Second language acquisition in Brazil since the social turn

Aquisição de segunda língua no Brasil desde a virada social

ABSTRACT

The present study is an appraisal of the literature that has been published in second language acquisition (SLA) in Brazil since what is widely known as the social turn in the field (BLOCK, 2003). The objective of such an examination is to assess the impact that the social turn has had upon SLA research in the country. One hundred forty-one texts (among articles from top-ranked journals, theses, dissertations, and books) were systematically reviewed and categorized based on three main criteria: a) main topics, b) SLA frameworks/approaches, and c) methods used. The results show that socially-guided research seems to have gained space in SLA in Brazil since the social turn; however, cognitivist work in the country is still very strong, and little dialog seems to occur among these different perspectives on SLA.

KEYWORDS:
second language acquisition; social turn; Brazil

RESUMO

O presente estudo traz uma apreciação da literatura publicada no Brasil na área de aquisição de segunda língua (ASL) desde o que é amplamente conhecido como a virada social da área (BLOCK, 2003). O objetivo de tal análise é avaliar o impacto que a virada social teve nas pesquisas do campo de ASL no país. Cento e quarenta e um textos (dentre artigos dos periódicos melhor avaliados do país, teses, dissertações e livros) foram revisados sistematicamente e categorizados de acordo com três critérios: a) tópicos principais; b) perspectivas sobre ASL; e c) métodos usados. Os resultados mostram que pesquisas que seguem perspectivas sociais na área parecem ter ganhado espaço no Brasil desde a virada social; no entanto, pesquisas cognitivistas continuam mais fortes no país, e parece haver pouco diálogo entre essas diferentes abordagens teóricas sobre ASL.

PALAVRAS-CHAVE:
aquisição de segunda língua; virada social; Brasil

1. Introduction

It has been two decades since the publication of Alan Firth and Johannes Wagner’s article entitled On Discourse, Communication, and (Some) Concepts in SLA Research in The Modern Language Journal (FIRTH; WAGNER, 1997). This paper has been considered by many (e.g., BLOCK, 2003; CANAGARAJAH, 2007; ZUENGLER; MILLER, 2006) as a landmark in the field of second language acquisition (SLA), due to the challenges that it brought to the ontological, epistemological, and methodological aspects of the area up to that point. In particular, Firth and Wagner (1997) called for a reconceptualization of SLA, one that would address the “imbalance between cognitive and mentalistic orientations, and social and contextual orientations to language, the former orientation being unquestionably in the ascendancy” (p. 295). To the authors, such an imbalance was “accompanied by an analytic mindset that conceives of the FL speaker as a deficient communicator struggling to overcome an underdeveloped L2 competence, striving to reach the ‘target’ competence of an idealized [native speaker]” (p. 295-296). In concluding their critique of the field, they made the case that SLA studies ignored the fact that people generally communicate successfully in an additional language, and thus the study of effective communication between L2 speakers in real-life settings (rather than in controlled ones) should be added to its agenda.

The publication of Firth and Wagner’s article triggered (or strengthened, at least) a new direction in SLA studies - one that would look less at cognitive factors associated with acquisition, and pay more attention to the sociocultural and discursive dimensions of such a phenomenon. This orientation was arguably consolidated a few years later, with the publication of Block’s The Social Turn in Second Language Acquisition¹ (BLOCK, 2003). In this book, Block brings a critique of the highly influential “Input-Interaction-Output” cognitive model (e.g., GASS, 1997), calling for work in the field that is more socially-oriented and interdisciplinary, and setting an agenda of socially-informed SLA research. This call was followed by a number of other volumes and articles that have since sought to address the need for studies of a social nature in the area (e.g., ATKINSON, 2011a; LANTOLF, 2000; MAY, 2013; ORTEGA, 2013).

However, it is arguable that the field of SLA has still been highly oriented by cognitive, rather than social, approaches - even if the latter have been growing in terms of strength and attention. As explained by Mota (2008), cognitive approaches to SLA are very strong in the international scenario, and considering that Brazilian studies in the area have generally followed international research (MOTA, 2008), the same is true for this particular context. This is not necessarily surprising, given the recent nature of the works of scholars such as Firth and Wagner (1997) and Block (2003). Still, given the growing importance of socially-informed studies in the area and the arguments for more integration between social and cognitive perspectives (ZUENGLER; MILLER, 2006), it is important to understand the impact that the social turn has had upon the field. In Brazil, such a need is arguably even stronger, given the still incipient nature of SLA studies in this context, which have only gained in strength in the past few decades (MOTA, 2008).

The present investigation addresses this need by bringing a systematic review of the literature that has been published in SLA in Brazil since the social turn (BLOCK, 2003). In particular, the study seeks to map out and categorize such work, in order to understand whether and how the social turn in SLA has impacted research in the country in terms of topics, theoretical perspectives, and research methods. It is also hoped that this review of the literature can offer further possibilities to strengthen SLA research in the country, especially as regards socially-oriented perspectives.

In what follows, I present a more detailed explanation of the social turn in SLA and how it has developed since the publication of Firth and Wagner’s seminal article (FIRTH; WAGNER, 1997). I then explain the method that was used to generate and analyze data. Finally, I present the research results and discuss them in light of the objectives that were established for the investigation.

2. What exactly is the social turn in SLA?

According to Block (2003), until the middle of the 1990s “explicit calls for an interdisciplinary, socially-informed SLA were notable by their absence” (p. 3). As Block states, such early calls in the 1990s (and I add that other previous ones from as early as the 1980s²) received responses that were largely dismissive. Block then goes on to explain that it was only with the publication of the special issue of The Modern Language Journal, in 1997, which featured Firth and Wagner’s aforementioned article (FIRTH; WAGNER, 1997), that more attention began to be given to the social nature of SLA.

