where citizens’ rights are entrenched in a constitutional text that constraints the monarch but, nonetheless, citizens are disenfranchised or the parliament lacks any meaningful legislative force?28 Now, that is the situation we have in most Western ﬁrms and the one WC recommends—a situation in which, somehow or other, workers enjoy civil rights (e.g., they cannot be discriminated on grounds of religion, race or gender) and social rights (e.g., they enjoy paid holidays or a minimum wage) but, unlike in a constitutional democracy, lack the right to have a binding say in managerial decisions. In most of these ﬁrms shareholders and their managers act, thus, as non-democratic constitutional monarchs.29 To the reasons why democracy and not only constitutionalism is as much needed in the ﬁrm as it is in the political sphere is that we turn now.

Following the Uniﬁcation of Germany in 1871, Otto von Bismarck rejected the British model. In the kind of constitutional monarchy established under the Constitution of the German Empire which Bismarck inspired, the Kaiser retained considerable actual executive power, and the Prime Minister needed no parliamentary vote of conﬁdence and ruled solely by the imperial mandate. 29 It might be objected that the analogy does not hold, for the constitution in a constitutional monarchy is not the result of citizens’ political participation while workers do participate, be it through their representatives in the parliament or at the sectorial level, in the making of the regulation in WC. However, the analogy still holds if we consider the case of an unelected town council constrained by the legislation passed at the state level by elected representatives, as it will become clear in section 8.4.

28

8.4. Workplace democracy

257

8.4. Workplace democracy
Freedom from arbitrary interference is not an all-or-nothing matter and the measures presented in the section above might well be necessary to ensure it in the workplace. This section aims to show that, however necessary, they turn out to be insuﬃcient in achieving this aim. To see why WD is also required to ensure this, consider one of the preconditions for republican freedom, according to which Political participation. X enjoys immunity against arbitrary interference only

if she is granted and regularly exercises a set of political rights that allow her to inﬂuence and contest the decisions aﬀecting her. Contrary to what some contemporary commentators have written, the reason why, within the republican tradition, political participation has been taken as a necessary, though not suﬃcient, condition for freedom is not that participation is intrinsically valued as part of a broader conception of the good life, as it is the case in the case of civic humanism. As Skinner recalls regarding classic republicanism, “the writers I am considering never suggest that there are certain speciﬁc goals we need to realise in order to count as being fully or truly in possession of our liberty.” “[They] merely argue that participation (at least by way of representation) constitutes a necessary precondition of maintaining individual liberty.” 30 On the contrary, the main rationale for political participation within the republican tradition is that, since the immunity against arbitrary interference can only be ensured institutionally (by the law), participation in its elaboration by those against whom the law is enforced turns out to be a necessary means—among other ones, such as the division of powers and the rule of law—to prevent that the law itself becomes arbitrary.31 This is so for two reasons. First, only political participation can ensure that all citizens’ interests are taken into account on an equal footing in the legislative process, even if only those of the majority are eventually turned into law. And second, active participation is required not only at the outset of the legislative process, but also during its implementation, as to ensure that citizens’ interests are adequately tracked and to prevent arbitrary enforcement and interpretation of the legislation when applied to particular cases and unforeseeable contingencies. Even if the ﬁrst reason were satisﬁed through citizens’ consent to a constitutional text, constitutional provisions are unavoidably to be applied to particular cases and contingencies that are not foreseeable ex ante—i.e., not fully speciﬁable in a constitutional document. That is why active and continuous participation of the citizenry is needed both in the law-making process and also in the day-to-day enforcement and interpretation of the law. As Rawls thoughtfully explained, Unless there is widespread participation in democratic politics by a vigorous and informed citizen body ... even the best-designed political institutions
30 Quentin Skinner, “The Paradoxes of Political Liberty,” in The Tanner Lectures on Human Values, VII (Salt Lake City: The University of Utah Press, 1986), 233; and Liberty Before Liberalism, pp. 74-75. 31 Pettit, “Law and Liberty,” pp. 53 ss.

258

Chapter 8. Republicanism at work

will eventually fall into the hands of those who hunger for power and military glory or pursue narrow class and economic interests ... If we are to remain free and equal, we cannot aﬀord a general retreat into private life.32 The drafting and enforcement of workplace regulation is not very diﬀerent. Consider the ﬁrst reason, according to which employees ought to take part in the elaboration of the norms regulating their workplaces. Otherwise, it is likely that such norms would turn into arbitrary themselves. As an instance of this, consider the case of guest workers that lack political rights in their host countries. For example, Philippine domestic workers have to sign a two-year employment contract to migrate to Hong Kong. Further, the law also requires that they must leave the country within two weeks after being ﬁred and that they cannot sign another contract in case they quit.33 As Hodson points out, these legal requirements are among the major reasons why Philippine maids accept continuous abuses, including those of the children they take care of. However, as we have seen in section 3, from a republican standpoint the freedom of these workers is diminished by the very fact of their being at the mercy of their employers, and not only by them being actually interfered. An adequate protection against the possibility of them being arbitrarily interfered should thus include workplace regulation. However, this regulation can turn into arbitrary itself unless those against whom it will be enforced also enjoy and exercise the right to inﬂuence its drafting, either directly or through their elected representatives, as it happens in the case of the Philippine maids. (To put it in a rather Schumpeterian fashion, since they lack voting rights, they also lack political purchasing power and, accordingly, is unlikely to emerge a political oﬀer to meet their demand.) In short, there is a clear republican case not only for workplace regulation but also for workers’ rights to take part in the elaboration of that regulation. Consider now the second reason, according to which active and continued participation is required in order to prevent arbitrary enforcement of the legislation. To be sure, many of the ways in which employers can arbitrarily interfere in the choices and actions of their employees are speciﬁable ex ante and are thus possible to legally prevent at the outset of the employment relationship. However, employment contracts are unavoidably incomplete, i.e. they do not specify “precisely what each of their obligations is in every conceivable state of the world.” 34 On the one hand, it would be impossible or prohibitively costly for the parties to the agreement to make their contract complete—much more so in large ﬁrms and complex and unpredictable economic environments. On the other hand, some ﬂexibility is desirable as to adequately address day-to-day unforeseen eventualities. For that reason, workers
J. Rawls, Justice as Fairness. A Restatement (Cambridge, Mass.: Harvard University Press, 2001), p. 141. It might seem somewhat odd to quote Rawls at this point. Yet it should be recalled that Rawls (Justice as Fairness, p. 144) himself cosidered classic republicanism and justice as fairness as fully consistent with each other. The idea is deeply rooted in the republican tradition. For example, according to Price: “If those in government are subject to no control from their constituents, the very idea of liberty will be lost.” And according to Prestley: “The more political liberty a people have, the safer is their civil liberty.” Both quoted in Pettit, Republicanism, p. 29. 33 Hodson, Dignity at Work, pp. 96-97. 34 O. Hart, “Incomplete Contracts and the Theory of the Firm,” Journal of Law, economics and Organizations, 4 (1988), 119-139, p. 121.
32

8.4. Workplace democracy

259

agree to subordinate to the command of the employer and the latter is given residual decision-making rights, that is, the ﬁnal say over decisions that have not been speciﬁed at the outset of the relationship. But the exercise of these rights opens the possibility of discretionary abuses taking place. WC attempts to reduce such abuses by means of regulation. However, exhaustive regulation is neither possible nor desirable given the very nature of the employment contract in large ﬁrms and complex economies. By contrast, WD attempts to reduce such abuses and arbitrary interferences by giving workers a say in the drafting of the decisions that aﬀect their lives in the workplace and also in the enforcement and interpretation of such decisions, which include managerial decisions at the three levels stated in section 3 and the more general labor legislation passed at the international, national or industry-wide levels, which no less than other pieces of law are often vague in their deﬁnition and uncertain in their implementation.35 Note that the reason for giving workers a binding say over the decisions made in the ﬁrm is not necessarily that there is a clash of interests between shareholders and workers in need to be balanced, so the latter’s interests are taken into account when decisions are to be made. Even if such a conﬂict did not arise, participation would still be necessary to track workers’ private knowledge and expertise and to close the epistemic gap between managers and workers. In ﬁrms in which division of labor and technical expertise are the norm, workers inevitably retain private information not only about their interests but also about how to improve their tasks and the overall functioning of the ﬁrm. To be sure, devices diﬀerent from binding participation, such as surveys or quality circles, can help track this knowledge. But, as theorists of descriptive representation often point out, those who are in the position of making decisions tend to understate this epistemic gap and take the interests and beliefs of those against whom decisions are enforced for granted (Phillips, 1995; Mansbridge, 1999). Binding participation in the decision-making process corrects that bias by forcing directors and managers to take such interests and beliefs into account. Further, it does so not only at the outset of the economic activity but also as the economic activity goes along, leaving room for unforeseeable contingencies to be addressed with ﬂexibility. Consider now two possible objections. First, it might be objected that WD adds little to WC, for in WC workers already enjoy control rights at the political level, having a say in the making and enforcement of workplace regulation. Put diﬀerently, it might be argued that the set of rights mentioned above—the right to inﬂuence and contest labor legislation, to bargain collectively or to unionize—are already included in the very idea of WC. Now, even if that were the case, a diﬀerence of scope between WC and WD would still exist. Consider as an analogy the diﬀerence between democratic rights at the state and municipal levels. It seems obvious that enjoying democratic rights at the state level is not a suﬃcient reason to prevent citizens from enjoying such rights at the municipal level as well, as long as there are decisions made at the latter level that aﬀect them. The same applies to the
35

See Bellamy, Political Constitutionalism, chap. 2.

260

Chapter 8. Republicanism at work

diﬀerence between WC and WD. Workers may well enjoy democratic rights at the political level, but that does not make democratic rights within the ﬁrm any less relevant, for there are decisions made in the ﬁrm whose potential arbitrariness cannot be prevented unless workers’ interests are properly tracked through their having a say in their making and enforcement. Further, given that people spend most of their daytime in the workplace, it is unreasonable to think that participation in the political sphere can replace participation in the workplace. Even further, participation might be more eﬀective in the workplace than in the political sphere, being the impact of each vote 1/n (where n = the number of participants in the decision) and being n much larger in the political sphere than in the workplace. In short, WD is to be seen as a multi-level arrangement, being the decisions made within democratic ﬁrms constrained by the decisions made at the industry-wide, legislative, constitutional and international levels, just as the decisions made at the municipal level are constrained by those made at the state and international levels. Secondly, it might be objected on pragmatic grounds that institutional devices other than workers’ participation in the ﬁrm might often better ensure in practice that their interests are rightly tracked and arbitrary interference ruled out. Consider judicial litigation and unionism as salient instances of such devices.36 To be sure, this is a complex and empirical issue that cannot be conclusively addressed independently of the background on which ﬁrms, courts and unions operate. The proper balance between the functions of democratic procedures in the ﬁrm, judicial litigation and unionism is thus inevitably context-dependent, as Nien-hê Shieh has pointed out.37 However, two responses may be advanced. As for judicial litigation, it presents two diﬃculties in contrast to WD. First, it addresses workers’ concerns ex post rather than incorporating them into the decision-making procedure itself. Second, it is carried out before courts, which, unlike boards in democratic ﬁrms, are not elected by workers and thus have fewer incentives to track their interests adequately. As for unionism, even though trade-oﬀs between WD and strong unionism can indeed arise, WD need not be inconsistent with the presence of strong unions. Actually, unions may be necessary for the proper functioning of WD, for their absence or weak presence may turn workers’ participation into ineﬀective, specially in large democratic ﬁrms, in which time-consuming decisions have to be made about complex issues such as investment policies, production engineering or budgetary planning.38 Further, unions are often the catalysts for the creation of workplace democracy when they buy failing conventional ﬁrms and re-organize them under the control of employees.

8.5. Ownership
Assume, for the sake of the argument, that everything said so far is correct and that a republican case for WD can be made. If that were the case, many questions would still remain unanswered. Just to mention a few, from a republican
36 37 38

See Hsieh, “Rawlsian Justice and Workplace Republicanism.” See Hsieh, “Rawlsian Justice and Workplace Republicanism.” See Sarryn Kasmir, The Myth of Mondragon (Albany, SUNY Press, 1997), esp. ch. 5.

8.5. Ownership

261

standpoint, on which issues and levels should workers be able to make decisions? By which procedures? Should workers participate directly or through their elected representatives? What status should part-time and temporary workers have in the decision-making process? What sort of input, if any, should stakeholders such as suppliers, consumers or the citizenry at large have? These are diﬃcult and important questions that are not addressed here. This last section focuses on another core issue regarding WD—namely, the ownership of the ﬁrm, which is deﬁned here as the ownership of non-human assets, such as ﬁnancial wealth or physical assets. Two models stand out in this regard, depending on whether ownership and control rights are separated or not. On the one hand, democratic ﬁrms in which capital is supplied by external investors but workers have control rights over the ﬁrm’s decisions. On the other hand, democratic ﬁrms in which workers have control rights but capital is also supplied by workers themselves, who ﬁnance collective asset ownership through debt contracting or by drawing upon their own savings. Thus, while ownership and control rights over the ﬁrm’s decisions are separated in the former case, they are kept together in workers’ hands in the latter case. I will take the German co-determination system (see section 5.2 below) and cooperativism as the most outstanding instances of the former and the latter, respectively. Now, which model is preferable from a republican point of view? The separation between ownership and control rights has been widely assumed in the literature on WD and has been put in practice extensively, as the German co-determination system has proven ever since 1976.39 However, this section attempts to show that cooperativism turns out to be preferable to co-determination due to bargaining power asymmetries and moral hazard problems arising from the formal separation of property and control rights. Let us ﬁrstly consider the republican conception of ownership as a source of material independence though.

8.5.1. Material independence Apart from slavery, republicanism has been traditionally concerned with many other forms of domination that, during the American colonial and early republican period (when republicanism was the mainstream political theory) had been referred to with the euphemism of relationships of “friendship.” Fathers were “friends” with their sons, men were “friends” with their wives and patrons were “friends” with their journeymen.40 Unlike the latter, the former were in a position to arbitrarily interfere in the choices and actions of the latter. As Sir Robert Filmer had declared in the seventeenth century regarding marriage and fatherhood, “The father of a family governs by no other law than by his own will.” What made the former dominate the latter is that, unlike the latter, the former owned property, which was seen as the main source of domination. “Domination” comes from “dominium,” which in classical Latin means both “property” and “the
See e.g. Hsieh, “Justice in Production,” p. 82. See G. S. Wood, The Radicalism of the American Revolution (New York: Vintage Books, 1991), p. 58.
40 39

262

Chapter 8. Republicanism at work

ability to make free use of this property.” 41 Thus it is hardly surprising that it was believed that those lacking property or dominium were in all likelihood to be dominated by those on whom they depended to make a living. Slavery, parenthood, marriage and wage labor were relationships where those who lacked property (slaves, children, wives and journeymen) were considered to be dominated by those on whom they depended materially (masters, fathers, husbands and patrons, respectively). Put brieﬂy, according to the republican tradition, Material independence. X enjoys immunity against the possibility of being

interfered arbitrarily only if she is materially independent. Republicans continuously used this condition to refuse granting full citizenship to those who lacked property. In Kant’s famous words, “The only justiﬁcation required by a citizen ... is that he must be his own master (sui iuris), and must have some property ... to support himself.” 42 It was thus argued that Requirement for full citizenship. she is materially independent. A number of reasons were provided to justify these two conditions. Two stand out. The ﬁrst and main one was that the have-nots were taken to be easily inﬂuenced by those on whom they depended. Had the propertyless been granted full citizenship, the only beneﬁciaries would have been those on whom they depended.43 The second reason was that (real estate) property served as a proxy for having stable interests in the community—a ‘stake in the society’—and thus an incentive to use political rights responsibly. This was so because owners had to feel directly the consequences of the laws passed, given that they were those over whom taxes were imposed. Two caveats are in order before seeing what these reasons have to do with the relationship between WD and ownership. First, it is certainly the case that traditionally most republicans used such reasons to exclude the majority from being granted full citizenship. However, a democratic branch can also be found within the republican tradition. The members of this branch argued for full citizenship without giving up neither the two conditions above nor the two reasons given to support them. Rather than defending that not everyone ought to be granted full citizenship due to widespread dependency, they argued the other way around—since everyone ought to be granted full citizenship, everyone should be granted material
See A. Domènech and D. Raventós, “Property and Republican Freedom: An Institutional Approach to Basic Income,” Basic Income Studies, 2 (2007), 1-8, at p. 3; and F. Lowett, A General Theory of Domination and Justice, Appendix 1. 42 I. Kant, Political Writings, ed. Hans S. Reiss (Cambridge: Cambridge University Press, 1970), p. 78. Hence, being propertyless was used as a suﬃcient condition for excluding someone from full citizenship, but not a necessary one, for many single women did own property and could meet property qualiﬁcations yet not enjoyed voting rights. Harrington had already expressed this non-biconditionality when he stated that “The man that cannot live upon his own must be a servant; but he that can live upon his own may be a freeman” (The Commonwealth of Oceana and A System of Politics, p. 269). 43 For example, according to Petty, a leveller spokesman in the Putney debates, “the reason why we exclude apprentices, servants and beggars is because they depend on the will of other men and they would be fearful of disgusting them,” quoted in Andrés de Francisco, Ciudadanía y democracia (Madrid, La Catarata: 2007), p. 139.
41

X is to be granted full citizenship only if

8.5. Ownership

263

independence as well.44 The second caveat is that few would nowadays accept that material independence relies on the ownership of real estate—rather than on enjoying a set of economic, cultural and social resources, as current neo-republicans often do.45 Neither would anyone consider nowadays wage labor as a relationship of domination per se.46 With these two caveats in mind, the remainder of the section attempts to show that the two reasons used to justify the requirement for full citizenship still apply nowadays, at least, regarding the possibility of formally separating ownership and control in the democratic ﬁrm. 8.5.2. Bargaining power asymmetries As we have seen above, republicans have traditionally taken those who lack material independence to be easily inﬂuenced by those on whom they depend. Within the ﬁrm, this can happen due to the inﬂuence of property ownership over bargaining power and, thus, over the actual exercise of rights. Put brieﬂy, if shareholders retain the ownership of the ﬁrm, it is very likely that ﬁnal decisions will be heavily inﬂuenced by their advantageous bargaining power. Consider the case of a ﬁrm in which control rights are extended to workers but, nonetheless, shareholders retain the right to sell their stocks and to collectively terminate the ﬁrm. In this instance, it is very likely that shareholders will determine workers’ decisions decisively through the ever-present threat of selling their shares en masse or terminating the ﬁrm. To be sure, bargaining power is to great extent determined by contextual factors such as the dependence of the ﬁrm on equity capital, the strength of unions or the unemployment rate. However, as long as they retain the ownership of the ﬁrm, shareholders will still enjoy a considerable bargaining power that workers, despite their formal control rights, will lack. And if, on the contrary, the right to terminate the ﬁrm is also transferred to—or, more precisely, shared with—workers, then it seems unclear whether we could still say that shareholders own the ﬁrm in any meaningful sense.47 Axel Gosseries has suggested that this problem could be avoided by using secret ballot, for it enables workers to freely express their actual will without being coerced.48 However, without neglecting its importance, secret ballot cannot by itself
As for the Early American Republic, some wings of the Jeﬀersonian Republican party are to be mentioned within this branch, e.g. the so-called Aurora-men of Pennsylvania. See Andrew Shankman, Crucible of American Democracy: The Struggle to Fuse Egalitarianism and Capitalism in Jeﬀersonian Pennsylvania (Lawrence, KS: Kansas University Press, 2004). The distinction between democratic and elitist republicanism is present, with diﬀerent terminologies, in Pettit, Republicanism; Domènech, El eclipse de la fraternidad; and De Francisco, Ciudadanía y democracia. For a contemporary critique of democratic republicanism as for the independence condition, see Robert E. Goodin, “Independence in Democratic Theory: A Virtue? A Necessity? Both? Neither?,” Journal of Social Philosophy, 24 (1993), 50-57. 45 See Pettit, Republicanism; and Domènech and Raventós, “Property and Republican Freedom.” 46 For two notable exceptions, see Pateman, “Self-Ownership and Property in the Person: Democratization and a Tale of two Concepts,” pp. 20-53; and D. Ellerman, “Translatio versus Concessio: Retrieving the Debate About Contracts of Alienation with an Application to Today’s Employement Contract,” Politics and Society 33 (2005): 449-480. 47 See López-Guerra, “Against the Parallel Case for Workplace Democracy,” pp. 20-21. 48 A. Gosseries, “Taking Workplace Democracy Seriously,” www.politika.lv/topics/civil_society/15968/ (accessed August 13, 2010).
44

