Multilingualism in early childhood: myths, facts, challenges

 

Rosemarie Tracy

Universität Mannheim

rtracy@rumms.uni-mannheim.de

 

 

While the Common European Framework for Languages calls for empowering European citizens with three languages each, many countries still need to wake up to the substantial multilingual potential already available right in their midst, even though the  languages involved, namely those of the immigrants of the past decades, are not part of Europe’s traditional educational curricula. One particularly intriguing consequence of this situation is that researchers interested in the psycholinguistic and sociolinguistic dimensions of language contact find a natural laboratory situation, with children of different ages and adults supplying us with new insights into the ways in which the human brain manages to juggle more than one language. In my presentation I will focus on the remarkable early linguistic abilities and analytical skills of children undergoing different acquisition types, such as first language acquisition (L1), the simultaneous acquisition of two first languages (2L1), and second language acquisition (L2) in preschool years; in addition there will be occasional reference to the verbal behavior of German immigrants to the USA who acquired their L2 English as adults. 

 

My contribution consists of three parts related to past, present and (hopefully) future views of multilingualism or bilingualism (terms which I will use interchangeably throughout). As I will explain below, I believe that linguists and practitioners in the field of language disorders should join forces and expertise in their efforts to foster multiple language acquisition in both typically and atypically developing children. My overall conclusion concerning what can be done will therefore be emphatically optimistic. As a point of departure I briefly address what kind of intellectual baggage we can safely discard, i.e. the myths alluded to in my title.

 

 

 

1.  Ideas to leave behind

 

Bilingualism as an exceptional cognitive state

 

The myth that monolingualism is more “normal” than bilingualism is easily disposed of once we take demographic facts into account. Numerically speaking, it is the monolingual who is the “rare bird”. From a linguistic point of view, of course, monolingualism is a fiction anyway, since speakers/hearers develop coexisting varieties from early on. Depending on interlocutors and contexts, even speakers who would be considered monolingual by most standards, may make use of various regional and social dialects, different registers and styles. Selecting and monitoring one’s choice of appropriate linguistic varieties is part and parcel of speech production. Strictly speaking, then, monolingualism may be not just rare but non-existent.

 

From a linguistic perspective, the notion of coexisting systems should actually be taken one step further. After all, each single utterance already involves mapping different types of representation onto each other, such as the levels of segmental and suprasegmental phonology, morphosyntax, argument structure and pragmatics, to name but a few. Second or foreign language learners may master these levels to different degrees, and in some domains, such as the lexicon and stylistic repertoires, we easily find differences among native speakers as well.

 

 

Double standards

 

Even those of us who are well aware of demographic facts may find it hard to refrain from prejudice when it comes to saying which languages are worth maintaining or acquiring in the first place. Attitudes clearly depend on the “market value” of a language, i.e. on the prestige of its speakers. At this point in time, European pediatricians, teachers and other educators are more likely to warn against the simultaneous acquisition of languages such as Berber, Urdu, or Akan than against high-status languages such as English and French. 

 

Double standards reveal themselves in other contexts as well. Most of us could probably produce anecdotal evidence of situations where we easily ignore a heavy foreign accent or deviant gender and case marking in the German of prestigious non-native speakers. At the same time we might feel considerably less tolerant toward the seven-year-old Turkish- or Russian-speaking child who, after having been exposed to German for two years or less, still errs in gender assignment or drops prepositions. Few of us might be willing to even consider the possibility that the child or young person whose German reveals ethnolectal features is quite capable of conforming to other variants of German, including articles and prepositions, should the need arise (cf. Keim 2004). Clearly, we need to re-evaluate our attitudes and overcome persistent deficit perspectives.

 

 

Bilinguals: two monolinguals in one

 

As other speakers at this conference have highlighted in their research, bilinguals do not necessarily behave like two monolinguals in one (if there were monolinguals, that is!). It is entirely unrealistic to expect bilinguals to gladly talk about any topic to any stranger happening to address them in one of their languages. Moreover, unless bilinguals have been exposed to the vocabulary needed for certain topics in both languages, it is unlikely that they can express themselves with equal ease in each. In many cases, bilinguals may not have cross-linguistic lexical equivalents in their languages for the simple reason that there are no equivalents available in the first place.  In particular, we have to stop convincing bilinguals (including most of us) that there is something wrong with them because their languages work for them according to some “division of labor” principle.