In brief, Firth and Wagner denounced the exclusively cognitive nature of SLA, which had, according to them, been “individualistic and mechanistic” (p. 285), meaning that it failed “to account in a satisfactory way for interactional and sociolinguistic dimensions of language” (p. 285). The authors claimed that SLA had generally reduced the nature of individuals to the status of “subjects,” with a preference for etic (rather than emic) perspectives of phenomena, and for quantitative, experimental methods. Concepts such as the idealized native speaker (NS) and interlanguage (IL) were strongly put into question by Firth and Wagner, based on their comprehension that misunderstandings and variations from idealized NS forms were not aberrations, but were rather “integral parts of the progression of normal, conversational discourse, regardless of the social identities of the actors involved” (p. 295).

It is thus arguable that Firth and Wagner’s article started what is now termed the social turn in SLA. However, as previously explained, such a moment was consolidated a few years later, with Block’s book-length critique of the Input-Interaction-Output Model (BLOCK, 2003). This critique was based on a close analysis and reconceptualization of each of the fundamental constructs in SLA; i.e., “second,” “language,” and “acquisition.” According to Block, “second” misrepresents the experiences and contexts of many individuals; “language” is only partially understood by traditional SLA; and “acquisition” is as social and external as it is individual and internal. It is essential to state that what Block was doing was not trying to dismiss cognitive understandings and methods of SLA, but making the case for a “broader, socially informed and more sociolinguistically oriented SLA that does not exclude the more mainstream psycholinguistic one, but instead takes on board the complexity of context, the multi-layered nature of language and an expanded view of what acquisition entails” (p. 4).³

2.1 The development of alternative frameworks

As discussed earlier, the development of this new socially-oriented SLA has become stronger with the works of other scholars and the development of other theoretical frameworks to understand the acquisition of a new language. One important example in that regard was the publication of Dwight Atkinson’s edited book entitled Alternative Approaches to Second Language Acquisition (ATKINSON, 2011a). In it, a number of researchers present frameworks that differ from cognitive orientations to SLA, and that seek to understand language acquisition from different perspectives. These perspectives are: a) sociocultural approach, b) complexity theory, c) language and identity, d) language socialization, e) conversation analysis, and f) sociocognition. As explained by Atkinson, “There is a persistent . . . body of scholars who do not follow mainstream SLA’s dominant cognitivist orientation, including four major scholars who formerly took cognitivist positions” (ATKINSON, 2011b, p. 16). His edited book is an attempt to “bring them into mutual dialogue and engagement” (p. 17), in an effort to strengthen such alternative approaches to cognitivism in the field.

In brief, the sociocultural approach to SLA is based on L. S. Vygotsky’s work, and thus focuses on how “sociocultural and mental activity are bound together in a dependent, symbolically mediated, relationship” (LANTOLF; PAVLENKO, 1995, p. 109). Concepts, such as mediation, action, and zone of proximal development, are central to the theory, while meaning creation - and consequently acquisition itself - “is a process that fundamentally arises in dialogue, either with others or with the self” (p. 110). Complexity theory, in its turn, focuses on how a multitude of interacting factors play a significant role in the acquisition of a new language, as well as in language itself. Therefore, the theory holds that no individual factor is determining for acquisition; “the interaction of them, however, has a very profound effect” on the process (LARSEN-FREEMAN, 1997, p. 151). Moreover, learning language items “is not a linear process-learners do not master one item and then move on to another. In fact, the learning curve for a single item is not linear either. The curve is filled with peaks and valleys, progress and backsliding” (p. 151). For some, there may be a question of how complexity theory may be understood as theory that looks into social factors of SLA, given its still strong cognitive basis. Still, it is included here because it “supports ecological accounts of learning that place its locus exclusively neither in the brain/body nor social interaction, but in their intersection” (LARSEN-FREEMAN, 2011, p. 66). It thus shares many views with other approaches that may be more explicitly associated with social accounts, such as the sociocultural approach (LARSEN-FREEMAN, 2011).

The main arguments of the language and identity approach to SLA are that the individual must be understood as integrated to the larger social world, and that relations of power “affect learners’ access to the target language community” (NORTON; McKINNEY, 2011, p. 73). Hence, some central concerns of the theory are “the multiple positions from which language learners can speak,” and “the ways in which opportunities to practice speaking, reading, and writing . . . are socially structured in both formal and informal sites of language learning” (p. 73). The concepts of investment (as opposed to motivation) and imagined communities - which relate to individuals’ personal desire, expected outcomes, and identification in relation to the target language and the social world - are central to the theory (for more, see also KANNO; NORTON, 2003; NORTON PIERCE, 1995; NORTON, 2000). Language socialization also has identity and power as a central concept. However, as explained by Duff and Talmy (2011), it differs from the language and identity framework, primarily in what regards its theory of learning. In brief, socialization research “seeks to account for and explain learning in much broader terms, examining not only linguistic development, but also the other forms of knowledge that are learned in and through language” (DUFF; TALMY, 2011, p. 95). These other forms include culture, social knowledge, ideologies, epistemologies and affect, among others. Furthermore, it has a stronger focus on longitudinal, ethnographic accounts - given its language anthropology origins - as opposed to the other methodological possibilities that are generally used in studies with a language and identity focus - such as the use of retrospection on the part of learners themselves (DUFF; TALMY, 2011).

The conversation analysis framework to SLA understands acquisition “as learning to participate in mundane as well as institutional everyday social environments” (KASPER; WAGNER, 2011, p. 117). Its object of inquiry is the interactional competence of participants in real-life conversations. This competence includes capacities such as understanding and producing social actions in sequential contexts, taking turns in an organized fashion, formatting actions and turns, drawing on different semiotic resources (including nonverbal ones), and repairing problems in communication (KASPER, 2006). Therefore, L2 speakers are conceived as competent participants of communicative events, rather than deficient communicators. It is important to note that conversation analysis does place emphasis on cognition; however, it conceptualizes cognition “as socially distributed between participants through their publicly displayed interactional conduct” (KASPER, 2006, p. 84). The sociocognitive framework also places emphasis on cognition (as suggested by its name), but it looks at both “inner” and “outer” cognitive processes - rather than exclusively at those that take place in interaction (ATKINSON, 2002, 2010). Cognition is thus seen as occurring both within the individual and as a continuum with the world, while learning takes place from participation in the world. Attention is given to alignment among the mind, world, and body - including one’s gaze, facial expression, gesture, body orientation, and available tools (ATKINSON et al., 2007).

