Cross-linguistic research on the development of tense-aspect marking has revealed a strong effect of lexical aspect. But the degree of this effect varies across languages. Explanation for this universal tendency and language-specific variation is still an open issue. This study investigates the early emergence and subsequent development of four grammatical aspect markers in Mandarin, -le(perfective),zai(progressive), -zhe(durative), and -guo(experiential), in the longitudinal speech data of four children acquiring Mandarin Chinese. It was found that the emergence of grammatical aspect marking generally follows the predictions of the aspect hypothesis, perfective -lepredominantly appearing with telic verbs, whereas progressivezaiis almost exclusively used with activity verbs. However, the typological features of Mandarin also affect the early uses of -le, which is used with stative predicates more frequently than predicted, and the input patterns play an important role in children's aspect marking. The results support a usage-based learning process in accord with a language-specific system of aspectual semantics, rather than a strong universal association of grammatical aspect and lexical aspect.
1 January 2010 in Journal of Cognition and Culture
In two eye-tracking studies we investigated the influence of Mandarin numeral classifiers – a grammatical category in the language – on online overt attention. Mandarin speakers were presented with simple sentences through headphones while their eye-movements to objects presented on a computer screen were monitored. The crucial question is what participants look at while listening to a pre-specified target noun. If classifier categories influence Mandarin speakers' general conceptual processing, then on hearing the target noun they should look at objects that are members of the same classifier category – even when the classifier is not explicitly present (cf., Huettig and Altmann, 2005). The data show that when participants heard a classifier (e.g., ba3, Experiment 1) they shifted overt attention significantly more to classifiermatch objects (e.g., chair) than to distractor objects, but when the classifier was not explicitly presented in speech, overt attention to classifier-match objects and distractor objects did not differ (Experiment 2). This suggests that although classifier distinctions do influence eye-gaze behavior, they do so only during linguistic processing of that distinction and not in moment-to-moment general conceptual processing.
This study investigates the developmental trajectory of relative clauses (RCs) in Mandarin-learning children's speech. We analyze the spontaneous production of RCs by four monolingual Mandarin-learning children (0;11 to 3;5) and their input from a longitudinal naturalistic speech corpus (Min, 1994). The results reveal that in terms of the syntactic role of the head noun in the matrix clause, isolated noun phrase RCs dominate, followed by those that modify the subject or object of the matrix clauses and predicate nominal relatives. This pattern differs from those observed in English (Diessel & Tomasello, 2000), German (Brandt, Diessel & Tomasello, 2008), and Japanese (Ozeki & Shirai, 2007). Regarding the syntactic role of the head noun inside the RC (i.e. subject, object, or oblique relatives), the early RCs are dominated by object relatives. This pattern also differs from those observed in English and Japanese. We propose a multifactorial usage-based learning account for the developmental patterns.
Mandarin categorizes cutting and breaking events on the basis of fine semantic distinctions in the causal action and the caused result. I demonstrate the semantics of Mandarin C&B verbs from the perspective of event encoding and categorization as well as argument structure alternations. Three semantically different types of predicates can be identified: verbs denoting the C&B action subevent, verbs encoding the C&B result subevent, and resultative verb compounds (RVC) that encode both the action and the result subevents. The first verb of an RVC is basically dyadic, whereas the second is monadic. RVCs as a whole are also basically dyadic, and do not undergo detransitivization
Two-way typological patterning between satellite- and verb-framed languages in construal of Manner of motion is well attested in speech (e.g. Slobin 2006) and gesture (e.g. McNeill 2001), but contradictory findings exist regarding a third category of equipollently-framed languages (Slobin 2004b). This study examined elicited descriptions of motion from 14 native speakers of Mandarin-Chinese (equipollently-framed), 13 native speakers of English (satellite-framed), and 16 native speakers of Japanese (verb-framed). Results showed that Mandarin and English speakers encoded Manner in speech significantly more than Japanese speakers, and rarely added Manner to gesture when it was absent from speech. However, Mandarin speakers differed from Japanese but also English speakers in gestural highlighting of Manner, where they frequently encoded Manner in speech but not in accompanying gestures, focusing instead on other aspects of the event, specifically Path. These results support a partial three-way typological distinction in construal of motion.