ERGATIVITY AND UNACCUSATIVITY

(to appear in the Encyclopedia of Chinese Language and Linguistics; please see my web

site for publication details)

Edith Aldridge, University of Washington

The main goal of this lemma is to summarize how the term ‘ergativity’ has been used in

analyzing aspects of Chinese syntax. In the typological literature, ‘ergativity’ typically

refers to the patterning of transitive objects with intransitive subjects for the purposes of

case-marking and certain syntactic operations like relative clause formation. The term

‘ergativity’ is also sometimes used to refer to an alternation between the transitive and

intransitive use of a verb in which the argument functioning as the direct object in the

transitive variant is the subject when the verb is used intransitively. It is in this latter sense

that ‘ergativity’ can be observed in Chinese.

1. Ergativity

An ergative-absolutive case-marking system is distinguished from a nominative-accusative

one in that the case of the subject in an intransitive clause receives the same marking as the

object in a transitive clause. This is referred to as ‘absolutive’ case. The subject in a

transitive clause receives a different case, which is termed ‘ergative’. In the Pama-Nyugan

language Dyirbal, absolutive case is phonologically null; the ergative case is realized as the

suffix -nggu.

Dyirbal (Dixon 1994:161)

(1) a. yabu banaga-nyu

mother.ABS return-NONFUT

‘Mother returned.’

b. nguma yabu-nggu bura-n

father.ABS mother-ERG see-NONFUT

‘Mother saw father.’

This pattern contrasts with an accusative language like English, in which transitive and

intransitive subjects are marked alike, while the object in a transitive clause is treated

differently. Case in English is indicated by position in the clause: subjects appear in

preverbal position, while objects in transitive clauses follow the verb.

(2) a. Mary saw John.

b. Mary came.

The accusative pattern is confirmed by examining the personal pronouns, which show a

morphological distinction between nominative and non-nominative case. Only the

nominative form of the pronoun is grammatical in the preverbal subject position, while the

non-nominative must be used in post-verbal object position.

(3) a. She/*her saw him/*he.

b. She/*her came.

In addition to the case marking system, some languages also exhibit ‘syntactic ergativity’,

whereby transitive objects and intransitive subjects are eligible to undergo certain syntactic

operations which transitive subjects are not. For example, in many Inuit, Mayan, and

Austronesian languages – as well as the Pama-Nyugan language Dyirbal – a relative clause

can be formed only on the absolutive argument. In the following Dyirbal examples, the gap

position in the relative clause in (4a) is the direct object; in (4b), it is the subject of an

intransitive clause. A relative clause cannot be formed directly on the subject of a transitive

clause. In order to extract a transitive subject, the embedded verb must have the antipassive

suffix –nga. An antipassive is semantically transitive in that the verb takes two arguments.

However, it is formally intransitive, since the object is not marked absolutive but rather has

oblique case. In Dyirbal this case is dative. The external argument is treated as an

intransitive subject and consequently can be extracted.

Dyirbal

(4) a. palan jukumpil [ ___ ngaja purangu] nyinanyu

there. ABS woman. ABS 1S.NOM see. REL.ABS sit-NONFUT

‘The woman whom I am watching is sitting down.’ (Levin 1983:282)

b. ngumai [ banaga-ngu] yabu-nggu bura-n

father.ABS return-REL.ABS mother-ERG see-NONFUT

‘Mother saw father, who was returning.’ (Dixon 1994:169)

c. yabui [ bural-nga-ngu nguma-gu] banaga-nyu

mother.ABS see-AP-REL.ABS father-DAT return-NONFUT

‘Mother, who saw father, was returning.’ (Dixon 1994:170)

Throughout their attested history, Sinitic languages have exhibited accusative alignment

with respect to both morphology and syntax. Like English, Chinese varieties generally

mark case positionally, placing subjects in clause-initial position and objects after the verb,

as seen in the following Standard Mandarin examples.

(5) a. 李四買了那本書。

Lǐsì mǎi le nà běn shū.

Lisi buy ASP DEM CL book

‘Lisi bought that book.’

b. 李四來了。

Lǐsì lái le.

Lisi come ASP

‘Lisi came.’

There is also no asymmetry between transitive subjects and objects in relative clause

formation. Relative clauses can be formed on either subject or object position.

