Notes by RBJ on

Studies in the Ways of Words

by H.P. Grice

by way of summary
Logic and Conversation
Further Notes on Logic and Conversation
Postwar Oxford Philosophy
In Defence of a Dogma
Retrospective Epilogue
by way of critique
Logic and Conversation
Further Notes on Logic and Conversation
Postwar Oxford Philosophy
In Defence of a Dogma
Retrospective Epilogue


This is a book which I came across late in life, after a great deal of frustration in my attempts to write philosophy, partly associated with my antipathy to "the linguistic turn" which dominated analytic philosophy through much of the twentieth century.

I did philosophy as an undergraduate in the 70's, and had hoped to do some philosophy after my part-retirement in 1998. I am a poor reader, and nothing was further from my mind than to take on another philosopher whose conception of philosophy was primarily rooted in the study of ordinary language. I might never have looked at Grice were it not for the enthusiasm of J.L.Speranza and others on the analytic discussion list. (For a sample of my antipathy to this kind of philosophy, see my notes on J.L.Austin's "A Plea for Excuses".)

I was delighted therefore, when I actually got my hands on this book, to discover myself agreeing with Grice a great deal more than I usually do with any philosopher (probably because it starts out with a critique of certain kinds of ordinary language philosophising, doubtless if I reach the more constructive parts of the work my capacity for disagreement will revive).

By way of Summary

The book begins with and hangs around Grice's William James lectures, delivered at Harvard in 1967, which were on two "closely linked ideas", viz. that of association and implication (on the one hand) and of meaning (on the other). This theme is supplemented by a methodological or programmatic theme, the approach to philosophy through a study of language, particularly of ordinary language.


Here Grice tries "to characterize a type of maneuver by which a conclusion is drawn about the meaning of a word or phrase from the inappropriateness of its application in certain sorts of situation, and to suggest that a method is needed for determining when such a maneuver is legitimate and when it is not". In general there is specified some condition C in the absence of which the usage in question is considered "inappropriate".

Suspect cases he enumerates (at much greater length) include:

Now Grice considers these to be suspect, if not illegitimate because these examples are all cases in which "the condition C, the presence of which is suggested as being required for the application of a particular word or phrase to be appropriate, is such that most people would, I think, on reflection have a more or less strong inclination to say that to apply the word or phrase in the absence of that condition would be to say something true (indeed, usually trivially true), however misleading it would be to apply the word or phrase thus".

There follows discussion of what these various philosophers (A-philosophers) may have held to be the consequences of the alleged inappropriateness of the usages at issue. Attention is given to what Searle had to say about this matter, culminating in the following concise account of one supposedly tenable version of "Searle's Thesis":

That an utterence or remark to the effect that p, will be inappropriate if it is pointless; that it will be pointless, in many situations, unless there is a real or supposed possibility that it is false that p; and that these facts can be used to account for some of the linguistic phenomena which have stimulated A-philosophers.

Finally Grice gathers together his threads and connects the discussion of inappropriateness by A-philosophers with the philosophical program which he intends to pursue in the remaining lectures. Grice considers that the "suspect conditions" cannot be considered as conditions of applicability of a usage if that is intended to mean that under these conditions the statement would lack a truth value. He considers that the kinds of inappropriateness under consideration are best explained by reference to some very general conditions of discourse or of rational behaviour.

The programme upon which Grice now proposes to set out is the investigation of these general conditions, not with particular reference to the problems of inappropriateness addressed by A-philosophers, but rather "with a focus on their capacity for generating implications and suggestions".

Logic and Conversation


Grice introduces a distinction between:
  1. What is said.
  2. What is conventionally implicated.
  3. What is conversationally implicated.
Where all these terms are introduced as "terms of art".
"In some cases the conventional meaning of the words will determine what is implicated, besides helping to determine what is said."

Grice gives the example: He is an Englishman; he is, therefore, brave, to illustrate this, indicating that the entailment English => Brave is conventionally implicated but not explicitly "said" (in his "favoured sense").

