請問間接言語行為indirect speech act和會話含義implicature的區別是啥?

作者：由西西弗斯說是發表于歷史時間：2021-08-02

西西弗斯說是2019-12-28 19:50:37

加粗的是個人認為的兩者間的重要區別

Indirect speech act

An utterance which achieves a certain illocutionary force without ‘wearing it on its sleeve’ is an indirect speech act。 For example， it is possible to request the butter by uttering the interrogative sentence， ‘Can you pass the butter？’ The grammatical form of this sentence is more closely associated with asking questions than making requests， so this is an indirect speech act。 Indirect speech acts are common， perhaps the norm。 The most obvious direct way of making a request is to say something like ‘I request that 。。。’。 In English this is uncommon； most requests at least are made by indirect speech acts。

The use of indirect speech acts depends partly on inference, partly on knowledge of how the language is typically used in a certain culture

。 ‘Can you pass the butter？’ could be a real question： the butter might be heavy and the addressee physically weak， for example。 So context-sensitive inference is required to determine the intended illocutionary force。 In some cases， the hearer will be helped by his knowledge of linguistic and cultural conventions： ‘Can you 。。。’， ‘Could you 。。。’ and ‘Would it be possible for you to 。。。’ are conventional ways of framing requests。 ‘Are you able to 。。。’ is not。

Implicature

A communicated implication of an utterance。 A speaker can intend to mean more by her utterance than what the words that she utters mean， as the philosopher Paul Grice pointed out。

Andy： I think we should get a pet。

Bess： Cats are my favourite animals。

Here Bess’s utterance is true if and only if cats are her favourite animals。 However， in the context， it is likely that she conveyed more， in making her utterance， than this （and that she intended to do so）。 She intentionally and openly implied that she and Andy should get a cat （or cats） as pets。 Pragmatic theorists would say that she implicated that she and Andy should get a cat （or cats） as pets。

Implicature and the related verb implicate are technical terms coined by Grice to cover what a speaker means in making an utterance beyond what the speaker actually says。 On Grice’s account， this is a broad area， covering indirect answers to questions （as above）， figurative speech such as irony and metaphor， non-logical readings of logical words （e。g。 an utterance of ‘Some Xs are Y’ may implicate that not all are） and much else。

In his theory of conversation， Grice divided implicatures into conventional and conversational， and subdivided conversational implicatures into generalized and particularized。 All of these types of implicatures are distinct from what is said in that they do not contribute to the truth-conditions of an utterance。

Conversational implicatures， like the one in the example above， are not part of what the words of an utterance mean， but are inferred from what is said。 More precisely， they are inferred from the speaker’s saying of what is said， that is from the fact that it is said and the way it is put。 In Grice’s theory of conversation the inference relies on a Cooperative Principle and conversational maxims， discussed below。

In contrast， conventional implicatures are supposed to be carried by particular words。 For example， the word ‘but’ conveys an idea of contrast， but this contrast does not affect the proposition expressed by the utterance。 Grice suggested treating such cases as conventional implicatures。 These differ considerably from conversational implicatures and are dealt with in their own entry。

Grice also mentioned the possibility that non-conventional implicatures might arise from non-conversational maxims relating to social or moral behaviour and aesthetics， but never elaborated on the point。 This entry focuses on conversational implicatures， and from this point on， that is what is meant by the word ‘implicature’ on its own， as in most writings in pragmatics。

Grice’s theory of conversation is an attempt to show how a speaker can implicate and how a hearer can work out what was implicated。 Briefly， the account is that rational agents should cooperate in conversation and other talk exchanges： this is the Cooperative Principle。 This means that （other things being equal） the speaker will conform to certain standards （the conversational maxims）， such as telling the truth and not giving too much nor too little information。 When what a speaker says would violate a maxim or maxims if taken as the whole content of what the speaker is communicating， the hearer is entitled to look for some extra meaning beyond what was explicitly said， in order to preserve the assumption that the speaker is being cooperative。 The speaker is able to exploit this in a systematic way to convey implicatures。

In the example at the beginning of this entry， what is said， taken on its own – that cats are Bess’s favourite animals – would violate Grice’s maxim of relation。 Therefore Bess must also have implicated something that was directly relevant to the conversation， namely that she and Andy should get a cat as a pet。

Conversational implicatures have several interesting properties， most of which were pointed out by Grice when he introduced the term ‘implicature’， including

calculability

，

cancellability

，

indeterminacy

，

non-detachability

and

non-conventionality

。 Among properties that other theorists have noted are

reinforceability

and

universality

。

Non-conventionality is part of the definition of conversational implicatures。 That is， they do not stem directly from the encoded/timeless （or as some say ‘conventional’） meanings of words。

Implicatures are calculable。 This property is also a necessary condition for a conversational implicature。 By definition， an implicature must be able to be worked out （inferred， in fact） from what is said， and how it was said， and the fact that it was said。 What is said does not in general logically entail the implicature or implicatures （although it may）： the relationship is not one of logical deduction。 The inference is of the kind called ‘inference to the best explanation’。 The speaker has said something that on the face of it is irrelevant （or false， or over/under-informative， long-winded etc。）。 What is the best explanation for this？ In many cases the best explanation will be that the speaker intended to convey something more， an implicature。 The implicature is worked out on the basis of the hearer’s knowledge of the speaker， the conversational situation and the world more generally。

Various pragmatic theories propose different principles from Grice’s Cooperative Principle and maxims， but almost all retain the key idea of implicatures as something intentionally conveyed beyond what is actually said， and the notion of calculability， that the relevant intentions of the speaker can be inferred non-demonstratively on the basis of the words uttered and the context。

