Conversational Basis of Some Presuppositions

[a]School of Foreign Languages， Changchun University of Science and Technology， Changchun， China.

* Corresponding author.

Received 27 August 2013； accepted 30 November 2013

Abstract

Through thinking about the derivation of presuppositions， this paper intends to demonstrate the assumption that at least some presuppositions have a conversational source. Moreover， it aims to provide a partial proposal for the conversational derivation and a multicomponent model of these presuppositions.

Key words： Presupposition； Conversational basis； Common ground； Contextual； Defeasibility； Nondetachability

MA Danni （2013）. Conversational Basis of Some Presuppositions. Studies in Literature and Language， 7（3）， 34-38. Available from： http：//www. /index.php/sll/article/view/j.sll.1923156320130703.2996 DOI： http：///10.3968/j.sll.1923156320130703.2996

INTRODUCTION

How should we define “presupposition”？ Linguists have proposed various definitions， and the debate about their relation to each other and their respective merits is far too extensive to summarize. Among them， two dominant kinds of explanation need to be mentioned here.

Presuppositions as conventional implicatures. According to this view， the presuppositions of a sentence are just another component of the total package of information that the sentence can be used to convey. Other components of this package include the content and the conversational implicatures. This view is proposed by Karttunen and Peters （1979） as well as Gazdar （1979）. Only Karttunen & Peters use the term “conventional implicatures”； Gazdar calls them “presuppositions”. But what matters here is not terminology， but the concepts behind it. On this view， certain lexical items have， in addition to their truth conditional content， a special presuppositional content， which is carried through the compositional process to produce a propositional presupposition. This theory is generally refuted by pragmatists， for its validity is too specific to cover sentences like “George has three children， and may have more for all I know”. Although the Karttunen and Peter’s model for treating presupposition has been rejected by many researchers， we still talk about the presuppositions of special words like know， too， even， and so on， as regard to their conventional meaning.

Presuppositions as admittance conditions. This view is best represented by Stalnaker’s work （1973， 1974， 1979） and also in Karttunen （1974）. It says that the propositions a sentence presupposes must be entailed by the common ground of any context admitting that sentence. This notion of presupposition relies on a notion of “common ground”： the common ground of a context of utterance is the combination of all those propositions that the interlocutors take for granted in that context （either because they are permanently shared beliefs in their community， or because they have been established in the course of the preceding conversation. Although mightier than the former， this theory is not authoritative enough to expel semantic aspects out of mention.

Till now， there seems to be no such a theory that is adequate and integrate enough to cover the whole area. The real exit for us to approach this kind of simple but vague pragmatic phenomena remains in a fog.

Presuppositions， however， might have a very different source， i.e.， might be conversationally derived， that is， they might be inferences regulated by general conversational principles. Stalnaker （1974） suggests that at least some presuppositions have a conversational source. He claims that：

One of the primary advantages of the move from a semantic to a pragmatic account of presupposition as being the possibility of explaining “some of the presupposition facts in terms of general assumptions about rational strategy in situations where people exchange information or conduct argument” （1974， p.205）.

My primary goal is to motivate the claim that at least some presuppositions have a conversational source， and to set out a partial proposal for the conversational derivation and multicomponent model of these presuppositions. Thinking about the derivation of presuppositions is worthwhile， because it gives us a different perspective on what presuppositions are， and perhaps could provide us a new perspective of solving the presupposition projection problem.

1 . E V I D E N C E F O R T H E CONVERSATIONAL BASIS OF SOME PRESUPPOSITIONS

The presuppositions having a conversational source possess two typical properties of conversational implicaturecontextual defeasibility and non-detachability.