Conversational implications are a subclass of non-conventional implications (hence the three above are disjoint).

To explain conversational implication Grice introduces his Cooperative Principle, which is:

"Make your conversational contribution such as is required, at the state at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged."

He then lists four categories of "conversational maxims" which spell out in greater detail what this involves:

  1. Quantity
    1. Make your contribution as informative as is required.
    2. And no more informative.
  2. Quality - try to make your contribution one that is true
    1. Stick to truths
    2. for which you have adequate evidence.
  3. Relation
    1. be relevant
  4. Manner - be perspicuous
    1. Avoid obscurity of expression.
    2. Avoid ambiguity.
    3. Be brief.
    4. Be orderly.
Grice believes that the standard type of conversational practice described by his maxims is in fact normally adhered to, and that it is reasonable to adhere to these principles. More specifically he would like to be able to show that:

"anyone who cares about the goals that are central to conversation/communication [...] must be expected to have an interest [...] in participating in talk exchanges that will be profitable only on the assumption that they are conducted in general accordance with the Cooperative Principle and the maxims"

Conversational implicature is then defined, along the following lines.

a speaker, by saying that p, has conversationally implicated that q, provided that:
  1. he is presumed to be observing the Cooperative Principle
  2. the supposition that he thinks that q is required in order to make his saying p consistent with that presumption
  3. the speaker thinks that the hearer is capable of concluding (2)
  4. the speaker would expect the hearer to think (3)

Examples of Conversational Implicature

Grice offers three groups of examples:
  1. Where it is not clear that any maxim is violated.
  2. Where a maxim is violated, but this is explained by the supposition of a clash with some other maxim.
  3. "exploitation", involving the flouting of a maxim in order to conversationally implicate by means of something like a figure of speech.

Generalized Conversational Implicature

Some propositions may be said to implicate without any special context, rather than as a result of some specific features of the context, this is called generalized implicature.

Five features are noted:

  1. Even generalized implicature can be explicitly or contextually cancelled.
  2. "one may expect a generalized conversational implicature that is carried by a familiar, nonspecial locution, to have a high degree of nondetachability."
  3. "initially at least, conversational implicata are not part of the meaning of the expressions to the employment of which they attach."
  4. Since there may be more than one possible explanation of an apparent violation of the Cooperative Principle, a conversational implicature may be a disjunction, and hence have a certain "indeterminacy".

Further Notes on Logic and Conversation

Total significance = The first two are part of the meaning but the third is not. Furthermore:
  1. The cooperative principle is usally observed.
  2. Necessary conditions of 1 are in systematic correspondence with conversational implications.




Postwar Oxford Philosophy

Here is a useful and interesting account of what Grice takes to be "the ordinary language approach to philosophy" as instantiated by his own.

He begins with the usual disclaimer that Oxford philosophers all practise the same kind of philosophy or adopt the same methods. Unfortunately, Grice is too aware himself of the unity which is captured by the phrase "ordinary language philosophy" and too honest to equivocate behind the pretence that there is none, so he doesn't make a lot of this disclaimer, making of it only an excuse for (the pretence of?) talking specifically of his own philosophy.

Grice subscribes to the following two propositions (as condensed by me):

  1. It is an important part of the philosopher's task to analyze, describe or characterize the ordinary uses of certain expressions. Grice particularly mentions that questions of linguisitic propriety may be of philosophical importance.
  2. A philosophical thesis which involves the rejection of some class of statements which would ordinarily be made and accepted as true is almost certain to be false.
Grice notes that he is not opposed to the use of technical terms.

He next considers three objections to his first credo, that part of the philosopher's task is to characterize the ordinary use of language.

  1. Concerning whether philosophy can be distinguised from sociology of language or lexicography.

    Response: Here Grice provides an account of the nature of conceptual analysis as a preliminary to indicating how this differs from sociology and lexicography. According to Grice:

    To be looking for a conceptual analysis of a given expression E is to be in a position to apply or withold E in particular cases, but to be looking for a general characterization of the types of cases in which one would apply E rather than withold it.
    Grice notes that only some concepts are of interest to philosophy, but attempts no characterization of those that are. He also notes that when engaging in conceptual analysis it is his own use which is subject to the analysis, and this is why no taking of polls is called for (taking of polls seems supposed by Grice to be the method of sociology).

