Third or Additional Language Acquisition

Gessica De Angelis

• Publisher: Multilingual Matters
• Year: 2007
• ISBN:
  9781788920568
• Topic: Sprachen & Linguistik
• Subtopic: Psycholinguistik

Third or Additional Language Acquisition examines research on the acquisition of languages beyond the L2 withing four main areas of inquiry: crosslinguistic influence, multilingual speech production models, the multilingual lexicon and the impact of bi/multilingualism on cognitive development. The book critically examines the evidence available keeping two main questions in mind. The first is whether multilinguals should be considered as learners and speakers in their own right and, consequently, whether the distinction between Second Language Acquisition and Bilingualism, and Third or Additional Language Acquisition and Multilingualism is fully warranted. The second is how proficient in a non-native language learners are supposed to be before they can begin to be classified as multilingual learners in empirical research

Chapter 1

The Multilingual Learner and Speaker

Introduction

Human beings are remarkable language learners who can easily learn and master several languages throughout their lives. Most of us have met people who can switch from one language to another within the same conversation, or children as young as four of five who can use one language with their mother, another with their father, and yet another with their kindergarten teacher. Multilingualism is, no doubt, a common achievement for many people around the word.

The increasing spread of multilingualism and the importance of language within society has led several scholars to investigate multilingual behaviour over the years, as evidenced by the strong tradition of work on sociolinguistic and educational aspects of multilingualism (Abu-Rabia, 1998; Baetens Beardsmore and Kohls, 1988; Bhatia, 2004; Bild and Swain, 1989; Brohy, 2001; Cenoz and Genesee, 1998; Cenoz et al., 2001; Clyne et al., 2004; Cummins, 2001; Dagenais and Day, 1998; Edwards, 1994; Jaspaert and Lemmens, 1990; Kramsch, 2006; Leman, 1990; Muñoz, 2000; Oksaar, 1983; Pandey, 1991). Research on the cognitive and psycholinguistic aspects of multilingualism has instead been much slower to appear. With the exception of a few early studies (Chamot, 1973; Chandrasekhar, 1978; Gulutsan, 1976; Haggis, 1973; Lococo, 1976; Tulving and Colotla, 1970; Vildomec, 1963), it is only in the 1980s that multilinguals’ processes begin to be examined closely and systematically, reaching the effect of raising a general awareness among scholars that multilinguals are learners and speakers of their own who should not be compared to L2 learners without some careful vigilance.

At present most studies on multilinguals’ acquisition and production processes can be found in academic journals, edited volumes, conference proceedings or unpublished M.A. or Ph.D. theses. This book intends to pull these references together and provide a comprehensive and up-to-date overview of research conducted within the following core areas of inquiry: crosslinguistic influence, multilingual speech production, the multilingual lexicon, and the impact of bi/multilingualism on cognitive development and the language acquisition process.

Discussions in each chapter reflect the fundamental belief that research on multilingual behaviour can offer some valuable insights about the process of non-native language acquisition and speech production as a whole. On the one hand, it can no longer go unnoticed that a large part of the world’s population speaks several languages on a daily basis, and a focus on L2 speakers alone is clearly too restrictive for future progress. On the other, most of today’s language learners go on to learn languages beyond the second one, and L2 learner behaviour cannot adequately inform us about phenomena related to multilingualism. An increased understanding of multilinguals’ processes can therefore help us develop theories and frameworks that are comprehensive and generalizable to wide groups of individuals. Most importantly, since all humans are capable of learning and speaking more than two languages, they are all actual or potential multilingual learners and speakers at any given time in their lives. In fact, humans can be argued to be multilingual by default, with the option of being monolingual or bilingual depending on factors such as educational and social context, personal interest, individual motivation and so forth.

