A Dialogic Approach to Pragmatics

Cover Page


This paper focuses on how the limits of pragmatics - as long as it is restricted to the analysis of one utterance at a time - are overcome by including the hearer not only as interpreter who tries to understand the speaker’s utterance but as an interlocutor who tries to come to an understanding with the speaker. The goal of the paper is not to describe and analyze the dialogue approach rather explain what inner developments in the pragmatics paradigm have made it necessary to move in a dialogic direction, specifically emphasizing the importance of evaluating speaker meaning from the perspective of the speaker rather than from the perspective of the hearer and the double role of the interlocutor (speaker-hearer).

1. Introduction Pragmatics is an utterance-based inquiry. However, recently there have been many attempts to expend the boundaries of the discipline to include dialogic sequences or discourse segments to make better understanding of utterance meanings, especially speaker meaning. This endeavor has become especially articulate with dialogic studies, intercultural pragmatics studies and computational pragmatics becoming strong and independent inquiries. Research in intercultural pragmatics (e.g. House 2002; Kecskes 2007, 2013) demonstrated that in intercultural communication participants are creative on discourse level rather than on utterance level. This is mainly due to limited language proficiency that may result in, among others, not-very-well formulated utterances. Consequently, intercultural interactions may require not only a bottom-up, sequential utterance by utterance analysis but also a top-down, holistic dialogue segment analysis if we want to make sure that we understand the message of interlocutors. The other field of inquiry that includes a dialogue approach is computational pragmatics that focuses mainly on two lines of research. First, it investigates the application of pragmatics to dialogue modelling, especially the development of spoken dialogue systems intended to interact with human beings in task-oriented scenarios, such as providing travel information. Second, computational pragmatics also shows how, and discusses why it differs from ‘linguistic’ pragmatics, and how pragmatics proper contributes to the computational analysis of dialogues (e.g. Morante et al. 2007; Bunt 2011). Recently several papers have talked about “narrow pragmatics” and “wide pragmatics” discussing the relationship of pragmatics and discourse analysis (e.g. Puig 2003; Taboada and Mann 2006; de Saussure 2007). The issue of beyond utterances analysis has also been on the agenda of dialogue studies (e.g. Weigand 2000, 2010a, 2010b; Carbaugh 2013; Cooren 2010). The importance of these studies is that they consider language as action that is always something that is shared. Whenever someone appears to act, others also proceed into action. They also emphasize that communicative function and communicative agenda are two separate things. 2. The dialogic approach Because “utterance” is hard to define, and its meaning is both in the building elements of an utterance (lexical units) and in the subsequent utterances produced in response, pragmatics, which is also defined as a theory of meaning in context, has been looking for meaning “ingredients” both inside the utterance and outside the utterance. As a result, there has developed an approach that can be called pragma-dialogue, which calls attention to the dialogic nature of communication by emphasizing that interlocutors are actors who act and react (e.g. Weigand 2010a; Kecskes 2012). So the speaker-hearer not only interprets but also reacts to the other interlocutor’s utterance. The basic dialogic principle is that human beings are dialogic individuals (social individuals) who communicate in dialogic interaction not only by producing and understanding utterances but also by acting and reacting (e.g. Weigand 2010a, 2010b). The dialogic principle defines dialogue as a sequence of actions and reactions. A sequence of speech acts can be considered a dialogue which is a highly structured activity involving (at least) two agents. The structure of real dialogues has been extensively studied by ethnomethodologists (Garfinkel, 1972; Psathas, 1979; Schenkein, 1978; Turner, 1974), who have advocated the necessity of non-quantitative, ethnic methods in the analysis of social interactions. Their work on naturally occurring conversation provides a significant amount of data on the way in which different types of dialogue actually evolve. A dialogue is a sequence of utterances, a reciprocal conversation between two or more entities. Buber believed that genuine dialogue can be obtained between two people “no matter whether spoken or silent” (Buber 1955). Buber further claimed that, “each of the participants really [must have] in mind the other or others in their present and particular being and turn to them with the intention of establishing a living mutual relation between himself and them (Buber 1955: 22)”. More simply, genuine dialogue can be understood as a momentary experience in which participants are consciously aware of the other/s. Preserving the original Buberian idea of dialogue being a reciprocal conversation in which interlocutors are consciously aware of the partners current researchers of dialogue studies (e.g. Weigand 2010a, 2010b; Cooren 2010, Tracy and Craig 2010) go one step forward and emphasize that dialogue is constituted by the interactive purpose of coming to an understanding, which is based on the sequence of actions and reactions as the following conversation demonstrates. (1) Action-directive Bill: Can I get a cup of coffee? Info-request Sara: Milk? Signal-non-understanding Bill: Hm? Info-request Sara: Do you want your coffee black? Agreement Bill: Oh yes, thanks. Weigand (2010b) argued that action and reaction are not two actions of the same type which are arbitrarily connected and only formally distinguished by their position in the sequence. Action and reaction are functionally different types of actions: action is initiative while reaction is responsive. According to Weigand