The Cognitive Model Of Temporal System In Modern Chinese And Its Calculation

The overall scope of this dissertation is to provide an account of temporal system in Chinese from the viewpoint of cognitive-functional linguistics and Derivatism, that is, to present a semantic-pragmatic explanation for correlations between tense and aspect, especially between deixical point and cognitive window, and between cognitive window and event. After a corpus-based study of what the limits are to the temporal information and the expressions of the information in Chinese, a unified temporal model of tense and aspect was set as the basic principle to the explanation of Chinese phenomena of time.The dissertation has 10 chapters with more than 277,000 characters in total.In chapter 1, first some preliminary remarks are given concerning the notion temporal system and its development, in particular defining the general model of time as 'cognitive window of time — event', that means mapping event into the cognitive window. The discussion of two ways of expression of time provides an evidence for the disparting of time from the event: time is confused with event and embedded in the expression of event in E-statement, which states an event and the time that the event occupies at the same time, but temporal constituent has independent status of theme in A-statement and the event acts as its rheme.Chapter 2 provides a further theoretical background on the Derivatism and the syntactic construction used in the processes of deriving. There are two ways of coding given information: a particular atomic unit is used for coding in atom-coding, but it is the whole net system, not any single unit, that codes the information in net-coding. The Derivatists justify the approach of net-coding taken in languages whereby aspect and other categories have meanings that are defined according to context. For the purpose of explanation, calculation must play an important role in the processes of deriving and so in the grammar. This dissertation differs from most other books on temporal system in that it is not only concerned with any one particular mark in language, nor with a comparison of various individual marks, but rather presents temporal system as a part of general inferential system.The body of the dissertation is composed of chapter 3-9.The discussion of the construction of cognitive window in Chapter 3 provides a corpus-based study of temporal entities and chains in the reference point construction (RC), which indicates the cognitive window as its target. The head of RC is the basicentity upon which the tense meaning depends. This chapter also provides a further look on the problems inherent in defining the meaning of a chain, with examples drawn from a corpus with more than 1,000,000 characters. There're six kinds of chains that have been distinguished in Chinese: null chain, entity indicater, region, part, range, and possession. Moreover, in a group with more than two E-statements, the former event with a prominent point of time establishes a heterochronous window for the latter, or a synchronous window if it has a prominent period of time.Chapter 4 presents a theory of tense and aspect that is an extensive revision of the one originally proposed in Cognitive Grammar. In view of the terminological, and conceptual, confusion of tense and aspect, it is worth ensuring now that an agreement has be done, before embarking on discussion of the term 'tense' and 'aspect': there must be a medial concept that connects both tense and aspect in a unified theory, which gives tense meaning in its relation with deixical point, or aspect meaning in its relation with event. It is 'point of reference' in Reichencach's and Johnson's, or 'time of viewing' in Interval Semantics, or 'immediate scope' in Langacker's Cognitive Grammar. The major proposal of this chapter is that the concept 'cognitive window of time', which is expressed by the temporal adverbial modifier of a sentence, should be used instead of 'immediate scope'. The idea is that tense categories (past-present-future) relate cognitive window to deixical point, and aspect categories (perfective-imperfective) relate event to window, that is, aspect is the appearances of event in window.Chapter 5 proposes a four-layers construction of events. In view of the general model of time, an event has itself temporal construction before being mapped into cognitive window, and should be called 'pre-event'. A further division is made between preliminary event and secondary event. The latter is the one with grammatical elements that regularly change the temporal construction of the preliminary event. After mapping, conceptualizer conceives the event in window, and rebuilds it as a new event, which could be called 'gen-event'. The marks of tense in a sentence indicate that an event has been mapped into a window.Chapter 6 investigates the temporal construction of preliminary events expressed by l,067adjectives (1,348 terms of meaning) and 258verbs (446 terms of meaning) in modern Chinese, with examples drawn from internet and a corpus with more than 20,000,000 characters. More importantly, three