There are still other theoretical frameworks that differ from cognitive orientations to SLA but that are not presented in Atkinson’s book (ATKINSON, 2011a). One of them is the study of beliefs in SLA (KALAJA; BARCELOS, 2003), which centers on “opinions and ideas that learners (and teachers) have about the task of learning a second/foreign language” (p. 1). Although research on beliefs and SLA have been influenced by cognitive psychology, and thus by cognitivism, newer research approaches - especially what Barcelos (2003) has called the contextual approach to the study of beliefs - have focused on students’ and teachers’ social contexts and cultural backgrounds, which are conceived as socially constituted. Beliefs are thus conceived as not only cognitive, but also “social constructs born out of our experiences and problems” (BARCELOS, 2003, p. 10). In that sense, the study of beliefs has certain similarities with language and identity and sociocultural theories.

Another theoretical approach that is not presented in Atkinson (2011a) is the dialogic model. This model is based on a combination between the works of Lev Vygotsky and the ideas of discourse proposed by Mikhail Bakhtin (PAIVA, 2014) - or sometimes even other theories about discourse, such as those proposed by Michael Halliday (e.g., HALLIDAY; MATTHIESSEN, 1985). As explained by Paiva (2014), the focus of this approach is on the identification, description, and explanation of the effects of discursive genres upon learners’ skills, including their ability to participate in a variety of sociocultural contexts (for more, see also Johnson, 2004). Learning is understood as a dialectic activity, and language is studied as discourse, rather than code. Learning a language thus involves learning the discursive practices of a given sociocultural context/community.

2.2 The contributions of the social turn

For Ortega (2003), the social turn in SLA “seems completed by now” (p. 4). She continues: “It has boosted the field’s capability to study social dimensions of additional language learning by adding at least six new theoretical frameworks to eight existing ones” (p. 4). In particular, Ortega highlights that an important achievement of the social turn has been the growth in the epistemological and methodological diversity of the field. In the words of Ortega herself:

Unprecedented in the social turn was the convergence of, on the one hand, efforts to dissociate from the quantitative, cognitive, positivist epistemologies dominant in SLA until the mid 1990s and, on the other hand, to refocus empirically on variation rather than universals and on individuals as much as on groups, with the allowance and even privileging of nontraditional explanations involving noncausal and probabilistic perspectives (ORTEGA, 2013, p. 3).

Ortega (2013) then goes on to state that the social turn has contributed “diverse empirical, qualitative and interpretive methodologies” (p. 5), which are “appropriate for the study of not just language development but also social dimensions of L2 learning” (p. 5). This includes methods such as ethnography, narrative inquiry, conversation analysis, dynamical description, biographical and autobiographical accounts, and interviews. Moreover, for Ortega, the new theoretical frameworks developed in the field have resulted in a “steady broadening of the contexts and populations that contemporary SLA researchers investigate, together with the consolidation of instructed SLA as a burgeoning subdomain in the field” (p. 5).

However, there are still some issues that deserve further consideration in relation to the social turn and SLA as a whole. One of them is presented by Ortega (2013) herself: the questioning of whether the disciplinary progress of the past few decades has been accompanied by transdisciplinary relevance. To her, the field has certainly contributed to the study of the ontogeny of language itself, which includes other fields, such as first language acquisition and early bilingualism (understood as different from SLA itself, in that the latter is framed by Ortega as the study of late bi/multilingualism). Moreover, with the social turn, SLA has certainly borrowed and adapted from other disciplines, which have contributed to the broadening of its theoretical basis and research methods. Nevertheless, according to Ortega (2013), it is still necessary to see how the field can contribute to other areas of knowledge.

The social turn has also been criticized by Canagarajah (2017) for still “working along the notion of competence for one language at a time” (p. 69), and not being able to account for super-diversity, which is a reality in many communities and has become even more relevant in our contemporary context of globalization - which involves more mobility and contact. Moreover, for Canagarajah, the new theoretical frameworks presented earlier “filter out the information on practices in order to focus on cognition as the locus of this competence” (p. 69). In other words, cognition has continued to play a central role, even within the new epistemological and methodological advances made by the social turn. In his view, therefore, cognition has yet to be “reconceived as embodied and embedded” (p. 70). Although this criticism is very important for the development of the field, my view is that at least some of the socially-oriented theoretical frameworks developed in the past decades - and reviewed here - have put the social (rather than the cognitive) at the center of discussions (especially language and identity, and language socialization). Even so, the criticism put forth by Canagarajah (2017) is of great relevance, and must not be taken lightly.

Addressing these and other concerns are beyond the scope of this investigation. Yet, as previously stated, the present study seeks to understand another important consideration in relation to the social turn: the impact it has had upon the SLA literature developed and published in different contexts. The context of Brazilian scholarship, in particular, is investigated here, and although the findings that will be presented later cannot be generalized and are aimed at an exclusive analysis of Brazil, they may serve as an illustration of how the social turn may have been influential for scholarship developed in different contexts worldwide - particularly those in periphery settings.

3. Method

The present study is a systematic literature review. There were three main steps for data generation and analysis: a) selection of studies for review (based on BORDINI; GIMENEZ, 2014); b) qualitative categorization of selected material; and c) quantification of categories, followed by a comparison between the number of studies in each one. In the remainder of this section, I detail each of these procedures.

3.1 Selection of studies

The selection of studies to be included in the present review followed the protocol used by Bordini and Gimenez (2014). The criteria for inclusion of a study in this systematic review were the following:

• Theme: Second language acquisition

• Chronological parameter: studies from 2003 to 2016. The year 2003 was chosen because it is the year of publication of Block’s The Social Turn in Second Language Acquisition (BLOCK, 2003), which, as previously stated, is arguably when the social turn was consolidated in SLA. Data generation took place in 2016, and thus many studies published during that year may not have been included here (it is important to note that many issues of journals in Brazil are published late).

