New perspectives on multilingualism and L2 acquisition: an introduction

Abstract

This article focuses on the description of one of the main features of current multilingualism, complexity, through a selection of issues related to its role in L2 acquisition, as the proper notion of multilingualism, multilingualism as a social phenomenon and multilingualism as a multidimensional phenomenon. We also present several aspects of multilingualism in its relationship to L2 acquisition, notably some involved in the notion of multilingualism as a holistic process, and the contexts of multilingual language acquisition, and illustrate new findings of multilingualism in these areas, both from a theoretical and research perspective.

KEYWORDS:

• multilingualism
• bilingualism
• multidimensionality
• holistic view
• multilingual contexts

View correction statement:

Erratum

Multilingualism: a complex phenomenon of our times

Multilingualism is one of the most relevant social phenomena of the present age and consequently it is an issue of utmost interest both in study and research, as seen in the proliferation of studies from different areas and disciplines. Their aim is to get to know and better understand this social, linguistic and individual phenomenon, which is fast becoming one of the core issues of current communities. Although multilingualism is a phenomenon that has always existed, current multilingualism differs from that of the past, the fundamental distinction being that multilingualism today affects society as a whole (Aronin, 2015). Multilingualism affects society but it is at the same time a reflection of society. In fact, the profound changes that current multilingualism is experiencing are a reflection of the ever more complex realities of the new world and the constant and rapid transformations that our communities are living through. As a reflection of present-day societies, multilingualism today can be considered a reality, one of whose main characteristics is complexity. We can say that complexity is a key feature of current multilingualism in all its dimensions.

The present Special Issue focuses on the perspective of multilingualism and second language acquisition and it seeks to illustrate some new findings of multilingualism in this area, both from a theoretical and research perspective. These new findings affect multiple and diverse aspects of current multilingualism and reflect its main feature: complexity. We will refer now to this complexity in a selection of aspects, related to its role in second language acquisition.

The first dimension that addresses the notion of complexity can be found in the proper definition of multilingualism, and its relationship to bilingualism, as the boundaries between both notions are not always clear. We find a wide range of definitions for multilingualism and bilingualism in the literature, based on different criteria concerning the number of languages used, the level of proficiency, the ways in which languages are used and learned, and so on. These criteria differ from one author to another, and we can say that there is no unanimity for the definition of multilingualism. Among all criteria, the definitions of multilingualism and bilingualism have predominantly been related to the individual's level of language competence: a very high level of competence can be seen in early definitions of bilingualism, as in that of Bloomfield (1933): ‘native-like control of two or more languages’ (p. 56). A much more open level of competence can be seen in definitions like the one by Comanaru and Dewaele (2015): ‘proficiency to various degrees in more than one language’. The traditional question which remains unanswered is: Which degree of language competence is necessary to be bi/multilingual? At what stage of competence can we speak of bi/multilingualism? There is no precise definition of the degrees of language competence. Besides, competence has to take into account the different language areas (lexis, phonetics, syntax, etc.) and the four language skills (reading, writing, listening and speaking). But this question is not easily answered because:

Even if we are able to gauge bilingual or multilingual capacities with some accuracy, there would remain problems of adequate labeling, for it is hardly to be expected that measured individuals would neatly fall into a small number of categories of ability. (Edwards, 2013, p. 13)

It is not our purpose here to select and discuss a number of definitions of multilingualism. We assume that multilingualism refers to the ‘coexistence, contact and interaction of different languages’ at the societal or individual level (Wei, 2013, p. 26). On the one hand, a society may have speakers of different languages and, on the other, within a society individuals may be bilingual or multilingual. Perhaps at a societal or individual level, contact and interactions between languages are necessary. In a more common usage of the term, a multilingual is ‘someone who can function in two or more languages in conversational interaction’ (Wei, 2013, p. 33). Some authors distinguish between individual and societal multilingualism and have proposed to call individual bilingualism ‘bilinguality’ (Hamers & Blanc, 2000). Aronin and Ó Laoire (2004, p. 17) have coined the notion of ‘multilinguality’ which refers to ‘a personal characteristic’. Moore and Gajo (2009) distinguish between multilingualism, the study of social contact, and plurilingualism, the study of individuals’ repertoires and agency in several languages, the term plurilingualism being mostly used in the French literature to refer to individual multilingualism.