1.1 Contextual Defeasibility

It is well known that all presuppositions can be canceled in certain sentential contexts. In addition， some， but not all presuppositions are prone to cancellation if the addressee realizes that the speaker is apparently ignorant of the proposition that is presupposed. Imagine a conversation taking place between two people who are meeting for the first time. One talks to the other：

（i） I notice that you keep chewing on your pencil. Have you recently stopped smoking？

In this situation， the addressee knows that the speaker is ignorant of her current or prior smoking habits， and in particular cannot assume that she was recently a smoker. In this situation， there is no implication that the speaker believes the addressee has been a smoker. The speaker would be understood by the addressee just as asking whether the addressee has undergone change from being a smoker to not being one， equivalent to：

（ii） Is it the case that you have recently been a smoker and have recently ceased to be one？

What is involved here is the choice of a nonpresupposing interpretation of the question over a presupposing interpretation （the addressee has been a smoker）. While the presupposing interpretation is usually preferred， we apparently have a stronger preference not to attribute to the speaker an unfounded assumption. So the presuppositional interpretation is not chosen.

Example （iii） is a variation on the same theme.（iii） I have no idea whether Jane ever smoked， but she has not stopped smoking.

Here， the speaker puts forth an explicit assertion of her ignorance about the relevant proposition， and thus the stop clause does not give rise to the presuppositional inference that Jane is or was a smoker. The clause is understood simply as denying that she has undergone the change of state.

Not all expected presuppositions fail to arise in explicit ignorance contexts. Especially， those presuppositions which have explicit lexical triggers such as even， too and again are not cancelable in this way. For example， suppose I came across a friend at a video rental store， whose history of video rental I know nothing about. I ask her：

（iv） Are you renting “Manhattan” again？

Despite the explicit ignorance context， my addressee has to reckon that I believe that she has rented “Manhattan”before， rather than the utterance in sentence （v）：

（v） Is it the case that you have rented “Manhattan” before， and are going to do so again？

The presupposition generated by again is not canceled even when the speaker explicitly asserts her ignorance， leading to anomaly in cases like （vi）：

（vi） I do not know if Jane ever rented “Manhattan” before， but perhaps she’s renting it again.

Examples （i） and （ii）， which show contextual defeasibility of a presupposition， both involve the presupposition of the change of state verb stop. Thus， we can make a conclusion that all presuppositions consisted by change of state predicates are contextually defeasible.

Contextual defeasibility also exists in another type of presuppositions： the presuppositions of factives. Suppose we are at a restaurant. A couple at another table are arguing with each other furiously. We are speculating what aroused the quarrel. I say to you：

（vii） Perhaps she just discovered that he’s having an affair.

As you know that I could not possibly know whether he has in fact been having an affair， you will not take me to presuppose that he has an affair， but just suppose that he has been having an affair and that she has discovered it. Another instance， sentence （viii） below could well be uttered as part of a conversation between two people who know that Henry is searching for Jane， but who do not themselves know where Jane is：

（viii） If Henry discovers that Jane is in New York， there’ll be trouble.

In this situation， there is no presupposition or implication that Jane is in fact in NY.

All of these examples above illustrate one point， that is， the presuppositions of change of state predicates and of factives display the same kind of contextual defeasibility as （generalized） conversational implicatures do. Although the implicatures generated by the Maxim of Quantity are robust， they are still defeated in explicit ignorance contexts. The next example is familiar to us. Ordinarily， if I say George has three children， you will understand me to mean that George has exactly three children. However， I can say：

（ix） George has three children， and may have more for all I know.

The assertion of ignorance in the second clause suppresses the usual scalar implicature， just as assertions of ignorance in the presuppositional examples suppressed the expected presuppositions. This could be a powerful evidence to prove that at least some presuppositions have a conversational basis.

1.2 Non-Detachability

Grice （1967） observes that conversational implicatures are by necessity non-detachable from the content of whatever utterance gives rise to them. This means that an implicature is attached to the semantic content of what is said， not to linguistic form. Therefore implicatures cannot be detached from an utterance simply by changing the words for synonyms or in other words by relexicalization or reformulation. Thus， if utterance of a sentence with a particular content generates an implicature in a given conversational context， then utterance of any other sentence with the same content will give rise to the same implicature. For example， consider the short exchange in （x）：

（x） Jane： Do you want to go out for a drink？ Julia： I have to finish writing my SALT paper.

