H. P. Grice: The conversational implicatum -- 1961

I shall embark on a discursus about certain aspects of the concept or concepts of irnplication, using some more or less well-worn examples. 3 (Implication) I shall introduce four different examples and use upon them four different ideas as catalysts. All are cases iri which in THE CAUSAL THEORY OF PERCEPTION ordinary parlance, or at least in philosophical parlance, something might be said to be implied as distinct from being stated. (l) " Smith has lcft off beating his wife ", where what is implied is that Smith has been bcating his wifc. (2) " She was poor but she was honcst ", whele what is implied is (vcry roughly) that there is some contrast between poverty and honesty, or between her poverty and her honesty. The first cxample is a stock case of what is sometimes called " prcsupposition " and it is often held that here 1he truth of what is irnplicd is a necessary condition of the original statement's beirrg cither true or false. This might be disputed, but it is at lcast arguable that it is so, and its being arguable might be enough to distinguish-this type of case from others. I shall however for convenience assume that the common view mentioned is correct. This consideration clearly distinguishes (1) from (2); even if the implied proposition were false, i.e. if there were no reason in the world to contrast poverty with honesty either in general or in her case, the original statement could still be false; it would be false if for example she were rich and dishonest. One might perhaps be less comfortable about assenting to its truth if the implied contrast did not in fact obtain; but the possibility of falsity is enough for the immediate purpose. My next experiment on these examples is to ask what it is in each case which could properly be said to be the vehicle of implication (to do the implying). There are at least four candidates, not necessarily mutually exclusive. Supposing someone to have uttered one or other of my sample sentences, we may ask whether the vehicle of implication would be (a) what the speaker said (or asserted), or (b) the speaker (" did he imply that . . . .':) or (c) the words the speaker used, or (d) his saying that (or again his saying that in that way); or possibly some plurality of these items. As regards (a) I think (1) and (2) differ; I think it would be correct to say in the case of (l) that what he speaker said (or asserted) implied that Smith had been beating this wife, and incorrect to say in the case of (2) that what te said (or asserted) implied that there was a contrast between e.g., honesty and poverty. A test on which I would rely is the following : if accepting that the implication holds involves one in r27 128 H. P. GRICE accepting an hypothetical' if p then q ' where 'p ' represents the original statement and ' q' represents what is implied, then what the speaker said (or asserted) is a vehicle of implication, otherwise not. To apply this rule to the given examples, if I accepted the implication alleged to hold in the case of (1), I should feel compelled to accept the hypothetical " If Smith has left off beating his wife, then he has been beating her "; whereas if I accepted the alleged implication in the case of (2), I should not feel compelled to accept the hypothetical " If she was poor but honest, then there is some contrast between poverty and honesty, or between her poverty and her honesty." The other candidates can be dealt with more cursorily; I should be inclined to say with regard to both (l) and (2) that the speaker could be said to have implied whatever it is that is irnplied; that in the case of (2) it seems fairly clear that the speaker's words could be said to imply a contrast, whereas it is much less clear whether in the case of (1) the speaker's words could be said to imply that Smith had been beating his wife; and that in neither case would it be evidently appropriate to speak of his saying that, or of his saying that in that way, as implying what is implied. The third idea with which I wish to assail my two examples is really a twin idea, that of the detachability or cancellability of the implication. (These terms will be explained.) Consider example (1): one cannot fi.nd a form of words which could be used to state or assert just what the sentence " Smith has left off beating his wife " might be used to assert such that when it is used the implication that Smith has been beating his wife is just absent. Any way of asserting what is asserted in (1) involves the irnplication in question. I shall express this fact by saying that in the case of (l) the implication is not detqchable from what is asserted (or simpliciter, is not detachable). Furthermore, one cannot take a form of words for which both what is asserted and what is implied is the same as for (l), and then add a further clause withholding commitment from what would otherwise be implied, with the idea of annulling the implication without annulling the assertion. One cannot intelligibly say " Smith has left off beating his wife but I do not mean to imply that he has been beating her." I shall express this fact by saying that in the case of (1) the implication is not cancellable (without THE CAUSAL THEORY OF PERCEPTION r29 cancelling the assertion). If we turn to (2) we find, I think, that there is quite a strong case for saying that here the implication ls detachable. Thcrc sccms quitc a good case for maintaining that if, instead of sayirrg " She is poor but shc is honcst " I were to say " She is poor and slre is honcst", I would assert just what I would havc asscrtcct ii I had used thc original senterrce; but there would now be no irnplication of a contrast between e.g', povery and honesty. But the question whether, in tl-re case of (2), thc inrplication is cancellable, is slightly more cornplex. Thcrc is a sonse in which we may say that it is non-cancellable; if sorncone were to say " She is poor but she is honest, though of course I do not mean to imply that there is any contrast between poverty and honesty ", this would seem a puzzling and eccentric thing to have said; but though we should wish to quarrel with the speaker, I do not think we should go so far as to say that his utterance was unintelligible; we should suppose that he had adopted a most peculiar way of conveying the the news that she was poor and honesl. The fourth and last test that I wish to impose on my exarnples is to ask whether we would be inclined to regard the fact that the appropriate implication is present as being a matter of the meaning of some particular word or phrase occurring in the sentences in question. I am aware that this may not be always a very clear or easy question to answer; nevertheless Iwill risk the assertion that we would be fairly happy to say that, as regards (2), the factthat the implication obtains is a matter of the meaning of the word ' but '; whereas so far as (l) is concerned we should have at least some inclination to say that the presence of the implication was a matter of the meaning of some of the words in the sentence, but we should be in some difficulty when it came to specifying precisely which this word, or words are, of which this is true. I may now deal more briefly with my remaining examples. (3) I am reporting on