时 间：2018年9月20日（周四）9:30-11:30
　　地 点：语言所大会议室
　　主持人：唐正大（副研究员）

　　主题报告一：
　　英汉数量结构中可数性的儿童语言获得研究
　　Child Language Acquisition of Countability in Numeral Classifier Phrases: Evidence from English and Mandarin Chinese
　　报告人：苑晓鹤（博士后）

　　主题报告二：
　　事件个体化与动量词
　　Event Individuation and Verbal Classifiers
　　报告人：张小倩（博士后）

英汉数量结构中可数性的儿童语言获得研究
Child Language Acquisition of Countability in Numeral Classifier Phrases:
Evidence from English and Mandarin Chinese
苑晓鹤

　　可数与不可数（count & mass）和计数与计量（counting & measure）是人类重要的认知方式。本研究的研究对象是数量结构中存在的可数与不可数、计数与计量的对立。通过儿童语言习得的实验，研究儿童是否能够像成人一样区分两者，并进行英汉对比。本文主要采用两种实验方法——理解性实验和诱导产出法开展了6个实验，首先对成人测试，然后对儿童进行测试，通过SPSS和R回归分析两种统计方法进行统计，得出以下结论。

　　首先，本研究以Chierchia（1998）的可数与不可数的区分标准之一复数化（pluralization）为理论依据，通过英语的名词加复数标记的产出实验得出，英语儿童能够像成人一样区分可数与不可数（p=.58）；当给予儿童特定的量词时，他们和成人的表现趋于一致，完全能够区分可数与不可数（p=.08）。

　　其次，Cheng & Sybesma（1998）认为可数与不可数的区分存在于量词句法层面，并将量词划分为可数量词（count classifiers）和不可数量词（massifiers）。本研究不完全同意此观点，采用了句法区分和语义区分相结合的方式，借助于“个＋可数名词／不可数名词”的理解性实验，证明了汉语的儿童（4; 02-5; 02，M= 4;08）虽然不能像成人一样完全已经习得了可数与不可数的区分，但表现出和成人一样的倾向性。从R的回归分析中我们也可以得出，虽然儿童和成人的反应有显著差异（p <0.05），但只有少数的儿童和成人的差异比较大，大多数儿童和成人趋于一致。

　　再次，本研究基于Rothstein（2009，2010，2017）有关量词的两大功能——计数和计量的区分，对20名中国海洋大学的在校学生进行了在线问卷测试。两种加“的”的实验与两种不加“的”的实验比较显示两者没有显著差异（p=0.568；0.476），两种中性语境和食谱语境相比较得出其并没有显著差异（p=0.688；p=0.752），这部分支持了Rothstein（2009，2010，2017）和李旭平（2013）的观点。

　　最后，本研究以Rothstein（2009，2010，2017）的计数与计量的区分为理论依据，通过汉语中“容器量词＋个体名词（countable）／集合名词（mass）／个体名词（countless）”的三种情况，以“多少”进行提问的产出实验得出，汉语的儿童（3;08-4;08，M=4;02）和成人一样具有计数与计量的区分，并以counting为默认值，儿童和成人的表现趋于一致，接受度都达到了70%以上，且R的回归分析显示只有少数儿童偏离了成人。

　　关键词：可数与不可数 计数与计量 数量结构 儿童语言获得 R回归分析

　　The count vs. mass distinction and the count vs. measure distinction are two crucial human cognitive mechanisms. This work treats the acquisition of these distinctions in English and Chinese numeral classifier phrases. Using experimental stimuli, we compare the responses of adults and children to investigate whether children act adult-like in using these distinctions. Mainly using comprehension and elicited production tasks, and a statistical analysis that relies on SPSS and R regression analysis, we here report on the results of six experiments.

　　Firstly, using the pluralization criteria by Chierchia (1998) as our theoretical basis, we adopted an elicited production task to investigate whether participants add the plural marker s to the English nouns. Results show that English-speaking children (3;07-5;11, M=4;05) can distinguish count and mass concepts just as adults do (p=.58), moreover, when provided with classifiers, children perform almost as well as adults in distinguishing count from mass targets (p=.08).

　　Secondly, our findings disagree with Cheng & Sybesma (1998)’s claim regarding a mere syntactic distinction of count and mass, one that relies on the division of count classifiers and massifiers. Instead, we adopted a combination of syntactic and semantic standards to test count and mass distinction in Chinese using a comprehension task of “ge + individual nouns/collective nouns”. Regarding the acceptance ranking of items in five groups, rather, Chinese children (4;02-5;02, M=4;08) and adults in fact display the same tendency. Our R regression analysis also shows that relevant difference in our data arises from selective individuals behaviors.