That said, what we can state is that the notions of multilingualism and bilingualism are related. Both terms are relevant in the literature about multilingualism and usually go together, although again a general agreement is lacking here. Although this could initially lead to the assumption that bilingualism is related to the use of two languages, and multilingualism to three, four, etc., current research makes use of these notions in different ways: bilingualism is sometimes used as a shorthand for multilingualism (Baetens Beardsmore, 1986); or bilingualism is distinguished from multilingualism, in the sense that two languages or more create two different phenomena with their own specific features (Herdina & Jessner, 2002), or multilingualism is considered the broad term which includes the concept of bilingualism as a particular case of multilingualism (Aronin & Singleton, 2012; Cenoz, 2013). Some scholars call for the need to make a distinction between bilingualism and multilingualism, although these boundaries once again do not seem easy to establish.

In the literature about bilingualism, there have been a number of classifications made in order to clarify and understand this field better. These classifications reflect the complexity and multidimensionality of bilingualism and they take into account different aspects (Goto Butler, 2013): the functional ability (receptive and productive bilinguals), the age of acquisition (simultaneous, sequential and late bilinguals), the relationship between language proficiencies in two languages (balanced and dominant bilinguals), the effect of L2 learning on the retention of L1 (additive and subtractive bilinguals), the organization of linguistic codes and meaning unit environment (elite/elective and folk/circumstantial bilinguals), etc. These classifications show the complexity of the phenomenon but at the same time they cannot give a thorough account of it, as they only partially cover all the possibilities and situations in which language is used by bilinguals and multilinguals. These are closed categories, but reality does not adjust easily to these labels; for instance, balanced bilinguals refer to bilinguals who have the same level of language competence in both languages, although bilinguals are very seldom equally proficient in all fields in both languages; their level of competence is influenced by the way each language is used, and normally their profile, like any bilingual profile, is constantly changing for different reasons. Concerning multilingualism, these categories are much more difficult to use because of its greater complexity.

Another dimension related to the complexity of multilingualism can be found in the consideration of multilingualism as a social phenomenon in itself. Multilingual societies arise in a number of ways: cohabitation of linguistic groups in a community; immigration or emigration, and the like. The expansion of multilingualism is attributed to the social, linguistic and cultural changes derived from globalization, geographical and social mobility, economic and political transformations and the omnipresence of technology in all areas of life. We are witnessing emerging multicultural and multilingual societies as a result of such mobility, and these societies are characterized by what Vertovec (2007) has called ‘super-diversity’. This term shows the amount and difference in nature of the parameters involved in the present population of our societies due to mobility: ethnicity, language, religion, race, age, economic and social class, to mention but a few. This results in new population groups living according to new forms of cohabitation where the concept of language linked to a territory has been overtaken.

New social models go together with new patterns of language use, and societal movements imply linguistic movement. Hence, multilingualism has become the backbone of these new societies, to the extent that Aronin and Singleton (2012) claim:

In fact, virtually every facet of life in the present era depends on multilingual social arrangements and multilingual individuals. (p. 51)

Indeed, multilingualism is considered a new social phenomenon in itself, not just the result of adding numbers of languages to individuals and societies. These authors call this new ‘status quo’ ‘new linguistic dispensation’, a term which applies to

language ideologies and policies, language education in all its dimensions, and the patterns of language use of communities and individuals. It also encompasses the systemic development and evolving status of the full spectrum of extant and emergent language varieties. (p. 43)

Multilingualism is a bidirectional process in its relation to this new ‘status quo’. The new world ‘status quo’ has created the phenomenon of multilingualism, but multilingualism contributes to the creation of the new ‘status quo’, giving rise to and making possible all kinds of relationships and interchanges through languages. Both processes continuously feed back into themselves. Our world comprises multilingual societies, and this multilingualism is leaving a deep mark on their present and orienting their future.