A first question rarely addressed about the multilingual mind relates to its capacity to retain and use linguistic information over time. We all know that individuals can learn a few foreign languages with ease, but we have hardly any knowledge of the possible number of languages that can be learned and maintained over short and long periods of time. The only information on the mind’s potential that we have amounts to occasional descriptive reports of polyglots who succeeded in acquiring and using an unusually large number of languages in their lives. For instance, Baker and Jones (1998) report on the achievements of three remarkable individuals. The first is a certain Harold Williams of New Zealand, who allegedly mastered 58 languages throughout his life. The second is Derick Herning of Lerwick, Scotland, who won the Polyglot of Europe Contest in 1990 thanks to his knowledge of 22 languages. The third is Alexander Schwartz, who worked for the United Nations from 1962 to 1986, translating from a total of 31 languages.

These are extraordinary language learners who are a world apart from the typical learner researchers encounter in their work. Nonetheless, their impressive achievements provide us with a measure of the mind’s potential to learn and maintain languages over time. From these three cases we can infer that the human mind is capable of handling an exceptionally large amount of linguistic information over long periods of time, which is a remarkable ability that theories and models of non-native language acquisition and speech production must necessarily be able to account for, regardless of how frequent or rare polyglots such as these may be in real life. While it is true that the average person will never learn thirty or fifty languages, it is the potential to learn and use language that concerns us the most here, as any model which aims to be comprehensive and generalizable must be able to describe how the mind works at its full potential, and not at its limited capacity.

Studies on multilingualism with specific reference to language acquisition and speech production are generally in short supply, but recent years has seen some positive changes in this regard. A noticeable growth of interest in these topics has emerged, as evidenced by the number of publications that have appeared in the literature within a fairly short period of time, particularly from the 1990s onwards. As is usually the case with any other emerging field of inquiry, there is a period of time in which the new field is conceived of as a mere extension of other well-established fields -in our specific case Second Language Acquisition (SLA) and Bilingualism. Any process of emergence inevitably sees opposing views fighting for space and clashing with one another, and the journey is rarely smooth. The study of multilinguals’ processes is no exception.

This chapter begins by exploring some of the issues that generally surround this process of emergence. The chapter is organized in four sections as follows. The first section introduces some basic differences between Second and Third or Additional Language Acquisition and overviews the ‘no-difference’ assumption that shapes so many of the SLA studies currently available. These initial remarks are followed in the second section by a description of some of the terminological issues in the field, and then in the third section by a discussion on the existence of a bilingual bias in multilingualism research. An outline of the book content concludes the chapter, with a brief introduction to the topics covered in each of the five chapters that follow.

From Second Language Acquisition to Third or Additional Language Acquisiton

For several decades, language acquisition research has attempted to uncover the mechanisms underlying the language acquisition process, aiming to provide a comprehensive account of how humans learn their first and their non-native languages. While much progress has been made, reviews of work conducted over the past fifty to sixty years (Cook, 2001; Gass and Selinker, 2001; Larsen-Freeman and Long, 1991; Mitchell and Myles, 1998) raise some concern with respect to the restricted focus that was applied. Most of the studies available focused on the acquisition of the first language or on the acquisition of the second language, while studies on the acquisition of languages beyond the L2 are rarely mentioned and are mostly missing. These reviews then tell us that most of what we know about language acquisition does not go beyond the L2, and this means that our understanding of how non-native languages are acquired is at best partial and incomplete.

Most scholars would agree that a general theory of non-native language acquisition cannot be based on L2 learner behaviour alone. A general theory must be able to explain how the mind operates when two, as well as more than two languages are involved, and must be based on the knowledge and understanding of how the mind acquires, treats, stores, organizes and uses all the linguistic information that is available to the learner, not just the information that belongs to the first or the second language.

If one were to state that learning a first language does not substantially differ from learning a second one, a chorus of objections would be raised in no time - and rightly so. Many arguments would be put forward, from the importance of learners’ age for acquisition, to learners’ different cognitive maturity, the presence or absence of prior knowledge in the mind and so forth. By contrast, stating, implying or assuming that the acquisition of a second language does not substantially differ from the acquisition of a third or additional language does not seem to cause much of a stir among scholars and goes frequently unchallenged. What are the reasons for this difference?