264

Chapter 8. Republicanism at work

prevent the inﬂuence of bargaining power asymmetries over actual decision-making. If workers are aware that a decision x against decision y can make shareholders sell their shares en masse or even terminate the ﬁrm, it is very likely that many of them will change their mind and vote for y. Secret ballot will serve for little in this case. Note that bargaining power asymmetries also arise from asymmetries in human capital among workers. Hence, even if the shares of the ﬁrm were equally distributed among workers and shareholders, the most skilled workers would still enjoy a powerful bargaining position due to their ever-present ability to quit and ﬁnd another job easily. Two responses to this problem stand out. First, the problem applies not only to democratic ﬁrms but in fact to all type of ﬁrms, to the point that it is often the case that top managers and most skilled workers rather than shareholders end up controlling decisions made in the ﬁrm. Second, the problem can be partially overcome by oﬀering eﬃciency wages to the most skilled workers, in order to provide them with disincentives to quit and curb their ability to successfully threaten shareholders and fellow workers when decisions are to be made. (It might also be argued that the necessity and eﬀectiveness of such wages depend on many additional factors, including the greater degree of commitment of highly skilled workers in cooperative ﬁrms, as it is frequently argued.) The closest actual model to formal separation between ownership and control is the German co-determination law. As we have seen above in section 7.2.1, the act, in force since 1976, makes compulsory for every ﬁrm with over 2,000 workers to have a two-tier board structure. In the management board, which directly runs the ﬁrm, only shareholders’ executives are represented, while in the supervisory board, which makes the more general decisions (such as the ratiﬁcation of important investments or the approval of the annual budget), both shareholders and workers are represented on a near-parity basis. The literature on co-determination, however, shows that it does not make a real comparative diﬀerence regarding, among other issues, its economic performance and workers’ wages.49 This of course could be so just because the German model is not completely paritarian, or because supervisory boards have always been rather weak in Germany. For these reasons, any conclusion as for this matter should be cautiously taken with a grain of salt, even if consistent with the concerns raised here. 8.5.3. Moral hazard Assume that bargaining power asymmetries were balanced due to a shortage of labor supply or the existence of the above-mentioned basic income. Now, even if that were the case, moral hazard problems would still arise from the separation of ownership and control rights. As we have seen above, republicanism has traditionally tended to impose the requirement of property for citizenship not only in order to
Benelli et al.’s study of co-determination in Western Germany showed, for example, that its implementation did not lead to less risky investments, smaller dividends, reduced ﬁrm leverage, higher and more stable salaries or more capital-intensive production. See G. Benelli, C. Loderer and T. Lys, “Labor Participation in Corporate Policy- Decisions: West Germany’s Experience with Codetermination,” Journal of Business 60 (1987): 553-575. For a summary of the ﬁndings, see Dow, Governing the Firm, pp. 86 ss.
49

8.5. Ownership

265

prevent the participation of those who can be easily inﬂuenced. But also because of owners’ tendency to have stable interests in their community and thus behave responsibly in their political decisions. This argument can be applied to the present context as follows. Consider again the case of a ﬁrm in which control rights but not ownership are transferred to or shared with workers. Further assume, not unrealistically, that information asymmetries exist between shareholders and workers, who now enjoy the opportunity to raise their concerns and proposals and have them taken into account through a binding procedure. Since workers do not have a share in proﬁts, they will have an incentive to use their control rights, in combination with their private information, opportunistically. To be sure, workers will still have, as employees in non-democratic ﬁrms do, a long-term interest in the survival of the ﬁrm—namely, not to get unemployed due to its bankruptcy. Contrary to the latter, however, these workers will also have a say in the running of the ﬁrm, thus having an incentive to free-ride and use their private information to make decisions in their own interests when such a possibility is available to them. As Hansmann points out, ‘”if those with control had no claim over the ﬁrm’s residual earnings, they would have little incentive to use their control to maximize those earnings, or perhaps even to pay out the earnings received.” 50 To put it clearly: by separating control rights and ownership, workers are given further rights to control the ﬁrm but not the proper incentives to use such rights in the ﬁrm’s best interests. In short, even though the possibility to formally separate ownership and control rights is often assumed, serious doubts arise about its feasibility. One solution to these problems is to put ownership and control rights together in shareholders’ hands. Another one, though, is to put them together in workers hands by promoting cooperatives—where workers retain both control and ownership—instead of other forms of WD such as co-determination. However, since many other normative and pragmatic considerations also need to be taken into account, this can only be a prima facie argument for cooperativism.

50

Hansmann, The Ownership of Enterprise, p. 12.

Concluding remarks to Part III
Part III of the dissertation has analyzed the extension of democratic decision-making to the workplace, a sphere that is usually taken to be beyond the scope of democracy. To be sure, workplace democracy is less pressing an issue if we compare it to constitutionalism, which expansion in the last decades has been astonishing. Even though workplace democracy was a central research topic in the 1970s and 1980s, it is without doubt that the topic has attracted less and less attention ever since the early 1990s (something that, as it has been said above in section 7.3.1, could change with the current economic crisis). That said, the democratization of the workplace is worth studying for present purposes because it provides a crystal-clear case-study of the limited scope of democracy for it is usually taken to be beyond the reach of democracy. The goal of Part III has been to analyze the main arguments for and against the extension of democratic decision-making to the workplace and to provide an original, republican case for WD. Chapter 7 has introduced and deﬁned the main forms of WD and has reviewed in some detail the core arguments for and against WD regarding its feasibility and eﬃciency, along with the so-called “epistemic” and “propertarian” arguments against WD and Bowles and Gintis’ theory of contested exchange, which they use to make a case for WD. Chapter 8, in turn, has put forward a novel, republican case for WD and has analyzed whether WD implies, from a republican standpoint, not only workers’ control rights but also workers’ ownership of the ﬁrm. It has been argued that neither the availability of an exit option from the ﬁrm nor workplace constitutionalism (i.e. regulation) are suﬃcient to ensure workers’ freedom from arbitrary interference in the workplace. From a republican standpoint, citizens’ active and continued political participation in the ﬁrm’s decision-making is required for two reasons—because workplace regulation can turn into arbitrary itself unless those against whom it is enforced also enjoy and exercise the right to inﬂuence its drafting and because workplace regulation can be arbitrarily enforced unless those against whom it is enforced also enjoy and exercise the right to inﬂuence its enforcement. Put simply, since it would be impossible or prohibitively costly for the parties of the employment relationship to make their contract complete and some ﬂexibility is also desirable as to adequately address day-to-day unforeseeable eventualities, workplace regulation turns out to be insuﬃcient. Workers are thus to be given a say not only at the outset of the employment relationship but also as the relationship goes along. Finally, it is

268

Concluding remarks to Part III

argued that WD requires, from a republican standpoint, not only workers’ control but also workers’ ownership of the ﬁrm due to bargaining power asymmetries and moral hazard problems arising from the formal separation of ownership and control rights. It is thus shown that cooperativism is ceteris paribus preferable to other forms of WD in which ownership and control rights, such as co-determination. The republican case for WD is connected to the previous parts of the dissertation in three ways. First, republicanism was the mainstream political though in the Early American Republic. Thus chapter 8 draws heavily upon Part I of the dissertation. Second, Part I shows how the employment relationship, which had been traditionally seen as a form of “limited slavery,” as Aristotle put it, turned out be seen as a free and voluntary exchange among juridical and how, as a consequence of this, the workplace turned out to be seen as a sphere in which power relations are absent and that is, accordingly, beyond the scope of politics, including democratic politics. 1.3.3. Third, even though Part II of the dissertation has shown that constitutionalism faces serious problems from a democratic point of view, it has not shown why democracy ought to trump other sorts of considerations and why it is legitimate at all as a decision-making procedure. Of course, there are many reasons why democracy may be legitimate as a decision making procedure. Chapter 8 precisely provides a republican rationale among the many available justiﬁcations of democracy.

General conclusion
The dissertation has analyzed the limits and scope of democracy both historically and normatively. First, it has explained historically how democracy turned from being despised into being cherished in less than half a century, and how this shift went hand-in-hand with another shift, according to which a number of institutional devices that had been traditionally seen as limits on the scope of democracy turned out to be seen as necessary for a good-working democratic system. Second, it has analyzed normatively two outstanding instances of such limits—namely, (a) constitutional constraints on legislative action and (b) decision-making in the workplace, a domain that is usually considered to be beyond the scope of democracy—, and has shown that there are good grounds to call into question the justiﬁcation of those limits from a democratic standpoint. The ﬁrst, and arguably most striking, problem I have addressed in the dissertation is strictly historical—namely, to explain the reasons why democracy, a form of government that had been despised for more than two thousand years, turned to be unanimously cherished in less than half a century. Part I of the dissertation has focused on the Early American Republic, a period in which the shift is outstandingly acute and which world-wide political inﬂuence has been enormous. Two trends are found behind the shift. First, an increasing acceptance of the idea of democracy due to the crisis of the theory of virtual representation and the acceptance of the fact of disagreement as a legitimate political phenomenon. Second, a shift in the very concept of democracy, which would come to accommodate elements that had been traditionally seen as constraints on democracy. These two trends help to explain why democracy, traditionally despised for being unstable, unfeasible, and incompatible with property rights and the division of powers, turned into a buzzword which legitimacy is nowadays rarely contested. Likewise, they help explain the reason why the constraints on the scope of democracy that have been analyzed in Part II and Part III are nowadays seen as perfectly consistent with democracy. As William Duane wrote to Jeﬀerson as early as 1824, “Democratic government is no longer Jacobinism; and those who formerly reprobated now use the language and profess the doctrine they reviled twenty four years ago.” A “revolution in speech,” as Duane labeled it, had been produced.51 Chapter 1 has shown that the idea of democracy gained increasing support as soon
W. Duane, “Letter to Jeﬀerson, October 19, 1924,” Proceedings of the Massachusetts Historical Society 20 (1906-1907): 381-383, p. 382.
51

270

General conclusion

as ordinary people assumed the fact of disagreement as an unavoidable and legitimate feature of the American society and the theory of virtual representation—including the organicist conception of the society as an entity with a “general will” that is beyond political struggle and partisanship— gradually lost its credibility and political eﬃcacy. Accordingly, some of the institutional devices traditionally attached to the idea of democracy were increasingly demanded and suﬀrage was eventually extended. Of course, the theory of virtual representation was not left aside once and for all and, in fact, its inﬂuence is still there nowadays. Thus, governments rarely admit that their actions beneﬁt some citizens at the expense of others, and open declarations of partisanship—such as the one by President Salvador Allende, who declared that “I am not the President of all the Chileans,” or the one by the former British Prime Minister Edward Heath, who openly declared to represent class interests—tend not to be without political costs.52 However, as the nineteenth century progressed, the use of the theory would lose its former credibility and the fact of disagreement would be increasingly assumed, giving way to the acceptance of political struggle and democracy as the legitimate means to make decisions in the face of deep disagreements. In a nutshell, democracy was called for only as long as the fact of disagreement was assumed as an unavoidable and legitimate feature of the American society. While the organicist conception of the society remained intact and no disagreement was acknowledged, no democratic decision-making mechanism was required and decisions could be made by an elite that allegedly represented the shared interests of the society. Chapter 2, in turn, has shown that, besides its increasing acceptance, the idea of democracy also went through a notable semantic shift in the ﬁrst half of the nineteenth century. Although some features of the concept remained the same, some others changed. Likewise, a number of institutional devices that had been traditionally taken to be constraints on democratic decision-making (e.g. representation, judicial review of legislation, the protection of property rights) turned out to be seen not only as consistent with democracy but actually as necessary for any good-working democratic system. This helps explain why the Federalists believed that the representative government they were putting forward was not only diﬀerent from democracy but in fact a very check against it. As Bernard Manin puts it, “for Madison, representative government was not one kind of democracy; it was an essentially diﬀerent and furthermore preferable form of government.” 53 The institutional upshot of these two trends is what we usually label as liberal democracy, a form of government in which sovereignty rests on the people and is exercised by the people—as a general rule, through their elected representatives—yet the scope of democratic decisions is heavily constrained by a number of issues that are removed from majority rule and protected by non-majoritarian bodies, such as
Allende and Heath are both quoted in B. Manin, A. Przeworski, and S. Stokes, “Introduction,” in Democracy, Accountability, and Representation, A. Przeworski, S. Stokes, B. Manin, ed. (Cambridge: Cambridge University Press, 1999), pp. 6-7. 53 B. Manin, The Principles of Representative Government, (Cambridge: Cambridge University Press, 1997), p. 2.
52

271 constitutional courts and independent central banks. Hence, democratic institutions have expanded over the last two centuries, but only as much as such constraints on democratic decision-making—which have in fact turned to be seen as necessary for a well-functioning democracy—have also expanded. (See, for example, the much-used indicators of the quality of democracy elaborated by agencies such as Freedom House, the Economist Democracy Index, or Polity IV).54 In short, the main contribution of Part I of the dissertation has been to explain the origins and rise of liberal democracy by analyzing historically the above-mentioned two trends. On the one hand, the way democracy, which had been traditionally despised, turned out to be unanimously cherished and demanded as the nineteenth century progressed. On the other hand, the way a number of institutional devices that had been traditionally seen as limits on democracy turned out to be included in the very concept of democracy, so that no well-functioning democratic system could go without them. Part II and Part III of the dissertation, by contrast, are less historical. They have normatively analyzed two outstanding instances of the limits on the scope of democracy and have called into question their justiﬁcation from a democratic point of view. Thus Part II has analyzed the justiﬁcation of constitutional constraints on legislative action—and, more precisely, constitutional rigidity and judicial review—from a democratic point of view. This is a very pressing issue, given the recent world-wide expansion of constitutionalism (to the point that nowadays three-quarters of the world’s states have some form of judicial review). Further, since judicial review originated in the Early American Republic as a check on democracy that only later turned out to be considered democratic, it provides a ﬁrst-rate case-study of the democratic legitimacy of the limits on the scope of democracy. Three prominent attempts to justify constitutionalism on democratic grounds have been analyzed—pure instrumentalist theories, precommitment-based theories, and proceduralist theories. According to pure instrumentalist theories, constitutionalism need not be at odds with democracy because constitutional constraints are necessary if democratic procedures are to bring about just outcomes (chapter 4). According to precommitment-based theories, constitutional constraints need not be at odds with democracy because they are the upshot of precommitments made by the people in constitutional moments (chapter 5). Finally, according to proceduralist theories, constitutional constraints need not be at odds with democracy because they set up the procedural rules and preconditions required for the proper functioning of the democratic system (chapter 6). A number of problems of these theories have been raised. However, one problem that is common to all of them is that they do not address adequately the fact of disagreement that, as we have seen in Part I, turns out to be a necessary condiSee among others the Freedom of the World Reports by Freedom House, the yearly Economist Democracy Index, the United Nations Development Program Human Development Report 2002, Deepening Democracy in a Fragmented World (Oxford and New York: Oxford University Press, 2002), or Polity IV reports, such as M, G. Marshall and B. R. Cole, Global Report 2009: Conﬂict, Governance, and State Fragility (Vienna: Center for Systemic Peace, 2009).
54

272

General conclusion

tion of democratic legitimacy. They all assume or end up assuming that there are a number of political issues that are beyond dispute and can be entrenched in a constitutional text and advanced by constitutional means. However, as it is carefully shown throughout Part II, even though constitutions tend to be introduced as complete contracts that are beyond dispute, they are inevitably pervaded by the fact of disagreement at a number of levels—e.g. their drafting, their ratiﬁcation, their interpretation, and their enforcement. Even further, the fact of disagreement is a necessary condition of politics, for if all citizens agreed on what ought to be done across the board no political means for decision-making would be required in the ﬁrst place. As Hannah Arendt put it, there is politics because “not man, but men inhabit the world.” 55 And we all disagree with one another—including constitutional judges, who use majority rule to reach their decisions as much as parliaments do. Even though this problem does not rule out the possibility of these theories being justiﬁed on non-democratic grounds, it shows that it is very problematic to do so on democratic grounds. This is the only point that I have tried to make in Part II, and of course it is far from being a completely novel point, for Waldron and Bellamy have argued similarly well before me. However, I hope that Part II has contributed to summarize and clarify the main theories that attempt to reconcile democracy and constitutionalism, to show why they fail to achieve this goal, and to add a number of original arguments to the cause, such as the application of Wollheim’s paradox to the critique of pure instrumentalism (section 4.3.2), or the use of path dependence theories to show the informal asymmetry between the constitutional founding and the constitutional amendment (section 5.4.4). Finally, Part III of the dissertation has addressed another outstanding instance in which the scope of democracy is heavily constrained—namely, decision-making in the workplace. In a liberal democracy there are number of spheres that tend to be insulated from political decision-making and enacted by non-majoritarian institutions such as constitutional courts and independent central banks. Consider, for example, the family and monetary policy. Even though for very diﬀerent reasons—the former for being purely private; the latter for being purely technical—, both of them tend to be considered unpolitical and, accordingly, tend to be insulated from ordinary politics. Now, the workplace oﬀers a ﬁrst-rate example of a domain that is usually taken to be beyond the scope of democratic decision-making precisely because it is seen as a sphere of voluntary relations among free and equal individuals. Accordingly, the democratization of the workplace would be redundant at best—for “persons may form their own democratically-run cooperative ﬁrms,” as Nozick put it—and an illegitimate interference in the contractual conditions which employers and employees have freely and voluntarily agreed upon at worst.56 After having addressed the most prominent arguments for and against workplace democracy, and having called into question the conclusiveness of the main arguments against extending democracy to the workplace, the main contribution of Part III
55 56

H. Arendt, Men in Dark Times (San Diego: Harcourt Brace & Co., 1968), p. viii. R. Nozick, Anarchy, State, and Utopia (New York: Basick Books, 1974), p. 250.

273 is to have provided a partly original—though only partly, for Nien-hê Hsieh has argued similarly—, republican argument for workplace democracy. Borrowing the normative deﬁnition of freedom from the republican tradition of political philosophy and thus heavily drawing upon Part I of the dissertation (for republicanism was the mainstream political thought in the Early American Republic), chapter 8 has shown that neither the availability of an exit option nor workplace regulation alone can ensure immunity from arbitrary interference in the workplace. The critique of the insuﬃciency of workplace regulation is key to the republican case for workplace democracy. Using ideas partly borrowed from incomplete contract theory, it has been shown that workplace regulation can become arbitrary itself unless those against whom it is enforced also enjoy and exercise the right to inﬂuence both its elaboration and enforcement. Put simply, since it would be impossible or prohibitively costly for the parties of the employment relationship to make their contract complete, or for the authorities to completely regulate the relationship, and since some ﬂexibility is also desirable as to adequately address day-to-day unforeseeable eventualities, discretion in the workplace is not only unavoidable but also desirable. Now, the existence of discretion in the enforcement of the employment contract and workplace regulation opens the door to disagreement among parties and to the possibility of arbitrary interference taking place. Thus, if immunity from arbitrary interference is to be ensured in the workplace, workers are to be given a say not only at the outset of the employment relationship but also as the relationship goes along. In short, the insulation of the workplace from democratic decision-making may be justiﬁed on a number of grounds (e.g. economic eﬃciency), but not because it is an unpolitical sphere in which the parties can settle the terms of the relationship in a complete contract, or the authorities regulate it completely through labor legislation. Just as allegedly complete constitutions turn out to be willy-nilly pervaded by the fact of disagreement, employment contracts and workplace regulation are also incomplete and thus equally pervaded by the fact of disagreement among parties. As it has been pointed out in the Introduction of the dissertation, the three parts of the dissertation are intended to be self-standing and independent from each other. However, the sources, role, and pervasiveness of the fact of disagreement and its relationship with democratic decision-making and democratic legitimacy is common to all of them. Thus Part I of the dissertation has historically shown how the fact of disagreement came to be acknowledged in early nineteenth-century America, and how this acknowledgment gave way to the crisis of the theory of virtual representation and to an increasing acceptance of democracy as a means to address such disagreement. In short, the demand for democracy only triggered once the fact of disagreement had turned to be seen as an unavoidable and legitimate feature of the American society. As it is put more analytically in Part II (sections 4.3.1 and 4.3.2), the fact of disagreement is a necessary condition of democracy being called for, for if all citizens agreed on what ought to be done across the board, no decision-making procedure would be required at all. In turn, Part II has focused on how constitutions, which are often presented as being beyond dispute, turn out to be unavoidably pervaded

274

General conclusion

by the fact of disagreement and has argued that, accordingly, there is no reason to insulate them from democratic decision-making. Finally, Part III has shown that the workplace, a sphere that is usually taken to be beyond the scope of politics, is equally pervaded by the fact of disagreement and has provided an argument for its democratization based on the republican notion of freedom as immunity from arbitrary interference.