a pupil at Collections. All I say is " Jones has beautiful handwriting and his Englishis grammatical." We might perhaps agree that there would here be a strong, even overwhelming, implication that Jones is no good at philosophy. It is plain that there is no case at all for regarding the truth of what is implied here as a pre-condition of the truth or falsity cf 130 H. P. GRICB what I have asserted; a denial of the truth of what is implied would have no bearing at all on whether what I have asserted is true or false. So (3) is much closer to (2) than (1) in this respect. Next, I (the speaker) could certainly be said to have implied that Jones is hopeless (provided that this is what I intended to get across) and my saying that (at any rate my saying /s/ that and no more) is also certainly a vehicle of implication. On the other hand my words and what I say (assert) are, I think, not here vehicles of implication. (3) thus differs from both (1) and (2). The implication is cancellable but not detachable; if I add o'I do not of course mean to imply that he is no good at philosophy " my whole utterance is intelligible and linguistically impeccable, even though it may be extraordinary tutorial behaviour; and I can no longer be said to have implied that he was no good, even though perhaps that is what my colleagues might conclude to be the case if I had nothing else to say. The implication is not however, detachable; any other way of making, in the same context of utterance, just the assertion I have made would involve the same implication. Finally, the fact that the implication holds is not a matter of any particular word or phrase within the sentence which I have uttered; so in this respect (3) is certainly different from (2) and, possibly different from (1). One obvious fact should be mentioned before I pass to the last example. This case of implication is unlike the others in that the utterance of the sentence " Jones has beautiful handwriting etc." does not standardly involve the implication here attributed to it; it requires a special context (that it should be uttered at Collections) to attach the implication to its uttgrance. (4) If someone says " My wife is either in the kitchen or in the bedroom " it would normally be implied that he did not know in which of the two rooms she was. This example might well be held to be very similar to the case under dispute, that if such statements as " This looks red to me " so I must be careful not to prejudge any issues to my objector's disadvantage. I think, however, that in the case of (a) I can produce a strong argument in favour of holding that the fulfllment of the THE CAUSAL THEORY OF PERCEPTION implication of the speaker's ignorance is not a precaution of the truth or falsity of the disjunctive statement. Suppose (c) that the speaker knows that his wife is in the kitchen, (b) that the house has only two rooms (and no passages etc.) Even though (a) is the casc, thc spcaker can certainly say truly " My wife is in the housc "; he is merely not being as informative as he could bc if nccd arose. But the true proposition that his wife is in thc housc together with the true proposition that the house consists entirely of a kitchen and a bedroom, entail the proposition that his wife is either in the kitchen or in the bedroom. But il to cxpress the proposition p in certain circumstances would bc to spcak truly, and p, togelher with another true proposition, crrtails q, then surely to express 4 in the same circvmstances must be to speak truly. So I shall take it that the disjunctive statement in (4) does not fail to be true or false if the implied ignorance is in fact not realized. Secondly, I think it is fairly clear that in this case, as in the case of (3), we could say that the speaker had irnplied that he did not know, and also that his saying that (or his saying that rather than something else, v2., in which room she was) implied that he did not know. Thirdly, the irnplication is in a sense non-detachable, in that if in a given context the utterance of the disjunctive sentence would involve the implication that the speaker did not know in which room his his wife was, this implication would also be involved in the utterance of any other form of words which would make the same assertion(e.g., "The alternatives are (1) . . . .(2) .. . ." or " One of the following things is the case: (a) . . . . (r) . . . . "). ln another possible sense, however, the implication could perhaps bc said to be detachable: for there will be some contexls of ruttcrance in Which the normal implication will not hold; e.g., thc spokesman who announces, " The next conference will be cither in Geneva or in New York " perhaps does not imply that lrc does not know which; for he may well be just not saying which. This points to the fact that the implication is cancellablg; :r nrarl could say, " My wife is either in the kitchen or in the bctlroorn " in circumstances in which the implication would rrornrally be present, and then go on, " Mind you, I'm not saying tlrrrt I don't know which"; this might be unfriendly (and grcr'lrrps ungrammatical) but would be perfectly intelligible, I2 131 132 H. P. GRICB Finally, the fact that the utterance of the disjunctive sentence normally involves the implication of the speaker's ignorance of the truth-values of the disjuncts is, I should like to say, to be explained by reference to a general principle governing the use of language. Exactly what this principle is I am uncertain, but L first sftol would be the following: "One should not make a weaker statement rather than a stronger one unless there is a good reason for so doing." This is certainly not an adequate formulation but will perhaps be good enough for my present purpose. On the assumption that such a principle as this is of general application, one can draw the conclusion that the utterance of a disjunctive sentence would imply the speaker's ignorance of the truth-values of the disjuncts, given that (a) the obvious reason for not making a statemcnt which there is some call on one to make is that one is not in a position to make it, and given (6) the logical fact that each disjunct entails the disjunctive, but not vice versa; which being so, the disjuncts are stronger than the disjunctive. lf the outline just given js on the right lines, then I would wish to say, we have a reason for refusing in the case of (4) to regard the implication of the speaker's ignorance as being part of the meaning of the word'or'; someone who knows about the logical relation between a disjunction and its disjuncts, and who also knew about the alleged general principle governing discourse, could work out for hirnself that disjunctive utterances would involve the implication which they do in fact involve. I must insist, however, that my aim in discussing this last point has been merelyto indicate the position I would wish to take up, and not to argue scriously in favour of it. My main purpose in this sub-section has been to introduce four ideas of which l intend to make some use; and to provide some conception of tlre ways in which they apply or fail to apply to various types of implication. I do not claim to have presented a systematic theory of implication; that would be a very large undertaking and one for another occasion.