The Foreign Language Classroom: Current Perspectives and Future Considerations

The Modern Language Journal has long been an important venue for the publication of research and reflection on the teaching and learning of foreign languages (FL) in classroom contexts. In this article, we offer a perspective on the contemporary FL classroom, informed by a descriptive survey of all studies that took place in FL classes that were published in The Modern Language Journal ( MLJ) between 2001 and 2014 inclusive ( N = 97). This yielded a profile of FL classrooms in terms of geographical locations, languages being taught, the amount and distribution of instructional time, and the age and language backgrounds of the students. The findings revealed that FL environments benefiting from research investigations in the MLJ typically involve older learners in on-site (rather than virtual) classes that afford limited exposure to the FL, which was typically English (in non-English-speaking countries) and French, German, or Spanish (in English-speaking countries). We consider the implications of these findings for the study of FLs in the future and identify aspects of FL classrooms that merit greater research attention as the MLJ moves into its second century. [ABSTRACT FROM AUTHOR]

insights on contemporary FL classrooms that emerged from the survey findings. In the second section, we look ahead to some of the consequences for the FL classroom of the diverse exposure opportunities that are increasingly available to students, including the expansion of instructional time, teachers' roles in managing learning inside and outside the classroom walls, and the FL norm adopted in the classroom.
The paper concludes with the identification of some aspects of FL classrooms that merit research attention as the MLJ moves into its second centenary. A number of additional central issues related to the FL classroom receive detailed treatment in other contributions to this volume, notably the roles of teachers (Kubanyiova & Crookes); applications of technology within and beyond the classroom walls (Chun, Smith, & Kern) and socio-institutional perspectives on classroom learners (Kibler & Valdés).

Modern Languages
The first issues of the MLJ focused on modern (as opposed to classical) languages being taught in American secondary and postsecondary institutions (primarily French, German, and Spanish). Over the years, the scope has broadened to include FLs being learned outside the USA and also second languages that are being used in the society immediately outside the classroom. We have opted to restrict our discussion to the FL classroom, not only because it reflects the original mandate of the journal, but also because it would not be possible in a single paper to also do justice to the various issues arising from the diversity of second language classroom contexts such as those that cater to newcomers to a host society, to international/exchange students, to heritage/community language learners, to children enrolled in bilingual or dual-language programs, and so forth. We acknowledge that the distinctions between 'second' and 'foreign' are increasingly blurred (Kramsch, 1993;Spolsky, 1999), notably in the case of a lingua franca language such as English (Cogo, 2008;Kachru, 1985;Nayar, 4 pedagogy; and it is populated by groups of people who share the common purpose of learning/using the target L2 (although their individual learning goals may differ). This definition also provides a point of reference when we refer to innovations that challenge or extend the reach of the classroom, because they tend to involve a reconceptualization of one or more of the three features in our definition, that is, the notion of space/time, the facilitator, and/or students' learning goals. Two examples serve to illustrate this point. In MOOCs, the classroom becomes a virtual space, populated by users with diverse goals (White et al., in press); in IRIs (Interactive Radio Instruction) the classroom is the site for the delivery of a distance education program monitored by a non-specialist instructor (for example, EFL in rural schools in Colombia, Centro Virtual de Noticias de la Educación, 2014).