Resumen en castellano
Durante las últimas décadas se han producido dos tendencias políticas aparentemente opuestas. Por una parte, la extensión global de las instituciones democráticas, así como de la ideología democrática, y por otra, la proliferación de instituciones no-mayoritarias, tales como bancos centrales independientes y tribunales constitucionales, con la consecuente limitación del alcance de la toma de decisiones democráticas. En suma, las instituciones democráticas se han extendido recientemente tanto como el alcance de la toma democrática de decisiones se ha visto limitado. Estas dos tendencias han ido de la mano de dos tendencias paralelas, si bien más ﬁlosóﬁcas y de más largo alcance. Por una parte, la crisis de la concepción metafísica de la sociedad como una entidad orgánica con una “voluntad general”, con la consiguiente aceptación del pluralismo y la confrontación política como fenómenos legítimos y, por otra, la crecientemente inﬂuyente idea de que hay un conjunto de cuestiones que han de ser extraídas a la política ordinaria y transferidas a organismos no electos, ya sea porque han de quedar fuera de la negociación política cotidiana (por ejemplo, los derechos humanos), o porque son puramente técnicos (por ejemplo, la política monetaria), o incluso porque son estrictamente privadas (por ejemplo, las relaciones familiares) y no deben ser sometidas a la decisión mayoritaria. Consideremos brevemente las dos primeras tendencias. Por una parte, la democracia se ha extendido rápidamente desde la Segunda Guerra Mundial, como muestra la ﬁgura 1 con respecto a la proporción de países con sufragio universal. Mientras que en 1900 sólo 6 de los 43 estados existentes eran mínimamente democrático, y sólo uno tenía sufragio universal, en 1946 20 de los 71 estados independientes eran ya democráticos. Y a principios de la década de 1990 la cifra pasó de 48 estados en 1989 a 77 en 1994, una cifra que aumento hasta 92 estados en 2009 y que es probable que continúe aumentando tras la llamada “primavera árabe”.57 La ideología democrática se ha extendido con igual rapidez. Hasta el punto de que es frecuente que gobiernos abiertamente no democráticos se reﬁeran a sí mismos como “democráticos” de manera oportunista, siendo los ejemplos más obvios el de la República Democrática de Alemania, la República Popular Democrática de Corea, la “democracia orgánica” de

57 Véase R. A. Dahl, On Democracy (New Haven: Yale University Press, 1998), p. 14; D. Potter, “Explaining Democratization,” en Democratization, ed. D. Potter, D. Goldblatt, M. Kiloh y P. Lewis (Cambridge: Polity Press, 1997); y M, G. Marshall y B. R. Cole, Global Report 2009: Conﬂict, Governance, and State Fragility (Vienna: Center for Systemic Peace, 2009), pp. 10-11.

276

Resumen en castellano

Franco, o el intento de Pinochet de “dar a la democracia una dimensión auténticamente representativa”.58 Por otra parte, el alcance de las decisiones democráticas ha sido limitado tanto como las instituciones democráticas se han visto extendidas. No sólo debido a la creciente presencia de facto de actores internacionales, como las empresas multinacionales o los llamados “gnomos de Zurich”, como el exprimer ministro británico Harold Wilson se reﬁrió célebremente a los especuladores ﬁnancieros. Sino también por la transferencia de iure de competencias parlamentarias a instituciones no-mayoritarias (es decir, instituciones cuyos miembros no son directamente elegidos por la ciudadanía, ni por los cargos electos), tales como el Banco Central Europeo en el ámbito supranacional, o los tribunales constitucionales a nivel nacional. Mark Tatcher y Alec Stone Sweet han descrito este proceso en el contexto europeo del siguiente modo: En campos como la regulación de los servicios, las telecomunicaciones, la competencia económica, y el pluralismo mediático, e incluso en la prestación de servicios sociales y sanitarios, se han creado un sinnúmero de organismos reguladores independientes que se han convertido en lugares en los que se crea nueva regulación, o se aplica la ya existente a nivel nacional. En el ámbito supranacional, los bancos centrales, aislados del control político directo, ﬁjan la política monetaria. En Bruselas, funcionarios de la Comisión Europea elaboran la legislación y la hacen cumplir en el ámbito cada vez más amplio de la Unión Europea. En Luxemburgo, el Tribunal de Justicia controla el cumplimiento del derecho europeo por parte de los Estados miembros con, incluyendo a los parlamentos nacionales, los gobiernos y las administraciones.

Figura 1. Proporción de países con sufragio universal por año.59

58 A. Pinochet, “Chile should Not Fall into the Vices of the Past,” in The Politics of Antipolitics, B. Loveman y D. Thomas, eds. (Wilmington, DE: Scholarly Resources, 1997), pp. 248-249. 59 A. Przeworski, “Conquered or Granted? A History of Suﬀrage Extensions,” British Journal of Political Science 39 (2008): 291-321, p. 292.

277 Estas dos tendencias han ido de la mano de dos tendencias paralelas, si bien más ﬁlosóﬁcas y de más largo alcance. Por una parte, la aceptación del desacuerdo y la confrontación política como fenómenos inevitables y legítimos frente a la concepción clásica de la sociedad como una entidad orgánica con intereses comunes que los partidos y la contienda política sólo podían socavar. Como veremos en el capítulo 1, la concepción orgánica de la sociedad había sido predominante hasta ﬁnales del siglo XVIII. Se trata de una concepción que puede ser divida en dos ideas básicas. La primera idea es que la sociedad es un cuerpo orgánico con un conjunto común de intereses y valores que están más allá de los intereses particulares. La analogía con el cuerpo apareció en la época medieval, pero incluso cuando el contractualismo llegó a ser dominante, las partes contractuales continuaban siendo vistas como partes de un todo. No es por casualidad que “el pueblo” siempre haya aparecido en singular, como le peuple, el pueblo, das Volk, el poble, herria, etc.: como una entidad orgánica con una voluntad uniﬁcada.60 La segunda idea es que las facciones sociales y políticas deﬁenden sus intereses particulares y, al hacerlo, ponen en peligro la unidad social. Como lo expreso Rousseau con claridad, “[c]uando los intereses particulares comienzan a hacerse sentir y las partes de la sociedad comienzan a ejercer una inﬂuencia (...) el interés común queda corrompido.” 61 En esta visión organicista la democracia queda descartada, ya que implica el desacuerdo político y la confrontación parlamentaria y éstos, a su vez, amenazan los intereses prepolíticos y orgánicos de la sociedad. Como lo expresó Hans Kelsen en 1929, [E]l ideal de un interés general superior y que trasciende los intereses de los grupos, y de los partidos, el ideal de una solidaridad de intereses de todos los miembros de la colectividad, sin distinciones de religión, nacionalidad, clase, etc. es una ilusión metafísica y, más exactamente, una ilusión metapolítica, expresada habitualmente en la forma (...) de una colectividad “orgánica” o una estructura “orgánica”. Ahora bien, según avanzó el siglo XIX, esta concepción organicista fue dando paso a una aceptación cada vez mayor del pluralismo y la discrepancia tanto al nivel de los intereses como de los valores y, por lo tanto, a una mayor aceptación de la confrontación política y el partidismo como medios inevitables y legítimos para la realización de dichos intereses y valores. Por supuesto, el reconocimiento de la existencia del desacuerdo y el partidismo es tan antiguo como Aristóteles, quien célebremente aﬁrmó que “los ricos y los pobres se consideran partes de una ciudad-estado. Además ... entre las partes de la ciudad-estado estos dos son opuestos”.62 Sin embargo, la oposición política y el partidismo no pasarían a ser tomados como políticamente legítimos —en lugar de como enemigos de la cohesión social— hasta mucho más
Véase A. Przeworski, “Consensus and Conﬂict in Western Thought on Representative Government,” Procedia Social and Behavioral Sciences 2 (2010): 7042–7055, p. 7046; y Democracy and the Limits of Self-Government (New York: Cambridge University Press, 2010), p. 19. 61 J.-J. Rousseau, Du contrat social, in Oeuvres complètes, vol. 3, B. Gagnebin and M. Raymond, eds. (Paris: Gallimard, 1961). Citado a partir de J.-J. Rousseau, The Social Contract, M. Cranston, trans. (New York: Penguin, 1968), p. 99. 62 Aristóteles, Politics, C. D. C. Reeve, trans. (Indianapolis: Hackett, 1948), 1291b.
60

278

Resumen en castellano

recientemente. Incluso Marx y Engels, quienes célebremente deﬁnieron “la historia de toda sociedad existente hasta el presente [como] la historia de la lucha de clases”, aún conﬁaban en “la fundación de una nueva sociedad sin clases”.63 Ahora bien, por razones que quedarán claras en el capítulo 1 de la tesis, la presencia del pluralismo y el desacuerdo social, económico, político y cultural se haría más acuciante a medida que avanzó el siglo XIX, de modo que la concepción organicista ya no podría defenderse con la credibilidad anterior. Ya a mediados del siglo XX, Schumpeter llevaría esta tendencia hasta el extremo al aﬁrmar que “[n]o hay tal cosa como un bien común en el que todas las personas puedan ponerse de acuerdo o sobre el que ponerse de acuerdo mediante la fuerza de la argumentación racional”.64 En consecuencia, la confrontación política y el partidismo ganarían legitimidad y la democracia, tradicionalmente despreciada por ﬁlósofos de todo tipo, conseguiría ﬁnalmente un pedigrí ﬁlosóﬁco respetable. Como lo ha expresado Habermas más recientemente, en “condiciones de pensamiento postmetafísico” no existen objetivos prepolíticos sustantivos; éstos sólo pueden deﬁnirse como resultado del proceso democrático, que es la única fuente de legitimidad de la autoridad política.65 Por otra parte, según la concepción orgánica fue quedando atrás y el desacuerdo y la confrontación política característicos de la política democrática ganaron aceptación, también se extendió una idea aparentemente opuesta. La idea de que existe un conjunto de cuestiones que no debe ser sometido a la regla de la mayoría, ya sea porque dichas cuestiones están fuera de discusión, o porque son puramente técnicas o porque son estrictamente privadas y deben ser protegida frente a la “tiranía de la mayoría”, en la célebre formulación de J. S. Mill.66 Consideremos los derechos de propiedad como un ejemplo de este desplazamiento. A pesar de que “las democracias [habían] sido consideradas incompatibles con la seguridad personal, o los derechos de propiedad”, como lo expresó James Madison, la protección de los derechos de propiedad ha pasado a ser considerada en la actualidad no sólo como compatible con la democracia, sino de hecho como necesaria para el correcto funcionamiento de todo sistema democrático.67 Como lo ha descrito John Dunn, “[n]adie en 1750 vio o pudo haber visto la democracia como un nombre apropiado o una forma institucional apta para la protección efectiva de la riqueza productiva. Pero hoy sabemos que no. Ante la probabilidad percibida ex ante, esto es exactamente lo que la democracia representativa ha demostrado a la larga”.68 Y, como veremos en el capítulo 2 de la
K. Marx y F. Engels, “Manifest der Kommunistischen Partei,” en Werke, vol. 4 (Berlin: Dietz Verlag, 1971). Citado a partir de K. Marx y F. engels, The Communist Manifesto, S. Moore y F. Engels, trans. (London: Pluto Press, 2008), p. 33; and “Rules of the Communist League,” art. 1. 64 J. A. Schumpeter, Capitalism, Socialism, and Democracy (New York: Harper & Brothers, 1942), p. 250. 65 Véase J. Habermas, Faktizität und Geltung. Beiträge zur Diskurstheorie des Rechtes und des demokratischen Rechtsstaats (Frankfurt am Main: Suhrkamp, 1992). 66 J. S. Mill, “On Liberty,” en Collected Works of John Stuart Mill, vol. 18, J. Robson, ed. (Toronto: University of Toronto Press). 67 J. Madison, “The Federalist no. 10,” en The Federalist, G. W. Carey y J. McCleallan, eds. (Indianapolis: Liberty Fund, 2001), p. 67. 68 J. Dunn, “Democracy before the Age of the Democratic Revolution,” conferencia presentada en Columbia University, 2003, p. 10.
63

279 tesis, este cambio en el alcance y los límites de la toma democrática de decisiones ha afectado también a muchos otros temas, incluida la representación, el constitucionalismo, la división de poderes o el empleo. En resumen, cuanto más se ha extendido la democracia y más legitimidad han adquirido las instituciones democráticas, más limitado se ha visto su alcance por todo un conjunto de restricciones institucionales que habían sido tradicionalmente consideradas como incompatibles con la democracia y que han pasado a ser incorporadas a la idea misma de democracia. El resultado de estas dos tendencias, tanto en el nivel político como en el ﬁlosóﬁco, es lo que solemos denominar como democracia liberal. Un sistema político en el que la soberanía reside en el pueblo y es ejercida por el pueblo, ya sea directamente o a través de sus representantes electos, y en que, no obstante, el alcance de las decisiones democráticas se encuentra fuertemente limitado por una serie de cuestiones que quedan extraídas a la decisión mayoritaria y son protegidas por institutciones no-mayoritarias, como los tribunales constitucionales o los bancos centrales. Los dos objetivos principales de esta tesis son explicar históricamente la aparición y difusión de estas dos características de la democracia liberal en el caso de la Joven República Norteamericana (Parte I) y analizar más normativamente dos tipos de limitaciones al alance de la toma de decisiones políticas en una democracia liberal: señaladamente, los límites constitucionales impuestos sobre la actividad legislativa (Parte II) y la toma de decisiones en las empresas, un ámbito que se considera ajeno al alcance de la democracia (Parte III). La Parte I se centra en la Joven República Norteamericana, un período en el que las dos tendencias mencionadas son especialmente acusadas y cuya inﬂuencia política posterior difícilmente puede ser exagerada. Tiene dos objetivos principales, a los que están dedicados el capítulo 1 y el capítulo 2, respectivamente. En primer lugar, mostrar cómo la democracia, una forma de gobierno que había sido criticada de manera generalizada durante más de dos mil años, ganó legitimidad y fue crecientemente demandada en la medida en la que el desacuerdo político fue aceptado como una característica inevitable y legítima de la sociedad estadounidense (debido, entre otras causas, al desarrollo de la economía de mercado y el surgimiento del primer y el segundo sistemas de partidos, como veremos en la sección 1.3.1). En segundo lugar, mostrar que la misma idea de democracia también sufrió un notable cambio semántico que hizo que incluyese una serie de elementos que habían sido tradicionalmente considerados como limitaciones a la democracia. Este cambio permite explicar un fenómeno bien conocido y aún así llamativo: que los fundadores del sistema constitucional estadounidense se reﬁriesen a éste como un sistema claramente no democrático y que, apenas medio siglo después, la gran mayoría de los norteamericanos —los herederos de los Federalistas incluidos— se reﬁriesen al mismo sistema constitucional como un sistema plenamente democrático. En pocas palabras, lo que los Federalistas entendían por "democracia" en 1787 no coincide con lo que el término pasaría a signiﬁcar medio siglo después. Asimismo, el cambio permite explicar por qué toda una serie de límites al alcance de la democracia que serán analizados en la Parte II y Parte III, y que habían sido tradicionalmente vistos como

280

Resumen en castellano

incompatibles con la democracia, pasaron a ser vistos (y son vistos hoy en día) no sólo como compatibles con la democracia sino, de hecho, como necesarios para el correcto funcionamiento de todo sistema democrático. La Parte II y la Parte III analizan de modo más normativo dos ejemplos especialmente relevantes de dichos límites, y muestran que hay buenas razones para poner en duda su justiﬁcación desde un punto de vista democrático. A pesar de que las democracias liberales se caracterizan por el gran número de ámbitos que quedan aislados al alcance de la toma de decisiones políticas (por ejemplo, la religión, la sexualidad, la política monetaria, los derechos de propiedad), la Parte II y la Parte III analizan dos ejemplos especialmente relevantes de estos límites: las restricciones constitucionales impuestas a la acción legislativa y la toma de decisiones en las empresas. Estos dos casos son relevantes tanto en el contexto actual como históricamente. En el contexto actual, puesto que tanto la rigidez constitucional y revisión judicial, por un lado, como el carácter no democrático y apolítico de la toma de decisiones empresariales, por el otro, han pasado a ser vistos como características perfectamente naturales y legítimas de nuestras democracias liberales. Históricamente, puesto que ambos están estrechamente relacionados con el período de la historia de Estados Unidos estudiado en la Parte I. Por un lado, la revisión judicial de la acción legislativa se originó en los primeros años de los Estados Unidos —a partir de la decisión Marbury c. Madison tomada por la Corte Suprema en 1803— como un freno a la democracia que sólo más tarde resultó ser considerado democrático, como veremos en la sección 1.2.2.2. Por otro lado, el trabajo asalariado, que había sido considerado tradicionalmente como una forma de “esclavitud limitada”, tal como lo expresó Aristóteles, pasó a ser visto como un intercambio libre y voluntario entre iguales en la medida en que la economía de mercado —y el trabajo asalariado— se extendió a lo largo del siglo XIX en los Estados Unidos y en otros países, como veremos en la sección 1.3.3.69 Así, la Parte II analiza las principales teorías normativas de las restricciones constitucionales impuestas sobre la actividad parlamentaria —y, más concretamente, la rigidez constitucional y la revisión judicial de las leyes— desde un punto de vista democrático. Ésta es una tarea de gran urgencia, dada la reciente expansión global del constitucionalismo, que se ha visto extendido tanto como la democracia (y de hecho más que ésta, como muestra la ﬁgura 2) y dado que, como ha aﬁrmado Cass Sunstein, sobre las razones empleadas para la constitucionalización de los mecanismos institucionales y los derechos sustantivos, “la teoría constitucional se encuentra en un estado sorprendentemente primitivo”.70 Se analizan los tres tipos principales de teorías empleadas para justiﬁcar la rigidez constitucional y la revisión judicial desde un punto de vista democrático: teorías puramente instrumentalistas, teorías basadas en el precompromiso y teorías procedimentalistas. Según estas teorías, el constitucionalismo no tiene por qué estar reñido con la democracia, y de hecho es necesario para el correcto funcionamiento del sistema democrático, bien porque las restricciones constitucionales resultan necesarias para garantizar que los procedimienAristóteles, Politics, 1260a-b. C. Sunstein, Designing Democracy. What Constitutions Do (Oxford: Oxford University Press, 2001), p. 97.
70 69

281 tos democráticos producen resultados adecuados (capítulo 4), o porque los límites constitucionales son el resultado de compromisos adoptados democráticamente por la ciudadanía (capítulo 5), o porque dichos límites establecen las reglas y precondiciones necesarias para el correcto funcionamiento del sistema democrático (capítulo 6). Como aﬁrmó Cherie Booth, esposa del exprimer ministro británico Tony Blair, La responsabilidad de un compromiso sustantivo con la democracia descansa en gran medida en los jueces (...) Los jueces en las democracias constitucionales son deﬁnidos como los guardianes de los derechos individuales (...) y se les otorga la oportunidad y el deber de hacer justicia a todos los ciudadanos mediante su compromiso con normas universales de decencia y humanidad (...) de un modo que nos enseña a los ciudadanos y el gobierno sobre las responsabilidades éticas de los participantes que en una verdadera democracia.71