(6) a. 買了那本書的人

[ mǎi le nà běn shū ] de rén

buy ASP DEM CL book DE person

‘person who bought that book’

b. 李四買的書

[Lǐsì mǎi ] de shū

Lisi buy DE book

‘book that Lisi bought’

In sum, Standard Mandarin, as well as the other Chinese varieties, is an accusative

language. However, the term ‘ergative’ has been applied in some research in Chinese

syntax to refer to one class of intransitive verbs, which I discuss in the next section.

2. Unaccusativity and ‘ergativity’ in Chinese

Since Perlmutter (1978), it is well known that intransitive verbs come in two varieties:

unergative verbs, which take an external argument like an agent as their sole argument; and

unaccusative verbs, which only take an internal argument like a theme. The term

‘unaccusative’ refers to the lack of canonical accusative licensing for the semantic object,

while ‘unergative’ refers to the lack of special licensing for the semantic subject. In most

languages, the single argument of any type of intransitive verb is marked uniformly as

either nominative (in an accusative language) or absolutive (in an ergative language).

Unaccusative constructions are treated as ergative by some linguists, because of the

mapping from argument structure to grammatical relations. Specifically, the grammatical

subject in these intransitive constructions has the thematic status of a direct object.

It is in this sense that Cikoski (1978) analyzes unaccusative verbs as ‘ergative’ in

Classical Chinese (approximately 500-200 BCE). The ‘ergative’ verbs in Classical Chinese

participate in alternations like that illustrated in (7). The intransitive variant in (7a) merely

asserts the proposition that the subject has escaped from death. The transitive variant in

(7b) is a causative construction which includes an agent (causer) responsible for the

servant’s being saved. Cikoski treats this alternation as a type of ergativity, because the

object in the transitive clause has the same semantic relationship to the verb in (7b) as the

subject does in the intransitive clause in (7a). Both are internal arguments, specifically

themes.

(7) a. 臣免於死。 (Zuǒzhuàn, Wén 左傳:文 10/3; Cikoski 1978:131)

Chén miǎn yú sǐ.

servant escape from die

‘Your servant has escaped from death.’

b. 君 … 免臣於死。 (Zuǒzhuàn, Chéng 左傳： 成 17/13; Cikoski 1978:130)

Jūn miǎn chén yú sǐ.

lord cause.to.escape servant from die

‘Your Lordship has … saved your servant from death.’

Unergative verbs are termed ‘neutral’ verbs in Cikoski’s analysis. These can also often be

either transitively or intransitively. In this alternation, the argument retained in the

intransitive clause is the semantic subject, i.e. the external argument.

(8) a. 王 … 避風雨。 (Zuǒzhuàn, Xī 左傳:僖 32/5; Cikoski 1978:131)

Wáng bì fēng yǔ.

king avoid wind rain

‘The king … retreated from the storm.’

b. 王 … 避。 (Zuǒzhuàn, Zhāo 左傳 昭公 12/9; Cikoski 1978:132)

Wáng bì.

king avoid

‘The king … retreated.’

Zhōu Fǎgāo 周法高 (1961), Yáng Bójùn 杨伯峻 and Hé Lèshì 何乐士(1992), Pulleyblank

(1995), and others characterize the alternation in (7) as an active/passive pair. This analysis

is ultimately credited to Mǎ Jiànzhōng 馬建忠 (1898), who classifies passives as

constructions in which the recipient of the action (shòuzhě 受者) functions as the subject.

Wáng Lì 王力 (2004:484) distinguishes this alternation from morphologically marked

passives by referring to it as ‘conceptual passive’ (gàiniàn bèidòngshì 概念被動式). Lǚ

Shūxiāng 呂叔湘 (1992:57) rejects the passive analysis and treats the unaccusative variant

as a stative construction (biǎotàijù 表態句), pointing out that the completion of an action

results in a state. More recently, Ōnishi 大西 (2004) and Wū 巫 (2008) adopt the

approach taken by Cikoski (1978), labeling the unaccusative verb class as ‘ergative’ (zuògé

作格).

Lǚ Shūxiāng呂叔湘 (1987) observes that the same two verb classes are found in

Modern Standard Mandarin. In the unergative class (his ‘pattern one’ dìyī géjú 第一格局),

the single argument in the intransitive variant corresponds to the subject in the transitive

version.