PROFILE OF THE FOREIGN LANGUAGE CLASSROOM IN THE MLJ: SURVEY FINDINGS
In preparation for writing this piece, we conducted a descriptive analysis of the FL classrooms that have figured in the articles published in the MLJ in the 21st century; that is, since the last set of anniversary articles published in the final issue of 2000 and the first issue of 2001. The purpose of this exercise was to use the contexts described in the articles to gain a perspective on the characteristics of FL classrooms, notably the following five factors: i) the geographical locations of the classrooms; ii) the age of the students in the classrooms; iii) the FLs being studied; iv) the language backgrounds of the students; and v) the classroom time (amount and distribution) available for instruction. Clearly, the portrait of the FL classroom that emerged from submissions to a single journal must be seen as a selective window on the contexts in which FLs are being studied around the world. The findings nevertheless serve as discussion points for reflecting on the 21st century FL classroom.
In keeping with the FL focus, we excluded any articles reporting on second or bilingual classrooms. We also excluded studies that drew students from existing classrooms (individually, in dyads or groups) to engage in researcher-created activities in lab-like settings, or that 'created' a temporary class of students for the purpose of the research. Articles meeting our criteria were then coded for the five factors mentioned above by three graduate students in applied linguistics. They began by independently coding the summer, fall, and winter issues of 2001. They then compared their coding, confirmed agreement on the inclusion criteria and the categories, and divided up the coding of the remaining articles. Two of the research assistants then verified 25% of the completed coding. This process had two purposes: to ensure that the inclusion criteria for the studies had been followed and that the relevant classroom information had been extracted accurately. The verification yielded five studies that required further discussion with the full team: four were rejected and one was retained, resulting in a final pool of 97 studies that took place in FL classrooms. For the information categories coded within each study, no discrepancies among raters were reported. This level of agreement is not surprising, given the low-inference nature of the factors (age, target language, location, etc.). Tables 1-5 in the Appendix summarize the findings from the survey. We will refer to the tables in the relevant sections below.

Overview of the MLJ Foreign Language Classroom
What does the FL learning environment reported on in the MLJ look like? First of all, it is almost always an actual classroom. Of the 97 studies that figured in our survey, the overwhelming majority (90%) took place in face-to-face contexts, with an L2 instructor. The remaining 10% included either combined face-to-face and virtual components, or contrasted the two contexts. The location of the FL classrooms spans a range of 23 countries (see Table 1). The highest concentration is in the USA, but over half of the articles reported findings from other FL contexts. It is thus clear that the physical space of a classroom continues to be a site in which contemporary FL learning takes place around the world, perhaps supplemented by different types of out-of-class exposure and interaction in the FL, but not supplanted by them. It is not possible, of course, to know the degree to which the numbers of real versus virtual classrooms in the MLJ reflect the 'typical' FL classroom orientation worldwide-one reason being that many regions of the world are not represented in the time frame we examined (no studies took place in FL classrooms in Africa or India, for example). In addition, researchers with a particular interest in the more virtual experiences afforded by technology-mediated FL learning have recourse to more specialized journals for reporting findings from these environments. It does appear, however, that research conducted in FL classrooms published in the MLJ tends to focus more on learning inside the classroom walls rather than on what students are doing outside the classroom, on their own or as part of the curriculum.
A second observation is that the FL classroom in the MLJ is largely populated by older learners.
The summary of participants' age profiles (see Table 2) demonstrates that 75% of the FL classrooms described involve adult students, of whom most are enrolled in university courses. One explanation for this demographic is likely practical: university researchers often have easier access to language classrooms in their own institutions than in other contexts. The challenges in gaining entry to schoolbased programs for children to conduct research can be formidable and are well documented (for example, Schachter & Gass, 1996). 2 The result of the focus on university learners is that we have comparatively less information about the characteristics of FL learning in school-based classrooms, which for many students, constitutes their first and most sustained classroom experience with a new language. This is particularly true for EFL classrooms, because the world-wide trend towards the earlier introduction of EFL in elementary and even pre-school has meant that the amount of time spent learning English for students under twelve in many educational systems is as great or greater than the amount of time spent learning the language at secondary levels, and much greater than the amount spent at tertiary 7 levels. We shall return to the need for greater attention to FL learning in classes for younger learners and across a broader range of geographical sites in the final section of the paper. 3 We now turn to the target languages being studied and the languages spoken by the students who are learning the new language.