    On the comparison with lexicography Grice concludes:

    .. dictionaries are designed for people who want to learn to use an expression correctly, whereas conceptual analyses are not.

  2. Ordinary language suffers from defects, namely ambiguity, misleadingness, vagueness, and the incorporation of mistakes or absurd assumptions, which render it unfit for conceptual analysis.

    Response (taking the defects in turn):

    1. ambiguity no bar to analysis
    2. if "misleading" is intended as philosophically misleading, then this is a reason for analysis, not against.
    3. vagueness may prevent a neat and tidy analysis, but does not prevent an analysis.

  3. "the sort of thing you say is an important part of philosophy is not worthy of the name `philosophy'. Philosophy is not just a matter of talking about words."


    1. Why speak of "words" with such contempt?
    2. There is a "fairly close connection" between modern conceptual analysis and discussions which form a central part of the writings of recognised "great philosopher" (my quotes).
    3. Grice himself does not claim that conceptual analysis is the whole of philosophy.

In Defence of a Dogma

In this paper Grice and Strawson aim to show that the criticisms offered by Quine of the analytic/synthetic distinction do not justify his rejection of it.

Sections 1-4 (of two dogmas)

They begin with a discussion of ways in which a distinction can be criticised, considering which of these Quine intends: Their view is that these kinds of criticism invite corresponding kinds of improvements rather than outright rejection, and that Quine's criticism is more radical.
"he declares ... that it is altogether illusory, that the belief in its existence is a philosophical mistake."
Such a position of radical scepticism is not (in general) in their view justified "merely by" by criticisms of attempts to clarify it. They observe that there are many distinctions which still await adequate elucidation but which are not rejected on that account. Quine's paper consists primarily of attempts at elucidation of the distinction, but include also an alternative which he represents as incompatible with belief in the distinction. If there is any presumption in favour of the distinction, they say, the critiques would not justify rejection, the rejection would then rest heavily on the latter. Grice and Strawson then discuss, by way of presumption in favour, a philosophical tradition which is "long, and not wholly disreputable", and present practice which suffices, they consider, to establish that there is a distinction. These considerations are sufficient to make them doubt that Quine really hold the extreme thesis which they have attributed to him, and they consider an alternative interpretation, viz. that he denies not that a distinction is marked, but rather that the distinction is the one which philosophers suppose it to be. However, Grice and Strawson then decide that even if this was Quine's position, a critique of elucidations would not suffice to establish it. Next we have an enumeration, following on from Quine's acceptance of the definability of analyticity in terms of "cognitive synonymy". The consequence of this is that the rejection of analyticity entails the rejection also of: They also conclude that if the notion of sentence synonymy is senseless, then so is that of sentence-significance. Now we come to discussion of the collection of terms which Quine alleges to be interdefinable with analyticity. They extract from Quine's discussion of these interelationships a criterion for what constitutes a satisfactory explanation which is "clearly unreasonable". They offer a similar example involving various moral concepts and conclude that such grounds are not in general sufficient to render senseless, and then go on to consider whether there are special conditions applying to the case in hand which make them sufficient in this case. The special consideration they consider is that the terms in question are "technical" terms, answering to the objection that some of them are not, that these non-technical terms have to be read in a special sense to belong here. Still, they don't think there is enough here. In defence of their denial that Quine's criteria are reasonable, they offer an illustration of the distinction between "logically impossibility" and "natural impossibility", by way of an informal explication of the difference.
definition and synonymy
Here is discussed Quine's denial that definition hold the key to synonymy and analyticity. Quine "argues" that except for explicit conventional introduction of new notations (these are called "extreme cases"), definitions rely on prior relations of synonymy. These cases Quine accepts as establishing synonymy. Grice and Strawson find Quine's position here to be incoherent. They seem to be taking Quine to be denying that the notion of synonymy makes sense in cases other than the extreme case, while accepting that it does in that case, and surely if it is senseless it would be in both cases.
everything green is extended
Grice and Strawson disagree with Quine on whether our doubts about the analyticity of this sentence are due to doubts about the meaning of the sentence or doubts about the meaning of "analytic". They observe that we have doubts not only about its analyticity but even about its truth, and trace these doubts to doubts about meanings, "should we count a point of green light as extended?".

