The results for topic persistence in the Tsou data are given in Table 23.
Table 23: Focus and persistence (narratives)
Focus AF NAF Total
A/S O A/S O
N % N % N % N % N % High (RD>=3) 30 45.4 2 11.7 33 60.1 14 31.8 79 43.4
Med (RD=2) 13 19.7 4 23.5 13 23.6 14 31.8 47 25.8
Low (RD<=1) 23 34.8 11 64.7 9 16.3 16 36.3 56 30.7
Total 66 100 17 100 55 100 44 100 182 100
(AF: χ2=7.19, p<.05; NAF: χ2=8.83, p<.05.
AF* NAF: χ2=4.34; p>.05 for A/S; χ2=4.51, p>.05 for O)
Table 23 shows that the results for persistence are consistent with Cooreman et al.’s prediction: AF agents are significantly more topical than patients; NAF agents are highly topical and patients are moderately topical. The overall differences between AF patients and NAF patients are not significant, however.
At this point, it is convenient to provide a summary of the pragmatic functions of agents and patients in AF and NAF clauses in Tsou as presented in the preceding sections.
Unlike in Seediq, the choice of NAF in Tsou is conditioned by topicality measures in that agents in NAF are consistently significantly more continuous than patients, as determined through the three topicality metrics. There is in addition no significant difference in continuity between agents in AF and those in NAF. The topicality values for agent and patient in these clauses can be ranked as follows.
4) A(AF) = A(NAF) >> P(NAF) > P(AF) These results suggest the following clause types:
5) AF: agent >> patient agent highly topical and often omitted; patient non- topical and rarely omitted
NAF: agent >> patient agent highly topical and rarely omitted; patient moderately topical and frequently omitted
What distinguishes Seediq from Tsou, then, is the behavior of their NAF clauses in discourse pragmatics. Recall that, as shown in Sections 5 through 7, there is a strong association between lexical transitivity and NAF clauses and between lexical
687
intransitivity and AF clauses in Tsou. Unlike in Seediq, there is also a strong association in Tsou between NAF clauses and identifiability of patients. Finally, as in Tagalog, NAF clauses in Tsou account for nearly 75% of the transitive clauses in the language, while they make up less than 30% of the transitive clauses in Seediq.
10. Conclusion
We have shown, based on a careful analysis of both narrative and conversational data, that the pragmatics of focus systems in Seediq and Tsou show considerable difference from each other, with the conversational data in Seediq in particular showing even greater divergence from the ‘expected’ behavior. It is clear that in many ways it is Tsou that has remained more ‘conservative’ and behaves much more like what is known about the Philippine languages (Wouk 1996, Starosta et al. 1982). No pragmatic difference appears to underlie the choice between AF and NAF clauses in Seediq, there being more no’s than yes’s in Table 24 under Seediq. Neither discourse transitivity nor grounding can be shown to be a significant factor in the deployment of focus forms.
Furthermore, the choice of NAF correlates with neither referential distance nor topic persistence. These results show that the intuitions regarding the relative pragmatic statuses of the core arguments in the clause, which form the very foundation of the
‘focus’ terminology, have been mistaken, at least as far as Seediq is concerned.
Table 24 summarizes the results we have presented thus far in the preceding sections, incorporating from other WAN languages in order to understand the pragmatics of focus in these two languages from a wider comparative perspective. Single question marks indicate non-availability of the relevant statistics; a double question mark in Seediq indicates ambiguity in interpretation, since the RD metric for narrative and conversational data has been shown to yield conflicting results.
688
Table 24: Pragmatics of focus in Seediq, Tsou and some WAN languages Features Seediq Tsou Philippine&W.
Indones (Wouk 1999)
Based on the preceding discussions and those reported in the literature, the pragmatics of focus characteristics of various WAN languages may be seen to form a continuum, with PAN at the top of Table 25 below representing a discourse-transitivity dominated language and English at the bottom representing a thematicity-dominated language, and all the WAN languages falling somewhere in-between the two points.