Languages
The FLs featured in the MLJ have not changed much in the past 100 years. Table 3 shows that English as a FL dominates, at 35%; after which, the most common languages studied remain Spanish (20%), French (14%) and German (13%). With the exception of three studies, anytime a language other than English was being studied, the classroom was located in an English speaking country (Australia, Canada, New Zealand, the UK and the USA). Thus non-English speakers featured in the studies were almost always learning EFL. English speakers were the most common student group whose FL learning was investigated (see Table 4); and these speakers were most typically studied in classrooms in which one of just three languages was being learned (Spanish, French and German). An additional finding is that while virtually every study reported the L1 of the participants, only 4% reported whether the target language was actually the first FL the students were learning. Moreover, no study in the period surveyed explicitly focused on the learning of a FL by students who already speak two or more languages. We return to the implications of this finding in the final section of the paper.
The dominance of EFL is not at all surprising, given the current status of English as the dominant international language, and its presence in the MLJ would undoubtedly be greater without the journal's editorial policy of encouraging the publication of articles on other FLs. There are a number of consequences of the situation with English that serve to distinguish it from other FLs being learned in classrooms around the world.
One well-documented consequence of the prominence of English is that it has an impact on whether and how other languages are taught and valued. In English-speaking countries, the worldwide status of English is increasingly seen as a key factor in the decline in interest in FLs, which is in turn reflected in educational policies that clearly do not favour FL learning. For example, the current lack of student demand for FLs in England at the university level may partly result from the fact that FL learning-beginning now at age 7-is only compulsory until age 14 (Godsland, 2010). According to the first European Survey on Language Competences (ESLC) (European Commission, 2012) only 9% of teenage students in the UK are capable of communicating in a FL on a basic level about straightforward, familiar matters without support (compared to an average of 42% across all countries in the survey). The economic and cultural risks of the resulting monolingualism have emerged as a current concern in the UK (British Council, 2012;Godsland;Tinsley, 2013), and recent initiatives aim to increase students' intrinsic motivation for studying FLs (e.g., Taylor & Marsden, 2014), to provide more appropriate FL curriculum and evaluation measures (UK Department of Education, 2014), 4 and to reinstate compulsory FL study at the secondary level (Adams, 2015). In the USA, one estimate of competency in a FL acquired at school puts the figure at just 10% (Robinson, Rivers, & Brecht, 2006). 5 Another challenge for English-speaking learners of other languages derives from the expansion of English itself, as it becomes increasingly challenging for these learners to use their additional languages when their interlocutors will frequently switch into English, a FL they may speak well (Komska, 2014).
Lo Bianco (2014) contrasts the mostly low-utilitarian demand for FLs in English speaking countries (with the exception of Spanish in the USA) with the hyper-utilitarian demand for English in non-English speaking countries. In fact, English is becoming "the additional language" by far, even in settings with language policies that actively promote multilingualism, such as the European Union.
According to the ESLC, this situation results in very different levels of competence in the first and second FL, respectively: whereas 42% of teenage students are capable of communicating in the first FL (i.e. in English in almost all cases) on a basic level about straightforward, familiar matters without support, only 20% may be considered similarly competent in their second FL. Furthermore, in some Member States, English has acquired the status of 'basic skill'; that is, an academic subject to be mastered to high levels of proficiency, with many more hours devoted to its learning than to other FLs (European Commission, 2012). This attribution to English of the status of additional rather than foreign language is increasingly common in other parts of the world (Lo Bianco, 2014; see also Graddol, 2006).
As a result, proficiency in the new lingua franca rather than multilingualism has become an essential commodity in the global marketplace (Dewaele, 2009).
The final factor examined was the amount and distribution of instructional time in the FL classrooms that figured in the studies during the time period of the survey.