Multilingualism is spreading all over the world and it affects almost every society and country, though in very different ways. We speak of multilingualism for realities which vary widely from one community to the next, and each multilingual community is multilingual in its own particular way (Aronin, 2015). This adds complexity to the phenomenon of multilingualism. Different social realities of language use are present in very different environments, and are put into practice by different language users. The combination of these elements, all related to the super-diversity mentioned above, create the particularity of each multilingual community. To illustrate this, we can select the case of South Tyrol (De Angelis, 2015). Most of the population in South Tyrol is of Austro-Bavarian heritage and speaks German or native Austro-Bavarian dialects of German; around a quarter of the population speaks Italian as their first language, and a small minority speaks Ladin as their first language; German, Italian and Ladin are official languages; every citizen has the right to use their own mother tongue and schools are often distributed according to each language (see De Angelis, this issue).

One of the fields most affected by the multilingualism of our societies is the field of education. The current educational panorama of our societies, besides having second or foreign languages in the school curriculum, includes bilingual or multilingual programmes in different parts of the world, under a diversity of models depending on the countries, in which instruction is realized in two or more languages. There exist many types of multilingual education, according to the characteristics of the languages involved, their use in society and educational factors (Cenoz, 2009). We can state that multilingual education is becoming the norm almost everywhere in the world. This multilingual education, together with technological development and deep social transformations, has changed teaching and learning methods, which in turn must respond to societal needs and prepare citizens to face present and future challenges. New and important pedagogical transformations are then required, more in compliance with today's society. Some of the premises that underlie these new transformations are, notably, the integrated approaches of content and linguistic development which seem to be favourable for education; or that, in the learning process, the individuals have to be active subjects and leaders of their own learning, critical and independent (Ruiz de Zarobe and Coyle, 2015), from the first stages of education to the last, even in multilingual education, so that they become subjects capable of constructing their learning during their whole lives.

In higher education one of the most outstanding factors that has contributed to developing multilingualism in Europe is the emergence of academic mobility programmes, such as Erasmus. One of the programme's major concerns from the outset has been the acquisition of a multilingual competence by students and the fostering of multilingualism as part of the strategy adopted in the construction of the European Union. This multilingual competence has not only allowed mobility and education in students, but it has also become a decisive factor in youth employability in Europe while contributing to the integration of European individuals into other member states. Multilingualism has then been a landmark in European policies, although this trend seems to be changing recently (Pérez Vidal, 2015).

A third domain of complexity in relation to multilingualism can be found when it is viewed as a multidimensional phenomenon. Multilingualism is a phenomenon which touches upon different dimensions in societies: language education, language use, language teaching and learning, language acquisition, language practices, language policies, etc. In each of these dimensions, there are a number of different factors involved, such as the number of languages or varieties of languages, the levels of mastery and the different types of use, among others. If we take only one dimension, for instance, language acquisition, there are many different parameters involved: the number of languages of instruction, the level of proficiency, the age factor, the interaction between L1 and L2, etc. In sum, multilingualism covers a broad range of areas and within each area, a broad range of dimensions.

Being a multidimensional phenomenon, multilingualism needs to be understood from a multidimensional viewpoint. In effect, multilingualism has become a topic of multidisciplinary research: linguistics, psychology, sociology, education, management and others. Scientific research on multilingualism is quite recent, especially since the 1970s, and from its beginning it has been multidisciplinary. From then on, there has been considerable variation in research into multilingualism, encompassing a range of perspectives and approaches in theoretical and empirical research, which are sometimes disparate. From an individual to a societal level, multilingualism calls upon many aspects that have been researched extensively by diverse research disciplines.

The current tendency for multilingualism is to be more and more interdisciplinary. On this theme, and in a new and creative manner, Ruxandra-S. Comanaru and Jean-Marc Dewaele, in their article ‘A bright future for interdisciplinary multilingualism research’ explore the potential directions which multilingualism research can take, concentrating mainly on the benefits of integrating psychological methods in mainstream multilingualism research. They propose that a closer collaboration between psychologists and applied linguists is necessary in order to make advances in the field of multilingualism. The integration of the themes and methodologies of the two social sciences will be advantageous to both. The diverse methodological approaches should not be viewed as a hindrance to the development of multilingual research, but rather as an asset. Linguists and psychologists can draw on each other's methodological expertise since many of the research questions asked are comparable and complementary.

It is our understanding that this correlation between disciplines will continue to evolve in the coming years in a symbiotic and advantageous way. Some of this understanding will become more evident in the next section, devoted specifically to multilingualism and L2 acquisition.

