Volume 4, Issue s1 - Special Issue: The acquisition of information structure / Issue Editors: Christine Dimroth and Bhuvana Narasimhan

The paper focusses on the early production of aber - (‘but’-) clauses in German L1-acquisition. It aims to explain the delay in the realization of finite clause structure in aber -clauses reported in a previous study. Delayed realization of finite clause structure was also reported for clauses containing the additive particle auch (‘also/too’) by other researchers. It has been attributed to the information-structural properties of auch -clauses. Following this approach, we investigated the information-structural properties of the aber -clauses documented in the longitudinal data of five monolingual children (age range: 1;11–2;10) from the first documented aber -production to the emergence of finite clause structure in aber -clauses. Information-structural complexity is measured in terms of number and quality of the information-structural patterns of contrastive information in the aber -clause and the explicitness of the contrast realization. Analyses reveal a change in preference from topic-including contrasts to non-topic-including ones, that is, comment-located contrasts as well as a change from explicit to implicit contrast realization. We argue that the delayed realization of finite clause structure in aber -clauses is caused by an initial concentration on the complex content relations raised by the requirement of information-structural alignment of at least two different constituents as well as the anticipating of an implicit proposition.

This study investigated Dutch-speaking four- to five-year-olds’ use of word order and prosody in distinguishing focus types (broad focus, narrow focus, and contrastive narrow focus) via an interactive answer-reconstruction game. We have found an overall preference for the unmarked word order SVO and no evidence for the use of OVS to distinguish focus types. But the children used pitch and duration in the subject-nouns to distinguish focus types in SVO sentences. These findings show that Dutch-speaking four- to five-year-olds differ from their German- and Finnish-speaking peers, who show evidence of varying choice of word order to mark specific focus types, and use prosody to distinguish focus types in subject and object nouns in both SVO and OVS sentences. These comparisons suggest that typological differences in the relative importance between word order and prosody can lead to differences in children’s use of word order and prosody in unmarked and marked word orders. A more equal role of word order and prosody in the ambient language can stimulate more extensive use of prosody in the marked word order, whereas a more limited role of word order can restrict the use of prosody in the unmarked word order.

The present study investigated pronoun resolution strategies in French native speakers and in German-speaking learners of French. French and German differ in antecedent preferences in ambiguous constructions such as The postman hit the pirate before he went home : while French shows a N2-preference, German shows a N1-preference. This difference is explained by effects of exposure to an unambiguous alternative construction referring to N1 that exists in French, but not in German (Hemforth et al. 2010, Language specific preferences in anaphor resolution: Exposure or gricean maxims? In Steallan Ohlsson and Richard Catarambone (eds.), 32nd Annual Conference of the Cognitive Science Society , 2218–2223. Austin, TX: Cognitive Science Society). In questionnaires, we tested French active and passive sentences to investigate (a) whether the L2-learners would apply the same strategy as natives in the active condition (i.e. N2-preference), and (b) whether the explicit topicalization of a referent as a consequence of passivization influences interpretation preferences in both groups (resulting in more N1 choices). The results show that German learners prefer the N1 more often than the French natives in the active condition. Crucially, the number of N1 choices increased in both groups in the passive condition. These results suggest that L2-learners might have difficulties acquiring strategies based on the frequency and availability of alternative constructions in the L2, and provide further evidence for the importance of information-structure-based strategies in L1 and L2 pronoun resolution.

Adult speakers typically order referents that have been previously mentioned in the discourse (“old” referents) before newly introduced referents (“new” referents). But 3–5-year-olds acquiring German exhibit a “new-old” preference in a task involving question-answer sequences (Narasimhan, Bhuvana and Christine Dimroth. 2008. Word order and information status in child language. Cognition 107. 317–329). Here we ask whether we can change 4–5-year-olds’ new-old preference by manipulating the context in order to encourage connected discourse. Findings show that discourse context changes children’s new-old preference. Children produce the new-old order in fluent utterances and the old-new order in non-fluent utterances. Adult controls overwhelmingly prefer the old-new order, even more so when the weight (number of syllables) of the old referent label is greater than that of the new referent label. Our study demonstrates that although cognitive and communicative biases may influence children’s ordering patterns in non-adult-like ways, such patterns are not categorical, but are flexibly influenced by factors such as discourse context.

The Given-before-New Principle holds in adult speech: Given information tends to precede new information. For instance, in the English dative alternation, the given-theme – i.e., the direct object [DO] – tends to precede the new-recipient – i.e., the indirect object [IO] – in the prepositional dative (e.g., John gave the books to some children ), while the given-recipient tends to precede the new-theme in the double object dative (e.g., John gave the children some books ). Likewise, in Korean datives, the given-recipient tends to occur earlier in the canonical [IO–DO] order, while the given-theme tends to occur earlier in the scrambled [DO–IO] order. This study investigates whether L1-English adult L2ers of Korean, who have knowledge of the Given-before-New Principle in their L1, automatically adhere to it in their interlanguage. L2ers’ choices between canonical and scrambled dative orders were tested using novel oral contextualized preference tasks. The native speakers of Korean overwhelmingly complied with the Given-before-New Principle. However, the intermediate-to-advanced L2ers exhibited a strong bias for the (default) canonical [IO–DO] order, which apparently overrode the Given-before-New Principle. The findings of analyses by group and by individual are discussed in terms of frequency, syntactic complexity, processing, and null arguments.

Comprehension of sentences in the non-canonical word order usually poses problems for preschoolers (e.g., Slobin, Dan I. and Thomas G. Bever. 1982. Children use canonical sentence schema: A crosslinguistic study of word order and inflections. Cognition 12. 229–265). These problems may be modulated by information structure, such as the presence of an appropriate context licensing the non-canonical word order and the type of referring expression. We examined the impact of the given-new order, induced by a context sentence, and the type of referring expression realizing the given referent (NP vs. pronoun) on the comprehension of SVO and OVS sentences in monolingual Russian-speaking 4- to 5-year-olds and adults. Children and adults showed high comprehension accuracy for SVO and OVS sentences, with accuracy rates above 80 % for OVS sentences. Context and the type of referring expression had no effect. Compared to a similar experiment conducted in German, Russian-speaking children outperformed their German-speaking peers. This difference may result from the earlier acquisition of the case system and a stronger given-before-new preference in Russian compared to German. Our data suggest that as soon as children rely more on morphological information during processing and employ adult-like processing strategies, their offline comprehension performance depends less on contextual and information structure factors.