A Unified Account of the Semantics of Discourse Particles

A uniﬁed account of the semantics of discourse particles

Markus Egg Humboldt-Universitat¨ Berlin [email protected]

Abstract This relation is described by the features KNOWN and CORRECTION in Thurmair (1989), The paper investigates discourse particles i.e., doch-utterances correct a previous utterance on the example of German doch, assigning by introducing old information. Karagjosova to them very specific semantic interpreta- (2004) regards doch-utterances act as reminders, tions that still cover a wide range of their which present old information to hearers. Ac- uses. cording to Konig¨ and Requardt (1997), doch- The analysis highlights the role of dis- utterances point out inconsistencies between old course particles in managing the common information and a new piece of information or ac- ground and crucially takes into account tion. that discourse particles can refer not only Such general descriptions of doch apply to cases to utterances they are a part of and to previ- like Karagjosova’s (1): B reminds A of Peter’s ously uttered utterances, but also to felicity illness, which seems inconsistent with A’s an- conditions of these utterances. nouncement and therefore can act as a correction of A: 1 Introduction (1)A: Peter wird auch mitkommen. B: Er ist This paper advocates very specific semantic in- doch krank. terpretations for discourse particles, concentrating ‘A: Peter will come along, too. B: But he is on German doch. There is a very wide range of ill.’ concrete usages of discourse particles in context While these general descriptions (excepting (which has motivated analysing them as polyse- Karagjosova (2004)) do not spell out in detail the mous, e.g., in Helbig (1988)). way in which doch contributes to the meaning of Assigning them a uniform semantic interpreta- larger discourses, they can capture a wide range of tion seems to be subject to two conflicting require- uses of these particles. ments: There remain a number of problematic cases, including discourse-initial uses of doch-utterances • the interpretation must be sufficiently specific like Konig¨ and Requardt’s (2), which functions as to allow deriving the interpretation of con- an opening line in a conversation, it neither cor- crete uses rects nor reminds the hearer, nor is there an incon- • it must be sufficiently general to cover a wide sistency between the utterance and the context: range of concrete uses (2) Sie sind doch Paul Meier. ‘You must be Paul Meier.’ So far, research on the interpretation of doch focusses on the second requirement (e.g., Thur- The proposed analysis of the particle doch is mair (1989), Konig¨ (1997), Karagjosova (2004), sufficiently general to account for a wide range of or Konig¨ and Requardt (1997)). uses yet being specific enough to specify the se- The meaning of doch emerges as a two-place mantic construction for discourses that comprise relation between the utterance doch is a part of and doch. a previous utterance to which the doch-utterance is I follow much previous work in developing my a reaction. analysis on the basis of simple examples like (1),

Proceedings of SIGDIAL 2010: the 11th Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 132–138, The University of Tokyo, September 24-25, 2010. c 2010 Association for Computational Linguistics

132 and then extending it to cases like (2). Most ex- problems of defining the semantics of the particles amples consist of two utterances, the second utter- in the literature, which is characteristically based ance comprises a discourse particle and is a reac- on the meanings of a- and p-utterance. tion to the first one. In the remainder of this pa- My claim is that there is a link between a- and p- per, these two utterances are called ‘involved ut- proposition and a- and p-utterance, respectively, in terances’. that the propositions can either be the meanings of In (1), the (propositional) semantic arguments the utterances or emerge through the felicity con- of the particle are the meanings of these two utter- ditions of the utterances. ances. But the semantic arguments of a discourse E.g., in (3) the doch-proposition reminds A of particle may differ from the meanings of the in- the fact that the first preparatory condition for a volved utterances, as illustrated by (3) (from Thur- question (that the speaker does not know the an- mair 1989): swer) does not hold, since A (as the one who gave the book to B) should know since when the book (3)A: Seit wann hast du denn den has been in B’s possession. ,,Zauberberg“? B: Den hast du mir doch vor The plan of the paper is to introduce back- zwei Jahren geschenkt. ground assumptions on discourse particles in sec- A: ‘Since when have you owned the tion 2, to apply the proposed approach to the (un- ‘Zauberberg’? B: But you gave it to me two stressed) particle doch in section 3, and to con- years ago.’ clude with an outlook on further research.