Multilingualism and L2 acquisition: new perspectives

As we mentioned in our previous section, complexity can be found in the relationship between multilingualism and L2 acquisition. It can be found in several aspects: first, this relationship is determined by the three components of multilingualism, the speaker, the language and the setting (Edwards, 1994) and the complex interactions among them; we know of the multiplicity and variety of multilingual situations created by different individuals who learn and use two or more languages in a variety of space–time scenarios. Second, most of the issues in the acquisition of multiple languages come from bilingual acquisition studies and continue to be discussed, although multilingualism has questioned some of the premises of the L2 field and bilingualism, in using, for instance, the monolingual native speaker as a reference, or in the separation of languages in their learning and teaching. Third, it seems undeniable that in the acquisition of multiple languages, there exists more variation, challenges and complexity than in the acquisition of two languages:

The essential difference between bilingualism and multilingualism is [ … ] the degree of complexity. Bilingualism is complex, but not to the same extent as multilingualism. (Aronin, 2015, p. 15)

Multilingualism is considered a specific phenomenon with its own particular characteristics. Most researchers agree that multilinguals have special characteristics which are different from those of monolinguals or even bilinguals. Multilingual speakers use languages as a resource in communication and they use the various languages in different ways according to their communicative needs and their interlocutors. They navigate among languages and use each of them, isolated or mixed, for distinct reasons or in a different manner according to situations (Cenoz, 2013). Monolingual speakers use one language for every situation and have fewer resources available.

A point of departure in the current trend of multilingual acquisition is the premise that acquiring two, three, four or more languages cannot be considered the sum of acquiring one language and then another language and another, in a consecutive and additive process of acquisition. A multilingual is not the sum of multiple monolinguals (Cook, 1992) and ‘a bilingual is not two monolinguals in one person’ (Grosjean's famous quote, 1989). As Goto Butler (2013) points out, under these premises the monolingual view of bi/multilinguals is called into question, as is the monolingual bias (De Angelis, 2007), and the fact that monolinguals should be considered the norm to measure bilingual or multilingual proficiency. Grosjean (1985, 1989) has emphasized the idea that a bilingual is a particular kind of speaker, with his or her own specificity, and should not be studied in reference to monolinguals. Bilinguals and multilinguals are qualitatively different from monolinguals. One of the bases of this approach is precisely the questioning of the native speaker as the ideal goal to reach. This is sustained, among other issues, by the idea that a multilingual has a specific competence, different from a monolingual, which is proved by the fact that multilingual discourse is different from that of a monolingual: it is more creative and includes instances of language interactions (Cenoz, 2013). Other authors have defended the bilingual bias which claims that from the psycholinguistic point of view, bilinguals are quite similar to multilinguals (De Angelis, 2007), although this needs further research to offer more robust evidence.

Another central question in multilingual acquisition is the development of the languages involved, notably if the languages evolve in a similar way, and what the relationship is between the different languages as they develop in a multilingual. It is often assumed that not all languages have the same status, nor do they develop in a similar way in a multilingual: generally, one language or some elements of one language can develop to a greater extent.

This raises a related issue in multilingual acquisition, namely whether languages develop in a separate or unified manner in a simultaneous multilingual. This debate, which has been called the ‘one-system-or-two-debate’, questions whether bilingual children begin with a single linguistic system and then differentiate the two languages, or if they have two differentiated systems from the beginning – although the frontiers between one system or two are once again not easy to establish.

In this debate about the separation or integration of the languages at the multilingual's disposal, some authors call for an integrative perspective where languages are considered a whole unit rather than separate entities. One of the most well-known positions of the integrative perspective is Cook's (1992, 1995) notion of ‘multicompetence’. In his website, he defines multicompetence as ‘knowledge of two or more languages in the same mind’ (2012) and ‘the overall system of a mind or a community that uses more than one language’ (2013). He upholds the idea of an integration continuum between the several languages in multicompetence. In line with this view, García (2009) refers to the idea of ‘translanguaging’, a term first coined by Williams (2002), to describe ‘multiple discursive practices in which bilinguals engage in order to make sense of their bilingual words’ (García, 2009, p. 45). Canagarajah (2011, p. 401) provides an enlightening definition of ‘translanguaging’: ‘the ability of multilingual speakers to shuffle between languages, treating the diverse languages that form their repertoire as an integrated system’. These authors and others, like Herdina and Jessner (2002), agree in what has been called a ‘holistic view’ of multilingualism, in which the independent conception of monolingual competences has been replaced by the assumption of an integrated bi/multilingual competence.

