DOCUMENT RESUME FL 023 645 ED 393 312 Zhang, Zheng-Sheng AUTHOR Towards a Grammar of Aspect for Mandarin Chinese. TITLE PUB DATE [96] NOTE 48p. Viewpoints Evaluative/Feasibility (142) PUB TYPE Reports (Opinion/Position Papers, Essays, etc.) (120) MF01/PCO2 Plus Postage. EARS PRICE Grammar; *Language Patterns; Language Research; DESCRIPTORS *Linguistic Theory; *Mandarin Chinese; *Morphology (Languages); *Second Languages; Semantics; Uncommonly Taught Languages; Verbs *Aspect (Verbs); Markedness IDENTIFIERS ABSTRACT A discussion of aspectual morphemes in Mandarin Chinese broadens the perspective on aspect to examine its grammar, or the relationship between aspectual and non-aspectual elements, between different aspectual elements, and between the related categories of tense, aspect, and modality. Previous assumptions about Mandarin aspectual markers are noted, evidence of distributional gaps are used to argue for the tense and modality functions of aspectual markers, two kinds of facts about the semantical compatibility between verbs and aspectual markers (situational restrictions and boundedness requirements) are used to illustrate the differentiation of functions of aspectual elements in different contexts, and two kinds of facts about the relationship between aspectual elements (co-occurrence and replacement patterns) are used to illustrate the differences and similarities in functions between these elements and between their variants. In addition, the variation in obligatoriness across contexts is used to argue for the variable extent of grammaticalization for different variants of aspectual elements. Results are summarized and some remaining problems are addressed. Contains 26 references. (MSE) *********************************************************************** Reproductionri supplied by EDRS are the best that can be made * from the original document. *********************************************************************** Gramnrar of Aspect for Mandarin Chinese Th,,-,rds -Sheng Zhang U S DEPARTMENT OF EDUCATION Office of Educational Reseatcn and Improvement E UCATIONAL RESOURCES INFORMATION "PERMISSION TO REPRODUCE THIS CENTER (ERIC) MATERIAL HAS BEEN GRANTED BY This document has been reproduced as received Irom the person or organization -`',D\rlenc originating it 0 Minor changes have been made to improve reproduction quality Points of view or opinions stated In this document do riot necessarily represent TO THE EDUCATIONAL RESOURCES official OERI position or policy INFORMATION CENTER (ERIC)." BEST COPY AVAILAr", 2 TOWARDS A GRAMMAR OF ASPECT FOR MANDARIN CHINESE Zheng-sheng Zhang ABSTRACT The functions of 'aspectual markers' in Mandarin are various, including tense, aspect and modality, and context-dependent. Our markers conclusion is based on examining the relationship between aspectual elements and tha and verb phrase semantics, between different aspectual Distributional between the related categories of aspect, tense and modality. modal environments suggest tense and modality functions. gaps in tense and in boundedness and Variation in compatibility with verb phrase semantics Co- situation types reveals semantic variation of markers across contexts. aspectual elements reveal their similarity occurrence and replacibility among Variation in obligatoriness reveals different and differences in function. extents of grammaticalization. the thorny problem of 1. INTRODUCTION. The present paper re-examines But it proceeds with a new aspectual morphemes in Mandarin Chinese. Past research on Mandarin aspect was largely methodological orientation. narrowly focused on aspect to divorced from grammatical context; it too was In its insistent and modality. the exclusion of the related categories of tense fails to give due recognition to the attempt to seek invariant meanings, it analyses of many-faceted nature of 'aspectuar markers. Homogenous brought into the picture; limited data lead to paradoxes when more data are data obscure the characteristics but abstract analyses attempting to cover more of aspectual morphemes in context. research is not to re-analyze In contrast, the basic strategy of the present examine aspect in the broader context of the stock examples in isolation but to 1 3 between aspectual and the grammar of aspect, which studies the relationship non-aspectual elements, between different aspectual elements and that This strategy is between the related categories of aspect, tense and modality. elements can best based on the Saussurian belief that the nature of linguistic linguistic system and that be seen in its relation with other elements in the be aided by the search for the elusive 'intension' of aspectual elements can looking at the 'extension' in which they manifest themselves. in this The following kinds of contextual information has been utilized distributional gaps in terms of tense and modal paper. Hitherto neglected so-called restrictions suggest that in some contexts the functions of some The variation in the aspectual elements are not purely aspectual at