• Terms used for search of studies: “second language acquisition” (in English); “aquisição de segunda língua” (in Portuguese). I acknowledge that the search for only these two terms may have been limiting for a number of reasons, including: a) the exclusion of studies that are in the field but do not use any of these specific terms, or use other similar terms (e.g., “foreign language acquisition” or “aquisição de língua estrangeira”); and b) the exclusion of studies in languages other than Portuguese or English. However, in my understanding, this was not a major issue, given that most studies in the area seem to use the term “second language acquisition” (either in English or Portuguese), and that Portuguese and English are the two most used languages of publication in applied linguistics journals, theses and dissertations in the country.

• Criteria for inclusion: Texts whose main field of study were SLA (texts with more than one main field of study, one of which was SLA, were included). The term “second language acquisition” was understood here based on Ortega (2013), who defined it as “the study of late bi/multilingualism” (p. 1), where additional languages are learned “subsequent to having acquired a language or languages from birth” (p. 8). Following a number of scholars (e.g., DE BOT et al., 2005), no distinction was made between “second language acquisition” and “foreign language acquisition” for two main reasons: a) the current understanding that the two terms are difficult to distinguish (and that the traditional distinction between them may be incomplete); and b) the fact that the term “second language acquisition” is the most commonly used one to refer to the research field of teaching/learning/acquiring an additional language (PAIVA, 2014), and encompasses the study of late bi/multilingualism in a number of contexts, including those traditionally associated with foreign language teaching/learning (MOTA, 2008).

• Criteria for exclusion: Texts in which one or both of the two terms used for search (“second language acquisition” and/or “aquisição de segunda língua”) were found, but which were not in this specific area were excluded. Following Ortega’s definition presented above (ORTEGA, 2013), texts about early bi/multilingualism were also excluded, although I acknowledge that some SLA researchers might include them within the scope of the field. Book reviews found were also excluded.

• Linguistic criterion: selected texts were in either English or Portuguese (see aforementioned limitations related to this methodological choice).

• Bibliographical sources: Brazilian applied linguistics journals, classified as Qualis A1 or A2 by CAPES (Coordination for the Improvement of Higher Education Personnel) (following BORDINI; GIMENEZ, 2014); Thesis/dissertation database provided by CAPES; Google Scholar; Books in the area (published in Brazil).⁴ It is important to note that at the time of data generation, the thesis/dissertation database provided by CAPES only contained studies from 2013 onwards. Google Scholar was used to find earlier theses and dissertations in the area (from 2003 to 2012; only those defended at Brazilian institutions were considered); however, given the vast number of studies found on Google Scholar, only those that had been cited at least once were included. Another important consideration is that the choice for articles in A1 and A2 journals, on the one hand, limits the possibilities of understanding published research on SLA in Brazil more broadly. On the other hand, a look at these highest-ranked journals may reveal other factors that need to be taken into consideration, such as trends in terms of publication, and what is most likely more valued in terms of topics, SLA frameworks and methods by these journals. Future research could address journals ranked lower by CAPES.

Following Bordini and Gimenez (2014), there were two readings for the selection of material: a) exploratory reading, which was fast and objective, in order to identify possible texts to be included in this review; and b) selective reading, which was used to choose which texts were actually to be included for review in this study.

3.2 Analysis of studies

Selected texts were categorized based on three main criteria: a) main topics; b) SLA frameworks/approaches; and c) methods used (quantitative, qualitative, mixed methods, or theoretical study). In terms of main topics, some texts could fit into more than one category. When this was the case, they were categorized based on the main theme that was addressed. This may be seen as limiting, given the subjective aspect of choices made for categorization; however, given the qualitative nature of the analysis conducted here, this was not understood as a problem. The case of frameworks was slightly different. When two frameworks were identified, I categorized the text as both. This was done with the intent of respecting the theoretical perspectives followed by the authors; at times, it posed a slight problem, as it made it harder to be exact in some counts. For instance, some of the identified studies not only followed cognitivist assumptions, but also brought understandings from sociocultural theory (e.g., Vygostkian theory); while this was not an issue in itself, it presented difficulties in some analyses, as in the case of identifying the number of cognitivist studies that used qualitative methods (the fact that such methods were used was probably a reflection of the sociocultural nature of these studies). Even so, considering that the number of studies that were identified under two theoretical perspectives was low (five, to be exact), this was not understood here as a major concern. The year of each publication was also taken into consideration. My intention in this case was to see if there were certain trends and changes over time, mainly in terms of theoretical frameworks/approaches and research methods.

An important explanation is necessary in terms of SLA frameworks and methods used. In both cases, the texts were categorized based on what was stated by the authors themselves. That is, if an author characterized his/her research as qualitative (method) and cognitivist (framework), his/her text was categorized accordingly, even if I identified other methods and frameworks within the study. Only in cases where characterization was not made explicit by the authors did I categorize the studies myself. It is also important to say that, as regards the methods used, the category “theoretical” was utilized for texts that discussed SLA theory/theories (e.g., cognitivism; complexity theory), rather than presenting an empirical study.

Once the categorization stage was over, the categories for each established criterion (main topics; SLA frameworks/approaches; and methods used) were quantified. The numbers of studies in each category were then compared.

4. SLA studies in Brazil since the social turn

A total of 141 texts were selected for inclusion in the present review. Of those, 52 were research articles (30 from Qualis A1 and 22 from Qualis A2 journals), 84 were theses/dissertations (65 master’s thesis, and 19 doctoral dissertations), and 5 were books. In this section, I present the results of the systematic review. The section is divided according to the 3 criteria that were used for the analysis of the studies: a) main topics, b) SLA frameworks/approaches, and c) methods used.

4.1 Main topics

The initial coding of main topics led to a total of 22 categories. Relationships among these categories were then established through axial coding, resulting in a final number of 9 topics. These were: a) SLA theory, b) four skills, c) instructed second language learning, d) psycholinguistic aspects, e) formal aspects, f) computer-assisted language learning (CALL), g) identity, h) literacy, and i) others. A definition for each of these categories is presented in Table 1.