B reacts to the implicit statement that A does not 2 Formal background know the answer to his question. This statement is an argument of doch in (3), even though it is not This paper follows much previous work in assum- the meaning of A’s utterance. This shows that the ing that discourse particles refer to the common semantic arguments of discourse particles must be ground (CG), e.g., Konig¨ (1997), Karagjosova distinguished from the meanings of their involved (2004), or Zimmermann (2009). utterances. Common ground and the interlocutors’ individ- Utterances with a discourse particle and pre- ual backgrounds are modelled as common or in- ceding utterances to which they react are called dividual belief (Stalnaker, 2002). Individual be- ‘p(article)-utterances’ (or ‘doch-utterances’) and lief is equated with the set of propositions that are ‘a(ntecedent)-utterances’. They are distinguished true in all possible worlds compatible with the in- from the semantic arguments of the particle, dividual’s beliefs; common belief, with the set of which are referred to as ‘p-proposition’ and ‘a- propositions believed by all members of the re- proposition’, respectively. spective group of believers. This is not just a question of nomenclature but Stalnaker notes that this is an idealisation in that reflects a fundamentally different view on the role the CG might comprise propositions not shared by of discourse particles. Instead of indicating the re- the background of every member of the group. But lation between two already identified propositions, this idealisation is not a problem for the analysis the meaning of the particle applied to its first ar- presented in this paper. gument (very often but not always the interpreta- Reasoning on CG and individual backgrounds tion of the p-utterance) determines the range of po- often uses defeasible deduction (Asher and Las- tential a-propositions in the context of utterance. carides, 2003). I.e., from statements of the form From this range, the hearer selects the appropriate ‘p defeasibly entails q’(p > q) together with p proposition. one can defeasibly deduce q. This resembles the intuition of Konig¨ and This defeasible Modus Ponens applies if ¬q Requardt (1997) that discourse particles are does not hold and ¬q cannot be deduced simulta- ‘metapragmatic instructions’ which tell hearers neously (Asher and Lascarides, 2003). Defeasible how to deal with the p-utterance in a communica- deducability of p from a set of propositions C is tive situation. written as ‘C |∼ p’. Consequently, a- and p-utterances do not deter- Reference to the common ground makes the se- mine the semantic arguments for all uses of dis- mantics of discourse particles context-dependent, course particles, which might account for some because the CG is relative to the interlocutor(s) of