Research into the discourse deployment of voice in Rukai, the only known active-passive language in Formosa, or in the Sulawesi languages, known to have lost their PF morphology (conjugated PF), has yet to be undertaken. Discourse grammarians interested in Austronesian languages are waiting with baited breath for news about these languages. Still, it is of considerable theoretical interest to note that Standard Jakarta Indonesian (SJI) is located roughly at the midpoint between the two ends of the continuum, since functionally its focus system is intermediate between Tagalog and English (Wouk l996) in that: (a) it has an agentless passive construction with high textural frequency; and (b) although its PF clauses continue to show correlation with high discourse transitivity, there is a strong correlation between focus choice and the relative salience of the argument NPs. AF clauses in SJI are used when agents are more thematic than patients and PF clauses used when patients are more thematic than agents.
689
Table 25: Pragmatics of focus in WAN languages5
*PAN
* Toba Batak, Tagalog (Wouk 1986), early modern Malay (Hopper 1986), Karao (Brainard 1994): transitivity, esp. individuation of patient the most important focus determinant
*Classical Malay (Cumming 1995:254): grounding is primary determinant for focus
*Tsou: shows innovations in focus morphology; functionally PF far more common, highly topical and is determined by discourse transitivity and topicality metrics.
*Cebuano (Payne1994): Topicality metrics condition choice of AF/NAF; some PF clauses have topicality pattern typically associated with passive (P>>A).
*Seediq: formal focus morphology still kept, but functionally the use of PF not determined by transitivity or topicality
*Standard Jakarta Indonesian (Wouk 1996):
(1) much of focus morphology is lost;
(2) functionally, focus system is intermediate between Tagalog and English in that (a) PF clauses are fairly frequent;
(b) It has agentless passives with high textual frequency
(c) PF clauses continue to show strong but incomplete correlation with high discourse transitivity (in punctuality, mode, and individuation of patient, but not in other parameters);
(d) But there is strong correlation between focus choice and the relative salience of the argument NPs (more salient patients correlate with PF; more salient agents correlate with AF)
*Sasak (Wouk 1999) (1) loss of PF morphology (conjugated PF)
(2) in oral clauses (PF) the patient generally much more topical than the agent
(3) focus system weakly determined by certain dimensions of transitivity, not topicality.
*Modern Malay (Cumming 1995): focus functions akin to English style voice system
*Rukai: complete loss of focus morphology; active-passive system (Li 1997)
*English: transitivity not a factor in determining clause structures.
5 The placement of the various Austronesian languages on the continuum relative to one another is motivated strictly in terms of their discourse pragmatics. Placing Toba Batak, Tagalog and Classical Malay rather than Tsou closer to PAN, for example, in no ways implies their genetic relationships, nor their chronological development in relation to PAN.
690
These and other findings suggest the following scenario of diachronic development from a transitivity-dominated language like Tagalog to a more thematicity-dominated language like Modern Malay or English. First, NAF clauses would have to show a weakening correlation with discourse transitivity or topicality, initially in just some of the parameters, but subsequently in more and more of the parameters, to the point where functionally the use of NAF is not determined by transitivity, as seen in Seediq and SJI.
Secondly, as the correlation between NAF and transitivity weakens, an increasingly strong association between the choice of NAF and thematicity of the patient would develop, resulting in the formation of a construction in which the patient is generally more topical than the agent, as seen in Sasak. Finally, with the function of NAF now largely replaced by the (agented or agentless) passive, the NAF morphology would be finally lost and the transition to a thematicity-dominated language would then ensue, again as seen in Sasak.6 And of course the path of development would be reversed for a thematicity-dominated language to change into a transitivity-dominated language.