all. compatibility between verb phrase semantics and aspectual elements in terms variation in of boundedness and situation type restrictions reveals the The co-occurrence and semantics of the same element in different contexts. similarity raid replacibility patterns among aspectual elements point to their The variation in obligatoriness across contexts differences in function. reveals the different extent of grammaicalization. assumptions and In contrast with the implicitness of methodological the present paper, modes of argumentation in much of previous research, in of the spedfic assumptions will be advanced concerning the interpretation above-mentioned contextual information. Although the paper inevitably points to a modular approach to many' debates, Mandarin aspect, it does not intend to add to the 'one or On the one itself. whose 'either...or' assumption may not be justifiable in contextual variants; on hand, a single marker may ilonetheless have different different contextual variants the other hand, as Comrie pointed out (1985), subsumed under an abstract semantic characterization may not be able to be 2 4 be related in terms of necessary and suffident conditions but may instead Even if ultimately there are more loosely as members of a prototype. unifying accounts, adequately characterizing the contextual variants is an important goal in itself. follows: The major conclusions of the present paper are as each has contextual variants. , -guo -Ie, -zhe a. not always purely aspectual. , -guo are b. -le, -zhe not be equally aspectual. , -guo may -le, -zhe c. function. d. -le, -zhe , -guo may simultaneously mark more than one grammaticalized to the same extent. e. -le, -zhe , -guo's variants are not The organization of the paper is as follows. After this introduction, in section 2, two previous assumptions explicit. concerning Mandarin aspectual markers will be made used to argue for the In section 3, evidence of distributional gaps will be markers. tense and modality functions of aspectual compatibility In section 4, two kinds of facts about the semantic and between verbs and aspectual markers, i.e. situation type restrictions differentiation of boundedness requirements, will be used to illustrate the functions of aspectual elements in different contexts. between aspectual In section 5, two kinds of facts about the relationship be used to show elements, i.e. co-occurrence and replacement patterns, will these elements and the differences and similarities in function between between their variants. is used to In section 6, the variation in obligatoriness across contexts of grammaticalization for different variants of argue for the variable extents aspectual elements. 3 5 functions of -le, In section 7, a summary of the results concerning the -zhe and -guo and some remaining problems will be presented. orientation In section 8, a brief recapitulation of the present research and some directions for future research will be indicated. challenges a rather 2. PREVIOUS ASSUMPTIONS. The present paper the aspect of prevalent but implicit underlying assumption concerning -le, -zhe and -guo are Mandarin Chinese, namely, that the aspectual markers in all equally and solely aspectual, grammaticalized to the same extent, specific assumptions: contexts. This is manifested in the following two This is the assumption of invariance and that 2.1. 'Once an X, always an X'. The assumption holds that an aspectual of generalizability across contexts. of one context marker has the same function in all contexts and what is true is true of all contexts. holds that a marker can bear 2.2. 'One form, one function'. This assumption in a given context; or, in other words, semantic no more than one function It does not entertain the possibility that in functions are coded analytically. encode the functions of tense, aspect and even some contexts a marker can from the invariance modality simultaneously. This assumption is distinct of a form in a single assumption in that it concerns the possible functions itself with functions of context whereas the invariance assumption concerns forms across contexts. 'Once an X, always The assumptions of 'One form, one function' and untenable, when we examine aspect in context. In an X' turn out to be rather of facts concerning aspect in the following sections, we will look at a number 4 homogeneous context, including some paradoxes that result from modular approaches, and suggest alternative analyses based on the new orientation. MODALITY FUNCTIONS. 3. DISTRIBUTIONAL GAPS AND TENSE AND the Outside of Chinese linguistics, many linguists have commented on of relatedness and the difficulty of separating the graxnmatica, categories TAM is used to refer to tense, aspect and modality, so much so that the term 1982). But the these categories as a whole (Dahl 1985, Givon 1982, Hopper invariably treated -le, Chinese linguistic literature, with few exceptions, has The difficulty with uncovering the tense -zhe, -guo as aspectual markers. functions and and modality functions lies in the abstract meanings of these In this section, we the lack of systematic contrasts exhibited in the language. of aspectual will investigate the possible tense and modality functions modal contexts. markers by examining their distributional gaps in tense and It has almost acquired the The tense function of aspectual markers. 