(9) a. 中國隊勝南朝鮮隊。 (Lǚ Shūxiāng呂叔湘 1987:1)

Zhōngguó duì shèng nán cháoxiān duì.

China team win south Korea team

‘The Chinese team defeats the South Korean team.’

b. 中國隊勝。 (Lǚ Shūxiāng呂叔湘 1987:1)

Zhōngguó duì shèng.

China team win

‘The Chinese team wins.’

In contrast, ‘pattern two’ (dìèr géjú 第二格局) verbs are unaccusative. The single argument

in the intransitive clause corresponds to the object in the transitive version.

(10) a. 中國隊敗南朝鮮隊。 (Lǚ Shūxiāng呂叔湘 1987:1)

Zhōngguó duì bài nán cháoxiān duì.

China team lose south Korea team

‘The Chinese team defeats the South Korean team.’

b. 南朝鮮隊敗。 (Lǚ Shūxiāng呂叔湘 1987:1)

nán cháoxiān duì bài.

south Korea team lose

‘The South Korean team loses.’

Lǚ considers whether the existence of the unaccusative/transitive alternation classifies

Chinese as an ergative language but quickly dismisses this possibility. He points out first

that Chinese lacks the morphological case alignment found in languages like Dyirbal, as

shown in (1) above. He further notes that only a subclass of Chinese verbs (specifically, the

unaccusatives) participate in the alternation in which a semantic object is the surface

subject.

Huáng Zhèngdé 黃正德 (1989) seconds the conclusion that Chinese is not an ergative

language, pointing out that the ‘ergative’ (i.e. unaccusative) type of verb is found in

accusative languages as well. As demonstrated by Burzio (1986), Italian unaccusative verbs

(which he terms ‘ergative’) behave very much like those in Standard Mandarin. The direct

object of the transitive verb in (11a) surfaces as the subject when the same verb is used

intransitively, as in (11b).

Italian (Burzio 1986:25)

(11) a. L’artiglieria affondò due navi nemiche.

the.artillery sank two ship enemy

The artillery sank two enemy ships.’

b. Due navi nemiche affondarono.

two ship enemy sank

‘Two enemy ships sank.’

A vast body of literature has explored the unaccusative/transitive alternation in Standard

Mandarin. Among those who do not use the term ‘ergative’ are Sybesma (1992, 1999),

Cheng and Huang (1994), Yu (1995), Gù Yáng 顾阳 (1996), Yuan (1999), Xú Jié 徐杰

(1999, 2001), Yáng Sùyīng 杨素英 (1999), Hán Jǐngquán 韩景泉 (2001), Tāng Tíngchí 湯

廷池 (2002), and Pān Hǎihuá 潘海华 and Jǐngquán 韩景泉 (2005). Here, I briefly

introduce some of the approaches which use the term ‘ergative’ to refer to unaccusative

verbs.

Y.-H. Li (1990) and Zhou (1990) examine existential and presentational verbs in

Standard Mandarin. These verbs are unaccusative (‘ergative’) and hence take an internal

argument as their subject. Interestingly, this argument can surface in its post-verbal base

position when it is indefinite, as in (12a). The subject may also appear in pre-verbal

position, but in this case it is interpreted as definite, as in (12b).

(12) a. 來了客人。

Lái le kèrén.

come ASP guest

‘There came some guests.’

b. 客人來了。

Kèrén lái le.

guest come ASP

‘The guests came.’ (Y.-H. Li 1990:137)

In contrast, only preverbal position is possible for the subject of an unergative verb. This is

because the sole argument of an unergative verb is an external argument, which never

occupies a VP-internal position during the course of the derivation.

(13) a. 客人哭了。

Kèrén kū le.

guest cry ASP

‘The guests cried.’

b. *哭了客人。

*Kū le kèrén.

cry ASP guest

‘There cried some guests.’ (Y.-H. Li 1990:137)

Another type of unaccusative construction which has been labeled ‘ergative’ in the

literature is the type which alternates with causative constructions, as in the following pair

of examples from Cheng and Huang (1994:188). As a theme internal argument, the subject

in the intransitive clause in (14a) merely undergoes the action expressed by the predicate.

In the transitive (causative) variant in (14b), this participant (now the direct object) is made

to undergo this action by the subject. This alternation is parallel to that observed in (7)

above for Classical Chinese.