133 a- and p-utterances. This shows up in the shifting Formally, doch states that the common ground effects observed in Doring¨ (2010). Consider e.g. C defeasibly entails q and the fact that q defeasi- what happens if one embeds (1) in a quotation like bly entails ¬p (which by defeasible Modus Ponens in (4): would allow one to infer ¬p, if the conditions for defeasible Modus Ponens are met): (4) A sagte, Peter komme auch mit. B entgegnete, er sei doch krank. (7) [[doch]](q)(p) iff C |∼ q ∧ C |∼ q > ¬p ‘A said Peter would come along, too. B ¨ retorted that he was ill.’ This analysis differs from the one of Konig (1997), who assumes that doch q points out a con- The shift in (4) arises because doch presents a tradiction in the CG, in that p is incompatible with proposition (here, that Peter is ill) as part of the a consequence of q. In contrast, I regard this common ground, and the relevant CG is calcu- incompatibility as a default only. The status of lated with respect to A and B, not with respect to q as derivable from the CG is also expressed in the interlocutors of (4). I.e., (4) does not express Karagjosova (2004) claim that doch introduces q that Peter’s illness is in the common ground of the as a reminder and in Thurmair’s (1989) feature speaker and hearer of (4). KNOWN. In (5) and (6), p and q are the semantics of the a- 3 The analysis and the doch-utterance. Being ill is a potential impediment for going out, so, by pointing out Peter’s The proposed approach to discourse particle is illness in (5), B expresses surprise or disbelief at now applied to doch, which introduces a notion A’s announcement but does not necessarily correct of tension between the a- and the p-proposition. it or refute it, because even ill people can go out in 3.1 Declarative a- and p-utterances principle. I will ﬁrst illustrate this notion with simple ex- Similarly, the speaker of (6) is surprised at his amples in which the a-utterance expresses the a- frequent illness, in spite of his healthy lifestyle. q q proposition, and the meaning of the declarative p- (6) shows that is only a default impediment: If p utterance provides the p-proposition. and were contradictory, (6) would be nonsensi- In (5) [= (1)] and (6), adapted from Karagjosova cal, but, intuitively, it is not. (2004), there is tension between being ill on the The use of defeasible implications to model the one hand and going out and living healthily on the tension between two propositions as indicated by doch other hand, respectively: is closely related to accounts of the discourse relation of CONCESSION in Grote et al. (1997), (5)A: Peter wird auch mitkommen. B: Er ist Oversteegen (1997), Lagerwerf (1998), and Knott doch krank. (1996). ‘A: Peter will come along, too. B: But he is They assume the same kind of defeasible impli- ill.’ cation for this discourse relation and model it as a presupposition, which is compatible with giving it (6) Ich bin oft krank. Dabei lebe ich doch common ground status. gesund. ‘I am often ill. But I have a healthy lifestyle.’ 3.2 Non-declarative a-utterances In (5) and (6), the a-proposition enters the CG The intuitive notion of tension between two as the meaning of an a-utterance. But the a- p q propositions and is formalised as defeasible proposition can also emerge as a felicity condition q > ¬p q entailment . I.e., given , one would ex- of the a-utterance. Consider e.g. doch-utterances p pect , but the propositions are not incompatible, as reactions to questions, as in (8) [= (3)]: even though q is a potential impediment for p. The effect of doch q as a reaction to an a- (8)A: Seit wann hast du denn den proposition p against the common ground C is to ,,Zauberberg“? B: Den hast du mir doch vor remind the hearer that C comprises a potential im- zwei Jahren geschenkt. pediment q for p, which expresses either surprise A: ‘Since when have you owned the at the fact that p nevertheless holds or puts doubt ‘Zauberberg’? B: But you gave it to me two on p. Still, p is not explicitly denied. years ago.’

134 Doch in (8) expresses surprise at the question This use of doch also shows up in reactions being asked, since A himself gave the book to B to declarative statements: The p-utterance of (10) and hence should know that B owns it. states no potential impediment for the proposition The proposed analysis reconstructs this intu- expressed by A. ition: B’s utterance expresses a proposition q (that Rather, B’s use of doch refers to A’s surprise, A gave the book to B) and points out that q is part suggesting that A should not be surprised at all. of the CG. It is also part of the CG that q is a po- The felicity condition of expressing surprise that tential obstacle for a speciﬁc a-proposition p (for- is cast into doubt by B is considering the fact mally, the CG entails q > ¬p). about which one is surprised as something extraor- Such p-utterances restrict the range of poten- dinary, which would not have happened in a nor- tial a-propositions p, and their hearers try to iden- mal course of events. tify the a-propositions in the given context. The a-utterance in (8), however, cannot directly con- (10)A: Peter sieht schlecht aus. B: Er war doch tribute p in any context, since its meaning is not a lange im Krankenhaus. proposition. ‘A: Peter does not look healthy. B: But he But due to the assumption that A is coopera- has been in hospital for a long time.’ tive, the question introduces into the CG the felic- Peter’s long stay in the hospital (= q) is no ity conditions for questions, among them the ﬁrst potential obstacle to looking unhealthy, on the preparatory condition, viz., that A does not know contrary, it entails defeasibly that his looking un- the answer to his question. This is a suitable p, be- healthy is quite normal (= ¬p). This would negate cause it is reasonable to assume that if A gave the the abovementioned felicity condition for A’s sur- book to B (= q), he should know the answer to the prise (= p), hence suggests that A should not be question (= ¬p). surprised. The intuition that the semantic arguments of discourse particles need not be identical to the 3.3 Non-declarative p-utterances meanings of the involved utterances is related to Another group of doch-utterances are imperative suggestions to let discourse relations relate ei- or interrogative (the latter adapted from Thurmair ther to the content of the discourse segments that (1989)): they link or to the corresponding intensions of the speaker or the intended effencts on the hearer, (11) Verklag mich doch! which is suggested by Sweetser (1990) and Knott ‘Go ahead and sue me.’ (2001). Doch-utterances in reaction to imperatives work (12) Komm doch nach Hause! analogously, e.g., (9): ‘Do come home.’