TABLE 1
Definitions of main topics

MAIN TOPICS	Definitions
SLA theory	Theoretical issues related to SLA (e.g., a theoretical take on interlanguage; a theoretical explanation of complexity theory).
Four skills	Studies addressing at least one of the four skills (reading, speaking, listening, and/or writing), whose main focus was not teaching/learning, but the skill itself.
Instructed second language learning	Studies that look into learning that occurs in a classroom context (GASS; SELINKER, 2008).
Psycholinguistic aspects	Studies addressing at least one psycholinguistic construct (GASS; SELINKER, 2008) involved in SLA (e.g., attention; working memory).
Formal aspects	Studies addressing formal constructs in SLA (e.g., syntax; phonetics; phonology).
Computer-assisted language learning (CALL)	Studies addressing the relationship between technology and language learning
Identity	Studies addressing perceptions of self and others in language learning
Literacy	Studies addressing the construct of literacy (broadly defined) in language learning
Others	Studies that did not fit one of the above categories, or that did not have representativeness in the data (e.g. immersion)

Figure 1 (below) shows the percentage of studies in each one of the topics. Again, as stated in the previous section, it is important to highlight the subjective nature of categorizations, which means that some studies could have been categorized differently by other researchers (and, in fact, even categories might have differed if another scholar had analyzed the data). Nevertheless, as also stated earlier, this was not understood as a problem, given the qualitative aspect of the present study.

As shown in Figure 1, overall, one of the topics with the highest percentage of studies is ‘instructed second language learning’. This is not surprising and, in fact, was expected, due to the specificities of our reality in Brazil, where many people learn an additional language in formal contexts (e.g., classrooms) and several researchers address issues related to these environments.

FIGURE 1
Percentage of each topic found in all studies

There are at least two other facts shown in Figure 1 that are interesting to highlight. The first one is that studies focusing on formal aspects have the same percentage of those focusing on instructed second language learning. This may suggest that much attention is being given to structural issues related to SLA. In fact, when one adds up studies whose focus is formal aspects to those which address psycholinguistic factors - generally associated with more traditional SLA research - one is able to observe that over a third of the total number of studies center on these issues. The second fact that is noteworthy is that topics that may be more associated with social issues - mainly identity, but also literacy - have received very little attention from research since 2003.

Of course, this look at main topics alone is not enough for us to have a clear picture of how social issues have started to take part in the research landscape of SLA studies in Brazil. However, a look at these topics in each of the bibliographical source types used to select studies for the review reveals other interesting facts. For instance, studies on formal aspects (37%) and psycholinguistics (27%) comprised 64% of the total number of texts published in A1 journals. Meanwhile, instructed second language learning - which, as shown earlier, seems to be a major topic overall in the field in Brazil - accounted for just 7% of studies in these journals. This may suggest a tendency for top-tier journals in the country to publish more structurally, cognitively oriented articles - an argument that gains strength when the other criteria analyzed (mainly theoretical frameworks/approaches) are taken into consideration (see 4.2 below). Still, as mentioned above, this look at main topics alone needs to be complemented by the analyses of the other two criteria, which are presented below.

4.2 SLA frameworks/approaches

The analysis of SLA frameworks/approaches is perhaps the most crucial for the objectives of the present study, as such an examination shows what theoretical perspectives (either from a cognitivist or more socially-oriented background) informed the studies reviewed here. The coding of texts based on this particular criterion revealed 8 different theoretical frameworks: a) cognitivist, b) sociocultural theory, c) complexity theory, d) language and identity⁵, e) conversation analysis, f) beliefs, g) dialogic, and h) others. Since the majority of these frameworks are reviewed in section 2, I feel there is no need to redefine them here. Figure 2 shows the percentage of studies that were based on each of these theoretical frameworks.

FIGURE 2
Percentage of each theoretical framework found in all studies

The first interesting finding in relation to the categorization of SLA frameworks (presented in Figure 2) is that no studies using the language and socialization or sociocognitive approaches were found. In my view, there are at least two main reasons for that, one for each theoretical framework. In the case of language and socialization, this lack of studies can probably be explained by the framework’s main focus - at least up to the present moment - on contexts of language acquisition in which the language that is learned is the language most commonly used in the local community. That is, the model may seem best suited to cases where individuals from other sociocultural contexts are learning Portuguese in Brazil (or one of the other languages spoken in local communities in the country). Studies of this type of situation (Portuguese as a foreign/second language) do exist, of course, and a number of them (n=9) are part of this review; yet - at least in the data analyzed here - they did not follow a language and socialization approach. This is already an interesting implication of the present analysis - that more studies focusing on Portuguese (or one of the other languages spoken in local communities in Brazil) as an additional language can use language and socialization as a theoretical framework. As for the case of the sociocognitive approach, it is likely that the lack of studies following this framework is due to its recent nature. It would be interesting to see more studies using this particular model in future studies conducted in Brazilian contexts.

Another important finding from the analysis of SLA frameworks is the vast predominance of cognitive studies in the data. As shown in Figure 2, 60% of the total number of texts analyzed here were based on cognitive orientations to SLA. The next most used theoretical model - sociocultural theory - only accounted for 12% of this total. This is not necessarily surprising, given the vast dominance of cognitivism in the field. Moreover, the fact that cognitivism was considered to be one single framework, while socially-oriented approaches were taken as separate, also helps to explain this large difference - after all, when put together, socially-informed studies accounted for almost 40% of the total number of analyzed texts. When read this way, it is actually possible to interpret these results as evidence of a substantial number of studies that focus on the social aspects of SLA (close to 40%).

A separate look at bibliographical sources brings some other interesting factors which are worth noting and which help problematize these results further. For instance, in spite of the almost 40% of studies that are socially-oriented, when we look at A1 articles alone, we observe that 90% of them are based on cognitivism. Studies following language and identity, the dialogic model, and complexity theory account for the other 10% of these data. This result is important because it supports the previously made claim that there may be a tendency for top-tier journals in the country to publish more structurally, cognitively-oriented articles. It is important to make it clear that, by tendency here, I do not mean intentional bias, but a propensity for A1 journals to either have cognitive orientations as their aim and scope (which I find more plausible) or value cognitive studies higher than socially-informed ones. An alternative explanation for such a discrepancy in cognitivist versus socially-oriented publications in A1 journals is that scholars working with social perspectives may be looking for other venues (e.g., journals ranked lower by CAPES) to publish their work (an explanation which would complement the possibility of A1 journals having a more cognitivist scope). Of course, this is only speculative here; it would be interesting to investigate this issue further.

