A Theoretical And Experimental Pragmatic Study Of Conversational Implicitures

There is a common and interesting linguistic phenomenon in our daily communication which has not yet been generally noticed by ordinary language users.That is,what the speaker means is not expressed explicitly or implicated indirectly by a given utterance,but is only implicit in that utterance.Ever since Grice introduced his concept of generalized conversational implicature into the study of pragmatics,pragmatic theorists have used different terms,like implicature,explicature(what is said)or impliciture,to designate this phenomenon.It is important to bear in mind that these are not simply diverse labels for the pragmatically determined aspect of utterance meaning but different ways of demarcating the boundary between what is said and what is meant,and of mapping from context to propositions,and from linguistic form to mental concept.Therefore,the definition and comprehension of this kind of utterance meaning have been the subject of intense debate within the philosophy of language and the psychology of language processing.Four main research questions were addressed in this dissertation:(1)Does the implicit meaning of an utterance contribute to what is said,or may it be counted as an implicature?(2)What category does it fall into,and so how can the distinction between this kind of meaning and other expressed propositions be accounted for?(3)How does it come about and how does the computation in one context differ from that in another context?(4)What role does it play in the recovery of what the speaker intended?Thus,this dissertation is made up of four main parts,each of which presents the evidence for the account relating to the one of the four research questions.To begin with the first question,three different dimensions that distinguish what is said from what is implicated were discussed in Chapter 3.They range from Semantic Minimalism,a theory of linguistic semantics in which what is said is the same as the minimal proposition and contextual information is not the prerequisite for its determination,to Radical Contextualism,a theory of pragmatic inferences in which what is said is entirely a matter of context with no contribution of the literal meaning to truth-conditional content of an utterance.Moderate Contextualism limits the amount of context that is allowed to enter into the determination of what is said.Based on the agreement and disagreement between the three theories,this chapter proposed a forth alternative one,so-called Moderate Minimalism.After a brief review of the terminological debates about the pragmatically determined aspect of utterance meaning in question,Chapter 4 offered some arguments for treating it as an instance of impliciture,similar to Bach's notion.However,his theoretical framework is imperfect.So,an attempt was made to refine the impliciture theory by making some corrections through a new definition of impliciture and discussion of the generative conditions,main properties and differentiations of implicitures.Next,the dissertation turned to the experimental work on impliciture comprehension.The study reported in Chapter 5 tested four empirical models about how implicitures are derived.The Literal-First Serial Model says that implicitures are derived from minimal propositions.Under the Default Model,implicitures are linked to the linguistic properties of utterances and are generated automatically and largely independent of contextual assumptions.By contrast,the Underspecified Model assumes that implicitures are context-dependent and more effort is required to access them.Finally,and more recently,the Constraint-Based Model predicts that the interpretation of implicitures is influenced by the context,but the inferential process is automatic.A mouse-tracking technique in a sentence verification and truth-value judgment task was used to examine the online processing of implicitures.Experiment I compared the impliciture interpretation and the minimal proposition expressed by an utterance under the neutral condition.Experiment ? compared semantically underspecified utterances and pragmatically underinformative utterances,and also compared the default impliciture and the alternative interpretation of the latter kind of utterances,in biasing contexts.Participants were adult native speakers of Chinese.They were presented with compute-delivered auditory stimuli and asked to judge whether the statements are true or false.Participants' judgment accuracy,processing times and hand-movement trajectories were recoded.Finally,Chapter 6 further reported findings of another two experiments which were intended to answer the last research question.One of the liveliest debates in recent years concerns the role of impliciture in the inferential process of intentionally conveyed meaning and the relationship between impliciture and implicature/what is said in terms of strength or priority.In Experiment III,participants were presented with a serious of conversational situations,each of which,consisting of a question,an answer and a statement,induces the relevant implicature.The experimental paradigm was composed of three different conditions:explicit condition(what is said),implicit condition(impliciture)and implied condition(implicature).In Experiment IV,each condition was presented in its negative form.And the task,in both experiments,was to judge the truthfulness of the statements in three different conditionsThe results show that:First,the Moderate Minimalism takes one further step to recognize a third level of utterance meaning,what is asserted(impliciture),between what is said and what is implicated(implicature).What is said,in a strict sense,can be understood solely on the basis of linguistic features,without recourse to contextually specified information;on the other hand,implicatures are essentially context-dependent,and are thus derived wholly by inferential pragmatic processes with an ineliminable appeal to the speaker's intention.What is asserted is a pragmatically strengthened meaning or truth-conditional content of an utterance.It is recovered prior to the operation of implicature,but that is not explicitly linked to the linguistic elements of the utterance and cannot be part of what is said.Second,impliciture is an embellished version of what is said,it probably includes,or at least overlaps with the minimal proposition under a given condition,and theoretically involves the figurative use of certain types of expressions.The generation of impliciture is the effects of the speaker's intention,the hearer's expectation,the linguistic meaning and the context of the utterance.The characteristic features of implicitures are that they are cancelable(somewhat controversially),calculable,detachable,indeterminate,and so on.Third,the experimental results show that impliciture is sometimes interpreted more accurately,quickly and smoothly than the minimal proposition or the alternative meaning of the target utterance.However,different types of implicitures differ in their degree of context-dependency.While providing evidence against the three classical models,these results lend support to the Constraint Based Model.Nevertheless,the results also pose a challenge to this model concerning the validity of multiple factors in on-line comprehension.Forth,these results also indicate that people are aware of the cognitive distinction between what is said,impliciture and implicature.In ordinary language use,impliciture could be accessed as rapidly as what is said,to influence further the comprehension of that utterance.But the recovery of impliciture in the uncooperative context could lead to increased processing time due to the assimilation of the conceptual information.Among all levels of meaning,what is said has the potential to be taken as the most basic meaning.On the other hand,the computation of implicature requires additional processing resources,thus increasing the processing load.Different types of implicitures/implicatures may differ in the ultimate selection of mental representation in different ways.The more frequently used pragmatic inferences are,the more difficult to remove they will be.In short,this dissertation demonstrates that impliciture,which is unarticulated in the surface form of the sentence,but which is not an instance of particularized implicature,is an independent level of meaning between what is said and implicature Implicitures are different from other meanings or expressions in terms of their definition,generative conditions and characteristics.Implicitures are computed immediately and locally,but the processing time and effort of inferences is constrained by numerous factors.Therefore,they play a distinct role in bringing about a communicative effectGenerally,the study offers some important insights into a better understanding of pragmatically determined aspects of utterance meaning.Theoretically,it raises questions for the traditional accounts of implicitures and suggests that alternative theoretical accounts or reformulations of these accounts may be worth considering.Therefore,this dissertation proposed a moderate minimalist approach to the levels of meaning,aiming to expand Bach's theory of implicitures.It also developed the Constraint-Based Model of impliciture computation.Methodologically,the present study poses a challenge to traditional experimental pragmatics to see evidence for the processing mechanism of impliciture interpretation in a new way.It makes a case for the claim that it is fruitful to apply a hand-movement tracking experiment in the online comprehension task to an area of inferential pragmatic research.