orthodox speech act theory that is exclusively based on illocution should be adjusted. She continues arguing that Searle’s theory of conversation is based on single illocutionary acts that are put together by what Searle (1995: 26) calls “collective intentionality” ignores the interdependence between action and reaction. Although Searle’s collective action aims at a communal goal it is not a dialogic act yet. Weigand claims that individual actions and reactions are organized into a dialogic sequence by their interdependence. Searle’s speech act theory focuses on the communicative functions of speech acts (assertives, directives, commissives, expressives, declaratives) and says less about the communicative agenda of interlocutors. To interpret an utterance properly, the interlocutor has to arrive at an understanding not only of its communicative function but also of the communicative agenda of his/her dialogue partner, i.e. what s/he really wants to achieve in the dialogue. The underlying interests of the dialogue partners become evident through dialogic sequences. This is in line with what the socio-cognitive approach (Kecskes 2010, 2013) claims: occasionally we should go beyond single utterance to understand the communicative agenda of interlocutors. Intercultural communication serves with several examples that support this point. In the following excerpt the Korean speaker is trying to explain the sandy wind they experience in Seoul in the spring. (2) Korean student is talking with an American student 10 NS: In Korea? = 11 NNS: =Yeah::. In (0.2) in ↑spring? 12 NS: Hm mm 13 NNS: There is a (0.5) um (0.7) how- how can I ‘spl- ah:: how can I say::? (0.5) Send wind? 14 (1.7) ((NS displaying mental effort)) 15 NS: Uh:::m 16 NNS: Sen::d (0.2) the wind? 0:30 17 NS: ((now nodding)) Yeah::, it would blow around [the pollen? 18 NNS: [Yes. 19 NNS: Yeah yeah [yeah (0.2) yeah. 20 NS: [(from) the trees? 21 NNS: Ye[ah 22 NS: [Yeah:: 23 NNS: Ah, the:: (0.7) the wind came from China? 24 NS: (0.7) 25 NNS: Do you know that wind? 26 (0.2) 27 NS: ↑No::. 28 NNS: The the:: (0.7) the- many sand? ((“sand” sounds like “send”)) 29 NS: (0.5) ((flashes little smile)) 30 NNS: Sen:d in a: (0.5) desert? 31 NS: Oh, (.) sa:nd. 32 NNS: Yeah, [sand ((shift in pronunciation to match NS)) 33 NS: [Oh:::, [god, yeah 34 NNS: [in deser::t. We’ve had- (0.2)wi:nd (0.2) go to- uh lan- (0.5) come- (0.2) over 35 to Korea: also. (0.5) So in spri:ng, it’s very difficult to (0.5) little bit difficult to (0.7) (?)hhh um:: In this dialogue the Korean speaker has a communicative agenda to explain to her American partner what Koreans call “morae baram” [sandy wind]. However, her pronunciation is not correct and the American partner does not understand what kind of wind the Korean speaker refers to. Not until the Korean connected “sand” with “desert” in line 30 did the American understand what her Korean partner really meant. The action-reaction sequence shows that they could work out the misunderstanding after all. The segment as a whole makes perfect sense and directs our attention to how hard the Korean speaker has worked on making her partner understand what exactly she wanted to say. Airenti et al. (1993) and Bara (2011) argued that a distinction must be made between a global and a local structure of dialogues. Local structure is about what binds utterances together and what makes one utterance coherent (or incoherent) when following another. Turn taking thoroughly studied by conversation analysts (e.g. Sacks, Schegloff, & Jefferson, 1974), is part of the local management system. The relationships among speech acts within a single turn also pertain to the local structure. The global structure determines the flow of conversation. It involves, in particular, the scheduling of dialogue phases, for instance, the opening and closing sections. In conversation analysis the case of telephone calls is often studied, because in those exchanges the general structure of the conversation is especially strictly determined (see, e.g., Schegloff, 1979). Dialogues share a global structure with all kinds of interpersonal activities. The global structure of these dialogues derives from mutual knowledge of an action plan, executed in the course of the activity. According to Bara (2011) cognitive pragmatics views the global structure of dialogues as a structure that derives from sharing the knowledge of an action plan. Consequently, the global structure of a dialogue does not derive from linguistic rules, but from behavior games. 3. Current pragmatic theories Kecskes (2008: 404) argued that in order to give an adequate explanation of communicative processes, we need a dialectical model of pragmatics that combines the perspective of both the speaker and hearer. This change is warranted because current pragmatic theories, both those that have grown out of Grice’s theory, such as the various neo-Gricean approaches and the approach proposed by Relevance Theory appear to be hearer-centered. They derive from the Gricean modular view that divides the interpretation process into two stages: what is said and what is implicated. Although the Gricean theory, with its cooperative principle and maxims, was supposed to embrace conversation as a whole, basically its further development has remained hearer-centered, with less emphasis on interest in the speaker’s position - a rather paradoxical turn, as Grice himself always emphasized speaker’s meaning. Even so, the Gricean divide of truth-conditional semantics and pragmatics has led to a somewhat impoverished speaker meaning, without much regard for the pragmatic features embedded in the speaker meaning. The division between what is said and what is implicated was made for the sake of utterance interpretation and for the sake of distinguishing the semantic meaning from the pragmatic meaning of an utterance. However, a theory that is concerned about the speaker’s [or speakers’] meaning should focus not only on the truth values of the speaker’s utterance, but also on its pragmatic elements and on the speaker’s commitment and egocentrism (in the cognitive sense of the term that will be explained later) deriving from