forms have been proposed in order to test which construction an event has: 1) the event has a prominent phrase iff the phrase is mentioned in 'when...' form; 2) the event has a start phrase if the 'before...'form means a period before the start point of the event; 3) the event has a end phrase if the 'after...' form means a period after the end of the event. Four types of typical events have been distinguished: 1) typical activity (c s e), with continuous meaning in 'when...' form, ingressive meaning in 'before...' form, and terminal meaning in 'after...' form; 2) typical state (c s s), with continuous meaning in 'when...' form, ingressive meaning in 'before...' form, and 'after...' form; 3) typical telic (e e e), with terminal meaning in all three forms; and 4) typical cause-effect (hc/xc hs he), with continuous meaning of the causal event or the resultant event in 'when...' form, ingressive meaning of causal event in 'before...' form, and terminal meaning of causal event in 'after...' form. Other events are not typical. The cognitive mechanisms behind the regularless phenomena have been explored in this chapter also.Chapter 7 provides a further investigation on the parameters that are relevant in the definition of terminal point of an event. The term 'terminal' in this dissertation, not necessarily indicating a real stop for an event, but rather some kind of boundaries the event has arrived at, should be further divided into two categories: 1) natural terminal, i.e. the event has built into it a terminal point, namely that point at which a boundary is complete; 2) arbitrary terminal, i.e. the event can stop at any point. The boundaries given in languages include quantities and results. In expressions referring to telic event with natural terminal it is important that there should be a further test with 'when...' form on whether the natural terminal is prominent. Unlike much recent work on telic-atelic category this dissertation discovers that only the events with prominent natural terminal can has the P property, i.e. they take in-phrases rather than for-phrases, and the progressive meaning of them must not entail the perfect (perfective) meaning. For the events with un-prominent natural terminal, the progressive entail the perfect sometimes. The relation between natural terminal and the pragmatic number of the entities involved in the event, and the question about how much and which entities have boundaries in a given telic event, are also important topics in this chapter.Chapter 8 first summarizes the history of study on grammatical meaning of 'Le₁' in Chinese linguistics. After statistical comparison of the sentences in corpus, we propose that the aspectual function of 'Le₁' is to make the event more complete or perfective. What degree of perfectiveness the event can arrive at could be derived from the constraints and context. So the aspectual meaning of the sentence may be perfective as well as ingressive or terminal. 'Making more complete' is actually the gestalt function of psychology. In order to explain the degree of perfectiveness indetail, i.e. when and what kinds of boundaries will be profiled, and why several boundaries appear synchronously in one time, but only one special boundary appears in another time, this chapter introduced three laws and principle in a fixed order: law of mapping, law of gestalt, and principle of temporal length. The aspectual meaning of a given 'Le₁' sentence can be derived from them. The law of gestalt is the most important one in these laws. It materializes the aspectual function of 'Le₁'. This is the reason why we call 'Le₁' as 'Gestalt Aspect'.Chapter 9 explores what features of the discussion in former chapters must be incorporated into a formal theory of time in order to extract semantic information of time from a given Chinese sentence. The temporal information is represented jointly by the units and their relations. Thus, an integral model should be established in order to calculate the temporal information. It includes four procedures: 1) reduces the concept of temporal properties into cognitive constituents as the three fundamental phrases (i.e. start, continuance, and end of an event), point and period of time, etc.; 2) describes and categorizes the types of temporal constructions of events with these constituents; 3) translates every possible units and relations in the modern Chinese sentences with emblems of the event-types and signs of temporal constituents. The result of the translation, namely, the meta-linguistic expressions (i.e. the translation forms) of the units or the relations, makes it clear what the units or the relations mean in the process of carrying temporal information; 4) establishes a rule-driven system of calculation in the cognitive model, simplifying the meta-linguistic expressions into the most economical expressions which is exactly the temporal information coded in the sentence as a whole. This dissertation suggests two schemas for formalization of temporal information.Finally, chapter 10 ties together the discussion of the dissertation and suggests what problems should be concerned in further study.