Figura 2. Número aproximado de países con constitución escrita (A) comparado

con el número aproximado de democracias constitucionales (B).72

A pesar de que la Parte II analiza toda una serie de problemas de cada una de estas teorías, un problema común a todas ellas es que no abordan adecuadamente el hecho del desacuerdo que, como se muestra en la Parte I en el caso estadounidense, es condición necesaria de la legitimidad democrática. Todos ellas asumen, o terminan por asumir, que hay una serie de cuestiones políticas que están fuera de discusión y que pueden ser aseguradas mediante su incorporación a un documento constitucional cuya supremacía es garantizada mediante la revisión judicial de las leyes. Sin
C. Booth, “The Role of the Judge in a Human Rights World,” Speech to the Malaysian Bar Association, July 26, 2005. Citado en R. Bellamy, Political Constitutionalim. A Republican Defence of the Constitutionality of Democracy (Cambridge: Cambridge University Press, 2007), p. 1. 72 D. S. Lutz, Principles of Constitutional Design (Cambridge: Cambridge University Press, 2006), p. 4.
71

282

Resumen en castellano

embargo, como se muestra detalladamente en la parte II, a pesar de que las constituciones suelen ser presentadas como contratos completos cuyo contenido queda fuera de toda disputa, éstas están inevitablemente atravesadas por el hecho del desacuerdo a todos los niveles (por ejemplo, en la etapa de su redacción, así como durante su ratiﬁcación, interpretación y aplicación). Y al dejar de lado el hecho del desacuerdo, estas teorías A pesar de que este problema no evita que el constitucionalismo pueda ser justiﬁcado por otros motivos, sí que muestra la diﬁcultad de ser justiﬁcado por su carácter democrático puesto que, como se analiza en detalle en las secciones 4.3.1 y 4.3.2, el hecho del desacuerdo es una condición necesaria para que surja la necesidad de establecer una autoridad política, ya sea democrática o no. La Parte III de la tesis aborda otro caso notable en el que el alcance de la democracia se encuentra fuertemente restringido: la toma de decisiones en el lugar de trabajo. La posibilidad de extender la democracia a las empresas es especialmente relevante para esta tesis porque en las democracias liberales las empresas son frecuentemente vistas como un ámbito de relaciones contractuales voluntarias entre individuos jurídicamente iguales y que, en consecuencia, queda fuera del ámbito de la política, incluida la política democrática. Y también es relavante porque esta concepción del empleo —como un ámbito en el que las relaciones de poder están ausentes y empleadores y empleados entran en relaciones contractuales en las que la coerción se halla ausente— no surgió hasta que la economía de mercado se extendió en los Estados Unidos y en otros países a lo largo del siglo XIX, como veremos en la sección 1.3.3, concepción que ha pasado a ser dominante desde entonces. Ésta es la famosa descripción que hicieron Alchian y Demsetz de la relación laboral: Es común ver la empresa caracterizada por el poder de resolver los problemas mediante órdenes, autoridad, o medidas disciplinarias superiores a las del mercado convencional. Pero se trata de una ilusión (...) La empresa (...) no tiene ningún poder de ordenar, ninguna autoridad, ninguna capacidad disciplinaria diferente en lo más mínimo a los contratos del mercado convencional entre dos personas (...) Decirle a un empleado que escriba esta carta en lugar de presentar ese documento es como decirle a mi tendero que me venden esta marca de atún en lugar de esa marca de pan.73 El capítulo 7 de esta parte de la tesis analiza los principales argumentos a favor y en contra de extender la democracia a las empresas. Sin embargo, la principal contribución de la Parte III es presentar un argumento original, de base republicana, a favor de la democracia en las empresas en el capítulo 8. Dado que el republicanismo fue la teoría política dominante durante el período revolucionario y posrevolucionario de los Estados Unidos (a pesar de tratarse de una tradición que se remonta a Aristóteles, Cicerón, Maquiavelo y Harrington, entre muchos otros), este capítulo se apoya en gran medida en la Parte I de la tesis al elaborar y emplear la noción republicana de libertad como inmunidad frente a la interferencia arbitraria. La principal cuestión de este capítulo es si la democracia en el lugar de trabajo resulta ser una condición
Alchian and Demsetz, “Production, Information Costs, and Economic Organization,” American Economic Review 62 (1972): 777-795, p. 777.
73

283 necesaria para garantizar la inmunidad frente a la interferencia arbitraria en el lugar de trabajo. Se emplea la teoría de los contratos incompletos para mostrar que los contratos de trabajo, así como los intentos públicos por regular las relaciones laborales por medio de la legislación laboral, resultan incompletos y por lo tanto permiten la interferencia arbitraria. Dicho en pocas palabras, puesto que sería imposible o prohibitivamente costoso para las partes de la relación laboral elaborar contratos completos anticipando todas las posibles contigencias futuras —así como para los organismos públicos regular exhaustivamente dicha relación contractual— y puesto que cierta ﬂexibilidad resulta conveniente para atender adecuadamente dichas contigencias, la discrecionalidad en el lugar de trabajo no sólo resulta inevitable sino también deseable. Ahora bien, la discrecionalidad en la aplicación de los contratos laborales y de la regulación laboral abre la puerta al desacuerdo entre las partes y a la interferencia arbitraria en el ejercicio de tal discrecionalidad. Ello justiﬁca que, a ﬁn de garantizar la inmunidad frente a la interferencia arbitraria, los trabajadores dispongan de derechos de participación sobre la aplicación de los términos de la relación contractual y su regulación legislativa en el día a día de la relación laboral. Para ﬁnalizar, es preciso indicar que, a pesar de que el principal objetivo de la tesis es analizar el alcance y los límites de la democracia histórica y normativamente, las tres partes en que se divide son en gran medida autónomas e independiente las unas de las otras. Sin embargo, el hecho del desacuerdo y su relación con la legitimidad democrática y la toma de decisiones democrática es común a todas ellas y recorre el conjunto de la tesis. En el siguiente sentido. La Parte I de la tesis muestra cómo el hecho de desacuerdo llegó a ser aceptado en Norteamérica del siglo XIX, y cómo dicha aceptación dio paso a la crisis de la teoría de la representación virtual y a una creciente aceptación de la democracia como un medio para la toma de decisiones en condiciones de desacuerdos profundos sobre qué cursos de acción tomar a nivel político. Dicho de otra manera, muestra que la demanda de la democracia sólo surgió una vez que el hecho del desacuerdo fue reconocido como una característica inevitable y legítima de la sociedad estadounidense. Como se analiza de modo más analítico en la Parte II (secciones 4.3.1 y 4.3.2), el hecho del desacuerdo resulta ser una condición necesaria de la toma de decisiones democráticas, ya que si todos los ciudadanos estuviesen de acuerdo en qué curso de acción tomar, no sería necesario emplear procedimiento alguno de toma de decisiones. Así, la segunda parte se centra en cómo las constituciones, que a menudo se presentan como contratos completos cuyo contenido se halla fuera de discusión, se hallan inevitablemente atravesados por el hecho de desacuerdo a diversos niveles. En consecuencia, se argumenta que no hay razones sólidas para sustraer las normas constitucionales al alcance del proceso democrático ordinario por medio de la rigidez constitucionales y la revisión judicial. Finalmente, la Parte III muestra que la relaciones laborales en el lugar de trabajo, una esfera generalmente considerada apolítica y sustraída al alcance de la toma de decisiones democrática, se halla igualmente atravesado por el hecho de desacuerdo. Y, sobre este supuesto, proporciona un argumento a favor de la democratización

284

Resumen en castellano

de la toma de decisiones en las empresas sobre la base de la noción republicana de libertad como inmunidad frente a la interferencia arbitraria. En el resto de este resumen se presentan los principales argumentos de las tres partes de la tesis de manera sintética. Parte I: Democracia La Parte 1 analiza históricamente el concepto de democracia, y su alcance, en los Estados Unidos de América en el periodo revolucionario y posrevolucionario. Se halla un doble desplazamiento. Primero, un incremento en la aceptación de la idea de democracia debido a la aceptación del desacuerdo político como un fenómeno legítimo y la consiguiente crisis de la teoría de la representación virtual, según la cual la sociedad es una entidad orgánica con un conjunto de intereses compartidos que deben ser aislados de la política democrática y el partidismo. Segundo, un desplazamiento del concepto de democracia, según el cual muchos de los mecanismos institucionales que habían sido tradicionalmente vistos como límites al alcance de la democracia pasaron a ser considerados no sólo consistentes con la democracia sino de hecho necesarios para su correcto funcionamiento. Se divide en dos capítulos en los cuales se analizan estos dos desplazamientos, respectivamente. El capítulo 1 analiza el incremento en la aceptación de la idea de democracia, junto a los mecanismos institucionales tradicionalmente asociados a ésta. Por su parte, el capítulo 2 muestra que, además de la creciente aceptación de la idea de democracia, el concepto mismo de democracia y su alcance también se vieron modiﬁcados, y reconstruye el concepto original y posterior de democracia. Una característica notable del sistema constitucional norteamericano es que sus fundadores se reﬁrieron abiertamente a él como un sistema no-democrático. Los Federalistas no sólo culparon a “la democracia: una bestia sin cabeza (...) el gobierno del populacho”. También concibieron la Constitución federal como un medio “para contener” y “para mantener bajo control la turbulencia de la democracia”.74 No menos sorprendente es que, sólo unas pocas décadas después de la ratiﬁcación de la Constitución, la práctica totalidad de los americanos —los herederos de los Federalistas incluidos— pasó a referirse al mismo sistema constitucional como un sistema plenamente democrático. Así, Jeﬀerson podría concluir en 1826 que “Nosotros, los norteamericanos somos (...) constitucionalmente y conscientemente demócratas”.75 Pues bien, ¿qué sucedió para que un sistema constitucional que había sido elaborado como no democrático pasase a ser considerado como el sistema democrático por excelencia en apenas medio siglo? La Parte I de la tesis muestra que, una vez que analizamos el signiﬁcado peyorativo del término "democracia" en la década de 1780, no debería sorprender que los Federalistas fueran reticentes a caliﬁcar de democrático el sistema político que diseñaron. Asimismo, no debería sorprendernos el hecho de que el término se emplease
W. Hooper, quoted in F. Dupuis-Deri, The Political Power of Words: The Birth of Pro-Democratic Discourse in the 19th century in the United States and France (PhD diss., University of British Columbia, 2001); R. Morris, in M. Farrand, ed., The Records of the Federal Convention of 1787, (New Haven: Yale University Press, 1911), vol. 1, pp. 58 and 517. 75 Quoted in Dupuis-Deri, The Political Power of Words, p. 227.
74

285 de manera generalizada para caliﬁcar el mismo sistema cincuenta años más tarde, cuando éste pasó a tener un sentido abiertamente positivo y fue utilizado de manera propagandística por Andrew Jackson primero y el Partido Whig después. Así, la Parte I se centra en las causas principales detrás de este cambio, que se analizan en detalle en el capítulo 1 y en el capítulo 2, respectivamente. El capítulo 1 muestra que, a pesar de su sentido peyorativo en el período colonial e inmediatamente posterior a la Guerra de Independencia, la idea de democracia, junto con algunos de los mecanismos institucionales tradicionalmente asociados a ella, pasó a recibir un apoyo cada vez mayor en las décadas posteriores. Esto sucedió en la medida en que los ciudadanos tomaron conciencia de la pluralidad de intereses y valores en la sociedad americana, se organizaron para defender dichos intereses y valores y la teoría de la representación virtual utilizada por los Federalistas en la década de 1780 perdió su anterior eﬁcacia política. De acuerdo con la teoría de la representación virtual, la sociedad era un cuerpo orgánico con un conjunto de intereses y valores compartidos anteriores a los intereses particulares de las partes de la sociedad. Por ello, se consideraba que aquellos ciudadanos que estaban excluidos de la toma de decisiones políticas (por ejemplo, del sufragio activo y pasivo), estaban no obstante representados, si no realmente, virtualmente mediante su participación en los intereses y valores compartidos de la comunidad. Ahora bien, a lo largo del siglo XIX, la ciudadanía tomo conciencia de la pluralidad de intereses y valores en la sociedad americana, se organizó para dichos intereses y valores, y la teoría de la representación virtual perdió gran parte de su credibilidad anterior. En consecuencia, la representación real a través de la extensión del sufragio y otros mecanismos tradicionalmente asociados a la idea de democracia fueron crecientemente demandados y el término "democracia" perdió parte de su anterior sentido peyorativo. El capítulo se divide en dos secciones. La sección 1.2 analiza en detalle la teoría de la representación virtual, que fue hegemónica durante el período colonial, revolucionario e inmediatamente posrevolucionario. La teoría puede ser divida en tres principios: un principio organicista, un principio elitista y un principio de independencia material, que son analizados en detalle. En primer lugar, se analiza el principio organicista y su uso generalizado durante la guerra de independencia, el llamado "período crítico" y la Convención Federal de 1787. En segundo lugar, se analiza el principio de la élite gobernante, según la cual las decisiones políticas deben ser tomadas por aquellas personas con el conocimiento experto y la imparcialidad necesarias para dejar de lado sus intereses particulares y representar el interés general de la comunidad. Se analizan las razones alegadas para justiﬁcar dicho principio, junto con los principales medios institucionales propuestos a ﬁn de ponerlo en práctica. En tercer lugar, se analiza el principio de independencia material, según el cual sólo aquellos que eran materialmente independientes (que en el período estudiado se identiﬁcaba con la posesión de bienes inmuebles) gozan de la imparcialidad necesaria para representar los intereses de la comunidad en detrimento de sus intereses particulares. Se identiﬁcan tres razones principales para justiﬁcar este requisito: (a) la ausencia de inﬂuencia de aquellos en quienes de lo contrario era inevitable depender

286

Resumen en castellano

para vivir, (b) la disponibilidad del tiempo necesario para participar en política de manera informada, y (c) la posesión de una “participación en la sociedad” a ﬁn de tener los incentivos adecuados para participar en política de manera responsable. En la misma época, Kant lo expresaría aﬁrmando que "[l]a única justiﬁcación requerida por un ciudadano (...) es que debe ser su propio amo (sui iuris), y debe tener alguna propiedad (...) para mantenerse a sí mismo”.76 Por su parte, la sección 1.3 analiza la crisis de estos tres principios de la teoría de la representación virtual y cómo ello afectó al sentido peyorativo de la idea de democracia. En primer lugar, se hallan tres causas principales detrás de la crisis del principio organicista: (a) el desarrollo de la economía de mercado, (b) la inﬂuencia de la Revolución Francesa y la posterior creación de sociedades democráticas a lo largo y ancho del país y (c) la aparición del primer y el segundo sistema de partidos. En segundo lugar, se analiza la crisis del principio de la elite gobernante, que fue sustituido en gran medida por la idea de que la gente común representa mejor los intereses de la ciudadanía que una élite de expertos, junto a los mecanismos institucionales necesarios para garantizar que los representantes respondan a los intereses de sus electores. Por último, se hallan dos causas tras la crisis del principio de independencia material. En primer lugar, los trabajadores asalariados —cuyos salarios habían sido tradicionalmente considerados como “una promesa de su esclavitud”, en la expresión de Cicerón—, pasaron a ser considerados como materialmente independientes —y por tanto dignos de ser concedidos la ciudadanía plena— ya que pasaron a ser considerados como propietarios: propietarios de su propia fuerza de trabajo, que podían intercambiar libremente en el mercado como si fuese una mercancía más.77 En segundo lugar, se extendió la idea —ampliamente difundida gracias a la Declaración de Independencia de Jeﬀerson— de que todos los miembros de la comunidad política eran dignos de ser concedidos la ciudadanía plena, con independencia de que gozaran de independencia material o no. En resumen, la combinación de todos estos factores hicieron que la teoría de la representación virtual perdiese su credibilidad anterior y la democracia, en consecuencia, fuese crecientemente demandada y ganase respetabilidad. El capítulo 2, por su parte, muestra que, además de su creciente aceptación, la idea de democracia también sufrió un notable cambio semántico durante las décadas posteriores a la ratiﬁcación de la Constitución federal. Aunque algunas de las características del concepto de democracia —y algunos de los mecanismos institucionales asociados a él— se mantuvieron estables, otros cambiaron. Esto permite explicar por qué los Federalistas creían que el gobierno representativo que diseñaron no sólo era diferente de la democracia sino, de hecho, un límite a ésta. En palabras de Bernard Manin, “para Madison, el gobierno representativo no era un tipo de democracia,
I. Kant, “Über den Gemeinspruch: das mag in der Theorie richtig sein, taugt aber nicht für die Praxis,” in Kant’s gesammelte Schriften, vol. VIII, Königlichen Preußischen Akademie der Wissenschaften, ed. (Walter De Gruyter: Berlin, 1923). Quoted from I. Kant, “On the Common Saying: ’This May Be True in Theory, but It Does Not Apply in Practice’,” Political Writings, H. S. Reiss, ed. (Cambridge: Cambridge University Press, 1970), p. 78. 77 Cicero, De Oﬁciis, W. Miller, trans. (Cambridge, MA: Harvard University Press, 1913), I. XLII.
76

287 sino que era una forma esencialmente diferente y, además, preferible de gobierno”.78 Asimismo, permite explicar por qué el mismo sistema pudo ser caliﬁcado de manera generalizada como perfectamente democrático apenas unas décadas más tarde. En pocas palabras, lo que los Federalistas entendían por “democracia” en la década de 1780 no coincide con lo que el término pasó a signiﬁcar medio de medio siglo después. Del mismo modo, los mecanismos institucionales asociados al concepto clásico no coinciden con los asociados a este último concepto. Lo que es más, un gran número de mecanismos institucionales que habían sido tradicionalmente considerados como limitaciones a la democracia pasaron a ser vistos no sólo como compatibles con la idea de democracia, sino de hecho necesarios para el correcto funcionamiento de toda sistema democrático. El capítulo se divide en dos secciones. La sección 2.2 traza la evolución del concepto desde la década de 1780 y a lo largo de los cincuenta años posteriores. Lo hace mediante el análisis de tres pasos clave. En primer lugar, muestra la inﬂuencia del republicanismo clásico de Aristóteles, Polibio y Cicerón —tradición en la que los Founders habían crecido intelectualmente— en su concepción de la democracia. En segundo lugar, la presidencia de Jeﬀerson y el debate explícito entre los jeﬀersonianos de Pensilvania sobre el signiﬁcado y las implicaciones institucionales de la democracia. Y, por último, la democracia jacksoniana y el segundo sistema de partidos, en el que la democracia fue utilizada de manera abiertamente elogiosa e incluso propagandística por primera vez. A continuación, la sección 2.3 compara los conceptos original y posterior de democracia en relación a cinco cuestiones especialmente relevantes: la representación y la soberanía popular, la viabilidad política, los derechos individuales y de propiedad, la estabilidad política y la división de poderes. Como veremos, el cambio en la idea de democracia en relación a estas cuestiones permite explicar por qué, apenas unas décadas después de la ratiﬁcación de la Constitución federal, el mismo sistema constitucional pasó a ser considerado de manera generalizada como perfectamente democrático. Como William Duane escribiría a Jeﬀerson en 1824: “El gobierno democrático ya no es el jacobinismo, y los que anteriormente lo habían reprobado ahora utilizan dicho lenguaje y profesan la doctrina que maldecían veinticuatro años atrás”. Se había producido, tal como lo caliﬁcó Duane, “una revolución lingüística".79 Parte II: Constitucionalismo La Parte I de la tesis muestra, entre otras cosas, cómo algunos de los mecanismos que fueron originalmente diseñados como límites a la democracia —y, en consecuencia, como incompatibles con ésta— pasaron a ser considerados no sólo como compatibles con ésta sino, de hecho, como necesarios para el correcto funcionamiento de todo sistema democrático. La Parte II se centra en dos de estos mecanismos: la rigidez constitucional y la revisión judicial de las leyes, así como en la noción misma
78 B. Manin, The Principles of Representative Government (Cambridge: Cambridge University Press, 1996), p. 2. 79 William Duane, “Letter to Jeﬀerson, October 19, 1924,” Proceedings of the Massachusetts Historical Society 20 (1906-1907), 381-383, p. 382.