To conclude, then, Seediq and Tsou, two primary-branch, Austronesian languages spoken in the central highlands of Taiwan, represent interesting case studies whose focus systems differ considerably from one another, on the one hand, and from other Western Austronesian languages, on the other, when the pragmatic functions of their focus forms are studied within the framework of quantitative discourse analysis, using naturally occurring interactional data. Indeed, an inspection of Table 25 should convince us that each of the languages given there exhibits unique discourse properties with respect to its focus system and that these properties at best form a family resemblance relationship among themselves. One can talk intelligibly about the nature of a ‘focus language’ only by making considerable simplifying assumptions. Since the linkage between language use and grammar is to be found in interaction, this process needs to be better understood and taken into account as it applies to the study of the discourse pragmatics of Formosan and other Austronesian languages. Hopefully, as discourse functions of the focus systems in other primary branch Formosan languages are investigated in comparable depth, we will be in a much better position to speculate on the nature of the focus system in the ancestor language and the diachronic break-up of the system into its present-day daughter languages.
6 Cumming (1988), cited in Wouk (1996), appears to make essentially the same proposal.
691
References
Baldi, Philip (ed.). 1990. Linguistic Change and Reconstruction Methodology. Berlin:
Mouton de Gruyter.
Brainard, Sherri. 1994. Voice and ergativity in Karao. Voice and Inversion, ed. by Talmy Givón, 365-402. Amsterdam: John Benjamins.
Blust, Robert. 1999. Subgrouping, circularity and extinction: Some issues in Austronesian comparative linguistics. Selected Papers from the Eighth International Conference on Austronesian Linguistics, ed. by E. Zeitoun and Paul Li, 31-94. Taipei: Institute of Linguistics (Preparatory Office), Academia Sinica.
Cooreman, Ann. 1982. Topicality, ergativity and transitivity in narrative discourse:
Evidence from Chamorro. Studies in Language 6:343-374.
Cooreman, Ann. 1987. Transitivity and Discourse Continuity in Chamorro Narratives.
Berlin: Mouton de Gruyter.
Cooreman, Ann, Barbara Fox, and Talmy Givón. 1984. The discourse definition of ergativity. Studies in Language 8:1-34.
Cumming, Susanna. 1988. Syntactic Function and Constituent Order Change in Malay.
Los Angeles: UCLA dissertation.
Cumming, Susanna. 1995. Multifunctionality and the realization problem in modelling discourse production. Discourse, Grammar and Typology: Papers in Honor of John Verhaar, ed. by W. Abraham, T. Givón and S. Thompson, 247-273. Amsterdam:
John Benjamins.
Cumming, Susanna, and Fay Wouk. l987. Is there ‘discourse transitivity’ in Austronesian languages? Lingua 71:271-296.
Dixon, R. M. W. 1995. Ergativity. Cambridge: Cambridge University Press.
Du Bois, John. 1987. The discourse basis of ergativity. Language 63:805-855.
Dyen, Isidore. 1990. Homomeric lexical classification. Linguistic Change and Reconstruction Methodology, ed. by Philip Baldi, 211-270. Berlin: Mouton de Gruyter.
Givón, Talmy. 1983. Topicality in Discourse: A Quantitative Cross-linguistic Study.
Amsterdam: John Benjamins.
Givón, Talmy. 1992. The grammar of referential coherence as mental processing instructions. Linguistics 30:5-55.
Givón, Talmy. 1994. The pragmatics of de-transitive voice: Functional and typological aspects of inversion. Voice and Inversion, ed. by Talmy Givón, 3-45. Amsterdam:
John Benjamins.
Givón, Talmy (ed.). 1994. Voice and Inversion. Amsterdam: John Benjamins.
Holmer, Arthur. 1996. A Parametric Grammar of Seediq. Lund: Lund University Press.
692
Holmer, Arthur. 1999. Structural implications of the function of instrumental focus in Seediq. Selected Papers from the Eighth International Conference on Austronesian Linguistics, ed. by E. Zeitoun and Paul Li, 423-453. Taipei: Institute of Linguistics (Preparatory Office), Academia Sinica.
Hopper, Paul. 1979. Aspect and foregrounding in discourse. Discourse and Syntax, ed. by Talmy Givón. New York: Academic Press.