In the context of Jane’s question， Julia’s utterance generates the implicature that she does not want to go out for a drink. However， the generation of the implicature is not dependent on the form of Julia’s utterance. Any other form which expresses the same content does not fail to produce the implicature. Consider the synonymous sentences in （xi）：

（xi） a. ... I need to finish my SALT paper.

b. ... My SALT paper needs to get finished tonight.

c. ... I have to work on my SALT paper.

Conversational implicatures are non-detachable because they are due to the expression of a particular content in a particular conversational content. They are not conventionally associated with any expression， so the implicature cannot be affected by the expression of the same content but in a different form. It seems that some presuppositions also have this kind of property. There’s another example with “stop”：

（xii） Jane did not stop laughing.

The presupposition of this sentence is that Jane had been laughing prior to the referent time of the sentence attributing to the presupposition of “stop”. But if we replace “stop” with any of its synonyms， the presupposition remains the same：

（xiii） a. Jane did not quit laughing.

b. Jane did not cease laughing.

c. Jane did not discontinue her laughter.

This kind of phenomenon also fits to other change of state predicates： Utterance of each synonymous sentence in （14） would normally give rise to the presupposition that Jane was in the house immediately prior to the reference time of the sentence.

（xiv） a. Jane did not leave the house.

b. Jane did not quit the house.

c. Jane did not go out of the house.

d. Jane did not exit the house.

e. Jane did not depart from the house.

The same point is illustrated with the factive realize and its synonyms in （xvi）.

（xv） Harry did not realize / come to know / become aware that he was a fool.

These observations suggest that the presuppositions of change of state predicates and of factives are nondetachable， that is， they attach to the content expressed， and not to any lexical item. But presuppositions or implications cannot attach to content by convention. Thus， these presuppositions must have a conversational source.

Here， someone could deny that the data show nondetachability， and argue that the presuppositions of these examples are due to a conventional property of one of the lexical items used， i.e.， the presuppositions triggered by even， too， again and so on. For example， the sentences in（xvi） all give rise to the presupposition that some （relevant） person other than Harriet is going to SALT.

（xvi） a. Harriet might go to SALT too. b. Harriet might go to SALT as well. c. Harriet might also go to SALT.

The same point can be made to again and its synonyms， but harder to even， for it does not have any obvious synonyms in English.

However， there is a difference between the cases of again and too， and the change of state and factive cases. It is possible to express the nonpresuppositional content of the sentences in （xvi） independently of the presupposition， by omitting the presupposition trigger. While in the cases of the change of state predicates and the factives， the two types of content are inseparable. Thus， we could say that the presupposition in （xvi） a-c is not non-detachable： the truth conditional content of these sentences is just that “Harriet might go to SALT”， and this content can be expressed without giving rise to the presupposition. Hence， the presupposition is not attached to the truth conditional content. Besides， my goal here is to demonstrate that at least some presuppositions have the properties of conversationally generated inferences. The presuppositions of change of state predicates and of factives are definitely non-detachable， as well as being contextually defeasible. This is enough.

2. A GENERAL PRINCIPLE

Several arguments have been given in this paper calling for that a general principle for generating presuppositional inferences is needed. But what kind of conversational principle would do the job？ Stalnaker had already been studying and carved out a short way. In speculating as to possible conversational sources of presuppositional constraints， Stalnaker （1974， p.205） suggests that：

The propositions that P and that Q may be related to each other， and to common beliefs and intentions， in such a way that it is hard to think of a reason that anyone would raise the question whether P， or care about its answer， unless he already believed that Q.

Later， he proposed a further Interpretation Principle which is much more precise and applicable. The idea may be amenable to a treatment in terms of logical ordering.

Interpretation Principle （tentative） Suppose that P entails but is not entailed by Q. A speaker who raises the question whether P indicates a belief that Q is true.