In reviewing the SLA literature it is clear that most researchers have been concerned with how second languages are acquired, and have not taken the time to place their findings and observations into the broader context of non-native language acquisition as a whole. Some may argue that this situation has arisen because a distinction between an L2 learner and an L3 or an L6 learner is in fact redundant, as the processes underlying the acquisition of all non-native languages is essentially the same. Others, on the other hand, may argue that prior knowledge and prior learning experience significantly affect the acquisition process and therefore that a distinction between types of acquisition is essential.

At present the onus of highlighting meaningful differences between the acquisition of a second language and the acquisition of third or additional languages rests upon those who actively work on multilingualism and language acquisition, who generally support the view that some differences between types of acquisition exist and should be accounted for. In contrast, SLA scholars appear more willing to embrace a ‘no difference’ assumption in their work, and it is not uncommon to read statements to this effect.

The ‘no-difference’ assumption probably finds its origin in the widespread tendency to overgeneralize the meaning of the word ‘second’ in the literature (see also discussion in Hufeisen, 2000). Most people understand SLA to be a field of research concerned with how second languages are acquired, and the term ‘second’ is usually taken to refer to a second language as well as to any other non-native language in the process of being acquired. From this broad interpretation of what a second language is, we can infer that a large number of scholars regard the process of acquiring a second language as sufficiently similar to that of acquiring additional languages, implicitly supporting the view that a distinction between types of acquisition is unnecessary. Moreover, the ‘no difference’ assumption is also openly stated in the literature. Singh and Carroll (1979: 51), for instance, explain that ‘there is, a priori, non reason to assume that L3 learning is any different from L2 learning. Learning a third language is [...] learning just another second language.’ More recently, Mitchell and Myles (1998: 2) wrote that the word ‘second’ is an umbrella term for them, arguing that ‘it is sensible to include “foreign” languages under one more general term of “second” languages, because [...] the underlying learning processes are essentially the same for more local and for more remote target languages, despite differing learning purposes and circumstances.’ These are two random quotes from the literature, but many more comments of this kind could be easily located. My primary objective, though, is not to list statements but to explore why scholars may take this position.

There are probably several reasons that concur in giving recognition to the ‘no difference’ assumption, including the two following. First, the general lack of research on multilinguals’ acquisition processes has made a systematic comparison between learners with and without prior knowledge of non-native languages difficult to carry out. Researchers do not have much information they can rely upon and, as a result, do not readily identify prior knowledge of non-native languages as a variable that can significantly affect and bias the results of their work. Insufficient evidence, on the other hand, also entails that scholars cannot easily engage in informed discussions on the similarities and differences between types of acquisition. Second, the field of SLA lacks a clear working distinction between those who are learning a second language and those who are learning third or additional languages. All learners are labelled as L2 learners - particularly when proficiency in the prior non-native language(s) is low - and it is usually up to the researcher to decide whether learners’ prior knowledge has the potential to bias the result of a study or not. Such freedom of choice, needless to say, conflicts with the most basic principles of methodological rigour in language acquisition research.

While it may seem obvious to many that the prior knowledge of a non-native language is a variable that needs to be properly controlled, the reality is that the control for this specific variable is often poor, inadequate, if not lacking altogether. Learners’ linguistic background is usually monitored with care only in the case in which learners are highly proficient in a non-native language. When, however, learners have some basic knowledge of a non-native language, the additional knowledge is typically ignored or minimized, as the following example can illustrate.

In a study on learning strategies, Nayak et al. (1990) compared monolingual and multilingual students, who were assigned to the monolingual or the multilingual group according to a seven-point self-rating scale of language proficiency. Those included in the monolingual group were described as being ‘native speakers of English, with very minimal or no proficiency (ratings of 3 or below) in any other natural language’ (Nayak et al ., 1990: 226). For Nayak and his colleagues, then, there is essentially no difference between having no knowledge, and having some knowledge of a non-native language. While one can argue against this position from various angles, on a broader level it is important to ask how one can decide who is a monolingual learner of an L2, and who is a bilingual learner of an L3. Are six months of instruction in a prior non-native language enough to be classified as an L3 learner? Are perhaps two or five years of instruction more suitable? Throughout the book it will become clear that we are not yet in the position to provide an answer to this question as there is hardly any evidence available on proficiency threshold levels in non-native language acquisition. Nonetheless, we will see that some studies have already shown that even as little as one or two years of formal instruction in a non-native language can affect the acquisition of another non-native language to a significant extent, hence some added caution with respect to subject selection procedures is indeed advisable.