　　Thirdly, we recruited 20 college students for the measure de experiments. This rested on the theoretical basis of distinguishing counting from measuring, as proposed by Rothstein (2009, 2010, 2017). We ran four experiments: experiments with/without de in neutral/recipe context, adopting one of the comprehension tasks: a preference task in the form of a covered test. Results show that the four experiments fail to result in a significant difference (p=0.568; p=0.476). Moreover, participants’ responses fail to reveal large differences in neutral (de) and recipe (de) contexts (p=0.688; p=0.752). Hence, some of our results support the proposals by Rothstein (2009, 2010, 2017) and Li(2013) but not all.

　　Lastly, for the form “container classifiers + count nouns (countable) / count nouns (countless) / mass noun (uncountable)”, we tested three conditions with an elicited production task, asking participants to answer duo-shao (much-little) items of some kind a given picture displays. We conclude that Chinese children (3;08-4;08, M=4;02) handle the count vs. measure distinction as adults do, where counting is the default value. As R regression analysis shows, the performance of children and adults tends to be alike.

Selected references:

　　Cheng, L. S. & Sybesma, R. 1998 Yi-wan tang, yi-ge tang: classifiers and massifiers. Tsing Hua Journal of Chinese Studies, New Series 28(3): 385–412.

　　Chierchia, G. 1998 Pluraity of mass nouns and the notion of ‘semantic parameter’. In S. Rothstein (ed.), Events and Grammar, 53–103.Dordrecht: Kluwer.

　　Li, X. 2013 Numeral Classifiers in Chinese: The Syntax-Semantics Interface. Berlin/NewYork: Mouton de Gruyter.

　　Rothstein, S. 2009 Individuating and measure readings of classifier constructions: Evidence from modern Hebrew. Brill Annual of Afroasiatic Languages and Linguistics 1 (1): 106–145.

　　Rothstein, S. 2010 Counting and the mass-count distinction. Journal of Semantics 27 (3): 343–397.

　　Rothstein, S. 2012 Measure readings of mandarin classifier phrases and the particle de. Language and linguistics 13(4): 693–741.

　　Keywords: count and mass; counting and measure; numeral classifier phrases; child language acquisition; R regression

事件个体化与动量词
Event Individuation and Verbal Classifiers
张小倩

　　本报告聚焦汉语动量词的功能，并依据不同标准将其进行分类。

　　首先，通过动量词修饰的事件单位将其分为两类。基于Cusic（1981）的观点，我们认为动量词既可作用于场合，也可作用于事件本身。这有别于Zhang（2017）的提议，即动量词修饰事件或是构成事件的单位。

　　其次，根据个体化事件的方式将事件量词分为两类。当事件量词修饰纯均质谓词时，他们可以选取谓词指称中任何事件实体来达到个体化事件的目的。相反，当事件量词修饰非纯均质谓词时，他们只能选取谓词指称的最小事件。此分析旨在区分事件量词的不同用法，这与Donazzan（2012）的提议不同，后者无法解释动量词与纯均质谓词搭配所触发的解读。

　　关键词：场合量词 事件量词 最小事件单位 时间非连续性

　　In this work, we focus on the functions of verbal classifiers in Mandarin Chinese and we aim to adopt different criteria to categorize them.

　　On the one hand, we divide verbal classifiers into two classes according to the event units that they modify. Based on Cusic (1981), we propose that verbal classifiers can apply to either occasions or events per se. This viewpoint is contrary to Zhang (2017) who assumes that verbal classifiers modify either events or “slices” of events.

　　On the other hand, we propose to classify verbal classifiers according to how they discretize events. When event-level classifiers modify strictly homogeneous predicates, they discretize events by selecting any event instance in the denotation of predicates. When they apply to non-strictly homogeneous predicates, they single out minimal event instantiations of predicates. Our analysis makes a distinction between the two different uses of event-level classifiers, which differs from the analysis in Donazzan (2012) that only accounts for the reading of minimal event instantiation.

Selected references:

　　Cusic, D. D. 1981 Verbal plurality and aspect. Ph.D dissertation, Stanford University.

　　Donazzan, M. 2012 On counting and measuring events. In E. Chemla, V. Homer and G. Winterstein(eds.) ， Proceedings of Sinn und Bedeutung, Volume 17, 219–236.

　　Zhang, N. N. 2017 The syntax of event-internal and event-external verbal classifiers. Studia Linguistica: 1–35.

　　Keywords: occasion-level classifiers; event-level classifiers; minimal event instantiation; temporal discontinuity