Implicatures are also non-detachable， with some exceptions。 Since they are conveyed by and inferred from what is said， it follows that saying the same thing in a different way will also give rise to the same implicature。 For example， if Bess had said ‘I prefer cats to dogs and birds’ or ‘Felines are better than all other animals，’ either would have conveyed much the same implicature as what she did say。 The exceptions to non-detachability are implicatures that rely on the way that something is said （in Grice’s framework， implicatures that rely on maxims of manner）。

‘Universality’ is an obvious extension of non-detachability。 Since what matters is what is said in the sense of what proposition is expressed， rather than what words were uttered， it does not matter what language it is said in。 Andy and Bess could have been having their conversation in any language from Acehnese to Zulu， and as long as they were expressing roughly the same propositions， Bess would have implicated roughly the same as she did。

Since implicatures are arrived at by inference to the best explanation， not linguistically decoded， they are cancellable。 On Grice’s account， the assumption that the speaker is being fully cooperative may not hold。 One type of cancellation is explicit： the speaker may realize or suspect that what she says will be taken to imply something she does not want to convey and explicitly say that it is not the case。 For example， a speaker might say ‘I broke a finger yesterday’ which would usually implicate that it was one of her own， and then go on by saying ‘But it wasn’t one of mine。’ It might be better to speak in such cases of deniability rather than cancellability， since it looks as though in these circumstances the speaker is simply denying the truth of something that she

implicated accidentally or mischievously。

The second kind of cancellation is contextual cancellation。 It is only clear with generalized or default implicatures， and is controversial， since it is controversial whether there are any generalized conversational implicatures。 Assume for the sake of argument that utterances that express propositions of the logical form ‘

’ （where

and

are propositions） normally implicate that the speaker does not know whether

（e。g。 an utterance of ‘John lives in Bloomsbury or Fitzrovia’ implicates that the speaker does not know which）。 In contexts in which it is clear that the speaker will not aim to be maximally informative the implicature may not arise。 For example， in a treasure hunt the organizer tells the participants ‘The prize is either in the house or in the garden。’ No implicature is conveyed that she does not know which is the case。

What is controversial here is whether it is ever better to treat such cases as involving cancellation of a normally present implicature or whether it is better simply to say that in this context no such implicature arises。 Grice allowed for the former with his category of generalized conversational implicatures。 A generalized conversational implicature of saying that

（where

is some proposition） is one that is normally carried by saying that

， that is in all but special circumstances。

For the same reason that implicatures （and other pragmatically inferred material） are deniable， they are reinforceable。 That is， a speaker can usually follow up an utterance that has an implicature by an assertion reinforcing what she implicated without much of a feeling of infelicitous redundancy。 In uttering the first sentence， the speaker communicates， but does not explicitly say that she broke her own finger：

I broke a finger yesterday。 One of mine， I mean。

Compare this with the odd effect of attempting to reinforce part of what is explicitly said：

？？ I broke a finger yesterday。 It was a finger。

The indeterminacy of implicatures is one final property that follows from the fact that implicatures are worked out by inference to the best explanation。 There may be several possible explanations， in which case what is implicated is one of these specific explanations。 The list of possible explanations may be indeterminately large。 In such cases what is implicated will be open in the sense that it is indeterminate。 This sort of pragmatic indeterminacy can be seen in poetic effects。

Implicature may be of great importance to the study of non-verbal communication， where most gestures， unlike words， do not have any intrinsic meaning。

In Grice’s work on meaning， he was clear that gestures that do not have any timeless or conventional meaning can nonetheless be used to convey particular speaker meanings on particular occasions。 For example， a little push might convey that the pusher wants the pushee to leave， if and only if the recipient of the push can work out that this is what the producer of the push meant by it。 At what level is this meaning conveyed？ It cannot be at the level of what is said， because a little push has no encoded/timeless meaning。 That apparently only leaves the possibility that what is conveyed here is implicated。 It is assumed in relevance theory that what is communicated by gestures with no encoded meaning is communicated at the level of implicature。

In retrospect， particularized （conversational） implicatures have been the most successful aspect of Grice’s theory of conversation。 All of the other major proposals of the theory – the maxims and the Cooperative Principle， generalized implicatures and conventional implicatures – have been seriously challenged， and no one of them is accepted in all major pragmatic theories。 Particularized implicatures are the only level of meaning beyond the proposition expressed that is accepted by just about everyone working in pragmatics。

Putting it this way brings out something important about the concept of implicatures that has not been discussed up to this point。

Implicatures are representational

。 Like what is said，

they are propositional: they have truth conditions

（although implicatures do not contribute to the truth-conditions of the utterance）。 So

to postulate implicatures is to postulate a level of representation; it is to claim that communication takes place on two levels

。 This is Grice’s extrapolation from the apparently obvious fact that speakers can mean more than – or something different from – what they say。

References

Allott， N。（2010）。

Key terms in pragmatics

（pp。 92-97， 102）。 London： Continuum International Publishing Group。

順便安利一發這本書，18年出了中文版

好書推薦 | 語用學核心術語

Leslie2020-01-04 00:46:14

Indirect speech act應該是和direct speech act 所對立而存在的，前者主要是指我們需要透過context （說話人之間的物理距離proximity，說話者的文化背景，說話者的認知能力；以及說話者之前的co-context，也就是之前討論的話或主題）來感受到說話者的動機；以及之後對接受者的驅動影響。與之相對的direct speech act 就比如說“把書還給我！” 這樣的所有文字意思的直接總和所產生的驅動影響。

implicature其實也是從context中解讀到言外之意，所以其實會又和indirect speech act 相重合的地方，只不過indirect speech act 強調語言有更進一步的驅動影響，即我們說一句話都是在表演一種行為，但是implicature 更多的是認知上理解言外之意。（以上都是俺根據我所學的解讀，希望可以幫到你）