(14) a. 張三嚇了一跳。

Zhāngsān xià-le yī tiào.

Zhangsan shock-ASP one jump

‘Zhangsan was taken by surprise.’

b. 李四嚇了張三一跳。

Lǐsì xià-le Zhāngsān yī tiào.

Lisi shock-ASP Zhangsan one jump

‘Lisi surprised Zhangsan.’

Dèng Sīyǐng 邓思颖 (2004) discusses an asymmetry between unaccusative and unergative

verbs in the ability to undergo passivization, which he terms ‘ergativization’ (zuògéhuà 作

格化). Specifically, ergativization derives a passive (unaccusative) verb from a transitive

verb and promotes the thematic object to subject status. The internal argument must then

occupy preverbal subject position.

(15) a. *被殺了張三。

Bèi shā le Zhāngsān.

PASS kill ASP Zhangsan

‘Zhangsan was killed.’

b. 張三被殺了。

Zhāngsān bèi shā le.

Zhangsan PASS kill ASP

‘Zhangsan was killed.’

Contrary to this, unergative verbs cannot be passivized, because there is no internal

argument in their argument structure to be promoted.

(16) *張三被哭了。

Zhāngsān bèi kū le.

Zhangsan PASS cry ASP

‘*Zhangsan was cried.’

Shěn Yáng 沈阳 and Rint Sybesma (2010) analyze what they refer to as ‘middle

constructions’ (zhōngdòng jiégòu 中动结构) employing the auxiliary gěi 给. They show

that the main verb in this construction must be unaccusative fēibīngé 非宾格 (which they

also term ‘ergative’ zuògé 作格). (17a) shows that gěi can precede an unaccusative verb but

not an unergative verb (17b).

(17) a. (犯人)给跑了.

Fànrén gěi pǎo le.

criminal GEI run ASP

‘(The criminal) was allowed to escape.’

b. *(她)给哭了.

Tā gěi kū le.

she GEI cry ASP

‘She was made to cry.’ (Shěn and Sybesma 2010:226)

Interestingly, they show that there is a difference between middle constructions and pure

unaccusative/ergative constructions in that the former has a causative sense that the latter

does not. Specifically, the middle construction has an implicit agent, while the

unaccusative/ergative construction is a simple intransitive.

(18) a. 米饭煮糊了.

Mǐfàn zhǔ hú le.

rice cook burn ASP

‘The rice burned.’

b. 米饭给煮糊了.

Mǐfàn gěi zhǔ hú le.

rice GEI cook burn ASP

‘The rice was burned (by someone).’ (Shěn and Sybesma 2010:231)

To summarize the discussion in this section, the term ‘ergative’ is often employed to refer

to unaccusative verbs in Classical Chinese and Modern Standard Mandarin. I have also

pointed out, however, that having an unaccusative verb class does not suffice to classify a

language as ergative. The next section considers an attempt in the literature to claim that

Standard Mandarin has ergative characteristics beyond the existence of an unaccusative

verb class.

3. Ergativity beyond unaccusativity?

Building on earlier work by Zēng Lìyīng 曾立英 and Yáng Xiǎowèi 杨小卫 (2005), Zēng

Lìyīng 曾立英 (2009) proposes that the surface positioning of subjects in unaccusative

constructions in Standard Mandarin provides evidence for split-ergativity in the language.

Specifically, she claims that case-marking in Standard Mandarin is sensitive to the animacy

hierarchy first proposed by Silverstein (1976) and expanded on by Dixon (1994).

According to this hierarchy, pronouns and NPs referring to humans are more likely to be

marked according to an accusative pattern, while common nouns, especially those referring

to non-humans, are more compatible with an ergative pattern.

(19) Nominal Hierarchy (Dixon 1994:85)

1st/2nd person pronoun 3rd person pronoun Proper noun Common noun

Nom/Acc marking ========================= Erg/Abs marking

This type of ‘NP split-ergativity’ is found in a number of ergative languages, including

Pama-Nyungan languages like Dyirbal. As shown above in (1), 3rd person NPs are marked

with ergative and absolutive case in Dyirbal. But marking on 1st and 2nd person pronouns

follows a nominative/accusative pattern. In (20), both intransitive and transitive subjects

appear without an overt case-marker, while the transitive object in (20c) takes the

accusative suffix -na. Morphologically, the null nominative marker is identical to the

absolutive marker shown in (1). The difference between (1) and (20) is the distribution of

this form, marking absolutives in (1) and subjects in (20).