Instructional Time
The time required to 'learn' a language depends on a number of factors, not the least of which is the targeted proficiency. An often cited figure for expert skill performance in any domain is the 10,000hour rule (Ericsson, Krampe, & Tesch-Römer, 1993)-which, for an L2, is still far below the estimated 17,000 hours of exposure a child has had to his or her L1 by the age of four (Roffwarg, Muzio, & Dement, 1966). 6 The Foreign Service Institute in the US Department of State provides estimates based on L1-L2 differences for times required for English speakers in their intensive programs to achieve the included information about the distribution of the available instructional time. In most of the reported FL contexts (90%) students spent an average of two to three hours a week in the FL class (see Table 5). The distribution of the time ranged from short daily periods to the concentration of the available weekly time into a single class period. The total length of the course was not always reported, but the university courses (the most frequent type) normally lasted a semester (12-16 weeks). Extrapolating from these findings, a 'typical' FL course in the MLJ consists of about 37.5 hours of class time (2.5 hours x 15 weeks). Classrooms offering extensive and concentrated exposure were not common sites for research in the survey. Just four studies took place in such contexts, offering students an average of 20 hours per week of classroom contact time, spread out over several months (an average of 478 total hours). 7 We do not know, of course, the total number of classes students in the MLJ survey of classrooms might take/have taken (this information was not usually provided); nor we do have information on how much teachers might supplement class time with additional out of class exposure tasks. However, at a couple of hours a week of 'drip feed' access (Stern, 1985), it would take a considerable number of courses and additional practice time to reach anywhere close to a threshold of exposure that could result in even an intermediate level of FL knowledge. This is not a novel observation: writing in one of the first issues of the MLJ, Hills (1919) included the increasing and intensifying of classroom time as one of the reforms needed to improve American students' FL proficiency. Decades later, Stern (1985) also advocated for the concentration of instruction into substantial blocks of what he labelled 'free flow' formats. More recently, scholars such as DeKeyser (2007a) draw from the insights of Skill Acquisition Theory to underscore the considerable amount of quality practice time needed for learning a new language. Yet classrooms providing extensive exposure to the FL were not common in the survey. The skewed distribution towards limited exposure contexts reflects what has been reported elsewhere as 'typical' for FL classrooms, especially in school-based programs (Lightbown, 2014;Mitchell, 2011). This is also a point we will return to in the final section.

Summary
The survey of FL classrooms figuring in research reports in the MLJ has revealed that, thus far in this century, the classes benefiting from research investigations typically involve older learners in programs that afford relatively limited exposure to the target languages. Students studied in non-English speaking countries tend to be learning EFL; in English-speaking countries they are frequently learning one of only three languages (French, German or Spanish). We have discussed the implications of these findings mainly in relation to the prominence of English and of the drip-feed distribution of instructional time. In the next section, we look ahead to how technological innovations may be able to address the time issue, providing greater and more varied exposure to target languages. We further consider the implications of the expansion of learning space and time for the role of the teacher, as well as the impact of diverse exposure and interactions in the FL on the target language norms in the classroom.