prior experience and dominated by salience. Kecskes (2013) argued that the dominance of the hearer’s perspective in current pragmatic theories can be explained by three facts: 1) the focus on what is said, and the truth values of the speaker’s utterance without due regard for its pragmatics features in the neo-Gricean approaches, 2) emergence of the powerful Relevance Theory, and 3) the partial misinterpretation of the Gricean approach. Although the neo-Griceans’ main concern is the speaker’s meaning, they still view communication as designed with a focus on the recipient and his/her recognition of speaker’s intention. In this approach the speaker designs his/her utterance for the hearer and the hearer’s task is to recognize the speaker’s intention. But what is recovered is not always what was intended because of the interlocutors’ differences in their private cognitive contexts and prior experience (see more discussion in Kecskes 2013). So an adequate account of interaction should consider interlocutors not only as common-ground seekers, but as individuals with their own agendas, with their own prior experience, with their specific mechanisms of saliency (based on prior experiences), and their individual language production systems. This issue is especially important in intercultural communication where interlocutors representing different L1s and cultures cannot be certain how much the speaker-hearers have in common with each other. They may have to create common ground almost from scratch. Consequently, the balance between the individual and societal factors in intercultural interaction is rather lopsided towards the individual factors. However, individuals in these cases also work under the constraints of societal factors but those happen to be different for each participant depending on their L1 and cultural background. This may lead to too much emphasis on individual factors like personality, experience, egocentrism, etc. in intercultural interaction. So we may have an intercultural exchange that is affected by societal factors that are different on each side about how much self-promotion you should do in a job interview (these aren’t individual factors), but it may be further affected by individual differences, such as condescension by one party or defensiveness by one party because of those societal differences (and these are entirely individual factors). The second reason for hearer-centeredness is Relevance Theory. Unlike the neo-Griceans, who attempt to give an account of the speaker’s meaning, relevance theorists focus on developing a cognitive psychological model of utterance interpretation, which does not address the question of how and why the speaker, given what s/he wants to communicate, utters what s/he utters. Saul (2002) said that the main difference between the neo-Gricean theory and Relevance Theory lies in ‘whose meaning’ they model. While the neo-Griceans follow the original perspective and consider utterance meaning, including implicature, to be the speaker’s intended meaning, relevance theorists discuss intentional communication from the perspective of the addressee’s reconstruction of the speaker’s assumptions. There is a third reason why current pragmatics is basically hearer-centered rather than interlocutor-centered. Grice’s implicature is often misinterpreted by pragmaticians. The Gricean implicature is an aspect of speaker meaning; what is said and what is implicated by the speaker. Although Horn (2004) and Bach (2001) called attention to the difference between implicating and inferring, not everybody has listened. Horn said: “Speakers implicate, hearers infer (Horn 2004: 6)”. Bach (2001) argued that “People sometimes confuse infer with imply.... When we say that a speaker or sentence implies something, we mean that information is conveyed or suggested without being stated outright. ... Inference, on the other hand, is the activity performed by a reader or interpreter in drawing conclusions that are not explicit in what is said”. Grice may (unwillingly) also have contributed to this misunderstanding by bringing the “audience” into the explanation of implicature. According to Grice, a speaker’s intending to convey that P by saying that Q is not enough for the speaker to implicate that P. The audience must also need to believe that the speaker believes that P to preserve the assumption of the speaker’s cooperativeness. With this claim Grice attempted to give some degree of intersubjectivity to the notion of conversational implicature. However, speakers have authority over what they utterer-implicate, but they can’t fully control what they conversationally implicate (Saul 2002). The following short conversation (example 3) demonstrates this very well. (3) Roy: - Is there something wrong, Susie? Susie: - I am fine, Roy. Roy: - I would have believed you if you hadn’t said “Roy”. Susie: - OK, OK, just stop…… Susie’s utterance “I am fine, Roy” gives way to different possible interpretations. The expression “I am fine” is a situation-bound utterance, a formulaic phrase. However, Susie added the name of her interlocutor to the formula, which basically broke the formula. It was not a formula any more. Roy had a choice to interpret it as a formula and say nothing. But he understood the speaker’s intention clearly. She wanted to indicate that her words should not be taken at face value. She was, in fact, not fine. Roy got that, and responded accordingly. So the speaker can direct the hearer to the right understanding of his utterance with something additional to the utterance such as discourse markers, gestures, intonation, sentence formulation, etc. From the speaker’s perspective what really matters is what the audience is required by the speaker to believe, not what the audience does actually believe. Davis argued that Grice was wrong to include audience-oriented criteria in his characterization of conversational implicature (Davis 1998: 122). The speaker’s intentions do not depend on what anyone else presumes. “To mean or imply something is to have certain intentions”. Saul (2002) found it important to distinguish between “utterer-implicatures” and “ audience-implicatures”. Utterer-implicatures are claims that the speaker attempts to conversationally implicate (intended by the speaker, but not necessarily recognized by the addressee). Audience-implicatures are claims that the audience takes to be conversationally implicated (recognized by the addressee but not necessarily intended by the speaker). In example (3) the utterer-implicature was recognized by the hearer so the utterer-implicature coincided with the audience-implicature. However, this is not always the case as example (4) demonstrates: (4) Allen and Sherry (of the sitcom “Two and a Half Men”) are sitting in a restaurant. Allan’s right eye is covered with a bandage so he does not see Sherry very well. Allen: - You know, Sherry, I would really like to see more of you. Sherry: - Maybe, we should wait and see how the night goes. Allen: - Oh, no. I mean I have only got one good eye. Can we change places? Sherry: - Sure. The conversation demonstrates that Sherry completely misunderstood Allan’s utterance “...I would really like to see more of you”. This may be due to the fact that she relied exclusively on linguistic salience and ignored perceptual salience in processing the utterance. The utterer-implicature is the literal meaning of the sentence, i.e. Allan expressed his desire to see more of Sherry right then. He wanted to change places hoping that from that angle he would see more of Sherry with his one eye. The hearer-implicature is significantly different because Sherry relied on the figurative meaning of “see” [meet someone romantically] and thought that Allan wanted to be more involved with her romantically. The discrepancy between utterer-implicature and hearer-implicature has caused the misunderstanding between the interlocutors. There have been several attempts in pragmatic theorizing to revise/correct the problems of the modular view and recognize pragmatic features of the speaker’s meaning (e.g. Sperber and Wilson, 1986; Carston 2002; Moeschler 2004: explicature/implicature; Burton-Roberts 2006: what-is-A-said/what-is-B-said; Bach 2001: what is said/ impliciture/implicature). However, they have not gone far enough because they still were interested primarily in utterance interpretation, without paying due attention to private/individual knowledge, prior experience, and the emergent, rather than the a priori only intentions of the speaker. The main problem with the hearer-centered views is that they want to recover speaker meaning from a hearer perspective. As a result, what is actually “recovered” is hearer meaning, in the sense of how the hearer interprets what the speaker said. The proposition the speaker produces will not be exactly the same as that which will be recovered by the hearer, because, as we said above, interlocutors are individuals with different cognitive predispositions, different commitments, different prior experiences, and different histories of use of the same words and expressions. 4. Why do we need the dialogue approach? As discussed above current pragmatic approaches struggle with two issues. First, it would be important to explain speaker intention and speaker meaning from the perspective of the speaker. Second, we need to treat the hearer as an interlocutor, i.e. a hearer-speaker, who is not only an interpreter but also a producer of utterances. The dialogue approach could help to solve both of these problems with its principle according to which human beings are dialogic individuals (social individuals) who communicate in dialogic interaction not only by producing and understanding utterances but also by acting and reacting. Bach (2007: 5) said that (actual situational) context does not literally determine, in the sense of constituting, what the speaker means. What the speaker really means is a matter of his communicative intention although what he could reasonably mean depends on what information is mutually salient for him/her and the hearer. Bach further argued that taking mutually salient information into account goes beyond semantics, for what a speaker means need not be the same as what the uttered sentence means. This claim can be best demonstrated in intercultural communication where “mutually salient information” is quite problematic. Interlocutors in intercultural interactions have different first languages and use a second (or Lx) language to communicate with each other sharing very little common ground (Kecskes 2007; 2013). Salience is based on familiarity, frequency, and common prior experience (Giora 1997, 2003). Mutually salient information is something lingua franca speakers usually lack because they speak several different L1s and represent different cultures. For them mutually salient information should be directly connected with the actual situational context, tied to some universal knowledge and/or encoded in the common linguistic code (lingua franca) so that it can be “extracted” by the hearer without any particular inference that relies on non-existing common prior experience. It appears that inferencing for the lingua franca interlocutor in intercultural interaction usually means something close to decoding. This can be demonstrated through the following dialogue. (5) The interaction below takes place between the clerk and a Korean student in the Office of Human Resources. Lee: - Could you sign this document for me, please? Clerk: - Come again... Lee: - Why should I come again? I am here now. The Korean student misunderstands the expression “come again”. He processes it literally in spite of the fact that the actual situational context is completely unsupportive to this interpretation. It is essential therefore that pragmatics for intercultural interlocutors cannot be something “...they communicate over and above the semantic content of the sentence” as King and Stanley (2005: 117) assumed. For lingua franca speakers the semantic content is usually the conveyed content. If this is not clear from their utterance they try to reinforce it with repetition, paraphrasing or other means. This can be seen by the analyst only if s/he goes beyond utterance and analyzes the dialogic sequence. So for nonnative speakers (especially