When we look at the books reviewed here, we see a similar picture. Of the five books included in the review, four followed cognitivist approaches - or centered mostly on them, with shorter accounts of socially-informed perspectives. The one book that was different (PAIVA; NASCIMENTO, 2009) focused on complexity theory - a theoretical approach which has gained much attention in Brazil, mostly due to the work of scholars such as Paiva and Nascimento themselves.

This dominance is lower in A2 articles and theses and dissertations, where cognitivist-informed studies account for 50% and 52% of the total number of texts, respectively. In these types of bibliographical sources, therefore, we have a higher balance between cognitivist and socially-oriented frameworks. Even so, it is important to highlight the fact that only three more socially-informed frameworks (complexity theory - 23%, sociocultural theory - 9%, and the dialogic model - 9%) appear in studies found in articles from A2 journals, which may be interpreted as further evidence for the little space that some theoretical perspectives have received thus far in prestigious journals in the country. We see more diversity, in this sense, in the case of theses and dissertations, where five socially-informed frameworks are represented: sociocultural theory (18%), complexity theory (9%), language and identity (8%), beliefs (7%), and conversation analysis (3%).

I also felt that it would be interesting to observe how the number of texts based on each of the categories for theoretical framework evolved over the chronological parameter set for the present investigation. My expectation here was that the number of texts with a more socially-oriented approach would gradually increase over the years. When looking at socially-informed frameworks individually, such an expectation was not met, although we can see certain peaks for at least three more socially-oriented frameworks after 2010: complexity theory, sociocultural theory, and language and identity. Figure 3 shows the evolution of all frameworks analyzed in the study.

FIGURE 3
Number of texts found for each theoretical framework (per year)

As shown in Figure 3, cognitivist studies also had peaks after 2010, and some before that date as well, as expected. It is important for us to consider that these peaks were at times caused by the publication of special issues in the field, which dedicated a number of articles to cognitivist studies in SLA. In spite of this consideration, it is still noteworthy that the number of cognitive studies in SLA remains much higher than that for any socially-oriented framework.

When we take the overall number of studies following socially-informed approaches together and compare it to the number of cognitive texts, we have a different picture (see Figure 4). It is true that once again the number of cognitive texts is higher than that of more socially-informed ones. However, over the time period under analysis here, and especially since 2013, we see that these numbers have become more balanced - even following similar trends in terms of peak. Therefore, it is arguable that my expectation that the number of texts with a more socially-oriented approach would gradually increase over the years was met, and such an increase seems to have been high enough to bring more balance to the cognitivist/socially-informed divide.

FIGURE 4
Number of texts found for cognitivist versus socially-oriented frameworks (per year)

It is still early to make any predictions, but this factor seems at least to indicate a promising trend in terms of more equilibrium between socially-informed and cognitivist perspectives, which, in my view, only benefits the field as a whole. It will be interesting to see how the field develops in that sense over the next few decades. It will also be interesting to see how one type of study (e.g., cognitivist) may dialog with the other (i.e., socially-informed), especially for those theoretical perspectives that put the social at the very center of discussions, which is especially the case, in my view, of language and identity and language socialization.

4.3 Methods used

As discussed earlier, one important aspect of the social turn is the diversity it brings in terms of research methods. Ortega (2013), for instance, explains that, with the social turn, more qualitative, interpretive investigations have become necessary. Therefore, an analysis of the methods used in the studies reviewed here was important. Such an analysis understandably revealed the expected 3 types of methods (quantitative, qualitative, and mixed methods); in addition to them, I also found texts that presented theoretical accounts, and one text which presented teaching methods. Figure 5 shows the percentage of texts that used each of the three types of methods, as well as those that bring theoretical accounts and the presentation of teaching methods.

FIGURE 5
Types of methods (as well as theoretical accounts and presentation of teaching methods)

It was interesting to observe that there were more qualitative than quantitative studies, given the expectation that cognitivist investigations generally favor the latter, as well as the larger number of these types of studies in the data. When looking further into this issue, I found that twenty (around 22%) of the eighty-eight texts that followed cognitivism used qualitative methods - a number that was considered high. Twelve (about 13%) other cognitivist studies used mixed methods. The remaining texts following cognitivist orientations (about 65%) were either theoretical (about 15%) or quantitative (around 50%).

It is important to remember here that, as stated earlier, some of the identified studies not only followed cognitivist assumptions, but also brought understandings from sociocultural theory (e.g. Vygostkian theory), in which case they were identified as both. This happened four times (out of the five in which a study was classified under two categories for SLA framework/approach). In all of these cases, the methods used were qualitative, and this was probably a reflection of the sociocultural nature of the investigations. Still, given the low number of studies under these two categories, this was not seen as a major issue; after all, even without them, the number of cognitivist texts using qualitative methods could still be considered high (at least for my expectations).

It is also important to say that over 90% of the quantitative studies were cognitivist, which is not surprising. Studies using complexity theory also used quantitative methods. Qualitative studies had a more diverse range of SLA approaches/frameworks, contemplating each of the identified perspectives. Mixed method studies also had a wide diversity of SLA perspectives; in fact, over 61% of them were cognitivist. This high number, combined with the number of qualitative studies that were also cognitivist, may suggest that cognitivist scholars in Brazil have looked for a variety of methods to conduct their investigations, which could be understood as an implication of the call from socially-oriented scholars for more diversity in terms of both methods and theoretical approaches. Further investigations would be necessary in order to verify whether this is indeed the case.

On the other hand, it was interesting to see that most theoretical texts from all bibliographical sources (over 90%) were cognitivist as well. This result is important because it shows a strong need for more theoretical accounts (in books, theses, dissertations, and articles) of frameworks other than cognitivism. These accounts, I believe, would certainly have an impact on the number and quality of socially-informed empirical research in SLA in Brazil.