THE EXPANDED MODERN FOREIGN LANGUAGE CLASSROOM
In the broader educational literature, increasing attention is being paid to the diversity of environments that support teaching and learning, which are "expanding the boundaries of space and time" (Trilling & Fadel, 2009, p. 139). In developed countries such as the USA, this may even include wireless internet access on school buses enabling students to engage in learning activities during long rural commutes, "part of a wider effort to use technology to extend learning beyond classroom walls and the six-hour school day" (Dillon, 2010; see also P. O'Dowd, 2010).
One consequence for the FL classroom of the future is that it need not be limited by the scheduled time allocated to classroom meeting times. Of course it never really has been, as good teachers have always sought ways to increase students' exposure to the FL. The very first issue of the MLJ contained articles on the provision of out of class experiences such as oral interaction activities with NSs in the community (Hervey, 1916) and reading for pleasure in German in which the teacher also participated (Mersereau, 1916). More recently, the proliferation of study abroad programs are seen as a way of breaking through the ceiling of 'drip-feed' programs to allow students to progress to advanced levels of proficiency (Rifkin, 2005). The change, however, is that with the 'anytime anyplace' learning that technology affords, what was once 'extra-curricular' or 'supplementary' practice can now in fact be built into a course syllabus. For example, in blended learning, a combination of face-to-face and computer-assisted learning in a single teaching and learning environment (Neumeier, 2005, p. 164), learning is stretched in time and integrated across modes. In the Flipped Classroom, a form of blended learning, students acquire topic knowledge on their own time outside the classroom (typically via video lectures), which allows class time to be used for exploring, refining and applying knowledge in more hands-on activities (Baker, 2000;Muldrow, 2013). There are many other uses of new technology (including chats, blogs, wikis, virtual worlds, gaming environments, and so on) that extend languagelearning opportunities in ways that are also transforming the classroom. These include the integration of technologies and task based language teaching, opening the classroom doors to authentic language environments (González-Lloret & Ortega, 2014); the incorporation of personal learning environments (PLEs) into the classroom (Buchem, Attwell, & Torres-Kompen, 2011) by means of electronic portfolios (e.g., Abrami et al., 2008;Attwell, 2007); and the use of multiple user domains objectoriented (MOOs) fully integrated into the syllabus to transform and enrich the language learning experience in the classroom (e.g., von der Emde, Schneider, & Kötter, 2001).
The extension of the classroom also provides richer and potentially more effective alternatives to increasing the hours of exposure than simply lowering the start of FL learning in elementary school. This is an increasingly common practice in EFL contexts around the world, but it has not necessarily resulted in the desired improvements to proficiency (see Muñoz, 2006), in large part because this is typically done by simply adding more years of drip feed access. In Québec, Canada, for example, EFL instruction for French-speaking students now begins at grade 1, but for the majority of children, the classroom time spent on English is a mere hour a week throughout the full six years of primary school. 8 Not only is the total amount of time insufficient for attaining reasonable levels of proficiency, it is often challenging for students to extract maximum benefit from the time that is offered. This is because when FL learning is treated like the learning of a specialized school subject in short infrequent sessions, it can be difficult for students to reactivate the dormant language in between lessons (Lightbown, 2014; see also Hawkins, 1978).
The expansion of learning opportunities also has a number of implications for the time spent in the classroom and consequently for the person who mediates language learning across time and space; that is, the FL teacher. This includes considerations for the organization of the actual physical layout of the classroom (see Trilling & Fadel, 2009, p. 140, for sample configuration of new learning environments/ flexible "learning studios"), so that it can accommodate the range of interactive activities that may take place during the classroom component of the learning experiences of the students. This may involve designated spaces for project meetings and presentations, in addition to whole class interactions with the teacher.