Dyirbal (Dixon 1994:161) (20) a. ŋana banaga-nyu

we.NOM return- NONFUT

‘We returned.’

b. nyurra banaga-nyu

you.PL.NOM return- NONFUT

‘You all returned.’

c. ŋana nyurra-na bura-n

we.NOM you.PL-ACC see- NONFUT

‘We saw you all.’

Zēng (2009) claims that case-marking in Standard Mandarin is subject to a similar animacy

hierarchy. As discussed previously for (12), the subject of an unaccusative verb can surface

in post-verbal position when it is indefinite. Zēng assumes that the post-verbal NP in an

unaccusative clause like (21b) functions as an object in the syntax. Since the intransitive

‘subject’ is treated as an object, the clause can be said to display an ergative pattern.

(21) a. 電報來了。

Diànbào lái le.

telegram come ASP

‘The telegram arrived.’

b. 來電報了。

Lái diànbào le.

come telegram ASP

‘A telegram arrived.’ (Zēng 2009:319)

In contrast to the alternation in (21), pronouns and personal names are restricted to

preverbal position when selected as the single argument of an intransitive verb. In other

words, this type of NP can only function as a subject in intransitive clauses. Since personal

names and pronouns are marked the same way (i.e. in preverbal position) in both transitive

and intransitive clauses, Zēng (2009) claims that they follow a nominative/accusative

pattern.

(22) a. 我來了。

Wǒ lái le.

I come ASP

‘The telegram arrived.’

b. *來了我。

Lái wǒ le.

come I ASP

‘A telegram arrived.’ (Zēng 2009:319)

Further consideration, however, reveals that the asymmetry between (21b) and (22b) does

not provide evidence for split-ergativity in Mandarin. Rather, it is merely an instantiation of

the definiteness restriction discussed for (12) in the previous section. Given that subjects in

post-verbal position must be indefinite, pronouns and personal names will never surface in

this position because they are inherently definite.

5. Concluding remarks

This lemma has introduced the two main ways in which the term ‘ergativity’ is used in the

field of linguistics. An ‘ergative’ language is uncontroversially a language which aligns

transitive objects with intransitive subjects in its case-marking system and possibly other

aspects of its grammar. Less commonly, the term ‘ergative’ is used interchangeably with

‘unaccusative’. This is the sense in which the term ‘ergative’ has been appropriately applied

in research on Chinese. In all other respects, Sinitic varieties are accusative languages.

References

Burzio, Luigi, Italian Syntax: A Government and Binding approach, Dordrecht: Kluwer

Academic Publishers, 1986.

Cheng, Lisa Lai-Shen and C.-T. James Huang “On the argument structure of resultative

compounds”, in: Matthew Y. Chen and Ovid J. L. Tzeng, eds., In honor of William S-Y

Wang: Interdisciplinary studies on language and language change, Taipei: Pyramid

Press, 1994, 187-221.

Cikoski, John, “Three essays on Classical Chinese grammar: (1) An outline of word-classes

and sentence structure in Classical Chinese”, Computational Analysis of Asian and

African Languages, 8, 1978, 17-152.

DeLancey, Scott, “Ergativity: Functional and formal issues”, Language 57, 1981, 626-657.

Dèng Sīyǐng 邓思颖 [Tang Sze-Wing], “Zuogehua he Hanyu beidongju 作格化和汉语被

动句 [Ergativization and Chinese passive sentences]”, Zhōngguó Yǔwén 中国语文, 301,

2004, 291-301.

Dixon, R.M.W, Ergativity, Cambridge: Cambridge University Press, 1994.

Gù Yáng 顾阳, “Shēngchéng yǔfǎ jí cíkù zhōng dòngcí de yīxiē tèxìng 生成语法及词库中

动词的一些特性 [Generative grammar and lexical properties of verbs]”, Guówài

Yǔyánxué 国外语言学 1996, 3, 1-16.

Hán Jǐngquán 韩景泉, “Yīng Hàn yǔ cúnxiànjù de shēngchéng yǔfǎ yánjiū 英汉语存现句

的生成语法研究 [Existential sentences in English and Chinese: Towards a generative

analysis]”, Xàndài Wàyǔ 现代外语, 24, 2, 2001, 143-158.