(9)A: Ubersetze¨ mir bitte diesen Brief. B: Ich (13) Wie heißt doch diese Kneipe in der kann doch kein Baskisch. Sredzkistraße? A: ‘Please translate this letter for me.’ B: ‘What is the name of this pub in the ‘But I do not know Basque’. Sredzkistraße?’ (14) Wie sagt Goethe doch so treffend? Here B’s lack of proficiency in Basque (= q) ‘What was this piercing remark of Goethe and A’s belief that B can translate a Basque letter again?’ (i.e., the first preparatory condition of the request, our p) are in tension. (15) Du kommst doch? Now q can be deduced from the common ‘I presume that you will be there.’ ground either because it has been explicitly in- troduced before or because it makes sense to as- Doch is used provocatively in imperatives like sume defeasibly that someone does not speak a (11); it suggests that the hearer cannot fulfil the less known language like Basque. In either case, A request. In cases like (12), doch signals that the re- should not take for granted that B speaks Basque, quested or suggested action is a very natural thing i.e., has a reason not to require B to translate letters to do. Doch-questions refer to a piece of knowl- written in Basque. edge that the speaker knows or is supposed to

135 know (Thurmair, 1989), e.g., (13) indicates that an option yet. Instead, visiting the speaker is pre- the speaker knows the answer at least in principle, sented as a very natural thing to do for the hearer, (14) announces a quotation, and (15) suggests that i.e., once more the second preparatory condition the answer can only be affirmative. of a request does not hold. There are two issues in interpreting these sen- Using doch in questions also indicates that a tences; the p-utterance does not denote a propo- felicity condition of the utterance does not hold, sition (which could be the semantic argument of even though its validity could be deduced defeasi- doch), and there need not be an a-utterance at all bly from the fact that the question has been asked. from which to derive the a-proposition. The relevant condition is the first preparatory con- But in all these utterances, speakers use doch to dition for questions (that the speaker does not point out that they are aware of evidence from the know the answer already). CG which suggests that a felicity condition of the Doch signals that this condition is not fulfilled, utterance itself does not hold. This can be mod- either because the answer escapes the speaker only elled by identifying the p-proposition q (the argu- momentarily, as in (13), because he obviously ment of doch) with the fact that the sentences were knows, as in the conventionalised announcement uttered, which can be (trivially) deduced from the (14), or because he would not accept a refusal, common ground C (the condition C |∼ q in (7)). which settles the question, like in (15). Then the felicity conditions associated with dif- The analysis predicts that doch is not acceptable ferent kinds of illocutionary acts emerge from the in ordinary questions, which is borne out e.g. by common ground C as default entailments from the (17), because in these questions there is no ten- utterance of the respective illocutionary type (the sion between uttering the question and potential condition C |∼ q > ¬p in (7); here ¬p refers to obstacles for its felicity conditions: one of the felicity conditions). (17)* Wer schreibt dir doch? I.e., using doch in these cases triggers a search ‘But who is corresponding with you?’ for a suitable a-proposition p in the CG which negates a felicity condition of the utterance. E.g., Rhetorical questions are also incompatible with doch in (11) shows that the first preparatory con- doch, but for a different reason. Consider e.g. dition of a request (the speaker believes that the the contrasting dialogue pairs (18a)/(18b) and hearer can do it) does not hold, even though this (18a)/(18c): condition follows defeasibly from the fact that the request was made. (18) (a)A: Ich werde meinen 30. Geburtstag In (12), doch addresses the second preparatory mit einem großen Fest feiern. condition of a request or advice (that it is not obvi- A: ‘I’ll throw a big party on the ous to speaker and hearer that the hearer complies occasion of my 30th birthday.’ with the request in a normal course of events). (b)B: Es wurde¨ doch keiner zu deinem Fest Thus, doch suggests that it is obvious that the kommen. hearer will do anyway what is requested or ad- B: ‘But no one would come to your vised, even though uttering (12) defeasibly entails party.’ the contrary. Consequently, (12) presents a request (c)B:* Wer wurde¨ doch zu deinem Fest or advice as a very natural thing to do. kommen? I.e., doch-imperatives do not correct unwanted B: ‘But who would come to your behaviour by the hearer (pace Thurmair (1989)), party?’ which is confirmed by examples like (16), which can be uttered between future lovers during their The inacceptability of (18a)/(18c) - and of the courtship to take the process of courting one step rhetorical doch-question in particular - is not due further: to the function of the rhetorical question as a negated statement: In this case, (18b) should be (16) Komm doch mal vorbei! an inacceptable response to (18a), too. ‘Just drop by.’ (18c) is inacceptable because rhetorical questions characterise statements as CG information (16) does not request the hearer to change his (Egg, 2007). This is also one of the effects of behaviour, because calling on the speaker was not doch; consequently, (18c) is as informative as