In this debate, Aronin and Singleton (2012) assume that there is an important interaction and interplay among the multilingual's languages but they take issue with the idea of a strict unitary linguistic system. Some arguments they put forward are: formal differences among languages, language selection according to their interlocutor, code-switching not as a counter-example to the idea of language separation, and selective recovery of language loss due to aphasia which affects only one of the multiple languages or as a result of neglect in favour of another language.

This holistic view is developed further in the field of education by Cenoz and Gorter (2011, 2015), when several languages are studied at school, either as subjects or languages of instruction:

This holistic approach aims at integrating the curricula of the different languages so as to activate the resources multilingual speakers have. In this way multilingual students can use their resources cross-linguistically and become more efficient language learners than when languages are taught separately. (Cenoz & Gorter, 2015, p. 4)

These authors adopt a ‘multilingual turn’ view (Cenoz & Gorter, 2011; May, 2014) which has called into question the separation of languages in educational settings. Their view tries to relate the way multilinguals learn second/foreign languages to the way they communicate in real life, making it possible for the learner to activate the use of metalinguistic awareness and communicative competence acquired in previously learned languages in order to learn more efficiently.

In general, the current trend today is the view of integration rather than separation of the languages at the bi/multilingual's disposal. An example of this integration can be seen in the article by Istvan Kecskes, which provides an enlightening discussion of a somewhat neglected area in the field of bi/multilingualism, that of bi/multilingual pragmatic competence. In his article entitled ‘How Does Pragmatic Competence Develop in Bilinguals?’, Kecskes discusses this issue of language development in bilinguals; more precisely, how the emerging new language with its own developing sociocultural foundation affects the existing L1-governed knowledge and pragmatic competence of adult sequential bilinguals. It is assumed that these bilinguals already have an L1-governed pragmatic competence in place, which will be adjusted to accommodate the sociocultural requirements of the new language, so there is no separate L2 pragmatic competence. According to Kecskes, the existing L1-governed pragmatic competence becomes bilingual pragmatic competence while changing dynamically under the influence of the new language and its sociocultural requirements and behaviour patterns, thus blending the emerging features and skills with the existing ones.

Kecskes argues that there is a basic difference between the development of pragmatic competence in L1 and the sequential development of bilingual pragmatic competence. While the former is controlled mainly by the sociocultural environment, the latter is predominantly motivated by individual will and preference. In the L1, language development and social development go hand in hand as a subconscious, automatic and instinctive process in which the individual consciousness and willingness to acquire social skills and knowledge play a limited role. The whole process depends mainly on exposure to and nature of sociocultural environment. Bilingual pragmatic competence, however, develops differently. When the new language is added gradually, the existing L1-governed pragmatic competence is affected (besides exposure to L2 and environment) by individual control, consciousness and willingness to modify existing skills and behaviour patterns and acquire particular social skills while ignoring others.

A fundamental issue in multilingual acquisition studies is the context of multilingual language acquisition; that is to say, the environment where individuals acquire and are exposed to more than two languages in their daily lives: family, school, community, working environment, residence in a foreign country and the domains where the language is used: tourism, business, education and others. Three articles of this Special Issue deal with some of these contexts.

In our societies, individuals may communicate in one language in the home, use another language at school and employ a third for use within the community. In each of these domains, there are several factors which affect multilingual acquisition, such as age, exposure to languages, the way in which the languages are used, duration of contact and frequency of use. These factors can be combined thus giving rise to a complex interplay of variables.

Yet some variables and their interplay may still require further consideration. The article ‘English L3 learning in a multilingual context: the role of parental education and L2 exposure within the living community’ by Gessica de Angelis represents a new strand of research in this area. In her research, she examines two factors in relation to English L3 proficiency development and school performance in a third language: parental education and second language exposure within the living community. Participants are Italian L1 students with German L2 and English L3 attending school in the multilingual region of South Tyrol, Italy. Final year written exams were examined in relation to parental education and the language most frequently spoken in the community (Italian L1 or German L2) aiming to identify which of the two factors (parental education or L2 exposure) best predicts school performance and English L3 proficiency development. Overall results show parental education to be a stronger predictor of English L3 development.