Huáng ZhèngDé 黃正德, “Zhōngwén de liǎngzhǒng jíwù dòngcí hé liǎngzhǒng bùjíwù

dòngcí 中文的兩種及物動詞和兩種不及物動詞 [Two kinds of transitive verbs and

intransitive verbs in Chinese]”, in Proceedings of the 2nd World Congress of Chinese

Language Studies 第二屆世界華文教學研究會議論文集, Taipei: World Chinese

Language Association, 1989,39-59.

Levin, Beth, On the nature of ergativity, Cambridge, MA: MIT dissertation, 1983.

Li, Y.-H. Audrey, Order and Constituency in Mandarin Chinese, Dordrecht: Kluwer, 1990.

Lǚ Shūxiāng呂叔湘,“Shūo ‘shèng’ hé ‘bài’ 说 ‘胜’和‘败’ [On ‘shèng’ and ‘bài’]”,

ZhōngguóYǔwén 中国语文, 196, 1987, 1-5.

Lǚ Shūxiāng 呂叔湘 (1942), Zhōngguó wénfǎ yàolüè 中國文法要略 [Outline of Chinese

grammar]. Reprinted in Taipei 台北: Wénshǐzhé 文史哲出版社, 1992.

Mǎ Jiànzhōng 馬建忠, Mǎ Shì wéntōng 馬氏文通 [Ma’s Guide to the written language],

Shànghǎi 上海: Shāngwù 商务印書館, 1898.

Ōnishi, Katsuya 大西, “Shīshòu tóngcí chúyì: Shǐjì zhōng de zhōngxìng dòngcí hé zuògé

dòngcí 施受同辭芻議: 《史紀》中的’中性動詞’和’作格動詞 [On transitivity

alternations: Neutral and ergative verbs in the Shiji]”, in: Ken-ichi Takashima and

Jiang Shaoyu, eds., Meaning and Form: Essays in Pre-Modern Chinese Grammar,

2004, 375-394.

Pān Hǎihuá 潘海华 and Jǐngquán 韩景泉, Xǐanxìng fēibīngé dòngcí jiégòu de jùfǎ yánjiū

显性非宾格动词结构的句法研究 [The surface syntax of unaccusative constructions]”,

Studies in Language and Linguistics 语言研究 25, 3, 2005, 1-13.

Perlmutter, David, “Impersonal passives and the unaccusative hypothesis”, in: Proceedings

of the 4th Annual Meeting of the Berkeley Linguistics Society, 1978, 157-190.

Pulleyblank, Edwin, Outline of Classical Chinese grammar, Vancouver: UBC Press, 1995.

Shěn Yáng 沈阳 and Rint Sybesma, “Jùfǎ jiégòu biāojì gěi yǔ dòngcí jiégòu de yǎshēng

guānxi 句法结构标记”给”与动词结构的衍生关系 [The derivational relation between

the syntactic marker gei and several verbal constructions]”, Zhōngguó Yǔwén 中国语文

2010, 3, 222-237.

Silverstein, Michael, “Hierarchy of features and ergativity”, in: R.M.W. Dixon, ed.,

Grammatical Categories in Australian Languages, Canberra: Australian Institute of

Aboriginal Studies, 1976, 112-171.

Sybesma, Rint, Causatives and Accomplishments: The case of Chinese ba, Doctoral

dissertation, Leiden University, 1992.

Sybesma, Rint, The Mandarin VP, Dordrecht: Kluwer Academic Publishers, 1999.

Tāng Tíngchí 湯廷池, “Hànyǔ fùhé dòngcí de shǐdòng yǔ qǐdòng jīaotì 漢語復合動詞的使

動與起動交替 [The causative-inchoative alternation in Chinese compound verbs]”,

Language and Linguistics 3, 3, 2002, 615-644.

Wáng Lì 王力 (1958), Hànyŭ shǐgǎo 漢語史搞 [Lectures on the history of Chinese].

Reprinted in Běijīng 北京: Zhōnghuá 中華書局, 2004.

