Among the tenseless languages, Cantonese as one of the Chinese dialects is well-known for its rich repertoire of sentence-final particles. With regard to temporal system, substantial Cantonese linguistic studies focus on two basic temporal parameters, namely tense and aspect. Two major issues are especially studied: i) if tense is linguistically manifested in such language and hence if Cantonese is a tensed or tenseless language (c.f. Sybesma 2005, Hu, Pan, Xu 2001) and ii) what kind of aspect the verbal suffixes are manifesting and what role aspect plays in Cantonese (c.f. Li et al. 1995, Cheung 2007, Matthews and Yip 1994, Zheng 2001 among others). It seems that if a grammatical morpheme is to be studied, and such morpheme conveys temporal information, it is manifesting either tense or aspect. Yet, is there the third possibility?
In this dissertation, I take Cantonese sentence final particle lei4 as a case to study. This particle is extensively studied for it is the grammatical morpheme which was argued to be either tense or aspect. And the debates over this particle have not yet ceased. The basic problem regarding this particle is that when it occurs in the simplex sentence, the sentence carries several pieces of temporal and aspectual information, which leads one to doubt if the particle conveys them all and what basic semantics of this particle is. Examine the following simplex sentence:
(1) Keoi5 heoi3 Jat6bun2 lei4.
S/he go Japan LEI
'S/he went to Japan.'
The above simplex sentence includes the bare sentence [Keoi5 heoi3 Jat6bun2] 's/he go Japan' and the sentence final particle lei4. With the presence of the particle, the sentence carries five pieces of temporal and aspectual information:
i) Pastness (Gao 1980, Li et al 1995, Tang 1998, Leung 2005, Sybesma 2005 Cheung 2007, Liu 2007 among others): the event of going to Japan happened in the past and the pastness reading is missing if the particle lei4 is absent from the sentence;
ii) Current Relevance (Sybesma 2005): the event of going to Japan is somehow relevant to the present moment;
iii) Perfective (Gao 1980: completion of the eventuality, Yiu 2001): the event of going to Japan has culminated/completed (c.f. the notion of 'on-going' or 'imperfective');
iv) Discontinuity (Yiu 2001): the result state of the event, e.g. be in Japan, has discontinued and thus the subject is no longer staying in Japan at the speech time;
v) Recency (Tang 1998, Lee and Yiu 1999, Yiu 2001, Leung 2005, Cheung 1972/2007): it is generally agreed that when lei4 is used, the sentence carries the meaning of recency, i.e. the event of going to Japan happened recently.
Focusing on the SFP lei4 in simplex sentences, three major proposals were made in the literatures to account for the particle in simplex sentences, namely
i) The recent past tense approach (Tang 1998, Sybesma 2004): it is claimed that lei4 is manifesting tense or is a 'tense element' (Sybesma 2004);
ii) The perfective approach (Yiu 2001): it is argued that the pastness is subject to two aspectual properties of the particle, namely perfectivity and discontinuity;
iii) The perfect approach (c.f. Sun 1995, He 1998, Zhang 2000, Chen 2006): when comparing lei4 with its Mandarin counterpart laizhe, I consider the analysis towards laizhe as insightful, especially the one proposing that laizhe is a perfect marker based on the notion of 'current relevance'. Sybesma (2004) while studying Cantonese, also states that lei4 is associated with the function to 'make the sentence especially relevant to some aspect of the conversational context' (ibid: 176 – 177) I thus consider that claiming lei4 as a perfect marker as one of the potential proposal towards the particle.
From the above proposal, the problem is clear: when the particle lei4 occurs, the sentence carries both temporal and aspectual information and this is, as I consider, the reason for the debates. Yet, the satisfactory solution has not come out yet. In this dissertation, I argue that none of the previous accounts can fully capture the semantics of the particle lei4. Consider the following sentence:
(2) Soeng6-go3-lai5baai3 lok6-jyu5 lei4. ?Kam4jat6 dou1 jau5 lok6.
Last-week fall-rain LEI Yesterday also have fall
'It rained last week and it also rained yesterday.'