Hopper, Paul. 1982. Aspect between discourse and grammar. Tense-aspect: Between Semantics and Pragmatics, ed. by P. Hopper, 3-18. Amsterdam: John Benjamins.
Hopper, Paul. 1986. How ergative is Malay? Studies in Austronesian Linguistics, ed. by Richard McGinn, 441-454. Athens, Ohio: Ohio University Press.
Hopper, Paul, and Sandy Thompson. 1980. Transitivity in grammar and discourse.
Language 56:251-299.
Huang, Lillian M. 1993. A Study of Atayal Syntax. Taipei: The Crane Publishing Co.
Huang, Lillian M. 1995. A Study of Mayrinax Syntax. Taipei: The Crane Publishing Co..
Huang, Shuanfan. (In preparation). A Functional Grammar of Tsou.
Li, Paul Jen-kuei, Cheng-hwa Tsang, Ying-kuei Huang, Dah-an Ho, and Chiu-yu Tseng (eds.). 1995. Austronesian Studies Relating to Taiwan. Taipei: Institute of History and Philology, Academia Sinica.
Li, Paul Jen-kuei. 1997. A syntactic typology of Formosan languages—Case markers on nouns and pronouns. Chinese Languages and Linguistics IV: Typological Studies of Languages in China, ed. by Chiu-yu Tseng, 343-378. Taipei: Institute of History and Philology, Academia Sinica.
Li, Paul Jen-kuei. 1999. The Linguistic History of Formosan Aborigines. Nantou: The Historical Research Commission of Taiwan Province. (in Chinese)
Payne, Thomas. 1994. The pragmatics of voice in a Philippine language: Actor-focus and goal-focus in Cebuano narrative. Voice and Inversion, ed. by Talmy Givón.
Amsterdam: John Benjamins.
Ross, Malcolm. 1995. Proto-Austronesian verbal morphology. Austronesian Studies Relating to Taiwan, ed. by Li et al., 729-765. Taipei: Institute of History and Philology, Academia Sinica.
Shibatani, Masayoshi. 1988. Passive and Voice. Amsterdam: John Benjamins.
Starosta, Stanley. 1988. A grammatical typology of Formosan languages. Bulletin of the Institute of History and Philology 59:541-576.
Starosta, Stanley. 1995. A grammatical subgrouping of Formosan languages.
Austronesian Studies Relating to Taiwan, ed. by Li et al., 683-726. Taipei: Institute of History and Philology, Academia Sinica.
Starosta, Stanley. 1997. Formosan clause structure: Transitivity, ergativity and case marking. Chinese Languages and Linguistics IV: Typological Studies of Languages
693
in China, ed. by Chiu-yu Tseng, 125-154. Taipei: Institute of History and Philology, Academia Sinica.
Starosta, Stanley, Andrew Pawley, and Lawrence Reid. 1982. The evolution of focus.
Papers from the Third International Conference on Austronesian Linguistics, vol. 2:
Tracking the Travelers, ed. by A. Halim et al., 145-170. Canberra: Pacific Linguistics.
Tseng, Chiu-yu (ed.). 1997. Chinese Languages and Linguistics IV: Typological Studies of Languages in China. Symposium Series of the Institute of History and Philology, Academia Sinica, No.2. Taipei: Institute of History and Philology, Academia Sinica.
Wouk, Fay. 1986. Transitivity in Toba Batak and Tagalog. Studies in Language 10:391-424.
Wouk, Fay. 1996. Voice in Indonesian discourse and its implications for theories of the development of ergativity. Studies in Language 20:361-410.
Wouk, Fay. 1999. Sasak is different: A discourse perspective in voice. Oceanic Linguistics 38:91-114.
[Received 7 November 2001; revised 31 March 2002; accepted 15 April 2002]
Graduate Institute of Linguistics National Taiwan University 1, Sec. 4, Roosevelt Road Taipei 106, Taiwan sfhuang@ms.cc.ntu.edu.tw
694
鄒語與賽德克語焦點系統的語用研究