Undoubtedly different learning contexts offer distinct opportunities for foreign language learning and may have an effect on learner progress, for instance in Study Abroad (SA) contexts in comparison to Formal Instruction (FI) contexts. It has often been claimed that mobility periods spent in the target language country are more favourable for language development, in comparison with FI contexts, particularly for the development of certain competences. This development has been put down to the nature of FI contexts, in terms of quantity and quality of input, interaction, output, etc., and to the effect of individual variables. Nevertheless, in these investigations much less attention has been paid to another factor: the role of practice in this domain, and the concept of automatization of explicit knowledge as an effect of practice during SA sojourns.

To this issue is devoted the article ‘Practice makes best: Contrasting learning contexts, comparing learner progress’ by Carmen Pérez Vidal. This author reports on the findings of the Study Abroad and Language Acquisition project, which has examined in depth the impact of these two learning contexts, SA within the ERASMUS scheme, and FI, on the progress made by advanced level higher education English as a Foreign Language (EFL) learners. The differential context effects on learners’ EFL relative gains in oral and written skills are thus measured. Following a description of the European multilingual policies in the Higher Education Area, each context is characterized from an SLA perspective. When such effects on oral skills are analysed for the dimensions of fluency and accuracy, higher gains seem to accrue in the SA context than in the FI context in both dimensions, with the exception of phonological perception. When written skills are examined, for complexity, fluency and accuracy, in a similar vein, benefits accrue to a larger extent after the period spent abroad, except for complexity. To conclude, there is a discussion of the overall superior effect on learners’ linguistic progress of the SA context over the FI context.

Another learning context that has received a great deal of attention in recent decades is content-based instruction, a learning context where both content and language are integrated in the curriculum (Content and Language Integrated Learning: CLIL). While research studies have often reported on the outcomes of integrated learning, analysing the benefits and areas for improvement of the approach (Ruiz de Zarobe, 2013), CLIL research increasingly calls for the study of the complexities of creating effective climates for learning that require new integrated approaches to learning and new pedagogic repertoires. The article written by Yolanda Ruiz de Zarobe y Do Coyle, entitled ‘Towards learner independence in bilingual educational contexts – creating conditions for reciprocity and pedagogic attention’, provides a highly innovative perspective on a number of bilingual educational contexts where CLIL is implemented as part of the curriculum. The authors reflect on some classroom-based research into effective practices across several bi/multilingual contexts with a focus on teachers and learners as co-researchers. This practitioner research involves reflexive processes which engage teachers and learners in the analysis of their own practices. These effective practices may emphasize the need to support processing reciprocity and pedagogic attention through a learning-orientation and critical incident technique as well as to provide methods for instructing learners in strategy use and metacognitive awareness. All these processes aim to promote learner independence in bilingual contexts where learning takes place through the medium of more than one language.

The field of multilingualism and second language acquisition is very broad and heterogeneous, as much as the field of multilingualism itself. Here we have brought together a selection of some issues and their innovative treatment, with the aim of paving the way for the development of multilingualism and to shed some light on the complexity of multilingualism. This complexity may feel like an impediment in our ability to encompass all the dimensions of multilingualism and the interactions of their many variables; nevertheless, it challenges us to grasp the true state of multilingualism today, which becomes ever more complex as it develops. We agree with Aronin and Singleton (2012) when they state:

Increasing awareness and explicit recognition of the complexity of multilingualism allows for a more comprehensive theoretical understanding of multilingualism and yields practical results in the teaching of multiple languages. (p. 185)

We hope then that this Special Issue will lead to a better understanding of the intricate trails followed in the acquisition of multiple languages, and will give rise to further research on multilingualism along the new pathways it has created.

Acknowledgements

We would like to thank all the contributors of this special issue for their willingness to participate in this venture. We warmly thank the reviewers of the manuscripts for their useful comments. Special thanks to Jasone Cenoz for her comments on an earlier version of this paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

We acknowledge the funding awarded by the Spanish Ministry of Economy and Competitiveness [grant number FFI2012-31811] and the Basque Department of Education, Research and Universities IT311-10 (UFI 11/06).

Notes

This article was originally published with errors. This version has been corrected. Please see Erratum (http://dx.doi.org/10.1080/14790718.2015.1112608).