<p> 1</p><p>Tenses and discourse connectives 1 Tenses and discourse connective as illocutionary force indicators It is common wisdom at least since the 1990’s that discourse connective do not merely bear on propositional contents, but also on more abstract objects of discourse, in particular speech acts. Jayez & Rossari (1997), Rossari & Jayez (2000) showed that consequence discourse connectives such as donc combined only with certain types of speech acts. In this sense, they can be seen as illocutionary force indicators (IFIs), to use Bierwisch’s (1980) terminology. Now Caudal & Roussarie (2002) argued that tenses should also be regarded as IFIs, because they did not allow for any type of illocutionary force. If indeed that is true, then one would expect tenses and discourse connectives to interact in terms of illocutionary force. The idea that tenses are IFIs can be shown to be related the distinction made by Benveniste (1966) between histoire and discours. Indeed, Benveniste used notably the distribution of the passé composé and the passé simple to characterize these two notions (see also Weinrich’s (1964) related proposals) – the passé simple being caracteristic of histoire discourses (narratives), while the passé composé is one of the dominating tenses in discours. Crucially, Benveniste’s opposition between histoire and discours can be shown to be at least weakly related to the notion of illocutionary force. For instance, to use some SDRT (Segmented Discourse Representation Theory, cf. Asher & Lascarides 2003) concepts, histoire involves mostly (narrative) discourse relations such as Narration or Elaboration, with relatively few discourse connectives. In contrast, discours narrative texts will rather rely on relations such as Explanation or Result, together with an abundance of discourse connectives. And since discourse relations are relational speech act functions (Asher & Lascarides 2003), the connection between histoire or discours and illocutionay force is rather obvious. I will not try and pursue a textual approach to tenses in the spirit of Benveniste, unlike Smith (2003). Instead, adopting the SDRT framework, I take the computation of discourse relations between utterances to be the main concern for a theory of discourse. So the question I would like to answer is whether tenses and discourse connectives interact with discourse relations in a way that makes sense with respect to their putative illocutionary content. However, I will retain the insight offered by textual strategies about the role of tenses. I will notably focus here on the passé composé and passé simple in French. The thesis defended by textual strategies (Benveniste 1966, Guillemin-Flescher 1981…) about these two tenses can be summarized as follows : the passé composé, as a so-called deictic tense, relates a propositional content to the hic et nunc of the speaker ; hence its so-called ’speaker-centered’ flavour, associated with a strong illocutionary content. In contrast, the passé simple as a so- called anaphoric tense, rather relies on propositional content information (e.g., causal relations) to structure discourse. But should we understand the distinction between the passé composé (PC) and the passé simple (PS) as one between a speech act-level operator and a propositional content-level operator ? I will try and show that the answer to the latter question is – no, by defining precisely the respective illocutionary function of these two tenses. Recent works about the passé composé in the SDRT framework, i.e., Molendijk & de Swart (2002), proposed an interesting account in terms of propositional contents solely ; although I believe these results to be important, I also believe that they do not suffice for us to be able to grasp the essence of the contrast between the PC and the PS. 2 Some elements about the illocutionary force of the passé simple/composé Some well-known facts about the PC/PS contrast take an altogether different meaning if we consider these tenses as IFIs. Thus, the PS is known to reject ‘genuine’ causally-reverse relations, while the PC accepts them, (0) (cf. Molendijk & de Swart 1999, de Saussure 2000). (0) a. La maîtresse a giflé l’élève. Il est arrivé en retard. b. La maîtresse gifla l’élève. #Il arriva en retard. 2</p><p>(0) shows that these tenses are not equally compatible with the Explanation relation (and the corresponding relational speech act type), and are therefore different IFIs. Caudal & Roussarie (2003) noted another related difference : unlike the PC, the PS rejects ‘genuine’ conditional/hypothetical readings, cf. (0b) vs. (0a). (0) a. Si Yannig est venu, Mona viendra aussi. b. *Si Yannig vint, Mona viendra aussi. Again, this shows that the PC and PS receive very different illocutionary functions. It seems that the semantics and pragmatics of the PS are more restrictive than that of the PC. 3 Donc and the passé simple vs. passé composé : evidence for tenses as IFIs But the most conclusive evidence concerning the analysis of PC and PS as IFIs is to be found in their distribution with consequence discourse connectives. According to Rossari & Jayez (2000), consequence discourse connectives such as donc are sensitive both to propositional contents and to illocutionary force. Thus while in (0), it is rather illocutionary information (i.e. the nature of the speech acts/discourse relations at stake) that blocks donc, it seems that propositional content information explains the impossibility of donc in (0).