with lower language proficiency) participating in intercultural interactions “what is said” often coincides with “what is communicated”. Most of the time what the utterance says is meant literally. This fact gives a strong support to Bach’s claim: “...it is a mistake to suppose that ‘pragmatic content is what the speaker communicates over and above the semantic content of the sentence’ (King and Stanley 2005: 117). Pragmatics doesn’t just fill the gap between semantic and conveyed content. It operates even when there is no gap. So it is misleading to speak of the border or, the so-called ‘interface’ between semantics and pragmatics. This mistakenly suggests that pragmatics somehow takes over when semantics leaves off. It is one thing for a sentence to have the content that it has and another thing for a speech act of uttering the sentence to have the content it has. Even when the content of the speech act is the same as that of the sentence, that is a pragmatic fact, something that the speaker has to intend and the hearer has to figure out (Bach 2007: 5)”. In intercultural communication the content of the utterance is very often the same as that of the sentence. That is a pragmatic act, as Bach says, something that the speaker intends to say and the hearer has to figure out. In L1 communication on which current pragmatic theories are built, there is much more of a gap between what is said and what is meant than in intercultural communication in which it is of utmost importance that the speaker should mean close to what s/he says otherwise the hearer may have difficulty to figure out the speaker’s intention because of limited core common ground, shared knowledge and mutual norms. 5. Salience effect in the dialogic sequence As discussed above there is always an intuitive possibility of a distinction between what a speaker says and what s/he actually implicates. The likelihood of the distinction is especially high when a communicative process is affected by salience. The socio-cognitive approach proposed by Kecskes (2010, 2013) argues that both cooperation (recipient design) and egocentrism (salience) are present in the communication process all the time in a varying degree, and the interplay of conscious recipient design (result of cooperation) and subconscious salience (driving force behind egocentrism) is what shapes speaker production. As a semiotic notion, salience refers to the relative importance or prominence of signs. It is the most probable out of all possible. The relative salience of a particular sign when considered in the context of others helps an individual to quickly rank large amounts of information by importance and thus give attention to that which is most important. We tend to overestimate the causal role of information (salience) we have available to us both perceptually and linguistically. The socio-cognitive approach (Kecskes 2010, 2013) claims that interlocutors should be considered social beings searching for meaning with individual minds embedded in a socio-cultural collectivity. Grice was right when he tied cooperation to speaker-hearer’s rationality. However, egocentrism must be added to speaker-hearer’s rationality. We human beings are just as egocentric (as individuals) as cooperative (as social beings). “Egocentrism” in the socio-cognitive approach refers to attention-bias that is the result of prior experience of individuals. It means that interlocutors activate and bring up the most salient information to the needed attentional level in the construction (by the speaker) and comprehension (by the hearer) of the communication. So there is nothing negative about egocentrism if the term is used in this sense. It should not be confused with ‘egotistic’ that refers to a person who is self-centered and focuses only on his/her agenda. In pragmatics, based on Giora’s graded salience hypothesis (Giora 1997, 2003), when we speak about salient information we mean given information that the hearer assumes to be in central place in the speaker’s consciousness when the speaker produces the utterance. Salience has two important features that affect the dialogic sequence and dialogic actions of interlocutors. First, as discussed above, salience affects not only comprehension (as the graded salience hypothesis of Giora says) but also production (Kecskes 2010, 2013). Second, salience is based on prior experience that is usually different for each individual. This may result in misunderstanding between interlocutors that needs to be repaired in the dialogic sequence which gives less chance to the speaker to manipulate meaning according to his/her needs and unfold his/her agenda because repair is generally dictated by the hearer’s needs that the speaker tries to meet. The following short dialogue demonstrates what happens when both speakers’ production is dominated by salience rather than recipient design (Excerpt is from the TV comedy “Two and a half man”). (6) Charlie: - Morning. Alan: - What’s so good about it? Charlie: - I ain’t say ‘good’. Neither Charlie nor Alan did much recipient design in this interaction. Charlie said “morning”, (which may be considered an elliptical form of “good morning”) automatically when noticing Alan. Although the utterance was produced subconsciously it was relevant in that actual situational context. However, Alan did not pay close attention to what exactly Charlie had said, rather relied on the situational frame (morning meeting). He was in a bad mood and may have been affected by some negative prior experience the previous evening. He assumed that Charlie had uttered the usual formula “good morning”, and his response was formulated accordingly. The analysis of dialogic sequences show very clearly what the difference is between an approach where we handle speaker and hearer separately, and where we handle dialogue participants as interlocutors who are speaker-hearers. If we talk about speaker-hearers we need to accept that salience (just like cooperation in the Gricean sense) will affect whatever the speaker-hearer does, let it be production or interpretation. But Kecskes (2008: 401) argued that there is a difference between scenarios