288

Resumen en castellano

de constitucionalismo, cuya reciente expansión global ha sido notable. Así, mientras que en 1802 no existía ningún sistema propiamente constitucional en el mundo y en 1978 sólo el 26% de las constituciones incorporaban alguna forma de revisión judicial, de las 106 constituciones ratiﬁcadas entre 1985 y 2008 101 han incluido un tribunal constitucional con poderes ﬁnales de revisión sobre la acción legislativa.80 Esta parte se divide en cuatro capítulos, que analizan en detalle las principales teorías normativas de la democracia constitucional. Según estas teorías, el constitucionalismo, incluyendo la rigidez constitucional y la revisión judicial de las leyes, no necesita entrar en contradicción con la democracia, puesto que las restricciones constitucionales resultan necesarias para garantizar que los procedimientos democráticos produzcan resultados adecuados, como la protección de los derechos humanos o la estabilidad ﬁscal (capítulo 4), o porque dichas restricciones son el resultado de compromisos adoptados democráticamente por los ciudadanos (capítulo 5), o porque dichas restricciones establecen los procedimientos y precondiciones previas necesarias para el correcto funcionamiento del propio sistema democrático (capítulo 6). Antes de pasar a analizar estas teorías, el capítulo 3 proporciona el marco analítico e histórico empleado a lo largo de la parte II. El constitucionalismo se deﬁne como la limitación de los poderes del órgano legislativo mediante la incorporación de dichos límites a un documento rígido y su aplicación mediante la revisión judicial de las leyes. La sección 3.2 presenta brevemente la historia del constitucionalismo, comparándolo con el constitucionalismo absolutista, el modelo de Westminster de soberanía parlamentaria y los llamados modelos “new commonwealth” de constitucionalismo, y distinguiendo entre restricciones sustantivas y procedimientales. A su vez, la sección 3.3 analiza brevemente la diferencia entre constituciones escritas y convenciones constitucionales, junto con el papel de éstas en el constitucionalismo. El resto del capítulo analiza los dos principales mecanismos institucionales del constitucionalismo, a saber: la rigidez constitucional y la revisión judicial de las leyes. La sección 3.4 deﬁne la rigidez constitucional, presenta los mecanismos especíﬁcos empleados para aplicar dicha rigidez y analiza con cierto detalle el clásico problema ﬁlosóﬁco de si realmente un soberano puede limitarse a sí mismo puesto que, como lo expresó Hobbes, “quien puede atarse, puede liberarse”.81 Se argumenta que en condiciones no ideales, las limitaciones constitucionales imponen costes sobre el legislador, por lo que resulta tan ingenuo creer que no impongan costes como considerar que la inclusión de ciertas restricciones en un documento constitucional garantiza su cumplimiento eo ipso. Finalmente, la sección 3.5 analiza la revisión judicial de las leyes y distingue entre revisión judicial fuerte y débil, entre modelos americano y europeo de revisión, y analiza su efectividad a la luz de la evidencia empírica que muestra que, en la práctica, los tribunales constitucionales tienden a no tomar decisiones contrarias a la opinión pública.
The ﬁgures come from D. L. Horowitz, “Constitutional Courts: A Primer for Decision Makers,” Journal of Democracy 17 (2006): 125-137; and A. Stone Sweet, “Constitutionalism, Rights, and Judicial Power,” in Comparative Politics, D. Caramani, ed. (Oxford: Oxford University Press, 2008). 81 T. Hobbes, Leviathan, R. Tuck, ed. (Cambridge: Cambridge University Press, 1996), p. 184.
80

289 El capítulo 4 analiza las teorías instrumentalistas de la legitimidad democrática y su justiﬁcación de las restricciones constitucionales. Dichas teorías actualizan la visión organicista analizada en la Parte I, de modo que deﬁnen la autoridad legítima en base a su capacidad para alcanzar un conjunto de objetivos sustantivos y previos al proceso político, tales como la protección de los derechos fundamentales, el crecimiento económico o la estabilidad política. Así, el constitucionalismo y la legitimidad democrática no tiene por qué entrar en conﬂicto, puesto que la legitimidad de la instituciones democráticas se basa exclusivamente en su capacidad para alcanzar una serie de objetivos que son consagrados en un documento constitucional y realizados por medios constitucionales. El capítulo analiza la estrctura del instrumentalismo en la sección 4.2, así como las dos teorías instrumentalistas principales del constitucionalismo, según las cuales el constitucionalismo (a) permite asegurar la protección de los derechos fundamentales y (b) produce resultados económicos superiores. Finalmente, se presentan tres problemas a los que se enfrenta el instrumentalismo. La sección 4.3.1 muestra que el instrumentalismo es incapaz de acomodar la llamada paradoja de Wollheim, que presenta la situación, habitual en las sociedades democráticas, en las que un ciudadano X está convencido tanto de que la política A debe ser aprobada (porque A es su opción preferida) como de que la política B, que es incompatible con A, debe ser aprobada (puesto que X es un demócrata y B es la opción respaldada por la mayoría).82 Se analizan las principales soluciones anteriores a la paradoja y se elabora una solución en dos niveles. De acuerdo con esta solución, X puede estar comprometido con ambas creencias puesto que proporcionan una respuesta a dos preguntas normativas muy diferentes: respectivamente, a "”qué se debe hacer” y a “qué se debe cuando discrepamos sobre qué hacer y, sin embargo, hemos de tomar un curso de acción común”. A continuación, se muestra que, una vez que aceptamos dicha solución, el instrumentalismo resulta ser una teoría normativa incompleta de la autoridad política, puesto que se muestra incapaz de acomodar la paradoja de Wollheim al reducir la segunda cuestión a la primera. Primero, la sección 4.3.2, por su parte, profundiza en las críticas presentadas al ﬁnal de la sección anterior, a saber: que las teorías instrumentalistas de la autoridad democrática resultan incompletas, puesto que no abordan adecuadamente las divergencias que precisamente hacen que sea necesario el recurso a una autoridad. Se divide en cinco parágrafos. Los cuatro primeros parágrafos deﬁnen las circunstancias de la política —que surgen cuando (a) hay una necesidad de un acuerdo vinculante colectiva sobre un curso de acción uniﬁcado pero (b) las partes discrepan sobre qué curso de acción tomar—, junto con el alcance, las causas y las implicaciones de tales circunstancias. El último parágrafo explica por qué el instrumentalismo resulta incapaz de acomodar las circunstancias de la política y, de hecho, termina por refutarse a sí mismo. Como lo ha expresado Waldron, “cualquier teoría que hace que la autoridad dependa de la corrección de los resultados políticos se refuta a sí misma,
R. Wollheim, “A Paradox in the Theory of Democracy,” en Philosophy, Politics and Society, Peter Laslett y Runciman, eds. (Oxford: Basil Blackwell, 1962).
82

290

Resumen en castellano

puesto que es precisamente porque las personas no están de acuerdo acerca de qué resultados son correctos que necesitan establecer y reconocer una autoridad”.83 Finalmente, la sección 4.3.3 muestra que el instrumentalismo resulta ser una teoría normativa de la autoridad política excesivamente radical cuando lo consideramos desde el punto de vista de los ciudadanos reales, que tienden a tener en cuenta no sólo los resultados de las decisiones políticas, sino también la imparcialidad de los procedimientos empleados para alcanzar dichos resultados. La sección revisa la literatura en psicología social sobre el tema, que converge en la centralidad de la imparcialidad procedimental para la percepción de la legitimidad de la autoridad política. El capítulo 6 analiza las teorías constitucionales basadas en el precompromiso, que justiﬁcan las restricciones constitucionales por ser el resultado de un compromiso adoptado democráticamente por los propios ciudadanos. Como aﬁrmó Hayek, “Sólo un demagogo puede presentar como ‘antidemocráticas’ las limitaciones que imponen sobre las mayorías temporales las decisiones a largo plazo y los principios generales celebradas adoptados por el pueblo”.84 El capítulo se divide en tres secciones. En primer lugar, se describe la estructura de los precompromisos institucionales democráticos. En segundo lugar, se analiza la democracia dualista de Ackerman como el ejemplo más elaborado e inﬂuyentes de un enfoque basado en el precompromiso.85 Por último, se presentan cinco problemas de este enfoque. La sección 5.4.1 sostiene que las teorías constitucionales basadas en precompromisos incurren en una confusión categorial, puesto que los arreglos políticos, incluyendo los arreglos constitucionales, no se convierten en democráticos por el hecho de haber sido adoptados por medios democráticos, puesto que una mayoría de ciudadanos podría aprobar legislación abiertamente antidemocrática sin que dicha legislación pasase a ser democrática por esa razón. La sección 5.4.2 analiza la analogía entre los compromisos colectivos con normas constitucionales y los compromisos individuales con normas personales. Esta es una analogía con gran fuerza normativa, puesto que las decisiones que regulan el comportamiento propio tienden a ser vistas como más autónomas que las desreguladas. Como aﬁrmó célebremente Montesquieu, “la libertad política no consiste en hacer lo que uno quiere (...) la libertad no puede ser otra cosa que ser capaz de hacer lo que uno debe querer”.86 Sin embargo, la analogía no se sostiene porque, a diferencia de los compromisos individuales, los compromisos colectivos adoptados en momentos constituyentes están atravesados por el hecho de desacuerdo, por lo menos, por tres razones. En primer lugar, porque si realmente existiese un consenso en el período constituyente, entonces podrían emplearse mayorías cualiﬁcadas como requisito de los procedimientos de ratiﬁcación constitucional. Sin embargo, se suele emplear un
J. Waldron, Law and Disagreement (Oxford: Oxford University Press, 1999), p. 253. F. Hayek, The Constitution of Liberty (Chicago: University of Chicago Press, 1960), pp. 106-107. 85 B. Ackerman, We the People. Foundations (Cambridge, MA: Harvard University Press, 1991). 86 C.-L. Montesquieu, De l’esprit des lois, Laurent Versini, ed. (Paris: Gallimard, 1995), p. 325.
84 83

291 requisito de mayoría simple, lo que indica que los compromisos constitucionales no disfrutan necesariamente de consensos más amplios que las leyes adoptadas durante los momentos de política ordinaria. En segundo lugar, porque las circunstancias en que las constituciones son aprobadas o modiﬁcadas (por ejemplo, transiciones desde regímenes autocráticos, crisis económicas, guerras civiles) no suelen ser las mejores para que los ciudadanos dejen de lado sus desacuerdos. Y en tercer lugar, porque el riesgo es demasiado alto, dado al enorme impacto y la diﬁcultad de enmendar las disposiciones constitucionales, como para que las partes dejen de lado sus intereses particulares y actúen sosegadamente. En lugar de amplios consensos en los que la ciudadanía habla con una sola voz, los momentos constitucionales pueden ser mejor descritos como procesos de negociación y las constituciones, a su vez, como el equilibrio resultante de la fuerza negociadora de las partes. La sección 5.4.3 muestra que, si bien las constituciones son equilibrios entre partes en desacuerdo, éstas aún pueden ser descritas como precompromisos democráticos siempre y cuando los equilibrios resultantes sean adoptados por una mayoría de ciudadanos. Sin embargo, se muestra que esta condición es incumplida con frecuencia tanto de manera sincrónica como diacrónica. Se presentan tres tipos de diﬁcultades para satisfacer esta condición. En primer lugar, cuando una minoría de los ciudadanos consigue imponer sus intereses en contra de los intereses de la mayoría de sus conciudadanos. En segundo lugar, cuando una mayoría de los ciudadanos —incluso una cualiﬁcada— pretende modiﬁcar la Constitución y, sin embargo, es incapaz de hacerlo debido a que la enmienda no es llevada a cabo o es bloqueada por las élites políticas. Y en tercer lugar, cuando un agente externo es capaz de imponer sus intereses en la constitución frente a los de la ciudadanía en su conjunto, como ocurrió en Japón tras la Segunda Guerra Mundial. A su vez, la sección 5.4.4 analiza diﬁcultades adicionales para cumplir con los criterios antes mencionados, sólo que en esta ocasión temporalmente, entre generaciones. Según el desiderátum de Thomas Paine, “Cada época y generación deberían ser tan libres para actuar por sí mismas, y en todos los casos, como las edades y generaciones que las precedieron”.87 Por consiguiente, para que una constitución sea plenamente democrática, los costes de ratiﬁcación y enmienda entre generaciones deben ser simétricos. Es decir, las disposiciones vigentes deben ser modiﬁcables por las generaciones siguientes, siempre que la enmienda cuente con el apoyo de la mayoría. Sin embargo, hay una asimetría entre los procedimientos de adopción y enmienda constitucional, tanto formal como informalmente. Formalmente, pues, como regla general, las constituciones sólo requieren mayorías simples para su ratiﬁcación pero mayorías cualiﬁcadas para su enmienda. De manera informal, puesto que las constituciones suelen ser mucho más resistentes al cambio de lo que formalmente se especiﬁca debido, entre otras causas, a fenómenos como la llamada dependencia de la senda, la preferencia por el statu quo o la función expresiva de la ley, que son analizados.
T. Paine, “The Rights of Man,” Political Writings, B. Kuklick, ed. (Cambridge: Cambridge University Press, 2000), p. 63.
87

292

Resumen en castellano

Finalmente, el capítulo 6 analiza las teorías procedimentalistas de las restricciones constitucionales. Según dichas teorías, para ser realmente democráticas, las instituciones políticas han de satisfacer un conjunto de condiciones constitutivas (por ejemplo, “elecciones generales, libertad de prensa, libertad de reunión y libertad de expresión”, como aﬁrmó Rosa Luxemburgo en su crítica a la suspensión de las libertades civiles por los bolcheviques rusos) y de precondiciones (por ejemplo, “salarios dignos y alfabetización universal”, en palabras de John S. Mill).88 Y, a menos que sean incorporadas a una constitución rígida y protegidas por un tribunal constitucional, dichas condiciones y precondiciones pueden resultar manipuladas por eventuales mayorías parlamentarias que pueden modiﬁcar los distritos electorales en su favor, socavar los derechos políticos de las minorías o prohibir partidos y medios de comunicación opositores. En resumen, a menos que sean protegida constitucionalmente, las democracias pueden acabar por ser destruidas desde dentro por mayorías parlamentarias temporales. El capítulo se divide en cuatro secciones. La sección 6.2 analiza brevemente el ejemplo más citado de un sistema democrático supuestamente destruido por medios democráticos: la caída de la República de Weimar a manos del NSDAP. Como a Goebbles mismo le gustaba decir, “será siempre uno de los mejores chistes de la democracia que proporciona a sus enemigos mortales los medios para destruirla.” 89 La sección 6.3 analiza las teorías procedimentalistas basadas en precondiciones y, más especíﬁcamente, el argumento estándar de Holmes y Sunstein y el más peculiar de Habermas, según el cual “la Corte Constitucional regida por una interpretación procedimental de la constitución no tiene por qué hacer peligrar su crédito de legitimidad”. La sección 6.4 presenta la toería puramente procedimentalista de John H . Ely, así como la de Stephen Holmes, si bien más brevemente.90 El principal argumento de Ely es que los tribunales cuentan con una mejor posición institucional que los parlamentos a la hora de garantizar la imparcialidad del proceso político, puesto que los cargos electos incurren en un conﬂicto de intereses a la hora de garantizar la imparcialidad y competitividad del mercado político, dado que tienen un claro incentivo para utilizar su posición de poder para incrementar su poder político. Por el contrario, los jueces son menos propensos a conﬂictos de intereses de este tipo debido a que su permanencia en el cargo es independiente del ciclo político (más aún cuando gozan de cargos vitalicios). A continuación se presentan cuatro problemas de las teorías procedimentalistas. La sección 6.5.1 muestra que las teorías basadas en precondiciones incurren en la paradoja de las precondiciones señalada por Carlos Nino.91 A continuación, la sección
Luxemburgo es citada en C. E. Schorske, German Social Democracy, 1905-1917: The Development of the Great Schism (Cambridge, MA: Harvard University Press, 1955), p. 324. Mill es citado en Przeworski, Democracy and the Limits of Self-Government, p. xiii. 89 J. Goebbles, citado en A. S. Kirshner, “Proceduralism and Popular Threats to Democracy,” Journal of Political Philosophy 18 (2010): 405-424, p. 405. 90 J. H. Ely, Democracy and Distrust. A Theory of Judicial Review (Cambridge, MA.: Harvard University Press, 1980); S. Holmes, “Precommitment and the Paradox of Democracy,” en Constitutionalism and Democracy, J. Elster and R. Slagstad, eds. (Cambridge: Cambridge University Press, 1988). 91 C. S. Nino, La constitución de la democracia deliberativa (Barcelona: Gedisa, 1997).
88

293 6.5.2 muestra que el hecho del desacuerdo atraviesa los supuestos acuerdos relativos a las condiciones y precondiciones de la democracia tanto como las disposiciones constitucionales sustantivas. La sección 6.5.3 analiza la concepción de las constituciones como reglas constitutivas. Según Holmes, mientras que las normas sustantivas son reguladoras y pueden ser limitantes de las decisiones democráticas, las normas de procedimiento son meramente constitutivas, puesto que posibilitan y organizan la propia toma de decisiones democrática. No es posible tomar decisiones democráticas sin reglas sobre cómo tomar dichas decisiones, al igual que no es posible jugar al ajedrez sin las reglas del ajedrez. Se muestra, sin embargo, que el argumento de Holmes incurre en un non sequitur. En pocas palabras, a pesar de que las normas de procedimiento —por ejemplo, las reglas electorales— son ciertamente necesarias para tomar decisiones democráticas, no hay ninguna razón por la cual esas normas deben ser incorporadas a una constitución rígida y protegidas por tribunales con capacidad para derogar leyes aprobadas en sede parlamentaria. Por último, la sección 6.54 cierra el capítulo analizando el argumento acerca de la mejor posición institucional de los tribunales en comparación a los parlamentos. Se analiza la decisión de la Corte Suprema de los EE. UU. en el caso Bush c. Gore y se concluye que, como parte integrante del sistema político, los jueces constitucionales se ven tan afectados como los parlamentos por la manera en que se organiza el sistema político y se distribuye el poder, como muestra el hecho de que estén tan en desacuerdo entre sí como los representantes en los parlamentos y los ciudadanos en la calle, y de que hagan uso de la regla de la mayoría para tomar decisiones tanto como éstos. Parte III: Las empresas La Parte III de la tesis analiza la extensión de la toma de decisiones democrática al lugar de trabajo, una esfera que es frecuentemente considerada ajena al alcance de la democracia. Está dividida en dos capítulos. El capítulo 7 introduce y deﬁne las principales formas de democracia en la empresa (DE), y revisa los principales argumentos a favor y en contra de esta forma de organización empresarial. El capítulo 8, por su parte, presenta un argumento original, de base republicana, en favor de la DE. Dado que el republicanismo era la ﬁlosofía política dominante durante la Joven República Norteamericana (a pesar de que se remonta a Aristóteles, Cicerón, Maquiavelo y Harrington, entre muchos otros ﬁlósofos), este capítulo se apoya en gran medida en la Parte I de la tesis. El capítulo 7 proporciona el marco analítico y normativo empleado a lo largo de esta parte. La sección 7.2 deﬁne la DE y distingue entre los derechos de control por el capital y por el trabajo y entre la propiedad pública y privada de la empresa. Esta parte de la tesis se centra únicamente en las empresas privadas democráticas, que son controlados por los trabajadores (en lugar de por el capital) y son de propiedad privada (en lugar de pública). A continuación, se establece una nueva distinción entre aquellas empresas democráticas que son propiedad de inversores externos o de los propios trabajadores. El modelo alemán de codeterminación y el cooperativismo son presentados como las formas más representatives del primer y el segundo caso, respectivamente, y son cuidadosamente deﬁnidos en las secciones 7.2.1 y 7.2.2. La