 

 

Language mixing:  a deficit

 

For many observers language mixing is indicative of a serious linguistic deficit or “semilingualism”. Thanks to intensive sociolinguistic research of the last decades, we now know that code-mixing is part of the natural stylistic repertoire of bilingual speech communities (cf. Myers-Scotton 2006). In general, it is surprisingly systematic and not due to the inability to converse in one language only. Situational and other variables (topic, discourse function) determine whether bilinguals inhibit their languages or not (Green 1998, Lattey & Tracy 2005). Clearly, whenever only one language is appropriate, considerable monitoring is called for in order to suppress undesirable interference from the other. In multilingual contexts, on the other hand, no effort needs to be wasted on suppressing any of the languages. What this also means is that multilingualism comes at a price: The wider the range of options, the harder it may be to make the right selection in every single instance.

 

Bilingual children are capable of choosing languages according to interlocutors and of controlling their output by the time they are 2-3 years old. In my talk I will show that the very nature of children’s language mixing reveals remarkable precision work and that they borrow lexical and structural patterns from one language in a highly precise manner to fill gaps in the other. Bilingual children also show us that children in general are capable of developing metalinguistic abilities at a much earlier age than hitherto assumed.

 

 

Early bilingualism:  (con)fusion and subtractive effects

 

There is still a strong belief that children’s first languages need to develop to a certain level before a second language should enter the scene for outcomes to be beneficial to both languages and to cognitive development in general (Cummins 1986). Research from the simultaneous acquisition of two first languages shows that this concern is unnecessary. We now know that children exposed to two languages from birth can differentiate and separate their languages very early, even though the ease with which this is possible in acquisition and in online production clearly depends on similarities and contrasts exhibited by their input languages. Also, as a consequence of the relative complexity of the systems acquired, the child’s languages need not develop in perfect tandem and one language may temporarily be faster or “stronger  than the other (cf. Gawlitzek-Maiwald & Tracy 1996, Genesee et al. 1995, Müller et al. 2005; Tracy & Gawlitzek 2000, 2005).

 

To be sure, just as there is no guarantee that children offered two first languages in the home will maintain both into adulthood, there is no a priori guarantee that a child’s L1 will survive in the context of a high-status L2. But this is an issue which is independent of the basic ability to cope with more than one language from early on and is clearly related to the issue of prestige mentioned above.

 

 

Languages as “objects”

 

One myth more or less implicitly intertwined with those already mentioned and one that – fortunately – linguists and experts in language and speech disorders are not likely to entertain is that languages are somehow acquired like real objects one can possess in an all-or-nothing fashion. Knowledge of language requires the reconstruction of abstract levels (or modules) of representation, which can be mastered to different degrees: Poor phonology may well exist alongside a fair range of vocabulary and elaborate stylistic repertoires; conversely, someone who has good pronunciation skills may never master specific syntactic or pragmatic domains. In preschool pedagogy this language-qua-whole-object view often goes hand in hand with the idea that programs fostering L2 acquisition should be “holistic”. Clearly, if this means that communication with small children should take into account what they are interested in and be based on the here and now, this is correct but trivial. If this means that one should not tailor one’s programs to specific states of learner grammars and learner needs, we run the risk of being ineffective. It seems that there still is a curious fear of overtaxing children by creating a structured environment. It is not overtaxing that is a problem, but lack of sufficiently specific challenges to the learner’s analytical skills.

 

 

2.         L1 and L2 acquisition:  the state of the art

 

We have good reasons for our fascination with L1 acquisition: In the typical case it is remarkably robust, fast (but, as I shall argue in my presentation, young L2-learners can be faster yet!!), as well as independent of intelligence and modality (signed or oral language). Learners proceed in a highly systematic fashion, go through very similar stages, and require neither explicit instruction nor corrective feedback. By the time they are 2-3 years old, children unaffected by SLI have mastered the most crucial features of their target grammars. German children of that age, for instance, have figured out the intricacies of verb placement and subject-verb agreement.

 

The same holds for children exposed to two languages from birth. I will discuss evidence from children growing up with German and English as two first languages and compare them with monolinguals of the same ages. I will argue that children’s code-mixing in utterances such as Soll ich hit it?, Kannst du move a bit? can be related in a straightforward way to the developmental state reached in each  language. Mixing patterns as well self-corrections reveal interesting details about children’s early ability to monitor their own speech and about their rudimentary awareness of cross-linguistic equivalence.