One obvious task for the classroom FL teacher in the expanded learning space is the promotion of self-directed and self-regulated learning, to allow students to extract maximum benefit from the different learning environments they may find themselves in. This echoes with the new understanding of learner autonomy as a set of "specific abilities to navigate different (learning) environments" (Reinders & White, 2011, p. 2) in which technology plays an important facilitative role. Examples include familiarizing students with resources and tools that they can use for individual practice, such as text-tospeech synthesizers (Soler Urzua, 2011), mobile speech recognition software (Liakin, Cardoso, & Liakina, 2013) and open source collections of language learning materials, such as FLAX (Flexible Language Acquisition; Fitzgerald, Wu, & Marin, 2015). It also extends to the actual content delivered by technology. Nielson's (2011) study of popular self-study CD-ROM packages with motivated adult learners found that a key factor in the high rate of attrition could be attributed to the lack of guidance and support for autonomous learning.
With the ever-expanding possibilities for students to interact with each other and others in the target language, it is possible to overlook that the language classroom is also the place where students can interact with the teacher (in addition to interacting with each other) and also where the teacher can help students prepare for, interpret, and learn from their experiences in other environments. The latter has been observed with study abroad, where the linguistic benefits are greater when students are provided with adequate pre-departure training as well as adequate follow up after returning home (DeKeyser, 2007b). Successful computer mediated communication also depends on appropriate guidance. Its absence can result in failed and frustrating learning encounters, as Thorne (2002) andRitter (2006) have documented. R. O'Dowd (2011) found that one explanation for unsuccessful telecollaboration activities in FL education is that teachers tend to receive much more training in dealing with the technology that supports out of class interaction than in the pedagogy that would better support students' in class integration of their interaction experiences. Although there is not yet research on the effectiveness of MOOCs for FL learning, including the percentage and profile of enrolled users who complete courses (the literature to date tends to describe the content rather than measure or document learning; see, for example, Wu, Fitzgerald, & Witten, 2014;Fitzgerald et al., 2015), the drop in/drop out behaviour of students enrolled in MOOCS in general  suggests that the type of exposure and practice they provide does not engage students in the same way as smaller groups of learners in actual classrooms with a teacher. These examples illustrate the key motivating role of the classroom FL teacher (Dörnyei, 2001), including the promotion of self-motivation, an essential component of self-regulated learning (Ushioda, 2006;Zimmerman, 2004).
Interaction with the teacher in the target language is also beneficial to students' actual linguistic development. There is evidence that whole-class teacher-led discourse can provide better exposure to and practice with some target features of the language than small group learner-led discourse (Toth, 2008), and that in typical pair and small group interaction students do not always generate contexts that would allow them to stretch their interlanguage (Collins & White, 2014). The teacher's contribution to students' language learning is not confined to simply managing activities, however; they also can provide students with rich exposure to language forms when, in addition to setting up opportunities for language practice, they also participate in the activities with the students, for example, taking a turn to recount an anecdote, or talking about weekend experience (Collins et al., 2009(Collins et al., , 2012. A final observation on the diverse interactions afforded by technology and global travel concerns their impact on the FL 'norm' that the classroom adopts. Most modern languages are taught with a 'native speaker' (NS) model in mind, with the expectation that students will be learning the language to use it with NSs. However, these norms are increasingly challenged. The widespread use of English as a lingua franca for communication among NNSs has been a major factor in the loosening of the NS model (Jenkins, Cogo, & Dewey, 2011) and of the privilege certain varieties of the FL have over others (e.g., Chapelle, 2009). The result is a broadening of the varieties of the language students may be exposed to and practice (including the out of classroom contact with the language that may be built into the syllabus) which is having an impact on teacher training (e.g., Davies & Patsko, 2013) and published teaching materials (e.g., Walker, 2010), not only for English but also for the teaching of other FLs, such as French (Guénette, Kennedy, Allard, & Murphy, in press;Kennedy, Guénette, Murphy, & Allard, in press;Stegu &Wachele, 2008) andSpanish (del Valle, 2014). The goal of these initiatives, as Kramsch (2014) notes is to encourage students to become "multilingual individuals, sensitive to linguistic, cultural, and above all, semiotic diversity, and willing to engage with difference" (p. 305), that is, to be able to make informed choices about and responses to language use.

