Cantonese reduplicative classifiers as quantifying determiner

Research output: Conference Papers (RGC: 31A, 31B, 32, 33)32_Refereed conference paper (no ISBN/ISSN)peer-review

View graph of relations

Author(s)

Detail(s)

Original languageEnglish
Publication statusPublished - Jul 2019

Conference

Title12th International Workshop on Theoretical East Asian Linguistics (TEAL-12)
LocationUniversity of Macau
PlaceChina
CityMacao
Period9 - 10 July 2019

Abstract

This paper studies prenominal reduplicative classifier in Cantonese. Two central claims have been proposed to account for the meaning of reduplicative classifiers: (i) with the obligatory presence of dou1 ‘all’/dou ‘all’, “CL + CL” is a distributive quantifier, considered to be on a par with English “every/each” or Mandarin mei ‘every’, as shown in (1) (see e.g. Cheng (2009), N. Zhang (2013) and Zhang & Tang (2013, 2018)); and (ii) it is a defective plural marker, which relies on the licensing of existential yi ‘one’ to give the plural reading of “many” or on the licensing of the distributive quantifier dou ‘all’ to give the distributive reading, as shown in (2), (see N. Zhang (2013)). (1) Go3-go3 hok6saang1 *(dou1) se2-zo2 bou3gou3. (Cantonese) CL-CL student all write-PERF report “All the students wrote a report.” (2) (a) Zhi-zhi mayi *(dou) kang-zhe dian shenme-dongxi. (Mandarin, cited from N. Zhang 2013) CL-CL ant all carry-IMP a-bit something “Every ant is carrying something.” (b) Yi-zhi-zhi mayi (*dou) pa-dao-le wo de beizi-li. (Mandarin, cited from N. Zhang 2013) one-CL-CL ant all climb-to-PERF my DE cup-in“Many ants climbed into my cup.” Moreover, besides the readings indicated above, it has been noted that prenominal “CL + CL” can convey a plural reading of “many” (see e.g. Steindl (2010)). (3) Tian-shang piao-zhe duo-duo yun. (Mandarin, cited from Steindl 2010) sky-on float-IMP CL-CL cloud “Many clouds are floating in the sky.” Questions related to the semantics of classifier reduplication and how the readings noted above are derived have not yet been fully answered, which include: (a) How can the distributive reading and the plural reading of “many”, be derived from prenominal “CL + CL” semantically? Are they attributed to two forms of “CL + CL”? (b) If distributivity is assumed to be coming from prenominal “CL + CL”, how can its co-occurrence with another distributive quantifier, Cantonese dou1 ‘all’/Mandarin dou ‘all’, be explained? This paper is to fill in the gap by providing a clearer semantics for classifier reduplication. The analysis I propose draws heavily on ideas introduced by Hoeksema (1983), Westerståhl (1984), Partee (2004) and others on quantifying determiners like “many” and “few”, and ideas introduced in the cover theory proposed by Schwarzschild (1996), Brisson (1998, 2003), and others for plurals. I will argue that prenominal reduplicative classifier, be it the distributive “CL + CL” or the “many” “CL + CL”, is a quantifying determiner, which is ambiguous between a quantifier type and a modifier type, with their semantics governed as follows: (4) (a) [CL + CL] --- reduplicative classifiers without dou1 Prenominal [CL + CL] serves as a quantifying determiner to convey the “many” reading, which is ambiguous between a modifier type and a quantifier type, with the former giving a weak cardinal reading and the latter a strong proportional reading. (b) [CL + CLdou1] --- reduplicative classifiers licensed by dou1 When prenominal [CL + CL] occurs with a distributive operator (D-operator) like dou1 ‘all’, the two form a licensing relation. As distributive quantification is performed by dou1, prenominal “CL + CL” serves as a modifier-type quantifying determiner. Its presence serves to regulate the quantification domain of dou1, with a maximizing effect imposed on the nominal modified by the prenominal “CL + CL”.(4) would account for the three readings of reduplicative classifiers: (i) [CL + CL] giving the strong proportional reading; (ii) [CL + CL] giving the weak cardinal reading; and (iii) [CL + CL] licensed by a distributive quantifier, viz. [CL + CLdou1] reading. When the prenominal reduplicative classifiers demonstrate the quantifier characteristics, it is a quantifying determiner of type <<e, t>, <<e, t>, t>>, which combines with a nominal argument of type <e, t>. This gives [[CL + CL] + NP] of type <<e, t>, t>, the type of a QP, which would be the case when strong proportional readings are conveyed by [CL + CL], and (4a) applies here. On the other hand, when the prenominal reduplicative classifiers demonstrate the modifier-like characteristics, it is a quantifying determiner of type <<e, t>, <<e, t>>, which as a type-preserving modifier, modifies a nominal argument of type <e, t>. This gives “[CL + CL] + NP” of type <e, t>, the type of a NP or DP, due to the ambiguity of “modifier-type” quantifying determiners between adjectives and determiners. This would be the case when the [CL + CL] gives a weak cardinal reading or when [CL + CL] is licensed by a distributive quantifier, viz. [CL + CLdou1] reading. The crucial difference between these two readings lies on whether a maximizing effect is imposed on the modified NP. If a maximizing effect is to impose on the modified NP, the reduplicative classifier has to be licensed by a distributive quantifier, and (4b) applies here. The maximizing effect comes from the reduplicative classifier, through its regulation of the context-dependent quantification domain of dou1 via covers (Cov). Without dou1 ‘all’, no Cov and no distributivity will be involved here. In such a case, [CL + CL] serves as a weak quantifier and gives a “many” reading to the NP, and (4a) applies here.If the above semantics of prenominal reduplicative classifier hold, the proposed analysis implies that Chinese may have determiners, which is at least true in Cantonese.

Citation Format(s)

Cantonese reduplicative classifiers as quantifying determiner. / Lee, Peppina.

2019. Paper presented at 12th International Workshop on Theoretical East Asian Linguistics (TEAL-12), Macao, China.

Research output: Conference Papers (RGC: 31A, 31B, 32, 33)32_Refereed conference paper (no ISBN/ISSN)peer-review