294

Resumen en castellano

sección analiza con cierto detalle los principales argumentos a favor y en contra de la DE. La sección 7.3.1 analiza su viabilidad, y la sección 7.3.2, su eﬁciencia. Las secciones 7.3.3 y 7.3.4 analizan, respectivamente, los llamados argumentos “epistémico” y “propietarista” en contra de la DE y, por último, la sección 7.3.5 analiza críticamente la teoría de Bowles y Gintis del intercambio disputado. El capítulo 8 se centra exclusivamente en el caso republicano en favor de la DE. La sección 8.2 deﬁne la noción republicana de libertad, en contraste con su noción liberal: es decir, como inmunidad frente a la interferencia arbitraria. La siguiente sección muestra las formas de interferencia arbitraria habituales en el contexto laboral y las formas principales de enfrentarlas. Se analizan dos enfoques. Según el primero, la disponibilidad de una opción de salida garantiza —al menos en los mercados libres y competitivos— la libertad de los trabajadores frente a la interferencia arbitraria; según el segundo, la interferencia arbitraria puede ser atajada por la vía de la regulación laboral. Se presentan tres objeciones al primer enfoque. En primer lugar, los mercados laborales tanto imperfectos como perfectos tienen desempleo involuntario —un “ejército industrial de reserva”, por usar la célebre expresión de Marx—, lo cual hace que sea costoso para los trabajadores abandonar su empleo (ya que no estarían en condiciones de encontrar otro trabajo con facilidad). En segundo lugar, incluso si los mercados de trabajo se vaciasen, existen costes de salida adicionales, tales como la inversión de los trabajadores en capital humano especíﬁco de la empresa de, la búsqueda y costes de transición o el coste psicológico. En tercer lugar, aunque la salida sin coste fuera viable, las alternativas de trabajo podrían ser tan despóticas, desreguladas y arbitrarias como el trabajo actual, en cuyo caso la opción de salida sería de poco valor. Por su parte, la sección 8.3.2 analiza el constitucionalismo laboral, y muestra por qué la regulación laboral resulta insuﬁciente a la hora de garantizar la inmunidad frente a la interferencia arbitraria en las empresas. A su vez esta crítica es fundamental para entender el argumento republicano a favor de la DE. Veámoslo brevemente. Empleando la teoría de los contratos incompletos, se muestra que los contratos laborales, así como los intentos públicos por regular las relaciones laborales por vía legislatival, resultan incompletos y por lo tanto permiten la interferencia arbitraria en su aplicación a casos concretos y contigencias futuras que resultan, por deﬁnición, no anticipables.92 Dicho en pocas palabras, puesto que sería imposible o prohibitivamente costoso para las partes de la relación laboral elaborar contratos completos anticipando todas las posibles contigencias futuras —así como para los organismos públicos regular exhaustivamente dicha relación contractual— y puesto que cierta ﬂexibilidad resulta conveniente para atender adecuadamente dichas contigencias, la discrecionalidad en el lugar de trabajo no sólo resulta inevitable sino también deseable. Ahora bien, la discrecionalidad en la aplicación de los contratos laborales y de la regulación laboral a casos concretos y contigencias no anticipables abre la puerta al desacuerdo entre las partes y a la posible interferencia arbitraria
Véase, por ejemplo, O. Hart, “Incomplete Contracts and the Theory of the Firm,” Journal of Law, economics and Organizations, 4 (1988), 119-139.
92

295 en el ejercicio de tal discrecionalidad. Como aﬁrmó célebremente el Conde de Romanones, “que otros hagan la ley, que yo escribiré los reglamentos”. Ello justiﬁca que, a ﬁn de garantizar la inmunidad frente a la interferencia arbitraria, los trabajadores dispongan de derechos de participación sobre la aplicación de los términos de la relación contractual y su regulación legislativa en el día a día de la relación laboral. Por último, la sección 8.5 analiza la separación formal entre derechos de control y derechos de propiedad en las empresas democráticas desde un punto de vista republicano. Se analizan las asimetrías negociadoras y los problemas de riesgo moral derivados de dicha separación y se deﬁnde que el cooperativismo resulta, ceteris paribus, preferible a otras formas de DE en el que los derechos de propiedad y de control son formalmente separados, como en el caso de la codeterminación alemana.

Bibliography
— M. Abramowicz, “Constitutional Circularity,” UCLA Law Review 49 (2001-2002): 1-91. — D. Acemoglu and J. A. Robinson, Economic Origins of Dictatorship and Democracy (Cambridge: cambridge UniversityPress, 2005). — B. Ackerman, “Beyond Carolene Products,” Harvard Law Review 98 (1985): 713-746. — B. Ackerman, We the People. Foundations (Cambridge, Mass.: Harvard University Press, 1991). — B. Ackerman, “Rooted Cosmopolitanism,” Ethics 104 (1994): 516-535. — B. Ackerman, We the People: Transformations (Cambridge, Mass.: Harvard University Press, 1998). — B. Ackerman and A. Alstott, The Stakeholder Society (New Haven: Yale University Press, 1999). — A. Alchian and H. Demsetz, “Production, Information Costs, and Economic Organization,” American Economic Review 62 (1972): 777-795. — E. Anderson, “An Epistemic Defense of Democracy: David Estlund’s Democratic Authority,” Episteme. A Journal of Social Epistemology 5 (2008): 129-139. — H. Arendt, Men in Dark Times (San Diego: Harcourt Brace & Co., 1968). — H. Arendt, “What Is Freedom?,” in Between Past and Future (New York: Penguin Books, 1993). — Aristotle, Politics, C. D. C. Reeve, trans. (Indianapolis: Hackett, 1948). — R. Arneson, “Democratic rights at national and workplace levels,’’ The Idea of Democracy, D. Copp, J. Hampton and J. E. Roemer, eds. (Cambridge: Cambridge University Press, 1993). — R. Arneson, “Defending the Purely Instrumental Account of Democratic Legitimacy,” Journal of Political Philosophy 11 (2003): 122–132. — R. Arneson, “Democracy Is Not Intrinsically Just,” Democracy and Justice, K. Dowding, R. E. Goodin and C. Pateman, eds. (Cambridge: Cambridge University Press, 2004). — R. Arneson, “The Supposed Right to a Democratic Say,” Contemporary Debates in Political Philosophy, T. Christiano and J. Christman, eds. (Oxford: Wiley-Blackwell, 2009). — J. L. Austin, The Province of Jurisprudence Determined (Cambridge: Cambridge University Press, 1995).

298

Bibliography

— B. Bailyn, “Political Experience and Enlightenment Ideas in Eighteenth-Century America,” The American Historical Review 67 (1962): 339-351. — B. Bailyn, The Ideological Origins of the American revolution (Cambridge, Mass.: Harvard University Press, 1967). — T. Ball, “A Republic—If You Can Keep It,” in Conceptual Change and the Constitution, T. Ball and J. G. A. Pocock, eds. (Lawrence: University Press of Kansas, 1988). — B. Barber, Strong Democracy. Participatory Politics for a New Age (Berkeley and Los Angeles: University of California Press, 1984). — B. Barry, Political Argument (London: Routledge & Kegan Paul, 1965). — B. Barry, Why Social Justice Matters (Cambridge: Polity Press, 2005). — L. Bartels, Unequal Democracy. The Political Economy of the New Gilded Age (Princeton, NJ: Princeton University Press, 2008). — J. C. Bayón, “Democracia y derechos. Problemas de fundamentación del constitucionalismo,” Constitucionalismo y derechos fundamentales, J. Betegón, F. Laporta, J. R. de Páramo, L. Prieto, eds. (Madrid: Centro de Estudios Políticos y Constitucionales, 2004). — C. A. Beard, An Economic Interpretation of the Constitution of the United States (New York: MacMillan, 1913). — C. A. Beard, The Supreme Court and the Constitution (Englewood Cliﬀs: Prentice-Hall, 1962). — C. Beitz, Political Equality. An Essay in Democratic Theory (Princeton, NJ: Princeton University Press, 1989). — R. Bellamy, “Introdution,” in Constitutionalism and Democracy, R. Bellamy, ed. (Aldershot: Ashgate, 2006). — R. Bellamy, Political Constitutionalim. A Republican Defence of the Constitutionality of Democracy (Cambridge: Cambridge University Press, 2007). — G. Benelli, C. Loderer and T. Lys, “Labor Participation in Corporate Policy- Decisions: West Germany’s Experience with Codetermination,” Journal of Business 60 (1987): 553-575. — J. Bentham, “Anarchical Fallacies,” in Works of Jeremy Bentham, vol. 2 (Edinburgh: William Tait). — N. Berggren and P. Kurrild-Klitgaard, “Economic Eﬀects of Political Institutions, With Special Reference to Constitutions” in Why constitutions Matter, N. Berggren, N. Karlson, and J. Nergelius, eds. (Stockholm: City University Press, 2000). — A. Berle and G. C. Means, The Modern Corporation and Private Property (New York: MacMillan, 1932) — I. Berlin, Two Concepts of Liberty (Oxford: Oxford University Press, 1958). — A. Bickel, The Least Dangerous Branch. The Supreme Court at the Bar of Politics (Indianapolis: Bobs-Merril, 1962). — M. S. Bilder, “Idea or Practice: A Brief Historiography of Judicial Review,” Journal of Policy History 20 (2008): 6-26.

299 — See J. Birchall and L Hammond Ketilson, Resilience of the Cooperative Business Model in Times of Crisis (Geneva: International Labour Organization, 2009) — N. Bobbio, The Future of Democracy (Cambridge: Polity Press, 1987). — J. Bohman, “Survey Article: The Coming of Age of Deliberative Democracy,” Journal of Political Philosophy 6 (1998): 400-425. — L. Boltanski and È. Chiapello, The New Spirit of Capitalism (London: Verso, 2005). — S. Bowles, and H. Gintis, “Contested Exchange: Political Economy and Modern Economic Theory,” American Economic Review 78 (1988): 145-150. — S. Bowles, and H. Gintis, “Contested Exchange: New Microfoundations of the Political Economy of Capitalism,” Politics and Society 18 (1990): 165-222. — S. Bowles, and H. Gintis, “Power and Wealth in a competitive Capitalist Economy,” Philosophy and Public Aﬀairs 21 (1992): 324-53. — S. Bowles, and H. Gintis, “A Political and Economic Case for the Democratic Enterprise;” “The Revenge of Homo Economicus: Contested Exchange and the Revival of Political Economy,” Journal of Economic Perspectives 7 (1993): 83-102. — S. Bowles and H. Gintis, “Is Workplace Democracy Redundant in a Liberal Economy?,” Democracy and Eﬃciency in the Economic Enterprise, U. Pagano and R. Rowthorn, eds. (London: Routledge, 1996). — G. Brennan, “’The Myth of Ownership.’ A Review Essay,” Constitutional Political Economy 16 (2005): 207-219. — J. Brennan, “The Right to a Competent Electorate,” The Philosophical Quarterly 61 (2011): 700-724. — C. Brettschneider, Democratic Rights. The Substance of Self-government (Princeton and Oxford: Princeton University Press, 2007). — J. Buchanan and G. Tullock, The Calculus of Consent (Ann Arbor: University of Michigan Press, 1962. — J. Buchanan, “Why constitutions matter,” in Why constitutions matter, N. Berggren, N. Karlson, and J. Nergelius, eds. (Stockholm: City University Press, 2000). — A. Caplin and B. Nalebuﬀ, “On 64%-Majority Rule,” Econometrica 56 (1988): 787-814. — C. Carman, “The Process is the Reality: Perceptions of Procedural Fairness and Participatory Democracy,” Political Studies 58 (2010): 731-751. — S. Carrillo, Demain l’Espagne (París: Seuil, 1974). — M. Carnoy and D. Searer, Economic Democracy. The Challenge of the 1980s (New York: Sharpe/Pantheon Books, 1980) — W. Carsel, “The Slaveholders’ Indictment of Northern Wage Slavery,” Journal of Southern History 6 (1940): 504-520. — N. Carter, “Political Participation and the Workplace: The Spillover Thesis Revisited,” The British Journal of Politics & International Relations 8 (2006): 410-426. — T. Christiano, “An Instrumental Argument for a Human Right to Democracy,” Philosophy & Public Aﬀairs 39 (2011): 142-176.

300

Bibliography

— T. Christiano, “Waldron on Law and Disagreement,” Law & Philosophy 19 (2000): 513-543. — J. A. Christophersen, The Meaning of "Democracy:” As Used in European Ideologies from the French to the Russian Revolution (Oslo: Universitetsforlagets Trykningssentral, 1968). — Cicero, De Oﬁciis, W. Miller, trans. (Cambridge, Mass.: Harvard University Press, 1913). — C. Clark, Social Change in America. From the Revolution to the Civil War (Chicago: Ivan R. Dee, 2006). — J. Cohen, “The Economic Basis of Deliberative Democracy,” Social Philosophy and Policy 6 (1989): 25-50. — J. Cohen, “Pluralism and Proceduralism,” Chicago-Kent Law Review 69 (1993-1994): 589-618 — J. Cohen, “Deliberation and Democratic Legitimacy,” in Democracy and Diﬀerence. Contesting the Boundaries of the Political, S. Benhabib, ed. (Princeton: Princeton University Press, 1996). — D. Christie, “Recent Calls for Economic Democracy,” Ethics 95 (1984): 112-128. — R. A. Dahl, “Decision-Making in a Democracy: The Supreme Court as a National Policy-Maker,” Journal of Public Law 6 (1957), pp. 279-295. — R. Dahl, “Procedural Democracy,” in Philosophy, Politics and Society, Peter Laslett and James Fishkin, eds. (Oxford: Blackwell, 1979). — R. A. Dahl, A Preface to Economic Democracy (Berkeley and Los Angeles: University of California Press, 1985) — R. A. Dahl, On Democracy (New Haven: Yale University Press, 1998). — R. A. Dahl, How Democratic is the American Constitution? (New Haven and London: Yale University Press, 2001). — R. A. Dahl, “A Right to Workplace Democracy? Response to Robert Mayer,” The Review of Politics 63 (2001): 249-253 — R. A. Dahl, “James Madison: Republican or Democrat?,” Perspectives in Politics 3 (2005): 446-447. — R. A. Dahl, A Preface to Democractic Theory, expanded ed. (Chicago and London: The University of Chicago Press, 2006). — K. Dau-Schmidt, “An Economic Analysis of the Criminal Law as a Preference-Shaping Policy,’’ Duke Law Journal 1 (1990): 1–38 — C. Davenport and D. A. Armstrong II, “Democracy and the Violation of Human Rights: A Statistical Analysis from 1976 to 1996,” American Journal of Political Science 48 (2004): 538-554. — L. Diamond and L. Morlino, “Introduction,” in Assessing the Quality of Democracy, L. Diamond and L. Morlino, ed. (Baltimore: The John Hopkins University Press, 2005). — A. V. Dicey, Introduction to the Study of the Law of the Constitution, E. C. S. Wade, ed. (London: MacMillan Press, 1945).

301 — R. J. Dalton, D. C. Shin and W. Jou, “Understanding Democracy: Data from Unlikely Places,” Journal of Democracy 18 (2007): 142-156. — A. Domènech, El eclipse de la fraternidad (Barcelona, Crítica, 2004). — A. Domènech and D. Raventós, “Property and Republican Freedom: An Institutional Approach to Basic Income,” Basic Income Studies 2 (2007): 1-8. — G. K. Dow, Governing the Firm. Workers’ Control in Theory and Practice (Cambridge: Cambridge University Press, 2003). — W. Duane, “Letter to Jeﬀerson, October 19, 1924,” Proceedings of the Massachusetts Historical Society 20 (1906-1907), 381-383. — J. Dunn, “Democracy before the Age of the Democratic Revolution,” Paper delivered at Columbia University, 2003. — F. Dupuis-Deri, The Political Power of Words: The Birth of Pro-Democratic Discourse in the 19th century in the United States and France (PhD diss., University of British Columbia, 2001). — F. Dupuis-Déri, “The Political Power of Words: The Birth of Pro-democratic Discourse in the Nineteenth Century in the United States and France,” Political Studies 52 (2004): 118-134. — R. Dworkin, Law’s Empire (Cambridge, Mass.: Belknap Press, 1986). — R. Dworkin, “What Is Equality? Part IV: Political Equality,” University of San Francisco Law Review 22 (1989): 1-30. — R. Dworkin, A Bill of Rights for Britain (Ann Arbor, MI: University of Michigan Press, 1990). — R. Dworkin, “Constitutionalism and democracy,” European Journal of Philosophy 3 (1995): 3-11. — C. L. Eisgruber, Constitutional Self-Government (Cambridge, Mass.: Harvard University Press, 2001). — D. Ellerman, Property and Contract in Economics. The Case for Economic Democracy (Oxford: Blackwell, 1992). — D. Ellerman, “Translatio versus Concessio: Retrieving the Debate About Contracts of Alienation with an Application to Today’s Employement Contract,” Politics and Society 33 (2005): 449-480. — J. Elster, Sour Grapes: Studies in the Subversion of Rationality (Cambridge: Cambridge University Press, 1983) — J. Elster, Ulysses and the Sirens (Cambridge: Cambridge University Press, 1984). — J. Elster, “From Here to There; or, If Cooperative Ownership is so Desirable, why are there so Few Cooperatives?,” Social Philosophy and Policy 6 (1989): 93-111. — J. Elster, “The Impact of Constitutions on Economic Performance,” Proceedings of the World Bank Conference on Development Economics 1994 (1995): 209-226. — J. Elster, “Forces and Mechanisms in the Constitution-Making Process,” Duke Law Journal 45 (1995): 364-396. — J. Elster, “Deliberation and Constitution Making,” in Deliberative Democracy, J. Elster, ed. (Cambridge: Cambridge University Press, 1998). — J. Elster, Ulysses Unbound (Cambridge: Cambridge University Press, 2000).

302

Bibliography

— Elster and Ove Moene, Alternatives to Capitalism (Cambridge: Cambridge University Press, 1989). — J. H. Ely, Democracy and Distrust. A Theory of Judicial Review (Cambridge, MA.: Harvard University Press, 1980). — D. Enoch, “Taking Disagreement Seriously: On Jeremy Waldron’s Law and Disagreement,” Israel Law Review 39 (2006): 22-35. — C. R. Epp, The Rights Revolution. (Chicago: The University of Chicago Press, 1998). — L. Epstein, J. Night and A. D. Martin, “The Supreme Court as a Strategic National Policymaker,” Emory Law Journal 50 (2001): 583-612. — D. Estlund, Democratic Authority. A Philosophical framework (Princeton: Princeton University Press, 2008). — R. J. Evans, The Coming of the Third Reich (New York: Penguin, 2005). — C. Fabre, Social Rights Under the Constitution. Government and the Decent Life (Oxford: Oxford University Press, 2000). — M. Farrand, ed., The Records of the Federal Convention of 1787, (New Haven: Yale University Press, 1911). — J. Ferejohn, “Judicializing Politics, Politicizing Law,” Journal of Law and Contemporary Problems 65 (2003): 41-68. — I. Ferreras, On Economic Bicameralism (master diss., MIT, 2004). — V. Ferreres, “Una defensa de la rigidez constitucional,” ed. F. Laporta, Constitución: Problemas ﬁlosóﬁcos (Madrid: Centro de Estudios Políticos y Constitucionales, 2000). — W. Findley, Address to the Assembly, March 31, 1786, in Debates & Proceedings of the General Assembly of Pennsylvania, Mathew Carey, ed. (Philadelphia, 1786). — S. E. Finer. V. Bogdanor and B. Rudden, Comparing Constitutions (Oxford: Clarendon, 1995). — D. H. Fischer, The Revolution of American Conservatism. The Federalist Party in the Era of Jeﬀersonian Democracy (New York: Harper & Row, 1965). — J. Fiske, The Critical Period of American History (Boston and New York: The Riverside Press, 1888). — A. de Francisco, Ciudadanía y democracia. Un enfoque repulicano (Madrid: Los Libros de la Catarata, 2007). — H. G. Frankfurt, “Freedom of the Will and the Concept of a Person,” Journal of Philosophy 68 (1971): 5-20. — S. Freeman, “Constitutional Democracy and the Legitimacy of Judicial Review,” Law & Philosophy 9 (1990): 327-370. — B. Friedman, “The History of the Countermajoritarian Diﬃculty, Part One: The Road to Judicial Supremacy,” NYU Law Review 73 (1998): 333-433. — B. Friedman, "The History of the Countermajoritarian Diﬃculty, Part Four: Law’s Politics,” University of Pennsylvania Law Review 148 (2000): 971-1064.