5. Concluding remarks

The present study sought to map out and categorize the literature that has been published in SLA in Brazil since the social turn (BLOCK, 2003) in order to assess the impact that such a turn has had upon the field in the country. To do so, a systematic review of literature was conducted, with a focus on the main topics, theoretical perspectives, and research methods used in articles, books, theses, and dissertations in the field of SLA in Brazil since 2003 (which is arguably the year when the social turn became consolidated internationally).

At least two main conclusions may be drawn from the review presented here. The first one is that cognitivist research in Brazil is still very strong; and such strength is perhaps best illustrated by the high number of cognitivist articles published in top-tier journals in the country (probably in detriment of articles that have socially-informed orientations) and the dominance of this perspective in theoretical accounts. Still, it seems that cognitivist researchers have begun to look for methods other than quantitative ones for their investigations, which can be interpreted as an increase in diversity in terms of their research.

The second main conclusion is that socially-guided research also seems to have gained space in SLA in Brazil. Two perspectives, in particular, seem to have been very prominent: sociocultural theory and complexity theory. It is particularly noteworthy to observe that since 2013, there has been a certain balance in the number of studies that were guided by cognitivist and socially-oriented approaches. Still, it would be important to see more of these socially-guided frameworks receiving more attention in the field, as they can bring perspectives on acquisition that will help us understand the immense complexity of this phenomenon.

Another issue that seems important here is that many times cognitivist and socially-guided work seem to be taking place in parallel, at least based on the overall account of the literature reviewed here. In other words, there seems to be little dialog between these perspectives and even little discussion over the tension that exists between them in the work reviewed. While the examination of such a dialog was beyond the scope of the present investigation, it could be observed in many of the studies that were reviewed. Yet it must be said that this observation was only superficial in the present study. Further research could look into this issue more closely (including the results and implications of key studies in the area, and how they relate to one another), as the dialog between different perspectives (especially between cognitivist and socially-oriented frameworks overall) may bring more richness to the area. After all, the call for the social turn was not one of substituting cognitivism, but rather of making the case for “a broader, socially-informed and more sociolinguistically oriented SLA that does not exclude the more mainstream psycholinguistic one, but instead takes on board the complexity of context, the multi-layered nature of language and an expanded view of what acquisition entails” (BLOCK, 2003, p. 4). In fact, even scholars who have more radically opposed this more mainstream understanding of SLA (e.g., Canagarajah, 2017) have engaged in the tensions that exist between such mainstream perspectives and what they themselves propose, and my view is that the field only gains from this engagement.

Finally, it is also crucial that SLA in Brazil engages more and more frequently with areas of inquiry and concepts that go beyond the traditional ones it has engaged with over the years, both within and outside linguistics and applied linguistics. A good example, more specifically in the case of English, is the study of global Englishes. As explained by De Costa (2010), there have been many calls for scholars working with global Englishes (in fields such as world Englishes and English as a lingua franca) to get involved in debates and research on SLA, and vice-versa, and several scholars have done work that attempts to bridge the gap between these two fields (for some examples of such work, see also Canagarajah, 2007; Jenkins, 2006; Y. Kachru, 1985, 2005; Sridhar; Sridhar, 1986). In fact, a recent round-table at the Conference of the American Association for Applied Linguistics in Orlando, in 2016, featured a number of prominent scholars in both areas trying to establish how they could collaborate with one another. It would be interesting to see work in Brazil also addressing this and other gaps - including the recent push for more transdisciplinary work in SLA (e.g., ORTEGA, 2013; THE DOUGLAS FIR GROUP, 2016) - more closely. The implications of these dialogs for teaching, empirical research, and theory can be of immensurable value.