Even though only time and further research will allow us to identify suitable proficiency threshold levels, in the absence of an agreed upon parameter we still have to question how most researchers have dealt with this decision so far. In reading the SLA literature, one is often under the impression that some of the second language learners used in research may have been exposed to some other non-native language in their lives. With this remark I do not intend to embark on isolating those studies which may have used multilingual learners rather than second language learners in the past, but rather to raise awareness about the possibility that third or additional language learners may have been used in place of L2 learners in some occasions. The implication of this error is that some hypotheses about SLA may turn out to be incorrect or inaccurate as learners’ prior linguistic knowledge was not properly accounted for. The following two examples illustrate how frequently this situation can arise.

If we come across a study with adult Italian L1 learners of German as an L2, for instance, we can safely assume that these subjects are third or additional language learners and not L2 learners for the simple reason that the study of foreign languages, usually French or English, has been compulsory in Italian schools for several decades. Italian L1 speakers could be true L2 learners of German only in the case in which they had failed to complete compulsory education in Italy, or if they were illiterate. Similarly, if a study examines English L1 learners of French as an L2 at a Californian University, it is reasonable to wonder whether these subjects are true L2 learners, as a large number of students in California, and the United States in general, study Spanish in high school.

Scenarios of this kind are undoubtedly quite common in the SLA literature, mostly because the majority of subjects are adult university students or individuals raised in bilingual or multilingual environments. But does having some knowledge of a prior non-native language truly make a difference?

Common sense generally tells us that an individual who has gone through the experience of learning one or more non-native languages has already gained much knowledge and experience that is likely to be put to use in later learning. The transfer of prior linguistic knowledge and prior learning experience is a strong force in human cognition (Pennington, 1999; Wilson and Sperber, 2006), and when an individual engages in a cognitive task as demanding as language learning, it is reasonable to presume that prior linguistic knowledge and prior learning experience will play a role in the learning task. Since the distinction between the processes that underlie the acquisition of second or additional languages must be based on sound empirical evidence, the aim of the various chapters of this book is precisely to examine the evidence available in support of the two positions outlined (the difference versus no-difference assumption), placing special emphasis on the range of phenomena that are only possible when more than two languages are in the mind. While SLA research will clearly form a useful background for each discussion, the book will not specifically review the SLA literature as the objective is not to compare Second with Third or Additional Language Acquisition directly, but to examine the uniqueness of multilinguals’ processes with respect to the acquisition and production of languages beyond the L2.

Terminological (In)consistencies

The newness of a field goes hand in hand with a period of uncertainty about the most appropriate terminology to use. Whenever there is a need to express a new concept or describe a new phenomenon, the most frequent tendency is to borrow the terminology already used in well-established fields such as SLA or Bilingualism and adapt it to multilingualism. These borrowing strategies are usually helpful and effective, but they can also be problematic if a term starts to be used with more than one meaning. In order to avoid unnecessary overlaps, sometimes researchers resort to coining new terms altogether, but these terms are also influenced by existing ones and semantic clarity can sometimes be affected.

This section intends to illustrate the kind of terminological problems that scholars have dealt with in recent years, and are in actual fact still in the process of dealing with. Since it would be impossible to provide an exhaustive list of borrowed and newly coined terms used in the literature on multilingualism and the problems associated with each of them, the focus is purposely narrowed to four general concepts which can illustrate the type of inconsistency that have emerged. We can begin by considering the use of the terms ‘bilingual/bilingualism’ and ‘multilingual/ multilingualism’ and the level of ambiguity that these terms can express.

Most people understand a multilingual person to be an individual familiar with three or more languages to some degree of fluency, and a bilingual an individual familiar with two languages, also to some degree of fluency. This is a clear and straightforward definition based on the number of languages an individual k...