On the Semantics of henduo ‘many’ and henshao ‘few’ in Mandarin Chinese

     Partee (2004) claims that English many and few are ambiguous between the cardinal reading and the proportional reading, differing in their morpho-syntactic roles as an adjective or a determiner. Many and few serve as adjectives and determiners under the cardinal readings and only as determiners under the proportional readings. Although many studies have been conducted on quantifying words in different languages (see e.g. Milsark 1977; Westerståhl 1984; Cohen 2001; Krasikova 2011), not much has been done on Mandarin.
     Many and few are generally translated as henduo ‘many’ and henshao ‘few’ in Mandarin, and this paper aims to study the semantics of henduo and henshao, in relation to their morpho-syntactic roles. Data have been collected from the CCL corpus, with targeted investigation conducted through experiments. Preliminary results reveal the following observations.
     Firstly, while henduo primarily functions as an attribute, henshao is found to serve dominantly as an adverbial, with the two demonstrating asymmetrical syntactic distributions. Subject and object asymmetry exists in henshao but not in henduo. For cases where henshao serves seemingly as an attribute, there is a covert existential marker you ‘have/exist’ converting henshao from an attribute to an adverbial in the subject position as in (1a), while not in the object position as in (1b). Contrarily, it is more natural for henduo to function as an attribute, with less obvious subject and object asymmetry.

(1) a. Henshao (you) ren jiang zhege gushi.
         few (have) people tell this story
         ‘Few people tell this story.’
     b. Zhege gushi sheji henshao (*you) ren.
         this story involve few (*have) people
         ‘This story involves few people.’

     Secondly, it is found that henduo gives priority to the weak cardinal reading (54.57%) over the proportional reading (22.29%) in the subject position. The more salient preference is found in the object position, with cardinal reading of 57.96%. This manifests that regardless of whether it is in the subject or the object position, henduo is adjective-like, namely a modifier in sentences. On the other hand, for henshao, a covert existential marker you ‘have/exist’ exists to convert henshao from an attribute to an adverbial in the subject position, and it is also inclined to the weak cardinal reading (45.72%) in the limited cases of being an attribute in the object position. The figures suggest that henduo and henshao give priority to the weak cardinal reading in both subject and object positions.
     Thirdly, topic leads to the proportional readings of henduo (46.28%) and henshao (59.14%), which manifests that henduo and henshao are quantifier-like in the topic position.
     Generalizing, these findings preliminarily reveal at least the following two points.
(a) Cardinal readings tend to be the dominant readings in henduo and henshao, regardless of whether the two occur in subject or object position, while the limited preference of proportional readings occur in the topic position. This shows that both henduo and henshao are weak quantifiers by nature. Proportional reading is only possible when they are licensed by the topic feature in the [Head, TopicP] position, which is a position assimilating QP.
(b) English many and few are lexically ambiguous between the cardinal reading and the proportional reading, which leads to their difference in morpho-syntactic roles as an adjective or a determiner. Chinese henduo and henshao count on syntactic movement from subject position to topic position, with the topic feature serving as the licencing condition for the proportional reading.