THE NEXT CENTENARY: FOREIGN LANGUAGE CLASSROOMS AND THE MLJ
To return to the two questions posed at the outset of this article, it is clear that the teacher and the classroom continue to be a crucial part of the FL learning experience. Engaging in this reflection on the FL classroom using the MLJ itself as our point of departure, however, has highlighted a number of aspects that merit attention in future issues of the journal, across the different types of article formats that currently exist (including special issues). Some have been mentioned in the previous section: the integration of out of class exposure, interaction and practice into course syllabi; the impact of anytime/anyplace learning on teachers' roles in managing and facilitating learning both within and beyond the classroom walls; and lingua franca approaches to FL pedagogy and assessment. In this final section, we conclude with proposals for two avenues for future research that stand to enhance our understanding of learning in the FL classrooms of the future.
The first is research reporting on a greater range of FL learners. As seen above, two notable absences in the studies conducted in FL classrooms were younger learners and multilingual learners for whom the target language was not the first FL being studied. School-based FL programs, particularly those at the elementary school level are increasingly common and yet under-represented, not only in the MLJ but in SLA research in general. Much of what we know about L2 acquisition in instructed settings (for both the learner internal and learner external factors) comes from studies involving either older learners (frequently university students) or younger learners in acquisition rich environments, such as immersion and second language contexts in which the classroom is not the only or necessarily the most significant exposure to the target language. The generalizability of these findings to children in FL classrooms is largely underexplored (see, for example, Butler's, 2015 state of the art overview of EFL at the elementary school level in East Asia for examples of under-researched issues in that context). On the topic of the effectiveness of early starts to FL learning, surprisingly little has been done in the limited exposure contexts in which it so commonly occurs (Muñoz, 2008).
In SLA research focused on the role of L1 influence the participants' knowledge of other L2s is often not adequately controlled for or even reported. Yet, as Jessner (2008) and others have pointed out, FLs are increasingly being learned as an L3 (if not L4, etc.), and a well-documented phenomenon in cross-linguistic influence research is transfer from a known L2 during the learning or use of an additional L2 (Falk & Bardel, 2010;Jarvis & Pavlenko, 2008). To date, the ways in which this prior linguistic knowledge may be used to facilitate the learning of new FLs (for example, helping students identify sources of erroneous hypotheses about the L3 based on influence from the L2) has not yet had much impact on mainstream pedagogical approaches. Several researchers have advocated the potential benefits of pedagogy that takes into consideration students' full language repertory, based on both theoretical accounts of multilingualism (reviewed in Jessner, 2008; see also the contributions to Cenoz & Gorter, 2011) and on observed behaviour among multilingual learners in FL classrooms (Carvalho, Friere, & da Silva, 2010). For example, there is evidence that adult bilingual speakers make connections on their own when learning an L3 that is typologically similar to a previously learned L2 (Carvalho & da Silva, 2006), and that children are more likely to do so if cross-linguistic reflection is encouraged and valued by the teacher (Moore, 2006). The potential advantages of having already learned an L2, especially in a FL classroom, are not restricted to the use of cross-linguistic similarities and differences, however; students may also benefit from their use of learning strategies and heightened metalinguistic awareness (Cenoz, 2003;Moore, 2006). Kramsch (2014) argues that the multilingual individual may become the model of instruction, which implies that teachers focus on awareness of Language with a capital L, facilitating reflection on language and language use in the classroom.
Our understanding of learning in the FL classroom would also be informed by research that examines the factor of time, more specifically how much time in language classrooms (however they may be configured) is needed to reach different learning goals and when in the learning process compact instruction may be more or most beneficial. The survey clearly demonstrated that limited exposure conditions continue to dominate in FL contexts. The drawbacks of this situation, as noted above, may be offset by technological innovations that permit increased exposure and practice in both formal and informal environments. White (2014) outlines potential research agendas for exploring this in the context of FL learning, including longitudinal case studies of the impact of multiple sources of exposure on different aspects of proficiency, the ways in which teachers mentor out of class learning, and learners' perceptions on the teacher's presence/absence during different types and phases of tasks.
Within the classroom itself, increasing and concentrating the available hours to provide substantial and significant exposure to the FL, where this is possible, has proven to be an effective and efficient way for leaners to make substantial gains (Muñoz, 2012), progress which may be maintained over time (French, Collins, & Gagné, 2014). However, we need more investigations of the impact on learning of different ways in which classroom time can be increased across a range of learners and contexts. For example, content and language integrated instruction such as CLIL in Europe has been extensively described, but overviews of the existing research in CLIL contexts consistently note the paucity of empirical evidence from studies that control for the different variables (including selection criteria) that affect learning outcomes (e.g., Cenoz, Genesee & Gorter, 2014;Dalton-Puffer, Nikula, & Smit, 2010;Muñoz, 2015;Pérez-Cañado, 2012).
A better understanding of the factors that interact with time is also needed. One such factor is age (Muñoz, 2008;Swain, 1981;Turnbull et al., 1998). On the one hand, a young age is believed to be a particularly suitable time to start learning a FL, especially in an immersion setting and as far as oral skills are concerned (see DeKeyser & Larson-Hall, 2005). On the other hand, older children and adolescents are faster and more efficient learners across settings with different intensity (Muñoz & Llanes, 2014;Muñoz & Singleton, 2011). We also do not have a good understanding of the interaction between proficiency and intensive instruction. The rapid learning that results from intensive experiences at the beginner level are well documented (e.g., Collins & White, 2011) but there is some evidence that advanced learners may not benefit as much from such an experience (Serrano, 2011). There is also evidence that substantial amounts of classroom time focused on communicative interaction or content instruction may not provide rich enough input and output practice that would allow FL students to push their interlanguage to more advanced levels (Collins et al., 2012;Collins & White, 2014;Lightbown & Spada, 1990;Swain, 1985), including opportunities to reflect on non-target use of language (Lyster, 2007). Thus the quality of the classroom time also matters.
The MLJ has played a crucial role in the past 100 years in disseminating information on the teaching and learning of FLs in some of the classroom environments in which they occur. We have identified some areas related to classrooms which the journal appears to be particularly well positioned to address in its ongoing mission of supporting FL instruction. We are pleased to have been invited to contribute to the marking of this historical milestone. feedback offered by two anonymous reviewers. Funding for this project was provided by a research grant to the first author from the Quebec Ministry of Education: Le Fonds de recherche du Québec: Société et culture (FRQSC) and to the second author by 2014SGR1089 and ICREA Academia.
Another example is the case of Spanish in the United States: widely spoken in certain regions of the country, but most often taught as a foreign language in the school system (Rhodes & Pufahl, 2014;Ruiz-Farjardo, 2012). Bigelow (personal communication, March, 2015), for example, reports delays of several months in obtaining ethics approval for a classroom literacy study with school children in the USA (Bigelow & King, 2014). The procedures were considerably longer and more complex than those required for working with university L2 students. For an example of the steps involved in seeking ethics clearance for working with school-aged children, see the Minneapolis Public Schools research guidelines (http://rea.mpls.k12.mn.us/research).