when the interlocutor acts as speaker and when s/he acts as hearer. When a lexical unit (labeled for private context) is used by a speaker to produce an utterance, private contexts (prior experience of the speaker) attached to this lexical expression are activated top-down in a hierarchical order by salience. This hierarchical order works differently for the speaker and the hearer. For the speaker, there is primarily an inter-label hierarchy, while for the hearer intra-label hierarchy comes first. The inter-label hierarchy operates in the first phase of production, when a speaker looks for words to express her/his intention. As a first step, s/he has to select words or expressions from a group of possibilities in order to express his/her communicative intention. These words or expressions constitute a hierarchy from the best fit to those less suited to the idea s/he is trying to express. To explain how this works we will analyze an excerpt from a movie. (7) Excerpt from the movie: “Angel Eyes”. Situation: A policewoman in uniform is driving the car, and the man sitting beside her is starring at her. PW: - What? M: - I was trying to picture you without your clothes on. PW: - Excuse me? M: - Oh no, I did not mean like that. I am trying to picture you without your uniform. PW: - Okaay? M: - I mean, on your day off, you know, in regular clothes. Excerpt (7) demonstrates a deductively built-up dialogic sequence. Salience effect usually result in deductive sequences where the speaker has something on his/her mind, and this intention is formulated abruptly, rather carelessly without specific planning, as seems to be the case in example (7). This dialogue appears to support the claim of cognitive psychologists according to which the initial planning of utterances ignores common ground (egocentric approach), and messages are adapted to addressees only when adjustments are required (Barr & Keysar, 2005; Keysar 2007; Barr 2004). It looks like salience effect generally appears in the form of a deductive process in the dialogic sequence that may contain repairs and adjustments. Why were the man’s attempts unsuccessful in the conversation in #7? Because, on the one hand we have the speaker with an intention to tell the woman why he was looking at her the way he did, and on the other hand there is subconscious salience that affected how he formulated his intention. So the utterance was not designed by the speaker to fit his words into the actual situational context, taking into account his partner. As a result, the word selection was wrong. Why was word selection wrong? Because it was not directed by recipient design but was prompted by salience. As argued above, salience (motivated by prior experience), which operates subconsciously and automatically, may affect word selection and utterance formation. “I was trying to picture you without your clothes on”. Is this what the speaker wanted to say and mean? Yes, this is exactly what he wanted to mean but not necessarily what he wanted to say. I agree with Bach (2005) that a speaker can always mean something distinct from the semantic content of the sentence he is uttering. But even if he speaks literally, and means precisely what his words mean, even that fact depends on his communicative intention (Bach (2005: 27). Wording, expressing intention in words is a tricky thing. Conceptualization is one thing, wording is another and meaning is a third one. There is no one-to-one relationship between any of the three. Now we should look at how salience works for the interlocutor when s/he is in the role of hearer. The hearer has to cope with a different type of hierarchy than the speaker from her/his perspective. Thus, an intra-label hierarchy is in force when the hearer processes a lexical unit(s) in an utterance (or even an entire utterance). The label (word) uttered by the speaker hierarchically triggers the history of that particular label as used by the hearer (but not by the speaker). This may also be a reason for misunderstanding in the communicative process, both in intracultural and intercultural interactions. Compare the following (source: American sitcom) interchange in (8): (8) Bob: Are you OK? Mary: I am fine. Bob: I know you are fine, but are you OK? Bob had several options to ask about Mary’s well-being: “Are you OK?”, “Are you fine?”, “Is everything all right?”, etc. His selection of “Are you OK?” caused a slight misunderstanding between the two because they interpreted “OK” differently. As we have seen, salience is both an individual and a societal phenomenon. Prior and actual situational experience is privatized/subjectivized and prioritized in the mind of interlocutors. Their different prior experiences, their different evaluations of the actual situational context, their dynamically changing intentions and individual degrees of salience result in a subjectivized process of production and comprehension. As a result, there may be no single point in the recovery process at which a speaker’s utterance fully matches a hearer’s interpretation. This is because both a speaker’s production and a hearer’s interpretation are “contaminated” by individualized pragmatic elements. Pragmatic enrichment processes work differently for speaker- hearers depending on which role they are in. Consequently, the match between the two sides keeps varying in the communicative process. This is why we think that, “we almost always fail. [...] Yet we almost always nearly succeed” (Rapaport 2003: 402). And this is why a pragmatic theory should be interlocutor-centered that take the individual as a speaker-hearer. Speaker’s production and hearer’s interpretation should be analyzed in their own rights. To get a full picture of what is going on in communication we need to look at a dialogic sequence where the role of salience is minimal and recipient design and intention recognition comes to the forth. (9) This is an excerpt from the film “Coogan’s Bluff”. Situation: A man and a young woman are sitting in a restaurant after meal. The woman stands up and with a short move reaches for her purse. W: - I have to be going. M (seeing that she reaches for her purse): - What are you doing? W: - Dutch. M: - You are a girl, aren’t you? W: - There have been rumors to that effect. M: - Sit back and act like one. W: - Oh, is that the way girls act in Arizona? When the girl wants to pay the man expresses his objection with asking “what are you doing?”