References

• ATKINSON, D. Toward a sociocognitive approach to second language acquisition. The Modern Language Journal, New York, v. 86, n. 4, p. 525-545, 2002.
• ATKINSON, D. Extended, embodied cognition and second language acquisition. Applied Linguistics, Oxford, v. 31, n. 5, p. 599-622, 2010.
• ATKINSON, D. Alternative approaches to second language acquisition. New York: Routledge, 2011a.
• ATKINSON, D. Introduction: cognitivism and second language acquisition. In: ATKINSON, D. (Org.). Alternative approaches to second language acquisition. New York: Routledge, 2011b. p. 1-23.
• ATKINSON, D.; CHURCHILL, E.; NISHINO, T.; OKADA, H. Alignment and interaction in a sociocognitive approach to second language acquisition. The Modern Language Journal, New York, v. 91, n. 2, p. 169-188, 2007.
• BARCELOS, A. M. F. Researching beliefs about SLA: a critical review. In: KALAJA, P.; BARCELOS, A. M. F. (Org.). Beliefs about SLA: new research approaches. New York: Springer, 2003, p. 7-33.
• BLOCK, D. The social turn in second language acquisition. Washington, DC: Georgetown University Press, 2003.
• BORDINI, M.; GIMENEZ, T. Estudos sobre inglês como língua franca no Brasil (2005-2012): uma metassíntese qualitativa. SIGNUM: Estudos da Linguagem, Londrina, v. 17, n. 1, p. 10-43, 2014.
• CANAGARAJAH, S. Lingua franca English, multilingual communities and language acquisition. The Modern Language Journal, New York, v. 91, n. S1, p. 923-939, 2007.
• CANAGARAJAH, S. A competence for negotiating diversity and unpredictability in global contact zones. In: DE FINA, A.; IKIZOGLU, D.; WEGNER, J. Diversity and super-diversity: sociocultural linguistic perspectives. Washington, DC: Georgetown University Press, 2017. p. 65-79.
• DE BOT, K.; LOWIE, W.; VERSPOOR, M. Second language acquisition: an advanced resource book. New York: Routledge, 2005.
• DE COSTA, P. Let’s collaborate: using developments in global English research to advance socioculturally-oriented SLA identity work. Issues in Applied Linguistics, Oxford, v. 18, n. 1, p. 99-124, 2010.
• DUFF, P.; TALMY, S. Language socialization approaches to second language acquisition: social, cultural, and linguistic developments in additional languages. In: ATKINSON, D. (Org.). Alternative approaches to second language acquisition. New York: Routledge, 2011. p. 95-116.
• FIRTH, A.; WAGNER, J. On discourse, communication, and (some) fundamental concepts in SLA research. The Modern Language Journal, New York, v. 81, n. 3, p. 285-300, 1997.
• GASS, S. Input, interaction, and the second language learner. Mahwah, NJ: Lawrence Erlbaum, 1997.
• GASS, S.; SELINKER, L. Second language acquisition: an introductory course. New York: Routledge, 2008.
• HALLIDAY, M. A. K.; MATTHIESSEN, C. M. I. M. Halliday’s introduction to functional grammar. New York: Routledge, 1985.
• JENKINS, J. Points of view and blind spots: ELF and SLA. International Journal of Applied Linguistics, Wiley Online Library, v. 16, n. 2, p. 137-162, 2006.
• JOHNSON, M. A. A philosophy of second language acquisition. London: Yale University Press, 2004.
• KACHRU, Y. Discourse analysis, non-native Englishes and second language acquisition research. World Englishes, Wiley Online Library, v. 4, n. 2, p. 223-232, 1985.
• KACHRU, Y. Teaching and learning of World Englishes. In: HINKEL, E. (Org.). Handbook of research in second language learning and teaching. Mahwah, NJ: Lawrence Erlbaum, 2005. p. 155-173.
• KALAJA, P.; BARCELOS, A. M. F. Introduction. In: KALAJA, P.; BARCELOS, A. M. F. (Org.). Beliefs about SLA: new research approaches. New York: Springer, 2003. p. 1-4.
• KANNO, Y.; NORTON, B. (Ed.). Imagined communities and educational possibilities. Journal of Language, Identity, and Education, Taylor & Francis Online, v. 2, n. 4, 2003.
• KASPER, G. Beyond repair: conversation analysis as an approach to SLA. AILA Review, John Benjamins, v. 19, p. 83-99, 2006.
• KASPER, G.; WAGNER, J. A conversation-analytic approach to second language acquisition. In: ATKINSON, D. (Org.). Alternative approaches to second language acquisition. New York: Routledge, 2011. p. 117-142.
• KRASHEN, S. Second language acquisition and second language learning. Oxford: Pergamon, 1981.
• LANTOLF, J. (Org.). Sociocultural theory and second language learning. Oxford: Oxford University Press, 2000.
• LANTOLF, J.; PAVLENKO, A. Sociocultural theory and second language acquisition. Annual Review of Applied Linguistics, Cambridge, v. 15, p. 108-124, 1995.
• LARSEN-FREEMAN, D. Chaos/complexity science and second language acquisition. Applied Linguistics, Oxford, v. 18, n. 2, p. 141-165, 1997.
• LARSEN-FREEMAN, D. A complexity theory approach to second language development/acquisition. In: ATKINSON, D. (Org.). Alternative approaches to second language acquisition. New York: Routledge, 2011. p. 48-72.
• LIGHTBOWN, P. M.; SPADA, N. How languages are learned. Oxford: Oxford University Press, 2013.
• MAY, S. The multilingual turn: implications for SLA, TESOL, and bilingual education. New York: Routledge, 2013.
• MOTA, M. B. Aquisição de segunda língua. Florianópolis: UFSC, 2008.
• NORTON, B. Identity and language learning: gender, ethnicity and educational change. Harlow: Pearson Education, 2000.
• NORTON, B.; MCKINNEY, C. An identity approach to second language acquisition. In: ATKINSON, D. (Org.). Alternative approaches to second language acquisition. New York: Routledge, 2011. p. 73-94.
• NORTON PEIRCE, B. Social identity, investment, and language learning. TESOL Quarterly, Wiley Online Library, v. 29, n. 1, p. 9-31, 1995.
• ORTEGA, L. SLA for the 21st century: disciplinary progress, transdisciplinary relevance, and the bi/multilingual turn. Language Learning, Wiley Online Library, v. 63, n. S1, p. 1-24, 2013.
• PAIVA, V. L. M. O. Aquisição de segunda língua. São Paulo: Parábola, 2014.
• PAIVA, V. L. M. O.; NASCIMENTO, M. (Org.). Sistemas adaptativos complexos: lingua(gem) e aprendizagem. Campinas: Pontes, 2011.
• SRIDHAR, S. N.; SRIDHAR, K. K. Bridging the paradigm gap: second language acquisition theory and indigenized varieties of English. World Englishes, Wiley Online Library, v. 5, n. 1, p. 3-14, 1986.
• THE DOUGLAS FIR GROUP. A transdisciplinary framework for SLA in a multilingual world. The Modern Language Journal, New York, v. 100, n. S1, p. 19-47, 2016.
• ZUENGLER, J.; MILLER, E. R. Cognitive and social perspectives: two parallel SLA worlds? TESOL Quarterly, Wiley Online Library, v. 40, n. 1, p. 35-58, 2006.

• 1
  Earlier volumes had already started to address the need for social and cultural aspects of SLA (e.g., LANTOLF, 2000). Yet Block’s book is seen as an actual call for an agenda of such studies.

• 2
  See, for instance, Sridhar & Sridhar (1986).

• 3
  It is also important to note that not only Block, but also others after him (e.g., ATKINSON, 2011b), include innatist perspectives of SLA – mainly Krashen’s monitor model (KRASHEN, 1981) – in what they call cognitivist, in spite of the fact that other scholars (LIGHTBOWN; SPADA, 2013) separate innatist and cognitive frameworks. This is mostly due to the fact that Krashen was responsible for the field’s emphasis on input, and thus current considerations of the construct can be traced back to him (ATKINSON, 2011b). This same understanding is taken here.

• 4
  Books also included electronic books and book-like material that was developed for the purposes of a course/discipline but that are widely available to the general public.

• 5
  It is interesting to observe that identity appeared both as a theoretical framework and as a topic of study in itself in the studies reviewed here.

Publication Dates

• Publication in this collection
  Mar 2018

History

• Received
  19 Aug 2017
• Accepted
  14 Nov 2017