3.1. It is true that -le has status of a truism that Chinese marks aspect, not tense. which has been been analyzed as a marker of 'anteriority' (Shi 1990), eschew the earlier, considered as relative tense; however, most linguists Ross 1995). 'naive' notion of -le as a marker of past tense (but see that the Li&Thompson (1981)'s position is fairly representative. They argue the following verbal suffix -le can sometimes occur in future contexts, as sentences show: 5 Ta mingtian chi le zao fan jiu zou. 1) he tomorrow eat LE early rice then go 'He will leave after breakfast tomorrow.' Dao mingnian ta jiu xue le wu nian Zhongwen le. 2) Chinese LE next year he then study LE five year till 'He will have studied Chinese for five years by next year.' The 'Once an X, general. Therefore -le cannot be a past tense marker, in this mode of argumentation. always an X' assumption is seen most clearly in assumption, however, Once we give up the 'Once an X, always an X' mark the possibility that in some contexts, -le may we are free to entertain While it is surely difficult to tense, instead of, or in addition to, aspect. elusive semantic content isolate the tense component by looking at -le's suggested by its distributional alone, a hint of its tense function is nonetheless and Although aspect by definition is independent of tense restrictions. however systematic gaps in therefore should freely combine with it, there are Despite conventional wisdom, their co-occurrence possibilities in Chinese. It is obvious that we cannot say: -le DOES occur primarily in past contexts. *Ta mingtian chi le zao fan 3) he tomorrow eat LE early rice perfective aspect, which -le is supposed to It is important to point out that the In Russian, perfective verb be a marker of, is not intrinsically tied up to past. forms are used for future as well: 6 Zdes' i poobyedaemP 4) here and have-lunch (perfective form) 'We will have lunch here.' of -le in future context is Most previous analyses fail to recognize that the use is the event only possible in limited, definable contexts. One such context Another context is the having the form of V1 le V2. sequence type sentences end. Where -le is the double le sentences with a sentential particle le at the be no future time sole TAM marker suffixed to the main verb, there can exactly because the reference. The Russian example serves as a good contrast, future reference. perfective form is used in a main verb context with tense', in general, We thus conclude that -le cannot NOT mean 'past time in some Thus, when we see that -le is restricted to occur in past either. indicates past tense, if syntactic contexts, we will interpret it to mean that it The contextual variation of such restrictions only in these syntactic contexts. accordingly entails the corresponding variation in function. function we attribute to -le is We want to make clear that the tense being that between really 'past' rather than 'anteriority', the distinction If the main verb -le indicates past tense, 'absolute tense' and 'relative tense'. double le sentences, where it can how about the -le in VI le V2 sentences and reference? As will be argued in section 4, where occur with future time examined, this boundedness requirement and situation type restrictions are Instead of the compound variant has a different kind of tense function. verb -le, it is a relative functions [perfective aspect+past tense] of the main without aspectual function. tense marker of 'anteriority', possibly 7 and a The differentiation of -le into an [perfective+pst] variant [anteriority] variant has a parallel in -guo. The marker of the so-called 'experiential aspect', -guo is also restricted to past contexts: *Wo yihou chi guo Riben cai. 5) I in future eat GUO Japan dish 'having the When -guo occurs in future, however, it no longer means experience of: Wo mingtian chi guo fan jiu zou. 6) I tomorrow eat GUO rice then go 'I will leave after the meal tomorrow.' the meaning of 'finish'. The -guo here rather is a verbal complement having will adopt the following To make our mode of reasoning explicit, we interpretive assumption: Restriction to particular tense should be interpreted concomitantly, of the same tense. as the marking, if only Chen (1988) sems to In his discussion of the role of tense in Chinese, have used the same assumption. He cites the restriction -le and verbal reduplication to past contexts and on the co-occurrence of The same concludes that -le in such a context indicates past tense. tense is more assumption allows us to conclude that the restriction to past pervasive than that. have said in no way We should hasten to add, however, that what we full-fledged past tense implies that -le is not an aspectual marker but a