. The girl perfectly understands what the man is referring to, so she tells him “Dutch”, which means she wishes to pay for her share of the bill. The man indicates his dislike in a very indirect but still expressive way: “You are a girl, aren’t you?” The girl’s response shows that she follows where the man is getting to. Then the man hints at what he expects the girl to do “Sit back and act like one”. This means that he does not want her to pay the bill. This inductively developed dialogic sequence is a good example for elaborated recipient design where nothing is said directly, still there is no misunderstanding because the speaker adequately alerts the hearer to what he means. According to the recipient design view in order to succeed speakers must correctly express intended illocutionary acts by using appropriate words, and make their attempt in an adequate context. In this process speakers relate propositional contents to the world (actual situational context; audience) with the intention of establishing a correspondence between words and things from a certain direction of fit. This is what happened in example (9). When looking at the dialogic sequence from the speaker’s perspective it appears that recipient design usually requires an inductive process that is carefully planned while salience effect generally appears in the form of a deductive process that may contain repairs and adjustments. The speaker has something on his/her mind, and this intention is formulated subconsciously, abruptly, rather carelessly without specific planning. From the hearer’s perspective intention recognition can also be driven by salience that may frequently lead to misunderstandings or misinterpretations because of the private and cultural specific nature of salience. The interplay of recipient design and salience demonstrates very well how cooperation (social) and egocentrism (individual) are intertwined in the communicative process. This gives strong support to the dialogic principle which claims that human beings are dialogic individuals (social individuals) who communicate in dialogic interaction not only by producing and understanding utterances but also by acting and reacting. 6. Conclusion The aim of this paper was to direct attention to how a dialogue approach can lead to better understanding of actions of interlocutors in communication. It was claimed that utterances in pragmatics research should be analyzed from the perspective of the interlocutor that comprises both roles: speaker and hearer. However, if we want to focus both on the speaker’s and hearer’s perspective, and want to figure out why the speaker said what s/he said the way s/he said it, and why the hearer opted for a given interpretation, we need to use a holistic, top-down approach combined with the bottom-up, utterance by utterance analysis. This means that we need to analyze the dialogic sequence or discourse segment rather than just the utterance in order to find cues that help us identify the real intention of the speaker and better understand the options of the hearer in interpreting the speaker’s utterance. Following Saul (2002) it was suggested that we make a difference between utterer-implicature and audience-implicature. In order for us to understand meaning in interaction we need to focus on the double role of interlocutors. While fitting words into actual situational contexts speakers are driven not only by the intent (conscious) that the hearer recognize what is meant as intended by the speaker (cooperation), but also by speaker individual salience that affects production subconsciously (egocentrism). Therefore it was proposed that speaker-hearer rationality should include not only cooperation but egocentrism as well (see also in Kecskes 2013). Through some examples it was pointed out that recipient design usually requires an inductive process in the dialogue that is carefully planned while salience effect generally appears in the form of a deductive process that may contain repairs and adjustments. The speaker has something on his/her mind, and this intention is formulated abruptly, rather carelessly without specific planning. The salience effect can lead to unwanted ways of expressing intention and occasionally misunderstanding, which requires the speaker to make repairs and adjustment in the communicative process and work together with the hearer to come to an understanding. On the other hand, recipient design gives the chance to the speaker to manipulate speaker meaning according to his/her needs: leave meaning conversationally open for interpretation or signal his/her intention with cues and markers. Here the goal of the speaker is to promote his/her own agenda by trying to manipulate the interpretation process. A dialogic, “wide pragmatics” approach may give us the chance to better understand the complexities of the communicative process. © Istvan Kecskes, 2016

Istvan Kecskes

State University of New York

Author for correspondence.
Email: ikecskes@albany.edu
1400 Washington Ave., 12222 Albany, USA

  • Airenti, Gabriella, Bruno G. Bara, & Colombetti Marco. 1993. Conversation and behavior games in the pragmatics of dialogue. Cognitive Science 17(2). 197-256
  • Bach, Kent. 2001. You don’t say? Synthese. 128: 15-44
  • Bach, Kent. 2005. “Context ex Machina”. In Semantics vs. Pragmatics, Szabó Zoltán Gendler (ed.), 15-44. Oxford: Oxford University Press
  • Bach, Kent. 2007. “Regressions in pragmatics (and semantics).” In Pragmatics (Advances in Linguistics), Burton-Roberts Noël (ed.), 24-44. Houndmills: Palgrave-Macmillan
  • Bara, Bruno. 2011. Cognitive pragmatics: The mental processes of communication. Intercultural Pragmatics 8-3 (2011), 443-485
  • Barr, Dale J. 2004. “Establishing conventional communication systems: Is common knowledge necessary?” Cognitive Science 28.6: 937-962
