Putting Interaction Back Into Child Language: Examples from Turkish

As in the case of other non-English languages, the study of the acquisition of Turkish has mostly focused on aspects of grammatical morphology and syntax, largely neglecting the study of the effect of interactional factors on child morphosyntax. This paper reviews indications from past research that studying input and adult-child discourse can facilitate the study of the acquisition of morphosyntax in the Turkish language. It also provides some recent studies of Turkish child language on the relationship of child-directed speech to the early acquisition of morphosyntax, and on the pragmatic features of a certain kind of discourse form in child-directed speech called variation sets. As in the case of other non-English languages, the study of the acquisition of Turkish has mostly focused on aspects of grammatical morphology and syntax reflected in the productions of native learners at different age periods. Descriptive linguists and psycholinguists have long regarded the properties of the Turkish morphological system and complex syntax interesting from a crosslinguistic point of view. These cross-linguistically interesting, even exotic, properties of the language led students of acquisition to prioritize their research focus on these aspects in this relatively recent area. Some of the well-known findings in Turkish child language involve the ease and the relative rapidity of the acquisition of case-inflectional and verbal-inflectional paradigms (Aksu-Koç & Slobin, 1985; Özcan, 1996; Topbaş, Maviş & Başal, 1997), the late emergence of the use of relative clauses (Slobin, 1986), the early mastery of flexible word order (Ekmekçi, 1979), and the protracted development of the different functions of the evidential marker –miş (Aksu-Koç, 1988). All of these features of Turkish child language presuppose some implicit comparison to the properties of the English language, either being non-existent or exhibiting different characteristics there. It is trivially obvious that linguistic categories such as case-marking, verbal inflections, word order, and evidentiality do not present themselves transparently to the child learner of a language. Both intuitively, and theoretically from a discourse-functional theoretical approach to language development (Budwig, 1995; Clancy, in press; DuBois, to appear), all these interesting components of the grammatical code come to the young learner in the give and take of everyday life, mostly embedded in early adult-child discursive interaction. However, as in most child language research, these “real” interactive events get reduced to textual transcripts that only represent interaction “in vitro.” For child language researchers, these transcripts constitute a mining source for grammatical elements that are combed through with painstaking eyeballing and more recently with sophisticated concordance software. Once the forms are picked out along with the propositional content surrounding them and made available for further statistical analysis, the real-time interactions that originally mediated these forms get entirely neglected. The tendency to decontextualize textual content or linguistic forms from discursive interactions troubles the entire field of child language, but it is accentuated in the study of the acquisition of non-Indo-European languages that have more recently Küntay & Slobin – Putting Interaction Back Into Child Language 2 come to the foreground of the field (see Slobin 1985 for a lucid discussion of the rise of crosslinguistic acquisition studies). Thus, especially in fields such as Turkish child language, which is relatively recent and which involves a language with salient morphosyntactic properties, research on topics such as child-directed speech and discursive interaction remains to be developed. This paper will review indications from past research that studying input and adult-child discourse can facilitate the study of the acquisition of morphosyntax in the Turkish language. It will also provide some recent examples from Turkish child language on the relationship of child-directed speech to the early acquisition of morphosyntax, and on the pragmatic features of a certain kind of discourse form in child-directed speech. Lessons from Past Research The first important chapter in the area of Turkish child language development was that of Aksu-Koç and Slobin (1985), providing a summary of the empirical data available at the time and a theoretical discussion of important language-specific features that pose interesting acquisition problems for the native learner. Apart from several parental diary studies conducted in the seventies, the monograph summarized the experimental studies carried out by Aksu-Koç, Slobin, and their associates, supplemented by tape-recorded naturalistic data, covering the range of 2 to 5 years. Verhoeven (1991) and Küntay and Slobin (1999) provided more recent overviews of selected research in Turkish language acquisition, the latter focusing on first-language acquisition by monolingual children while the former included discussion of the large literature on the language of bilingual Turkish children in France, Germany, the Netherlands, Denmark, and Sweden. Aksu-Koç and Slobin (1985) have a section titled ‘Input and Adult-Child Interaction’ where they briefly discuss variable word order, types of inflections, discourse scaffolding, and politeness norms in adult-child discouse. They propose that the input offers ample opportunity for the child to read off pragmatic word order rules, the discovery of inflectional paradigms, “leading the child to linguistic structures from surface cues...” (p. 868). As a clear example of discourse scaffolding, the authors mention the acquisition of causal connectives, where they emerge first as a response to adult questioning and only later in self-constructed utterances (Aksu, 1975). Norms and rules about polite forms such as the usage of second person plural and lexical markers such as lütfen ‘please’ were also found to involve explicit modeling and reinforcement on the part of adult interactants of child language learners. These empirical observations aside, both at the beginning (p. 844) and at the end of their chapter (p. 876), Aksu-Koç and Slobin (1985) identify adult input as one of the crucial gaps in studying the acquisition of Turkish.