 

In comparison with what we know about first language acquisition (including the simultaneous acquisition of two first languages) our knowledge of early L2 acquisition is still in its infancy – which is highly unfortunate, given the current need for information in this domain. The few longitudinal case studies currently available of children first exposed to German as their L2 between the ages of three and four  suggest that their development matches what we know from L1-acquisition in crucial respects (cf. Rothweiler i.Dr., Thoma & Tracy 2006). On the basis of L2 data from children with Russian, Turkish and Arabic as L1, it will be shown that early L2 acquisition is surprisingly robust, even under less than perfect conditions.

 

 

3.         Challenges

 

In more than one way researchers interested in studying and promoting multilingualism and experts in the profession of language and speech therapy find themselves in a similar position. First of all, they all face the challenge of differentiating what is a “typical” product of the coexistence of two or more languages in a single head (including some degree of interference and more or less intensive code-mixing) from what is a situation calling for remediation. In addition to this common interest in diagnostic puzzles, they should be able to join forces in eliminating the myths mentioned above. Also, because of the heightened awareness in both fields of the selective nature of many language and speech disorders, there is an common understanding of the need to finely tune therapeutic or educational measures to quite specific problems.

 

In countries with a high number of immigrant children we are currently faced with a kind of hybrid acquisition type cutting across the traditional categories of L1, 2L1, natural L2, and institutional or foreign language learning. This hybrid calls for a new and well-designed combination of both early L2-immersion and the deliberate exposure to very specific linguistic patterns and contrasts, as a way of challenging the learner’s natural problem-solving skills. What speech therapists can contribute to this discussion is their knowledge that even small, highly specific, and temporarily sparse interventions and remediation steps can make a big difference and trigger off important changes. This awareness of what can be achieved by children once their linguistic input is systematically enriched is certainly one of the most important messages a conference like this one can help generate.

 

 

References

 

Cummins, Jim (1986). Linguistic intedependence: A central principle of bilingual education. In: Jim Cummins & Merrill Swain (eds.). Bilingualism in education. Aspects of theory, research and practice. London: Longman, 80-95.

Gawlitzek-Maiwald, Ira & Tracy, Rosemarie (2005): The multilingual potential in emerging grammars. International Journal of Bilingualism 9 (2), 277-297.

Genesee, Fred, Nicoladis, Elena & Paradis, Johanne (1995): Language Differentiation in Early Bilingual Development. Journal of Child Language 22 (1995), 611-631.

Green, David (1998): Mental Control of the Bilingual Lexico-Semantic System. Bilingualism: Language and Cognition 1, 67-81.

Keim, Inken (2004):  Kommunikative Praktiken in türkischstämmigen Kinder- und Jugendgruppen in Mannheim. Deutsche Sprache 32, 198-226.

Lattey, Elsa & Rosemarie Tracy (2005). ’Well, I tell you, das war’n Zeiten!’ – ein deutsch-amerikanisches Sprachporträt. In: Hinnenkamp, Volker & Meng, Katharina (eds.):  Sprachgrenzen überspringen. Sprachliche Hybridität und polykulturelles Selbstverständnis.  Tübingen: Narr, 345-380.

Müller, Natascha, Kupisch, Tanja, Schmitz, Katrin & Cantone, Katja (2006). Einführung in die Mehrsprachigkeitsforschung. Tübingen: Narr.

Myers-Scotton, Carol (2006). Multiple Voices. Malden (Mass.): Blackwell.

Rothweiler, Monika (in press). The acquisition of V2 and subordinate clauses in early successive acquisition of German. Journal of Bilingualism.

Thoma, Dieter & Tracy, Rosemarie (1986). Deutsch als frühe Zweitsprache: zweite Erstsprache? In: Bernt Ahrenholz (ed.). Kinder mit Migrationshintergrund. Freiburg: Fillibach, 53-75.

Tracy, Rosemarie (1996): Vom Ganzen und seinen Teilen: Überlegungen zum doppelten Erstspracherwerb. Sprache und Kognition 15, 1/2, 70-92.

Tracy, Rosemarie & Gawlitzek-Maiwald, Ira (2000): Bilingualismus in der frühen Kindheit. In:  Grimm, Hannelore (ed.). Enzyklopädie der Psychologie. Band 3. Göttingen: Hogrefe, 495-535.