English Determinatives: a Distributional Proposal Based on Determinative Valence

Abstract

[eng] A determinative is understood as a type of word that gives reference to the noun (Quirk et al., 1985). Therefore, the determinative word class is understood to comprise items such as the articles a and the, the demonstratives that/those and this/these and the possessives like my and your, among others. Moreover, the determinative grammatical category is often divided in the literature—as is the case in Quirk et al. (1985)—into 3 subgroups, namely predeterminatives (e.g. half in half the time), central determinatives (e.g. the in all the three victims) and postdeterminatives (e.g. first in the first time). The idea behind these subgroups is that the is a central determinative whenever three determinatives co-occur. Therefore, if first takes place after the, it must be a postdeterminative. On the contrary, if half is located preceding the, it has to be a predeterminative. Since relevant authors as Quirk et al. (1985) are behind this syntactic approach to determinatives, most recent works seem inclined to follow this subgroup classification or at least, to create highly similar divisions. While this syntactic subgroup distribution is useful to account for the grammaticality of certain instances in the language (e.g., all the three victims), it shows inconsistencies with other possible utterances in English, especially when taking into account the following two rules that govern the subgroups: a. Determinatives belonging to the same group or position cannot co-occur.

b. Where there is more than one determinative, the order is fixed as I + II + III (i.e., predeterminative + central determinative + postdeterminative). For instance, no and more are considered position II (i.e., central determinatives) by Huddleston (1984) and they should not be able to combine according to rule a. However, their combination in an utterance like no more apologies is grammatical.

Regarding rule b, Quirk and Greenbaum (1990) consider many to be a postdeterminative and such a predeterminative. This is problematic since it does not account for the grammaticality of an instance such as she has shared many such moments. The present thesis explores such problematic occurrences. As we shall see, the unstable nature of the determinative word class in the literature leads one to consider that a reworking is needed. The present proposal aims at classifying determinatives according to the amount of different types of nouns they can determine. The three new subgroups I create are called Single Determinatives (SDs), Dual Determinatives (DDs) and Free Determinatives (FDs). SDs can only determine one type of noun (e.g., a only combines with singular count nouns, as in a man), DDs are limited to determine two types of nouns (e.g., this can only go with singular count nouns and mass nouns, as in this man and this blood) and FDs may determine any type of noun (e.g., the can combine with singular count nouns, plural count nouns and mass nouns, as in the man, the tables and the blood). The particular types of nouns a determinative may combine with are described as well. Ultimately, a division of determinatives allows for the creation of a table to describe determinative co-occurrence. In conclusion, the present thesis aims at providing a possible solution to the determinative problems found in the literature, offering a novel distributional system.

[cat] S’entén com a determinant un tipus de paraula que dona referència al nom (Quirk et al., 1985). Alguns exemples d’aquest tipus de paraula en anglès són a, the, this/that, these/those i els possessius com my. Aquesta categoria gramatical té tendència a dividir-se en la literatura en tres subgrups segons la seva posició sintàctica: predeterminatives, central determinatives i postdeterminatives. Un exemple de la coocurrència de determinants abans d’un nom nucli tables seria all the three tables. No obstant, hi ha normes: a. Els determinants del mateix subgroup no poden coocórrer. b. Quan hi ha més d’un determinant, l’ordre és fixe I + II + III (predeterminative + central determinative + postdeterminative). En principi, aquestes normes tenen l’objectiu de justificar la gramaticalitat de combinacions com all the three tables, però presenten inconsistències en altres casos. Per exemple, autors com Huddleston (1984) consideren que no i more són determinants centrals, tanmateix això és problemàtic si es té en compte la regla a. D’altra banda, autors com Quirk and Greenbaum (1990) que en la frase she has shared many such moments, la paraula many és un postdeterminant i such un predeterminant. Aquest ordre també és problemàtic si es considera la regla b. Finalment, la naturalesa inestable de la categoria gramatical de determinant en la literatura em porta a considerar que una reestructuració és necessària. La meva proposta organitza els determinants en funció dels tipus de nom que poden determinar, així creant 3 tipus diferents de subgrups. A més a més, es descriuen les particularitats de l’ítem determinat i al final s’ofereix una taula descriptiva per mostrar la possibilitat de coocurrència dels determinants.

En conclusió, la meva tesi intenta oferir una possible solució als problemes trobats en la literatura en relació als determinants, oferint un nou sistema de distribució