Wū Xuěrú 巫雪如, “Cóng rènzhī yǔyìxué de jiǎodù kàn Shànggǔ Hànyǔ de ‘zuògé

dòngcí’ 從認知語義學的角度看上古漢語的「作格動詞」 [An investigation on the

‘ergative verbs’ in Pre-Qin Chinese from the viewpoint of cognitive semantics]”,

Qīnghuá Zhōngwén Wuébào 清華中文學報 2008, 2, 161-198.

Xú Jié 徐杰, “Liǎngzhǒng bǎoliú bīnyǔ jùshì jí xiāngguān jùfǎ lǐlùn wèntí 两种保留宾语

句式及相关句法理论问题 [Some theoretical issues of the two types of Chinese

‘retained object’ constructions], Dāngdài Yǔyánxué 当代语言学 1999, 1, 16-29.

Xú Jié 徐杰, “’Jíwùxìng tèzhēng yǔ xiāngguān de sìlèi dòngcí ‘及物性’特征与相关的四类

动词 [‘Transitivity’ and four classes of relevant verbs]”, Yǔyán Yánjiū 语言研究 2001,

3, 1-11.

Yáng Bójùn 杨伯峻 and Hé Lèshì 何乐士, Gǔ Hànyǔ yǔfǎ jí qí fāzhǎn 古汉语语法及其发

展 [Old Chinese Grammar and its development], Běijīng 北京: Yǔwén 语文出版社,

1992.

Yáng Sùyīng 杨素英, “Cóng fēibīngé dòngcí xiànxiàng kàn yǔyì yǔ jùfǎ jiégòu zhī jiān de

guānxi 从非宾格动词现象看语义与句法结构之间的关系 [On the relationship

between syntax and semantics from the perspective of unaccusative verbs]”, Dāngdài

yǔyánxué 当代语言学 1999, 1, 30-62.

Yu, Ning, “Towards a definition of unaccusative verbs in Chinese”, in: Jose Camacho and

Lina Choueiri, eds., Proceedings of the Sixth North American Conference on Chinese

Linguistics, vol. 1, 1995, 339-353.

Yuan, Boping, “Acquiring the unaccusative/unergative distinction in a second language:

Evidence from English-speaking learners of L2 Chinese”, Linguistics 37, 2, 275-296,

1999.

ZēngLìyīng 曾立英, Xiàn Dài Hànyǔ Zuògé Xiànxiàng Yánjiū现代汉语作格现象研究

[Ergativity in Mandarin Chinese], Běijīng 北京: ZhōngyāngMínzúDàxué 中央民族大

学出版社, 2009.

ZēngLìyīng 曾立英 and YángXiǎowèi 杨小卫, “Cóng ‘zuògé’ jiǎodù tán zhǔyǔ xìtǒng de

xuǎnzé 从”作格”角度谈主语系统的选择[On the choice of subject system from the

ergative perspective]”, HànyǔXuébào 12, 4, 2005, 22-30.

Zhōu Fǎgāo 周法高, Zhōngguó gǔdài yǔfǎ: Zàojù biān 中國古代語法: 造句編 [A

Historical grammar of Ancient Chinese, Part 1: Syntax], Taipei 台北: Academia Sinica,

Institute of History and Philology, 1961.

Zhou, Xinping, Aspects of Chinese Syntax: Ergativity and Phrase Structure, Urbana:

University of Illinois dissertation, 1990.

Aldridge, Edith, is associate professor in the Department of Linguistics at the University

of Washington. She has published on Old and Middle Chinese syntax in a variety of

journals, including the Journal of East Asian Linguistics, Language and Linguistics

Compass, Studies in Chinese Linguistics, The Linguistic Review, as well as collected

volumes, such as Movement Theory of Control (Benjamins, 2010), and Historical Syntax

and Linguistic Theory (OUP, 2009). Her work on ergativity begins with her (2004)

dissertation from Cornell University and continues in subsequent work published in

Language and Linguistics Compass, Lingua, and numerous collected volumes.

Keywords: alignment, case, causative, ergativity, external argument, internal argument,

passive, transitivity, unaccusative, unergative

Teaser: This lemma summarizes two uses of the term ‘ergativity’ in linguistic research and

shows that’ ergativity’ in Chinese is limited to the mapping of a thematic object to

grammatical subject in unaccusative clauses, a universal syntactic phenomenon which is

unrelated to the typological classification of languages according to their case-marking

pattern.