  • Barr, Dale J. and Boaz, Keysar. 2005. “Making sense of how we make sense: The paradox of egocentrism in language use.” In Figurative Language Comprehension: Social and cultural influences, Colston Herbert L. and Albert N. Katz (eds), 21-43. Mahwah, N. J.: Lawrence Erlbaum
  • Buber, M. 1955. Dialogue. In between man and man (R. G. Smith, Trans.; pp. 1-39). Boston, MA: Beacon Press
  • Bunt, H.C. 2011. The Semantics of Dialogue Acts. Proceedings 9th International Conference on Computational Semantics (IWCS 2011). Bos, J. & Pulman, S. (eds.). Oxford, UK: Oxford University, p. 1-14
  • Burton-Roberts, Noël. 2006. “Cancellation and intention.” Newcastle Working Papers in Linguistics 12: 1-12
  • Carbaugh, Donal. 2013. On Dialogue Studies. Journal of Dialogue Studies 1.1 (2013): 9-28
  • Carston, Robyn. 2002. Thoughts and Utterances: The pragmatics of explicit communication. Oxford: Blackwell
  • Cooren, F. 2010. Action and agency in dialogue: Passion, incarnation, and ventriloquism. Amsterdam/Philadelphia: John Benjamins
  • Davis, Wayne. 1998. Conversational Implicature: Intention, convention and principle in the failure of Gricean theory. Cambridge: Cambridge University Press
  • de Saussure, Louis. 2007. Pragmatic Issues in Discourse Analysis. Critical Approaches to Discourse Analysis. Across Disciplines 1 (1): 179-195
  • GarfinkeI, H. 1972. Remarks on ethnomethodology. In J.J. Gumperz & D.H. Hymes (Eds.) Directions in sociolinguistics. New York: Holt, Rinehart & Winston
  • Giora, Rachel. 1997. “Understanding figurative and literal language: The graded salience hypothesis.” Cognitive Linguistics 8.3: 183-206
  • Giora, Rachel. 2003. On Our Mind: Salience context and figurative language. New York: Oxford University Press
  • Horn, Laurence R. 2007. Neo-Gricean pragmatics: a Manichaean manifesto, in N. Burton-Roberts (ed.) Pragmatics, Basingstoke: Palgrave
  • House, Juliane. 2002. Developing pragmatic competence in English as a lingua franca. In Lingua Franca Communication, K. Knapp and C. Meierkord (eds.), Frankfurt am Main: Peter Lang, 245-267
  • Jaszczolt, K.M. 2005. Default Semantics. Foundations of a Compositional Theory of Acts of Communication. Oxford: Oxford University Press
  • Kecskes, Istvan. 2007. Formulaic language in English Lingua Franca. In Kecskes, I. & L. Horn (eds.) Explorations in Pragmatics: Linguistic, Cognitive and Intercultural Aspects. Berlin/New York: Mouton de Gruyter: 191-219
  • Kecskes, Istvan. 2008. Dueling context: A dynamic model of meaning. Journal of Pragmatics 40 (3): 385-406
  • Kecskes, Istvan. 2010. The paradox of communication: A socio-cognitive approach. Pragmatics and Society 1(1). 50-73
  • Kecskes, I. 2012. Is there anyone out there who really is interested in the speaker? Language and Dialogue. Vol. 2. No. 2: 285-299
  • Kecskes, Istvan. 2013. Intercultural Pragmatics. Oxford, UK: Oxford University Press
  • Kecskes, I. & F. Zhang. 2009. Activating, seeking and creating common ground: A socio-cognitive approach. Pragmatics and Cognition. 17(2). 331-355
  • Keysar, Boaz. 2007. “Communication and miscommunication: The role of egocentric processes.” Intercultural Pragmatics 4.1: 71-84
  • King, Jeffrey C. and Jason, Stanley. 2005. “Semantics, pragmatics, and the role of semantic content.” In Semantics versus Pragmatics, Szabó Zoltán Gendler (ed.), 111-164. Oxford: Oxford University Press
  • Moeschler, J. 2004. Intercultural Pragmatics: a cognitive approach. Intercultural Pragmatics, 1, 1: 49-70
  • Morante, R., Keizer, S. & Bunt, H.C. 2007. A dialogue act based model for context updating. Proceedings of the eleventh international conference on the semantics and pragmatics of dialogue (DECALOG 2007). Trento, Italy: [s.n.], p. 9-16
  • Psathas, O. (Ed.). 1979. Everyday language. Studies in ethnomethodology. New York: Irvington
  • Puig, Margarida Bassols. 2003. Pragmatics and discourse analysis. Noves SL. Revista de Sociolingüística http://www.gencat.cat/llengua/noves Winter 2003
  • Rapaport, William J. 2003. “What did you mean by that? Misunderstanding, negotiation, and syntactic semantics.” Minds and Machines 13.3: 397-427
  • Sacks, H., Schegloff, E. A., & Jefferson, G. (1974). “A simplest systematics for the organization of turn-taking for conversation.” Language, 50, 696-735
  • Saul, Jennifer. 2002. What is said and psychological reality: Grice’s project and relevance theorists’ criticisms. Linguistics and Philosophy 25: 347-72
  • Schegloff, Emanuel A. 2007. Sequence Organization in Interaction: A Primer in Conversation Analysis, Volume 1, Cambridge: Cambridge University Press
  • Schenkein, J. (Ed.). 1978. Studies in the organization of conversational interaction. New York: Academic
  • Schiffer, S.R. 1972. Meaning. Oxford: Oxford University Press
  • Searle, John. 1995. The Construction of Social Reality. London: Allen Lane, The Penguin Press
  • Sperber, Dan and Deirdre, Wilson. 1986/1995. Relevance: Communication and cognition (2nd ed.). Oxford: Blackwell
  • Taboada, M. and W.C. Mann. 2006. Rhetorical Structure Theory: Looking Back and Moving Ahead. Discourse Studies. 8(3): 423-459
  • Tracy, Karen and Robert T. Craig. 2010. “Studying interaction in order to cultivate communicative practices: Action-Implicative discourse analysis.” In New Adventures in Language and Interaction, Streeck Jürgen (ed.), 145-166. Amsterdam: John Benjamins
  • Turner, R. (Ed.).1974. Ethnomethodology: Selected readings. Harmondsworth, England:Penguin
  • Weigand, Edda. 2001. Negotiation and Power in Dialogic Interaction, ed. by Edda Weigand and Marcelo Dascal. Amsterdam/Philadelphia: Benjamins (Current Issues in Linguistic Theory 214)
  • Weigand, Edda. 2004. Emotions in Dialogic Interaction. Advances in the Complex, ed. by Edda Weigand. Amsterdam/Philadelphia: Benjamins (Current Issues in Linguistic Theory 248)
  • Weigand, Edda. 2006. Argumentation - The mixed game. Argumentation 20/1, 59-87
  • Weigand, Edda. 2010. Language as dialogue. Intercultural Pragmatics. Vol. 7. No. 3: 505-515
  • Weigand, Edda. 2010b. Dialogue: the mixed game. Amsterdam/Philadelphia: Benjamins


Abstract - 814

PDF (English) - 408


Copyright (c) 2016 Кечкеш И.

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.