</p><p>(0) Ouvre la fenêtre ! ??Donc / ??Alors / ??Par conséquent / ??Du coup / ??De ce fait / (??Aussi) il fera (fera-t-il) moins chaud. (0) Je n’ai pas lu le journal ce matin. ??Donc / ??Alors / ??Par conséquent / ??Du coup / ??De ce fait / ??Aussi est-ce que la SNCF est encore en grève ? Moreover, donc has been shown to convey an evaluation made by of the speaker – contrast donc with du coup in (0) (Degand & Pander Maat 1999) (du coup lacks such an evaluation). (0) J’étais en retard, donc/du coup j’ai pris le sens interdit. This speaker-centered property of donc is highlighted by certain discourses, such as mathematical demonstrations. Interestingly, the PS is incompatible with the illocutionary force associated with utterances such as (0b), suggesting that it is not speaker-centered : (0) a. On a vérifié HR(1),HR(2),HR(3), donc on a fondé la récurrence. (web corpus) b. ??On vérifia HR(1),HR(2),HR(3), donc on fonda la récurrence. Given that donc allows for so-called abductive discourses, one would expect it to be compatible with abductive discourses regardless of tense information if tenses are not IFIs. But it is not the case, since only the PC clearly allows for abductive readings (cf. (0) and (0)). (0) a. Il reçut un pot de fleurs sur la tête. #/ ??Donc il passa sous un balcon. b. Il a reçu un pot de fleurs sur la tête. Donc il est passé sous un balcon. (0) a. Le meurtrier n'a pu entrer que par la porte donnant sur le hall commun, reprit-il. Elle n'a pas été fracturée. Il a donc sonné. Angus a reconnu son visiteur et l'a lui-même introduit. (web) b. Le meurtrier ne put entrer que par la porte donnant sur le hall commun, reprit-il. Elle ne fut pas fracturée. ??Il sonna donc. Angus reconnut son visiteur et l’introduisit lui-même. Converselly, certain constructions involving donc and receiving a very special kind of (narrative) illocutionary force require the PS and reject the PC : (0) a. Arriva donc ce qui devait arriver: le canot frappa de plein fouet une grosse épinette. (web) b. ??Est arrivé donc ce qui devait arriver… Starting from such data, I intend to show that the PC and the PS differ not only with respect to their respective propositional (semantic) content, but also with respect to their illocutionary (pragmatic) function, these two interpretational levels of tenses constraining each other (see Vetters (2001) for a related conception concerning the PS). I will also offer an SDRT implementation of what I take to be essentially an issue at the semantics/pragmatics interface. 3</p><p>References Asher, N. & A. Lascarides (2003), Logics of Conversation, Cambridge University Press, Cambridge. Benveniste, E. (1966), Problèmes de linguistique générale, Gallimard, Paris. Bierwisch, M. (1980), "Semantic structure and illocutionary force", in J.R. Searle, F. Kiefer & M. Bierwisch (eds.), Speech Act Theory and Pragmatics, Reidel, Dordrecht, 1–35. Carlier, A., V. Lagae & C. Benninger (eds.), Cahiers Chronos 6, Rodopi, Amsterdam/Paris. Caudal, P. & L. Roussarie (2002), "Aspectual viewpoints, speech act functions and discourse structure", communication à NSF-funded workshop on the syntax, semantics & acquisition of aspect, University of Iowa, Iowa City (IA), U.S.A ; to appear in P. Kempchinsky & R. Slabakova (eds.), Kluwer, Dordrecht. Caudal, P. & L. Roussarie (2003), "Sémantique et pragmatique des propositions en Si", paper presented at Chronos’5; to appear in Cahiers Chronos. Degand, L. & Pander Maat, H. (1999), "Scaling causal relations in terms of speaker involvement", LORID 99, 45-53. Guillemin-Flescher, J. (1981), Syntaxe comparée du français et de l’anglais, Ophrys, Gap. Jayez, J. & C. Rossari (1997), "Connecteurs de conséquence et portée sémantique", Cahiers de Linguistique Française, 19, 233-266. Molendijk, A. & H. de Swart (1999), "L’ordre discursif inverse en français", in W. de Mulder & C.Vet (eds.), Temps Verbaux et Relations Discursives, Travaux de linguistique, 39, pp. 77-96. Rossari, Corinne & Jacques Jayez (2000), "Du coup et les connecteurs de conséquence dans une perspective dynamique", Linguisticae Investigationes XXIII:1, pp. 303-326. de Saussure, L., (2000), "Quand le temps ne progresse pas avec le passé simple", in Carlier et al. (eds.), pp. 37-48. Smith, C. (2003), Modes of Discourse, Cambridge University Press, Cambridge. de Swart, H. and A. Molendijk (2002), "Le passé composé narratif: une analyse discursive de L’Etranger de Camus", in B. Laca (ed.), Temps et Aspect – De la morphologie à l’interprétation, Presses Universitaires de Vincennes, Saint-Denis, pp. 193–211. Vetters, C. (2001), "L’aspect global – un effet secondaire d’un contenu procedural ?", in S. Mellet et M. Vuillaume (eds.), Modes de repérages temporels, Cahiers Chronos 11, pp. 113-131. Weinrich, H. (1964), Tempus. Besprochene und erzählte Welt, Kohlhammer, Stuttgart</p>