136 (18b) but more complex, hence, its use would Finally, the sincerity condition of a statement not comply to conversation maxims (Grice, 1975; can also be targeted by doch: Krifka, 1989). To sum up, non-declarative doch-utterances re- (23) Da sagt er doch im letzten Moment ab! fer to their own felicity conditions; since they do ‘I can’t believe that he cancelled the not denote propositions, the first semantic argu- appointment at the last moment.’ ment of doch cannot be the meaning of the doch- In (23), doch expresses disbelief of the speaker, utterance. he cannot believe what he is saying. This violates Instead, doch applies to the fact that the speaker the sincerity condition for statements. The effect uttered the sentence. In contrast, declarative doch- of doch here is one of expressing surprise. utterances like in (8) or (9) refer to a felicity con- The same effect shows up in exclamative wh- dition of the non-declarative a-utterance. sentences: This analysis of non-declarative doch- utterances also applies to the hitherto extremely (24) Wie schon¨ Amelie´ doch ist! problematic group of discourse-initial doch- ‘How beautiful Amelie´ is!’ utterances: Following analyses of these sentences like (19) Morgen fahre ich doch nach Wien. Zanuttini and Portner (2003) or Rett (2009), (24) ‘Well, I’ll go to Vienna tomorrow.’ characterises the degree of Amelie’s´ beauty as un- expectedly or surprisingly high. Hence, doch nat- (20) Du hast doch ein Auto. urally occurs in these exclamatives to deny a belief ‘Well, you have a car.’ of the speaker in what he is stating. These examples are characterised by doch as a In sum, I offered a uniform semantic analysis reminder. This means that the p-utterance (the of doch, which still covers a wide range of its us- speaker’s travel plans or the fact that the hearer has ages. Doch relates two propositions p and q iff q is a car) contributes information semantically that is derivable from the common ground as well as the already in the CG. However, this information is fact that q defeasibly implies ¬p, i.e., q presents a not obviously in tension to any other information. potential impediment for p. This raises the question of what the semantic ar- The correlation of p and q with utterances is guments of doch are in these cases. flexible, however: Often q is the meaning of Here doch addresses the first preparatory condi- the doch-utterance, but for non-declarative and tion for statements, viz., that it is not obvious to discourse-initial declarative doch-utterances, q is the speaker that the hearer already knows what the the fact that this utterance has been made. speaker will say. Uttering the statement (= q) de- The proposition p can be the meaning of a pre- feasibly implies this condition (= ¬p), but accord- ceding a-utterance to which the doch-utterance is ing to the CG the speaker knows that the hearer a reaction. But especially for non-declarative a- knows (= p). utterances, p can also be one of its felicity con- (21) [= (2)] instantiates this case, too: ditions, or, for discourse-initial doch-utterances, a felicity condition of the utterance itself. (21) Sie sind doch Paul Meier. ‘You must be Paul Meier.’ 4 Conclusion and outlook

Telling someone his name obviously violates The paper outlined a research programme for dis- the first preparatory condition for statements, course particles that captures their meanings in whence the use of doch. very specific semantic descriptions that neverthe- Another such case is the use of doch in expres- less account for the wide range of their uses. These sions of outrage. Here doch signals that it is com- two competing goals can be pursued simultane- mon knowledge that the hearer knows that the sit- ously because doch-utterances can be integrated uation or action to which the speaker refers is out- flexibly into the meaning of the surrounding dis- rageous: course. While discourse particles like doch uniformly (22) Das ist doch die Hohe!¨ relate two propositions semantically, the meaning ‘That is the limit!’ of the utterance of which the particle is a part, and