303 — B. Friedman, "The History of the Countermajoritarian Diﬃculty, Part Three: The Lesson of Lochner," NYU Law Review 73 (2001): 1383-1455. — B. Friedman, "The History of the Countermajoritarian Diﬃculty, Part Two: Reconstruction’s Political Court," Georgetown Law Journal 91 (2002): 1-65. — B. Friedman, "The Birth of an Academic Obsession: The History of the Countermajoritarian Diﬃculty, Part Five," 112 Yale Law Journal 153 (2002): 153-259. — P. Frymer, “Labor and American Politics,” Perspectives on Politics 8 (2010): 609-616. — P. Funk, “Is There An Expressive Function of Law? An Empirical Analysis of Voting Laws with Symbolic Fines,” American Law and Economics Review 9 (2007): 135-159. — W. B. Gallie, “Essentially Contested Concepts,” Proceedings of the Aristotelian Society 56 (1956), 167-198. — S. Gardbaum, “The New Commonwealth Model of Constitutionalism”, American Journal of Comparative Law 49 (2001): 707-760. — S. Gardbaum, “Reassesing the New Commonwalth Model of Constitutionalism,” International Journal of Constitutional Law 8 (2010): 167-206. — R. Gargarella, La justicia frente al gobierno. Sobre el carácter contramayoritario del poder judicial (Barcelona: Ariel, 1996). — D. Gauthier, Morals by Agreement (Oxford: Clarendon Press, 1986). — A. Gewirth, “Are All Rights Positive?,” Philosophy & Public Aﬀairs 30 (2002): 321-333 — T. Ginsburg, “Public Choice and Constitutional Design,” in Research Handbook in Public Choice and Public Law, D. Farber and A. J. O’Connell, ed. (Cheltenham: Elgar, 2010). — Robert E. Goodin, “Independence in Democratic Theory: A Virtue? A Necessity? Both? Neither?,” Journal of Social Philosophy 24 (1993): 50-57. (accessed August 13, 2010). — A. Gourevich, “The Unﬁnished Revolution: Freedom and Property in the Nineteenth Century Republicanism,” unpublished draft, Brown University. — T. P. Govan, “The Rich, the Well-born, and Alexander Hamilton,” The Mississippi Valley Historical Review 36 (1950), pp. 675-680. — K. Graham, “Democracy, Paradox, and the Real World,” Proceedings of the Aristotelian Society 76 (1976): 227-245. — E. Greenberg, L. Grunberg and K. Daniel, “Industrial Work and Political Participation: Beyond ’simple spillover’,” Political Research Quarterly 49 (1996): 287-304. — C. G. Griﬃn, “Democracy as a Non-Instrumentally Just Procedure,” Journal of Political Philosophy 11 (2003): 111-121. — R. M. Gummere, “The Classical Politics of John Adams,” Boston Public Library Quarterly 9 (1957): 167–182.

— A. Gosseries, “Taking Workplace Democracy Seriously,” www.politika.lv/topics/civil_society/15968

304

Bibliography

— R. M. Gummere, “The Classical Ancestry of the United States Constitution,” American Quarterly 14 (1962): 3-18. — J. R. Gundersen, “Independence, Citizenship, and the American Revolution,” Signs 13 (1987): 59-77. — J. Habermas, Faktizität und Geltung. Beiträge zur Diskurstheorie des Rechtes und des demokratischen Rechtsstaats (Frankfurt am Main: Suhrkamp, 1992). — J. Habermas, “Constitutional Democracy: A Paradoxical Union of Contradictory Principles?,” Political Theory 29 (2001): 766-781. — V. Haksar, “The Alleged Paradox of Democracy,” Analysis 37 (1976): 10-14. — M. H. Hansen, The Athenian Democracy in the Age of Demosthenes, trans. J. A. Crook (Oxford and Cambridge, Mass.: Blackwell, 1991). — M. H. Hansen, “The Tradition of the Athenian Democracy A. D. 1750-1990,” Greece & Rome 39 (1992): 14-30. — H. Hansmann, The Ownership of Enterprise (Cambridge, Mass., Harvard University Press, 2000). — J. Harrington, The Commonwealth of Oceana and A System of Politics, J. G. A. Pocock, ed. (Cambridge, Cambridge University Press, 1992). — R. Harrison, “No Paradox in Democracy,” Political Studies 18 (1970): 514-17. — O. Hart, “Incomplete Contracts and the Theory of the Firm,” Journal of Law, economics and Organizations, 4 (1988): 119-139. — L. Hartz, The Liberal Tradition in America (New York: Hartcourt, Brac & Co., 1955). — D. M. Hausman, “Are Markets Morally Free Zones?,” Philosophy & Public Aﬀairs 18 (1989): 317-333. — F. Hayek, The Constitution of Liberty (Chicago: University of Chicago Press, 1960). — D. Herzog, “Democratic Credentials,” Ethics 104 (1994): 467-479. — J. Hiebert, “Parliamentary Bills of Rights: An Alternative Model?,” Modern Law Review 69 (2006): 7-28. — R. Hirschl, Towards Juristocracy. The Origins and Consequences of the New Constitutionalism (Cambridge, Mass.: Harvard University Press, 2004). — R. Hirschl, “The New Constitutionalism and the Judicialization of Pure Politics Worldwide,” Fordham Law Review 75 (2006): 721-754. — A. M. Honoré, “Ownership,” Oxford Essays in Jurisprudence, A. G. Guest, ed. (Oxford, Clarendon Press, 1961). — S. Hix and B. Hoyland, The Political System of the European Union (London: Palgrave Macmillan, 2011). — T. Hobbes, Leviathan, R. Tuck, ed. (Cambridge: Cambridge University Press, 1988). — G. L. Hodgson, “Organizacional Form and Economic Evolution: A Critique of the Williamsonian Hypothesis,” Democracy and Eﬃciency in the Economic Enterprise, U. Pagano and R. Rowthorn, eds. (London: Routledge, 1996). — R. Hodson, Dignity at Work (Cambridge: Cambridge University Press, 2004).

305 — W. N. Hohfeld, Fundamental Legal Conceptions (New Haven: Yale University Press, 1919). — R. Hofstadter, The American Political Tradition and the Men Who Made It (New York: A. A. Knopf, 1948). — R. Hofstadter, The Idea of a Party System: The Rise of Legitimate Opposition in the United States, 1780–1840 (Berkeley: University of California Press, 1969). — S. Holmes, “Precommitment and the Paradox of Democracy,” in Constitutionalism and Democracy, J. Elster and R. Slagstad, eds. (Cambridge: Cambridge University Press, 1988). — S. Holmes and C. Sunstein, “The Politics of Revision,” in Responding to Imperfection, ed. Sanford Levinson (Princeton, NJ: Princeton University Press, 1995). — S. Holmes and C. Sunstein, The Cost of Rights. Why Liberty Depends on Taxes (New York and London: Norton. 1999). — T. Honderich “A Diﬃculty With Democracy,” Philosophy & Public Aﬀairs 3 (1973-4): 221-26. — D. L. Horowitz, “Constitutional Courts: A Primer for Decision Makers,” Journal of Democracy 17 (2006): 125-137. — M. J. Horwitz, The transformation of American law, 1780-1860 (Cambridge, Mass.: Harvard University Press, 1977). — N. Hsieh, “Rawlsian Justice and Workplace Republicanism,” Social Theory and Practice 31 (2005): 115-142. — N. Hsieh, “Survey Article: Justice in production,” Journal of Political Philosophy 16 (2008): 72-100. — D. Hum and S. Choudry, “Income, Work and Marital Dissolution: Canadian Experimental Evidence,” Journal of Comparative Family Studies 23 (1992): 249-265. — D. Hume, A Treatise of Human Nature, L. A. Selby-Bigge, ed. (Oxford: Oxford University Press, 1978 [1739-1740]). — D. Hume, “Of Parties in General,” in Perspectives on Political Parties, S. E. Scarrow, ed. (New York: Palgrave MacMillan, 2002). — International Labour Organization, World of Work Report: Income Inequalities in the Age of Financial Globalization (Geneva: International Institute for Labour Studies, 2008) — T. Jeﬀerson, The Political Writings of Thomas Jeﬀerson, M. D. Peterson, ed. (Chapel Hill, NC: University of North Carolina Press, 1993). — M. Jensen, “Democracy and the American Revolution”, The Huntington Library Quarterly 20 (1957): 321-341. — M. C. Jensen, and W. H. Meckling, “Rights and Production Functions: An Application to Labor-managed Firm and Codetermination,” Journal of Business 52 (1979): 469-506. — I. Kant, “Über den Gemeinspruch: das mag in der Theorie richtig sein, taugt aber nicht für die Praxis,” in Kant’s gesammelte Schriften, vol. 8, Königlichen Deutschen Akademie der Wissenschaften, ed. (Walter De Gruyter: Berlin, 1923). — I. Kant, Grundlegung zur Metaphysik der Sitten, in Kants gesammelte Schriften,

306

Bibliography

vol. 8, Königlichen Deutschen Akademie der Wissenschaften, ed. (Berlin: Walter De Gruyter, 1923) — S. Kasmir, The Myth of Mondragon (Albany: SUNY Press, 1996). — A. Kavanagh, “Participation and Judicial Review: A Repy to Jeremy Waldron,” Law & Philosophy 22 (2003): 451-486. — H. Kelsen, Vom Wesen und Wert der Demokratie (Mohr: Tübingen 1920). — R. Ketchman, ed., The Selected Writings of James Madison (Indianapolis, Heckett, 1996). — A. Keyssar, The Right to Vote. The Contested History of Democracy in the United States (Philadelphia: Basic Books, 2009). — A. S. Kirshner, “Proceduralism and Popular Threats to Democracy,” Journal of Political Philosophy 18 (2010): 405-424. — J. Kis, “Constitutional Precommitment Revisited,” Journal of Social Philosophy 40 (2009): 570-594. — L. Kramer, The People Themselves. Popular Constitutionalism and Judicial Review (Oxford: Oxford University Press, 2004). — T. Kuran Private Truths, Public Lies: The Social Consequences of Preference Falsiﬁcation (Cambridge, Mass.: Harvard University Press, 1995). — P. Kurrild-Klitgaard and N. Berggren, "Economic Consequences of Constitutions: A Theory and Survey," Journal des Economistes et des Etudes Humaines 14 (2004): article 1. — A. Lahera, La participación de los trabajadores en la democracia industrial (Los Libros de la Catarata: Madrid, 2004). — B. Laniel, Le mot "democracy" aux Etats-Unis de 1780 a 1856 (Saint-Etienne: Publications d e l’Universite d e Saint-Etienne, 1995). — R. D. Lansbury, “Workplace Democracy and the Global Financial Crisis,” Journal of Industrial Relations 51 (2009): 599-616. — W. Laurens Walker and J. Thibaut, Procedural Justice: A Psychological Analysis (Hilsdale, NJ: L. Erlbaum Associates, 1975). — E. Lazarus, Closed Chambers: The First Eyewitness Account of the Epic Struggles Inside the Supreme Court (New York: Times Books, 1998). — A. Lerner, “The Economics and Politics of Consumer Sovereignty,” American Economic Review Proceedings 62 (1972): 258-66. — M. Levi, “A Model, a Method, and a Map: Rational Choice in Comparative and Historical Analysis,” ed. M. I. Lichbach and A. Zuckerman, Comparative Politics: Rationality, Culture, and Structure (Cambridge: Cambridge University Press, 1997) — H. M. Levin, “Worker Democracy and Worker Productivity,” Social Justice Research 19 (2006): 109-121. — A. Lijphart, Patterns of Democracy. Government Forms and Performance in Thirty-Six Countries (New Haven, CT: Yale University Press, 1999). — J. J. Linz, Problems of Democratic Transition and Consolidation: Southern Eu-

307 rope, South America, and post-Communist Europe (Baltimore: Johns Hopkins University Press, 1994). — J. J. Linz, Obras escogidas. Vol. 4. Democracias: quiebras, transiciones y retos, T. J. Milley and J. R. Montero, eds. (Madrid: Centro de Estudios Políticos y Constitucionales, 2009). — S. Linares, La (i)legitimidad democrática del control judicial de las leyes (Madrid: Marcial Pons, 2008). — R. N. Lokken, “The Concept of Democracy in Colonial Political Thought,” The William and Mary Quarterly 16 (1959): 568-580. — C. López-Guerra, “Against the Parallel Case for Workplace Democracy,” Revue de Philosophie économique 8 (2008): 11-28. — A. Lorenz, “How to Measure Constitutional Rigidity. Four Concepts and Two Alternatives,” Journal of Theoretical Politics 17 (2005): 339-361. — F. Lowett, A General Theory of Domination and Justice (Oxford: Oxford University Press, 2010). — D.S. Lutz, “Toward a Theory of Constitutional Ammendment,” American Political Science Review 88 (1994), pp. 355-370. — D. S. Lutz, “Introductory Essay,” in Colonial Origins of the American Constitution: A Documentary History (Indianapolis: Liberty Press, 1998). — D. S. Lutz, Principles of Constitutional Design (Cambridge: Cambridge University Press, 2006). — S. Lynn, “The Mechanics in new York Politics, 1774-1785,” in Class Conﬂict, Slavery, and the United States Constitution (New York, Cambridge University Press, 1967). — G. Mackie, Democracy Defended (Cambridge: Cambridge University Press, 2004). — C. B. MacPherson, The Political Theory of Possessive Individualism: Hobbes to Locke (Oxford: Oxford University Press, 1963). — J. Madison, A. Hamilton and J. Jay, The Federalist, G. W. Carey and J. McCleallan, eds. (Indianapolis: Liberty Fund, 2001). — J. K. Martin, “The Whiskey Rebellion Redivivus,” in., The Whiskey Rebellion. Past and Present Perspectives, S. Boyd, ed (Westport, Ct.: Greenwood Press, 1985). — B. Manin, The Principles of Representative Government (Cambridge: Cambridge University Press, 1996). — B. Manin, A. Przeworski, and S. Stokes, “Introduction,” in Democracy, Accountability, and Representation, A. Przeworski, S. Stokes, B. Manin, ed. (Cambridge: Cambridge University Press, 1999). — J. T. Main, “Government by the People: The American Revolution and the Democratization of the Legislatures”, The William and Mary Quarterly 23 (1966): 391-407. — J. M. Maravall, The Transition to Democracy in Spain (New York: St. Martin’s Press, 1982).

308

Bibliography

— M, G. Marshall and B. R. Cole, Global Report 2009: Conﬂict, Governance, and State Fragility (Vienna: Center for Systemic Peace, 2009). — J. L. Martí, La república deliberativa. Una teoría de la democracia (Madrid: Marcial Pons, 2006). — A. Marx and H. Hansen Peeters, “An Unconditional Basic Income and Labor Supply: Results From a Pilot Study of Lottery Winners,” The Journal of Socio-Economics, 37 (2008), 1636-1659 — K. Marx and F. Engels, “Manifest der Kommunistischen Partei,” in Werke, vol. 4 (Berlin: Dietz Verlag, 1971). — R. Mason, Participatory and Workplace Democracy (Carbondale, IL: Southern Illinois University Press, 1982). — F. Maultzsch, “Morals and Markets: The Signiﬁcance of Pre-Tax Ownership,” The Modern Law Review 67 (2004): 508-523 — K. O. May, “A Set of Independent Necessary and Suﬃcient Conditions for Simple Majority Decision,” Econometrica 20 (1952): 680-684. — B. H. McDonnell, “Employee Primacy, Or Economics Meets Civic Republicanism at Work,” Stanford Journal of Business and Finance 13 (2008): 334-393. — J. Meade, “The Theory of Labour-managed Firms and Proﬁt Sharing,” Economic Journal 82 (1972): 402-428 — M. Merrill, “Putting “Capitalism” in Its Place: A Review of Recent Literature,” The William and Mary Quarterly 52 (1995): 317-326. — M. Meyers, The Jacksonian Persuasion. Politics & Belief (Stanford: Stanford University Press, 1957). — E. A. Miles, “The Young American Nation and the Classical World,” Journal of the History of Ideas 35 (1974): 259-274. — J. S. Mill, “On Liberty,” in Collected Works of John Stuart Mill, vol. 18, J. Robson, ed. (Toronto: University of Toronto Press, 1963). — A. A. Montenegro, “Constitutional Design and Economic Performance,” Constitutional Political Economy 6 (1995): 161-169. — C.-L. Montesquieu, De l’esprit des lois, Laurent Versini, ed. (Paris: Gallimard, 1995). — R. A. M. Morantz, “Democracy” and “Republic” in American Ideology, 1787-1840 (PhD diss., Columbia University, 1971). — J. J., Moreso, Constitución: Modelo para armar (Madrid: Marcial Pons, 2009). — G. Mosca, The Ruling Elite, A. Livingston, ed. (New York: McGraw-Hill, 1939) — D. Mueller, Constitutional Democracy (Oxford: Oxford University Press, 2000). — T. Nagel, The Myth of Ownership: Taxes and Justice (Oxford, Oxford University Press, 2002). — G. B. Nash, B. G. Smith and D. Hoerder, “Labor in the Era of the American Revolution: An Exchange,” Labor History 24 (1983): 414-439. — F. Neumann, The Democratic and the Authoritarian State: Essays in Political and Legal Theory (Glencoe, IL: Free Press, 1957).

309 — C. S. Nino, La constitución de la democracia deliberativa (Barcelona: Gedisa, 1997). — D. C. North and B. R. Weingast, “Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in Seventeenth-Century England,” Journal of Economic History 49 (1989): 803-832. — R. Nozick, Anarchy, State and Utopia (New York: Basick Books, 1974). — G. O’Donnell, P. C. Schmitter, L. Whitehead, Transiciones desde un gobierno autoritario, vol. 4 (Buenos aires: Paidós, 1988). — G. O’Donnell, “Notas sobre la democracia en América Latina,” La democracia en América Latina. El debate conceptual sobre la democracia, Programa de Naciones Unidas para el Desarrollo, ed. (New York: PNUD, 2004). — J. P. Olsen, “Coping with Conﬂict at Constitutional Moments,” Industrial and Corporate Change 12 (2003): 815-842. — M. Olson “Autocracy, Democracy and Prosperity,” in Strategy and Choice, R. J. Zeckhauser, ed. (Cambridge: MIT Press, 1991). — E. Ostrom, Governing the Commons. The Evolution of Institutions for Collective Action (New York: Cambridge Univeristy Press, 1990). — T. Paine, “The Rights of Man,” Political Writings, B. Kuklick, ed. (Cambridge: Cambridge University Press, 2000). — R. R. Palmer, “Notes on the Use of the Word ’Democracy’ 1789-1799,” Political Science Quarterly 68 (1953): 203-226. — R. R. Palmer, The Age of Democratic Revolution, 2 vols. (Princeton: Princeton University Press, 1959-1964). — D. Parﬁt, Reasons and Persons (Oxford: Oxford University Press, 1984). — R. Parks, D. Kruse and J. Sesil, “Does Employee ownership Enhance Firm Survival?,” Employee Participation, ﬁrm Performance and Survival, V. Perotin and A. Robinson, eds. (Amsterdam: JAI Press, 2004). — C. Pateman, Participation and Democratic Theory (Cambridge: Cambridge University Press, 1970) — C. Pateman, “Self-Ownership and Property in the Person: Democratization and a Tale of Two Concepts,” The Journal of Political Philosophy 10 (2002): 20-53. — R. Pennock, “Democracy Is Not Paradoxical: Comment,” Political Theory 2 (1974): 88-93. — P. Pettit, Republicanism (Oxford: Oxford University Press, 1997). — P. Pettit, “Keeping Republican Freedom Simple. On A Diﬀerence with Quentin Skinner,” Political Theory 30 (2002): 339-356. — P. Pettit, “Republican Liberty: Three Axioms, Four Theorems,” in Republicanism and Political Theory, ed. Cecile Laborde and John Maynor (Oxford: Blackwell, 2008). — “Law and Liberty,” in Law and Republicanism, ed. S. Besson and J. L. Martí (Oxford: Oxford University Press, 2009). — P. Pierson, “Increasing Returns, Path Dependence, and the Study of Politics,” American Political Science Review 94 (2000): 251-267.