However， the Interpretation Principle is too strong， predicting presuppositional implications which do not occur. Thus， this suggestion still needs further refinement.

3. A SUPPOSITION

According to Verschueren （2000）， whether the presupposition of a referring expression can be said to be satisfied depends as much on the sentence in which the expression is used as on the expression itself. That is， presuppositions are relations between a form of expression and an implicit meaning which can be arrived at by a process of （pragmatic） inference. Karttunen & Peter also support that the cooperative participants have the obligation to “organize their contributions in such a way that the conventional implicate of the sentence uttered are already part of the common ground at the time of utterance” （1974， p.269）. Levinson modified the suggestion into requirement “that the so-called conventional implicate are consistent with the common ground” （2000， p.209）. Thus， we can assume that presupposition is an area both semantic and pragmatic. Presuppositions are better to be predicted from the semantic content of presupposition-triggers， by means of general pragmatic principles. First， there always seem to be conventionally close relations between the semantic content of presupposition-triggers and their corresponding presuppositions. Second， the introduction of “common ground” solves the headachy defeasibility and projection problem. And what the conversational basis of presupposition does is to build a bridge connecting between the semantic and pragmatic perspective.

CONCLUSION

What we have established in this discussion is that at least some presuppositions must have a conversational basis， that is， they must be inferences derivable from some general conversational principle or principles. On the view of presupposition now most standard in the literature， presuppositions are thought to be propositions which must be entailed by the presumed common ground of the discourse participants. However， if at least some presuppositions are derived by the kind of mechanism which gives rise to other conversational inferences （i.e.， conversational implicatures）， it is more appropriate to view them as propositions which the addressee can infer the speaker to believe on the basis of what the speaker has said， （plus the assumption that the speaker is behaving cooperatively）， which is more informative and explainable than the common ground. Although current treatments of presupposition in the dynamic semantics literature have a great deal of appeal， it is still impossible to demonstrate presupposition involving both semantics and pragmatics， merely by means of semantic approach.

Thus， the point is how to connect the two approaches together. The conversational basis of some presupposition discussed in this paper explicates that at least some presupposition has the properties of contextual defeasibility（pragmatics） and non-detachability （semantics）. Perhaps the gap between two dimensions could be filled up.

REFERENCES

Bach， K. （1999）. The myth of conventional implicature. Linguistics and Philosophy， 22（4）， 327-366.

Chierchia， G.， & McConnell-Ginet， S. （1990）. Meaning and grammar. Mass.： MIT Press.

Gazdar， G. （1979）. Pragmatics： Implicature， presupposition and logical form. New York： Academic Press.

Karttunen， L.， & Peters， S. （1979）. Conventional implicature. In C.-K. Oh & D. A. Dinneen （Eds.）， Syntax and semantics Vol. 11： Presupposition. NY： Academic Press.

Kuppevelt， J. V. （1995）. Discourse structure， topicality and questioning. Journal of Linguistics， 31， 109-147.

Levinson， S （2000） Pragmatics. Beijing： Foreign Language Teaching and Research Press.

Levinson， S. （2000）. Presumptive meanings. Cambridge， MA： MIT Press.

Roberts， C. （1996）. Information structure in discourse： Towards an integrated formal theory of pragmatics. OSU Working Papers in Linguistics， 49， 91-136.

Simons， M. （2000）. Issues in the semantics and pragmatics of disjunction. NY： Garland.

Simons， M. （2001）. Why some presuppositions are conversational implicatures. Unpublished manuscript. Carnegie Mellon University.

Stalnaker， R. （1974）. Pragmatic presuppositions. In M. K. Munitz & P. K. Unger （Eds.）， Semantics and philosophy. New York： New York University Press.

Verschueren， J. （2000）. Understanding pragmatics. Beijing： Foreign Language Teaching and Research Press.

Wang， J. Y. （2002）. Presupposition. Rhetoric Learning， 06， 9-24.

Xu， S. M. （1998）. Presupposition. Journal of Yangzhou University， 03， 56-59.