Previous Studies on Specific Languages

國

立 政 治 大 學

‧

N a tio na

l C h engchi U ni ve rs it y

(6) a. Plural: three books [3 book*1]

b. Classifier: 三 本 書 [3 *1 書]

san ben shu three CL book

‘three books’ (Her 2012a: 1674)

Although Her (2012a) agreed with Borer’s (2005) statement that classifiers and plurals are the same category, he proposed that it is possible for classifiers and plurals to co-occur in a noun phrase.

(7) a. Chinese:三 個 學生 們 san ge xuesheng-men 3 CL student PL

‘three students’

b. Japanese: san-nin-no gakusei-tati 3 CL-NO student -TATI

‘three students’ (Her 2012a:1684)

The two examples are well-formed and widely used by native speakers, but they violate the strict CPCD principle. So, are the examples exceptions? Or, is the CPCD principle incorrect? We will take a closer look at this issue in Section 4.1.

2.3 Previous Studies on Specific Languages

In this paper, there are 22 languages which deserve a detailed examination in the relationship between their classifiers and plurals. Some of the languages have been surveyed by several linguists, but they have not reached a consensus. This section

‧ 國

立 政 治 大 學

‧

N a tio na

l C h engchi U ni ve rs it y

presents some viewpoints on Chinese in Section 2.2.1, and on Japanese in Section 2.2.2.

2.3.1 Chinese

In Chinese, –men is considered to be either a plural marker (Li and Thompson 1981, Li 1999, Huang 2009) or a collective marker (Lu Shuxiang1947, Chao1968,

Norman1988, Iljic1994, Cheng and Sybesma1999). If –men is a collective marker rather than a plural marker, Chinese is not a counter example to Borer’s generalization.

If –men is a plural marker, it is worthy finding out the degree to which it violates the generalization.

Li (1999) suggested that plural markers and classifiers are different heads which project NumP and ClP, respectively. She proposed that -men and -s in English both generate under Number. Unlike -s as realized in nouns, the application of -men as

realized in determiners is more limited Therefore, san ge xuesheng men ‘three students’ is ungrammatical because the classifier ge blocks the head movement of the

noun (N) xuesheng to D position.

(8) a. 三個學生 b. *三個學生們

san ge xuesheng san ge xuesheng-men 3 CL student 3 CL student PL

‘three students’ ‘three students’

‧ 國

立 政 治 大 學

‧

N a tio na

l C h engchi U ni ve rs it y

(Li 1999: 87) (Li 1999: 87)

Huang (2009) further supported Li’s claim by the cases of pronouns and proper

names which are generated in D.

(9) a. 我 對 他們 三 個 人 特別 好 wo dui ta-men san ge ren tebie hao I to he-PL 3 CL people especially good

‘I am especially nice to them three’

b. 我 對 小強們 三 個 人 特別 好 wo dui XiaoQiang -men san ge ren tebie hao I to XiaoQiang -PL 3 CL people especially good ‘I am especially nice to XiaoQiang them three persons’

(Huang and Li 2009: 313)

Because -men should attach to a pronoun and a proper name, -men must be realized in the D position.

From Huang and Li’s (2009) perspective, Borer’s (2005) generalization is incorrect. Classifiers and plurals are in complementary distribution because of the head movement constraint rather than being the same category.

‧ 國

立 政 治 大 學

‧

N a tio na

l C h engchi U ni ve rs it y

Unlike Huang and Li (2009), Her (2012a) indicated that san ge xuesheng men

‘three students’ is grammatical which is supported by real data from Google search

engine. Thus, Chinese violates not only the strict CPCD principle, but also the statement of Huang and Li (2009).

2.3.2 Japanese

In Japanese, there are two kinds of word order in noun phrases with classifiers as following:

(10) a. gakusei san-nin -ga kita student 3-CL -NOM came

‘Three students came.’

b. san-nin gakusei -ga kita 3-CL student -NOM came

‘Three students came.’ (Yasuo Ishii 2000: 2)

When the noun is pluralized by -tati, gakusei-tati san-nin -ga kita ‘Three students came’ is grammatical in Japanese. Ishii (2000) suggested that classifiers and

number plurals are different heads. And the plural marker -tati in Japanese is a phrasal affix attaching to NP which reveals in DP spec via feature checking. In addition, the assertion of Ishii (2000) convinced another linguist, Kurafuji, who had once thought that the co-occurrrence of classifiers and plurals is unacceptable in Japanese.

Although we are not sure about the relationship between classifiers and plurals, we do

‧ 國

立 政 治 大 學

‧

N a tio na

l C h engchi U ni ve rs it y

know that classifiers and plurals can co-occur in a noun phrase in Japanese.

Downing (1996) adopted Greenberg’s concept in which the singulative/collective

system often exists in classifier languages and the singular/plural system in non-classifier languages. Furthermore, Downing (1996) suggested that the singulative/collective and singular/plural systems were combined in Japanese, so that -tati can refer to either a plural meaning in common nouns or a collective meaning with proper names. However, the different word order of the nominal construction will influence the acceptability when the nouns are attached by -tati, as following.

(11) a. Taro-tati san-nin Taro-PL 3-CL

‘Taro and his friends’

b. *san-nin Taro tati 3-CL Taro-PL

‘Taro and his friends’

c. gakusei-tati san-nin student-PL 3-CL

‘three students’

d. san-nin gakusei-tati 3-CL student-PL

‘three students’ (Downing 1996 )

Taro tati means Taro and his friends which is a collective usage. And proper names must be realized in D, so ‘san-nin Taro tati’ is ungrammatical. gakusei-tati on the other hand is a plural reading and realized in N, so both gakusei-tati san-nin and

‧ 國

立 政 治 大 學

‧

N a tio na

l C h engchi U ni ve rs it y

san-nin gakusei-tati are acceptable. Although –tati has both plural and collective usage, Downing (1996) proposed that this kind of mixed system is in fact not a stable one. So, Japanese might become a language with either a singulative/collective system or a singular/plural system in the future.