310

Bibliography

— A. Pinochet, “Chile should Not Fall into the Vices of the Past,” in The Politics of Antipolitics, B. Loveman and D. Thomas, eds. (Wilmington, DE: Scholarly Resources, 1997). — K. Polanyi, The Great Transformation. The Political and Economic Origins of our Time (Boston: Boston: Beacon Press, 1944). — R. Plant “Socialsim, Markets and End States,” Market Socialism, J. Le Grand and S. Strin, eds. (Oxford: Clarendon Press, 1989). — D. Plotke, “Representation Is Democracy,” Constellations 4 (1997): 19-34. — J. G. A. Pocock, The Machiavellian Moment. Florentine Political Thought and the Atlantic Republican Tradition (Princeton: Princeton Unievrsity Press, 1975). — K. Polany, The Great Transformation. The Political and Economic Origins of Our Time (Boston, Mass.: Beacon Press, 1957). — D. Potter, “Explaining Democratization,” in Democratization, ed. D. Potter, D. Goldblatt, M. Kiloh and P. Lewis (Cambridge: Polity Press, 1997). — A. Przeworski, “Minimalist Conception of Democracy: A Defense,” in Democracy’s Value. I. Shapiro, ed. (Cambridge: Cambridge University Press, 1999). — A. Przeworski, “Self-Enforcing Democracy,” in The Oxford Handbook of Political Economy, Donald Wittman and Barry Weingast, eds. (New York: Oxford University Press, 2006). — A. Przeworski, “Conquered or Granted? A History of Suﬀrage Extensions,” British Journal of Political Science 39 (2008): 291-321. — A. Przeworksi, “Democracy, Equality, and Redistribution,” in Political Judgement: Essays in Honour of John Dunn, R. Bourke and R. Geuss, eds. (Cambridge, Cambridge University Press, 2009). — A. Przeworski, “Consensus and Conﬂict in Western Thought on Representative Government,” Procedia Social and Behavioral Sciences 2 (2010): 7042–7055. — A. Przeworski, Democracy and the Limits of Self-Government (Nueva York:Cambridge University Press, 2010). — A. Przeworski and F. Limongi, “Political Regimes and Economic Growth,” Journal of Economic Perspectives 7 (1993): 51-69. — A. Przeworski, J. A. Cheibub, M. E. Alvarez and F. Limongi, Democracy and Development: Political Institutions and Material Well-being in the World, 1950-1990 (Cambridge: Cambridge University Press, 2000). — E. B. Rasmusen and R. McAdams, “Norms in Law and Economics,” in Handbook of Law and Economics, vol. 2, A. Mitchell Polinsky and S. Shavell, eds. (Amsterdam: Elsevier, 2007). — D. Raventós, Basic Income: The Material Conditions of Freedom (London, Pluto Press, 2007). — J. Rawls, A Theory of Justice (Cambridge, Ma.: Cambridge University Press, 1971). — J. Rawls, Political Liberalism (New York: Columbia University Press, 1993). — J. Rawls, Justice as Fairness. A Restatement (Cambridge, Mass.: Harvard University Press, 2001).

311 — J. Raz, Ethics in the Public Domain (Oxford: Clarendon Press, 1994). — J. P. Reid, The Concept of Representation in the Age of the American Revolution (Chicago: The University of Chicago Press, 1989). — C. J. Richard, The Founders and the Classics: Greece, Rome, and the American Enlightenment (Cambridge, Mass., Harvard University Press, 1994). — A. J. Reichley, The Life of Parties. A History of American Political Parties (Maryland: Rowman & Littleﬁeld, 1992). — N. Rescher, Pluralism. Against the Demand for Consensus (Oxofrd: Oxford University Press, 1993). — A. Rieger, “Voting on Voting Systems, or the Limits of Democracy,” Analysis 71(2011): 641-642. — N. Riemer, “James Madison’s Theory of the Self-Destructive Features of Republican Government,” Ethics 65 (1954), 34-43. — W. Riker, Liberalism Against Populism: A Confrontation Between the Theory of Democracy and the Theory of Social Choice (San Francisco: W. H. Freeman, 1982). — J. E. Roerner, A Future for Socialism (Cambridge, MA: Harvard University Press, 1994). — P. Rosanvallon, "The History of the Word ’Democracy’ in France," Journal of Democracy 6 (1995): 140-154. — N. L. Rosenblum, On the Side of the Angels. An Appreciation of Parties and Partisanship (Princeton: Princeton University Press, 2008). — A. Ross, “On Self-reference and a Puzzle in Constitutional Law,” Mind 78 (1969): 1-24. — J.-J. Rousseau, Du contrat social, in Oeuvres complètes, vol. 3, B. Gagnebin and M. Raymond, eds. (Paris: Gallimard, 1961). — A. Ruiz Miguel, “Modelo americano y modelo europeo de justicia constitucional,” Doxa 23 (2000): pp. 145-160. — P. Samuelson, “Wages and Interests: A Modern Dissection of Marxian Economics.” American Economic Review 47 (1957): 884-912. — I. Sánchez-Cuenca, Más democracia, menos liberalismo (Madrid: Katz, 2010). — J. Searle, “How to Derive ’Ought’ from ’Is’,? Philosophical Review 73 (1964): 195-240. — S. M. F. Sauls, The Concept of Instability and the Theory of Democracy in the Federalist (Ph.D. dissertation, Duke University, 2008). — F. Schauer, “Incentives, Reputation, and the Inglorious Determinants of Judicial Behavior,” University of Cincinnati Law Review 68 (2000): 615-636. — J. S. Schacter, “Ely and the Idea of Democracy,” Stanford Law Review 57 (2004): 737-760. — C. E. Schorske, German Social Democracy, 1905-1917: The Development of the Great Schism (Cambridge, Mass.: Harvard University Press, 1955). — D. Schmidtz, “Property and Justice,” Social Philosophy and Policy 27 (2010): 79-100.

312

Bibliography

— J. A. Schumpeter, Capitalism, Socialism, and Democracy (New York: Harper & Brothers, 1942). — D. Schweickart, “Economic Democracy: A Worthy Socialism That Would Really Work,” Science and Society 56 (1992): 9-38. — M. Schwartzberg, Democracy and Legal Change (New York: Cambridge University Press, 2007). — J. Sebo, The Personal is Political (PhD dissertation, NYU, 2011). — A. Sen, Collective Choice and Social welfare (San Francisco: Holden-Day, 1970). — A. Sen, Development as Freedom (New York: Knopf, 1999). — R. Sennett and J. Cobb, The Hideen Injuries of Class (New York: Knopf, 1972) — A. Shankman, “Malcontents and Tertium Quids: The Battle to Deﬁne Democracy in Jeﬀersonian Philadelphia,” The Journal of the Early Republic 19 (1999): 43-71. — A. Shankman, Crucible of American Democracy. The Struggle to Fuse Egalitarianism & Capitalism in Jeﬀersonian Pennsylvania (Lawrence: University of Kansas Press, 2004). — C. Shapiro and J. Stiglitz, “Unemployment as a Worker Discipline Device,” American Economic Review 74 (1982): 433-444. — M. Shapiro and A. Stone Sweet, “The New Constitutional Politics of Europe,” Comparative Political Studies 26 (1994): 397-420. — S. Stokes, “Pathologies of Deliberation,” in Deliberative Democracy, J. Elster, ed. (Cambridge: Cambridge University Press, 1998). — R. Sieder et al, The Judicialization of Politics in Latin America (London: Palgrave, 2005). — Q. Skinner, “The Paradoxes of Political Liberty,” in The Tanner Lectures on Human Values, vol. 7 (Salt Lake City, The University of Utah Press, 1986). — Q. Skinner, Liberty Before Liberalism, (Cambridge: Cambridge University Press, 1999). — B. F. Snowden, The Feasibility of Workplace Democracy in Large, Publicly Traded Corporations in the United States (PhD diss., Union Institute and University of Cincinnati, 2005). — J. P. Sterba, “Review of The Myth of Property,” Ethics 114 (2004): 628-631. — R. Stacey, “Democratic Jurisprudence and Judicial review”, Oxford Journal of Legal Studies 30 (2010): 749-773 — J. Steinfeld, The Invention of Free Labor. The Employment Relation in English and American Law and Culture (Chapel Hill: The University of North Carolina Press, 1991). — J. Steinfeld, “Property and Suﬀrage in the Early American republic.” Standford Law Review 41 (1989): 335-376. — J. A Stimson, M. B. Mackuen and R. S. Erikson, “Dynamic Representation,” American Political Science Review 89 (1995): pp. 543-565. — H. J. Storing, The Anti-Federalist (Chicago: University of Chicago Press, 1985). — A. Stutzer and B. S. Frey, “Political Participation and Procedural Utility: An Empirical Study,” European Journal of Political Research 45 (2006): 391-418.

313 — C. Sunstein, “Incompletely Theorized Agreements,” Harvard Law Review 108 (1995): 1733-1772. — R. Sudgen, “Justiﬁed to Whom?,’’ The Idea of Democracy, D. Copp, J. Hampton and J. E. Roemer, eds. (Cambridge: Cambridge University Press, 1993). — C. Sunstein, “On the Expressive Function of Law,’’ University of Pennsylvania Law Review 144 (1987): 2021-53. — C. Sunstein, Designing Democracy. What Constitutions Do (Oxford: Oxford University Press, 2001). — C. Sunstein “The Law of Group Polarization,” Journal of Political Philosophy 10 (2002): 175-195. — A. Stone Sweet, “Path Dependence, Precedent, and Judicial Power,” in A. Stone Sweet, On Law, Politics, and Judicialization (Oxford: Oxford University Press, 2002) — A. Stone Sweet, “Why Europe Rejected American Judicial Review—And Why It May Not Matter,” Michigan Law Review 101 (2003): 2744-2780. — A. Stone Sweet, “Constitutionalism, Rights, and Judicial Power,” in Comparative Politics, D. Caramani, ed. (Oxford: Oxford University Press, 2008). — H. Strum, “Property Qualiﬁcations and Voting Behavior in New York, 1807-1816,” The Journal of the Early Republic 1 (1981): 347-371. — M. Tatcher and A. Stone Sweet, “Theory and Practice of Delegation to Non-Majoritarian Institutions,” West European Politics 25 (2002): 1–22. — N. Tate and T. Vallinder, The Global Expansion of Judicial Power (NYU Press, 1997). — N. Tate and T. Wallinder, eds. The Global Expansion of judicial Power (New York: New York University Press, 1995). — T. Theorell, “Democracy at Work and its Relationship to Health,” Emotional and Physiological Processes and Positive Intervention Strategies, en P. L. Perrewé and D. C. Ganster (Amsterdam and London: Elsevier, 2004). — A. de Tocqueville, Democracy in America (Chicago, The University of Chicago Press, 2002). — L. H. Tribe, “The Puzzling Persistence of Process-Based Constitutional Theories,” Yale Law Journal 89 (1980): 1063-1080. — H. A. Turner, Hitler’s Thirty Days To Power: January 1933 (Basic Books, 1997). — M. Tushnet, Weak Courts, Strong Rights. Judicial Review and Social Welfare Rights in Comparative Constitutional Law (Princeton and Oxford: Princeton University Press, 2008). — T. R. Tyler, “Social Justice: Outcome and Procedure,” International Journal of Psychology 35 (2000): 117-125. — United Nations Development Program Human Development Report 2002, Deepening Democracy in a Fragmented World (Oxford and New York: Oxford University Press, 2002). — N. Urbinati, “Unpolitical Democracy,” Political Theory 38 (2010): 65-92.

314

Bibliography

— P. van Parijs, “Justice and Democracy: Are They Incompatible,” Journal of Political Philosophy 4 (1996): 101-117. — P. van Parijs, “Contestatory Democracy versus Real Freedom,” Just Democracy. The Rawls-Machiavelli Program (Colchester: ECPR, 2011). — S. Verba, K. L. Schlozman and H. E. Brady, Voice and Equality. Civic Voluntarianism in American Politics (Cambridge, Mass.: Harvard University Press, 1995). — J. Waldron, “Rights in Conﬂict,” Ethics 99 (1989): 503-519. — J. Waldron, “Participation: The Right of Rights,” Proceedings of the Aristotelian Society 98 (1998): 307-337. — J. Waldron, Law and Disagreement (Oxford: Oxford University Press, 1999). — J. Waldron, The Dignity of Legislation (Cambridge: Cambridge University Press, 1999). — J. Waldron “The Core of the Case Against Judicial Review,” Yale Law Journal 115 (2006): 1346-1406. — S. Wall, “Democracy, Authority and Publicity,” Journal of Political Philosophy 14 (2006): 85-100. — J. Waluchow, A Common Law Theory of Judicial Review: The Living Tree (Cambridge: Cambridge University Press, 2007). — W. Waluchow, “Constitutionalism,” The Stanford Encyclopedia of Philosophy (Fall 2008 Edition). — M. Walzer, Spheres of Justice (New York: Basic books, 1983) — M. Weber, The Protestant Ethic and the Spirit of Capitalism, trans. T. Parsons (New York: Scribner, 1958). — N. Webster, “Government. The Practice of Instructing Representatives Absurd and Contrary to the True Principles of Liberty,” American Magazine 1 (1787-188). — L. E. Weinrib, “The Postwar Paradigm and American Exceptionalism,” The Migration of Constitutional Ideas, Sujit Choudhry, ed. (New York: Cambridge University Press, 2006). — A. Weale, Democracy (Basingstoke: Macmillan, 1999). — D. Weiss, “Wollheim’s Paradox: Survey and Solution,” Political Theory 1 (1973): 154-70. — T. Whateley, The regulations Lately Made concerning the Colonies and the Taxes Imposed upon Them, considered (London: J. Wilkie, 1765). — S. White, The Civic Minimum (Oxford: Oxford University Press, 2003). — D. Viehoﬀ, “Debate: Procedure and Outcome in the Justiﬁcation of Authority,” Journal of Political Philosophy 19 (2011): 248-259. — S. Wilentz, The Rise of American Democracy. Jeﬀerson to Lincoln (New York and London, W. W. Norton & Co., 2005). — C. Wing-hung Lo, China’s Legal Awakening (Hong Kong: Hong Kong University Press, 1995). — R. Wollheim, “A Paradox in the Theory of Democracy,” in Philosophy, Politics and Society, Peter Laslett and Runciman, eds. (Oxford: Basil Blackwell, 1962).

315 — G. S. Wood, The Creation of the American Republic, 1776-1787 (Chapel Hill, NC: The University of North Carolina Press, 1969). — G.S. Wood, “Interest and desinterestedness in the Making of the Constitution,” in Beyond Confederation. Origins of the Constitution and American National Identity, R. Beeman, S. Botein and E. Carter II, eds. (Chapell Hill: University of North Carolina Press, 1987). — G. S. Wood, “Comment on Galston Paper,” Chicago-Kent Law Review 66 (1990): 69-72. — G. S. Wood, The Radicalism of the American Revolution (New York: Vintage Books, 1991). — G. S. Wood, The Empire of Liberty. A History of the Early Republic, 1789-1915 (New York: Oxford University Press, 2009). — D. Wootton, ed., Essential Federalist and anti-Federalist Papers (Indianapolis, IN: Hackett, 2003). — E. O. Wright and M. Burawoy, “Coercion and Consent in Contested Exchange,” Politics and Society 18 (1990): 251-266.

Contents
a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Abstract/resumen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction and summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 5 7 9

I.

Democracy

Chapter 1. From demophobia to demophilia . . . . . . . . . . . . . . . . . . . . 31 1.1. Introduction . . . . . . . . . . . . . . . . . . . . . 1.2. Virtual representation . . . . . . . . . . . . . . . 1.2.1. Organic society . . . . . . . . . . . . . . . 1.2.2. Ruling elitism . . . . . . . . . . . . . . . . 1.2.3. Material independence . . . . . . . . . . . 1.3. Actual representation . . . . . . . . . . . . . . . . 1.3.1. Plurality of interests . . . . . . . . . . . . 1.3.2. Ruling anti-elitism . . . . . . . . . . . . . 1.3.3. Transformations of material independence 1.3.4. Republican democracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 34 35 42 50 53 54 60 63 69

Chapter 2. Conceptual shift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 2.1. Introduction . . . . . . . . . . . . . . . . . . . . 2.2. Brief history of the concept . . . . . . . . . . . 2.2.1. The inﬂuence of the classics . . . . . . . 2.2.2. Jeﬀersonian democracy . . . . . . . . . . 2.2.3. Jacksonian democracy . . . . . . . . . . 2.3. Early and later concepts compared . . . . . . . 2.3.1. Representation and popular sovereignty 2.3.2. Feasibility . . . . . . . . . . . . . . . . . 2.3.3. Individual and property rights . . . . . . 2.3.4. Stability . . . . . . . . . . . . . . . . . . 2.3.5. Division of powers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 76 76 78 83 87 87 90 91 93 94

Concluding remarks to Part I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

II.

Constitutionalism

Chapter 3. Analytical and normative background . . . . . . . . . . . . . . . . 101 3.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 3.2. History and deﬁnition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 3.3. Constitutional conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

318

Contents

3.4. Constitutional rigidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 3.5. Judicial review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 3.6. The counter-majoritarian diﬃculty . . . . . . . . . . . . . . . . . . . . . . . . 115 Chapter 4. Instrumentalist theories . . . . . . . . . . . . . . . . . . . . . . . . . 117 4.1. 4.2. Introduction . . . . . . . . . . . . . . . . . . The structure of instrumentalism . . . . . . 4.2.1. Outcomes 1: fundamental rights . . 4.2.2. Outcomes 2: economic performance . 4.3. Criticisms . . . . . . . . . . . . . . . . . . . 4.3.1. Wollheim’s paradox . . . . . . . . . . 4.3.2. Political disagreement . . . . . . . . 4.3.3. Procedural minds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 117 120 124 128 128 140 148

Chapter 5. Precommitment-based theories . . . . . . . . . . . . . . . . . . . . . 151 5.1. 5.2. Introduction . . . . . . . . . . . . . . . . . . . . . . . The structure of democratic precommitment . . . . . 5.2.1. Commitments . . . . . . . . . . . . . . . . . . 5.2.2. Constitutions . . . . . . . . . . . . . . . . . . 5.2.3. Reasons . . . . . . . . . . . . . . . . . . . . . 5.3. Ackerman’s dualist democracy . . . . . . . . . . . . . 5.4. Criticisms . . . . . . . . . . . . . . . . . . . . . . . . 5.4.1. Categorial confusion . . . . . . . . . . . . . . 5.4.2. Constitutional disagreements . . . . . . . . . 5.4.3. Constitutions as other-binding devices . . . . 5.4.4. Future generations and constitutional change 5.4.5. Private citizenship and participation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 152 152 153 154 156 161 162 164 170 175 181

Chapter 6. Proceduralist theories . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 6.1. Introduction . . . . . . . . . . . . . . . . . . 6.2. Excursus: was Hitler democratically elected? 6.3. Preconditions-related constraints . . . . . . 6.4. Pure procedural constraints . . . . . . . . . 6.4.1. Holmes’ constitutional grammar . . . 6.4.2. Ely’s procedural constitutionalism . 6.5. Criticisms . . . . . . . . . . . . . . . . . . . 6.5.1. Paradox of preconditions . . . . . . . 6.5.2. Disagreements, yet once again . . . . 6.5.3. Constitutive rules and rigidity . . . . 6.5.4. Nemo Iudex in Sua Causa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 186 188 190 192 192 196 196 198 203 205

Concluding remarks to Part II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209

III.

The workplace

Chapter 7. Analytical and normative background . . . . . . . . . . . . . . . . 213 7.1. Introduction . . . . . . . 7.2. Deﬁnition and models . . 7.2.1. Co-determination 7.2.2. Cooperativism . 7.3. Normative arguments . . 7.3.1. Feasibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 215 216 217 220 221

Contents 7.3.2. 7.3.3. 7.3.4. 7.3.5. Eﬃciency . . . . . . . . . . . . Expertise . . . . . . . . . . . . Ownership . . . . . . . . . . . . Power and contested exchange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

319
223 227 230 235

Chapter 8. Republicanism at work . . . . . . . . . . . . . . . . . . . . . . . . . . 247 8.1. Introduction . . . . . . . . . . . . . . 8.2. Republican freedom . . . . . . . . . . . 8.3. Republicanism in the workplace . . . . 8.3.1. Exit option . . . . . . . . . . . 8.3.2. Workplace constitutionalism . . 8.4. Workplace democracy . . . . . . . . . . 8.5. Ownership . . . . . . . . . . . . . . . . 8.5.1. Material independence . . . . . 8.5.2. Bargaining power asymmetries 8.5.3. Moral hazard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 248 251 252 255 257 260 261 263 264

Concluding remarks to Part III . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267 General conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 Resumen en castellano . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297