3
Another under-represented population is low-literacy adults who may need to learn a FL (frequently English) to fulfill their employment duties, through work-sponsored, community or extended education courses. As Tarone and Bigelow (2005; have pointed out, this a neglected demographic in SLA research in general. Studies that do exist, however, have tended to involve students in second language classrooms (for example, Bigelow, Delmas, Hansen, & Tarone, 2006). There may be similarities in some of the learning challenges in the second and foreign contexts, but also potentially important differences with respect to exposure, motivation, and needs.

4
It is interesting to note that the three FL languages targetted for curriculum reform are the same three most commonly reported on (after English) in our survey: French, German, and Spanish.

5
It is difficult to locate comparable information for other English speaking countries, because existing surveys in the form of census data normally focus on languages spoken in the home, not those learned as part of FL education in school programs. In the USA, for example, census data for 2011 indicates that 18% of school-aged children spoke a home language other than English (Ryan, 2013), but the assumption is that these are languages that were learned in the home, not at school. The survey data reported on in Robinson et al (2006) were based on a very small sample of respondents-just 1,398 (p.

459).
6 Roffwarg, Muzio, and Dement (1966) reported that the total amount of daily sleep for 2-to 3year-old children is close to 12 hours, which implies that they are exposed to their L1 for about 12 hours a day. This adds up to approximately 17,520 hours by the time children reach the age of four.

7
There were three additional studies that took place in the more extensive exposure provided by concentrated instruction (one content-based immersion program, two language-focused intensive programs), but the total amount and distribution of time were not reported.

8
Although Canada is officially a bilingual country, French is the only official language of the province of Québec. In many regions of the province, including the suburbs of Montréal (a city where English is present), students have limited opportunities to interact in English beyond their classrooms.
Consequently, researchers working with this population often characterize English as more of a foreign than a second language in these contexts. Note. The total is less than 97 because many studies did not provide information on class time.