What this sentence is illustrating is this: there happened two raining events in the past, one in the last week and another yesterday. Lei4 is associated with the raining of last week. And such raining event is not the last token event by the speech time, for the sentence is continued by another clause indicating that there exists another raining after last week, i.e. the raining which happened yesterday. However, the conjunction is infelicitous. This suggests that when lei4 is used, the eventuality must not be followed by another token eventuality, i.e. must be the last token eventuality. Put it in another way, the eventuality has to be the closest one to the speech time, keeping the shortest distance from the speech time. Based on the observation of sentence (2), limitations of each of the previous studies are found:
i) For the recent past tense approach, it cannot explain why the conjunction is infelicitous even if the raining associated with the particle in the sentence is located in the past. This shows that the eventuality does not only need to locate in the past but it also needs to satisfy the event occurrence depicted above. To account for the even occurrence as well as the temporal distance, one may resort to the notion of 'recency'. However, this approach leaves this notion undefined. I argue that the definitional accuracy of the notion is essential to respond to the criticism that the notion is vague (c.f. Yiu 2001) on one hand and to provide a better understanding of such temporal parameter (c.f. Comrie (1985)) on the other hand.
ii) For the perfective approach, it is puzzling why the second occurrence of the same token event is prohibited while this property is not aspectual.
iii) Similarly, it is doubtful why a perfect would not allow the second occurrence of the same token event, while lei4 could imply that the raining of last week is relevant to some present state, i.e. current relevance.
I argue in this dissertation that the sentence final particle lei4 is neither a tense marker nor an aspect marker. Rather, it is a temporal modifier indicating the shortest temporal distance, i.e. recency. I define recency in terms of two notions, namely 'measurement' and 'comparison'. These two notions are the essential components of recency. Based on these two notions, I propose that the meaning of pastness, perfectivity and discontinuity are subject to the requirement of 'measurement'; current relevance is the derived meaning from 'recency' (c.f. Comrie 1976); and 'comparison' differentiates the recency meaning of lei4 from the recency meaning of recency adverbs, e.g. ngaamngaam 'just' and adverbial particle sin 'not until'.
On the ground of the analysis proposed above, I also address another issue regarding the particle lei4 – its semantics in the bi-clausal construction which I term as dak…lei4 construction. The construction consists of a main clause and a subordinate clause in which the post-verbal modal dak occurs with the particle lei4. Examine:
(3) Keoi5 zou6-dak1-jyun4 fan6 gung1fo3 lei4,
S/he do-DAK-finish CL assignment LEI
ngo5 dou1 tai2-jyun4 tou3 kek6 laa3.
I already watch-finish CL drama SFP
'Shortly before the time s/he starts doing the assignment, I will have finished watching the drama.'
'Shortly before the time s/he finishes doing the assignment, I will have finished watching the drama.'
'At the time s/he starts doing the assignment, I will finish watching the drama.'
'At the time s/he finishes doing the assignment, I will finish watching the drama.'
'Shortly after the time s/he starts doing the assignment, I will finish watching the drama.'
'Shortly after the time s/he finishes doing the assignment, I will finish watching the drama.'
There are two reasons to study this construction in particular. First, the bi-clausal construction is expressing information related to time, i.e. when the second event may occur. It is thus believed that lei4 in this construction is used to convey temporal information. Thus, an in-depth analysis of the particle in bi-clausal sentences is called for. Notably, while comparing to simplex sentences containing lei4 which must be interpreted past, the bi-clausal sentence above which also contains lei4 is not necessarily interpreted past. Instead, while no explicit temporal reference is given, the sentence is interpreted as future. This leads to the question: while both are temporal, are they polysemous? Second, this construction is intriguing in its various temporal interpretations. Take the sentence (3) as an example. There are at least six temporal locations in which the second events may occur. It is doubtful what gives rise to these temporal interpretations.
I extend the proposed analysis of lei4 in simplex sentences to the analysis in bi-clausal sentences. I argue that lei4 indicates recency in simplex sentences as well as bi-clausal sentences, but it does not indicate temporal relations, which is the function of tense and aspect (c.f. Klein 1994). The ambiguities of the temporal interpretations can be accounted for by the basic semantics of lei4, i.e. recency and the interactions between lei4 and other elements in the sentence, e.g. the post-verbal dak, the situation aspect of the eventualities denoted by the subordinate clause and the main clause and the particle used in the main clause.
In brief, by studying the Cantonese particle lei4, this dissertation makes the following contribution. First, a unified account of the particle, in terms of its temporal use, provides a better understanding of the clause final particle lei4 in both simplex and bi-clausal sentences. In particular, the temporal information codified in it is clarified. Second, arguing that the particle is a temporal modifier advances the knowledge of temporal modification in Cantonese linguistics. Unlike the previous studies, this dissertation proposes that the clause final particle is not manifesting tense or aspect, despite of the fact that it is 'semi-lexical' (terminology in Sybesma 2004) and it conveys temporal information. Instead, it is the other parameter, temporal distance, which the particle is realizing. This proposal hence introduces such parameter to the Cantonese temporal system, offering a new perspective to the analysis of the role of clause final particles in temporal system.
| Date of Award | 3 Oct 2014 |
|---|
| Original language | English |
|---|
| Awarding Institution | - City University of Hong Kong
|
|---|
| Supervisor | Po Lun Peppina LEE (Supervisor) |
|---|