137 the meaning of the utterance to which this ﬁrst ut- Ekkehard Konig¨ and Susanne Requardt. 1997. A terance reacts are not the only feasible semantic relevance-theoretic approach to the analysis of arguments of the particles: They can also have fe- modal particles. Multilingua, 10:63–77. licity conditions of these two utterances as seman- Ekkehard Konig.¨ 1997. Zur Bedeutung von Modal- tic arguments. partikeln im Deutschen: Ein Neuansatz im Rah- This research programme was illustrated by in- men der Relevanztheorie. Germanistische Linguis- tik, 136:57–75. vestigating the particle doch. The next steps now are to extend the coverage of this analysis to other Manfred Krifka. 1989. Nominalreferenz und Zeitkon- particles, in particular, schon, and to contrast ‘min- stitution. Fink, Munchen.¨ imal pairs’ of discourses which differ only by dis- Luuk Lagerwerf. 1998. Causal connectives have course particles (e.g., Komm schon! as opposed presuppositions. Effects on coherence and discourse to Komm doch!, which both require the hearer to structure. Ph.D. thesis, Universiteit van Tilburg. come). Leonoor Oversteegen. 1997. On the pragmatic nature This analysis can also be used for investigations of causal and contrastive connectives. Discourse of stressed and unstressed forms of discourse par- Processes, 24:51–85. ticles and of the relation between them. Here it is Jessica Rett. 2009. A degree account of exclamatives. particularly interesting to take prosody seriously In Proceedings of SALT XVIII, pages 601–608. CLC and to look into the semantic effects of emphasis- Publications. ing a discourse particle. Robert Stalnaker. 2002. Common ground. Linguistics & Philosophy, 25:701–721.

References Eve Sweetser. 1990. From etymology to pragmat- Nicholas Asher and Alex Lascarides. 2003. Logics ics. Metaphorical and cultural aspects of semantic of conversation. Cambridge University Press, Cam- structure. Cambridge University Press, Cambridge. bridge. Maria Thurmair. 1989. Modalpartikeln und ihre Kom- Sophia Doring.¨ 2010. On context shift in German dis- binationen. Niemeyer, Tubingen.¨ course particles. BA thesis, Humboldt-Universitat¨ Raffaela Zanuttini and Paul Portner. 2003. Exclama- Berlin. tive clauses: At the syntax-semantics interface. Lan- Markus Egg. 2007. Meaning and use of rhetorical guage, 79:39–81. questions. In Maria Aloni, Paul Dekker, and Floris Roelofsen, editors, Proceedings of the 16th Amster- Malte Zimmermann. 2009. Discourse particles. dam Colloquium, pages 73–78. Universiteit van Am- In Claudia Maienborn, Klaus von Heusinger, and sterdam, ILLC. Paul Portner, editors, Semantics: an international handbook of natural language meaning. Mouton de Paul Grice. 1975. Logic and conversation. In Peter Gruyter, Berlin. In press. Cole and Jerry Morgan, editors, Syntax and semantics 3: Speech acts, pages 41–58. Academic Press, New York. Brigitte Grote, Nils Lenke, and Manfred Stede. 1997. Ma(r)king concessions in English and German. Dis- course Processes, 24:87–118. Gerhard Helbig. 1988. Lexikon deutscher Partikeln. Verlag Enzyklopadie,¨ Leipzig. Elena Karagjosova. 2004. The meaning and function of German modal particles. Ph.D. thesis, Univer- sitat¨ des Saarlandes. Alistair Knott. 1996. A Data-Driven Methodology for Motivating a Set of Coherence Relations. Ph.D. thesis, University of Edinburgh. Alistair Knott. 2001. Semantic and pragmatic relations and their intended effects. In T. Sanders, J. Schilper- oord, and W. Spooren, editors, Text representation: linguistic and psycholinguistic aspects, pages 127– 151. Benjamins, Amsterdam.

138