Sections 5-6 (of two dogmas)

Grice and Strawson focus on just two assertions drawn from Quine's "positive account".
  1. there are not statements whose truth is immune from revision
  2. no statement can on its own be confirmed or disconfirmed
Of these two they say, Quine takes the first to be incompatible with the existence of the analytic/synthetic distinction and the second he takes to be a barring one way of explaining the distinction. Both of these contentions they reject. Grice and Strawson deny that either point has the supposed consequences. In the first case they point out that a revision in the status of an analytic statement involves a change of meaning, but this causes no difficulty for the tenability of the distinction. In the second, the don't even think that this poses an insuperable problem for someone who takes meaning to be determined by verification conditions.


In conclusion Grice and Strawson find that though the paper contains many valuable points, if fails to provide a justification for the rejection of the analytic/synthetic distinction.

Retrospective Epilogue

This essay is offered as a review of the "deeper aspects" of unity in the essays in this volume. Of these there are three.
  1. The connections between the topics in the essays are sometimes stronger than the essays make clear.
  2. The topics addressed are ones which Grice considers important, is still interested in, and on which Grice's view still needs to be made more thoroughly and clearly.
  3. They are applications of a certain method of philosophising in which interest in ordinary language plays an important part.
Grice lists the main strands in the work represented by the essays in this volume, then he goes on to talk about these strands..

They are (bracketed entries are interests of Grice not covered in this book):

  1. Philosophy of Perception
    1. The causal analysis of perception.
    2. The experential quality of different senses.
    3. (The analysis of statements describing objects of perception.)
  2. Defence of the Analytic/Synthetic distinction.
  3. Defence of the rights of the ordinary man and common sense vis-a-vis the professional philosopher.
  4. Meaning:
    1. "necessary to distinguish between a notion of meaning which is relativised to users of words or expressions and one that is not"
    2. un-relativised meaning must be understood in terms of the relativised meaning
  5. Further meaning distinctions:
    1. conventional and non-conventional meaning
    2. what is asserted and what is implicated
  6. Parallels between language and other rational activities.
  7. That phrases like "the King of France" should be considered as genuinely rather than ostensibly referential.
  8. That formal logic can be amended to meet the above requirement for phrases to be genuinely referential.
Grice closes with a discussion of "Philosophical Method and Ordinary Language".

Notes on the strands:

  1. Grice is looking for an aspect of the significance which has a claim to being considered central and around which other aspects of significance might cluster.

    He considers two candidates, "formality" and "dictiveness", and mentions two criteria, firstly that the central notions should be simple rather than complex and secondly that they should be direct rather than indirect. Now another pair of candidates are offered "conventional" (in contrast with more "informal or indirect relationship with the signifying expression") and "said" (by contrast with "implies", "suggests" or "hints"). The "conventional" is offered as the "formality" candidate "said" as the "dictiveness" candidate. These are said to be two distinct, logically independent, criteria of "centrality".

    In connection with both of these candidates, it emerges, Grice regards truth conditions as essential (though he does not appear to remark on this common feature).

Philosophical Method and Ordinary Language

Here Grice discusses the "general character" of his attitude towards ordinary language. This involves him in saying something about "linguistic botanising" as a foundation for conceptual analysis in general and philosophical analysis in particular, in which connection he notes that Oxford was unique in going beyond the mere association of philsophy with linguistic analysis by an "unswerving association" of philosophy with the study of ordinary language.

Two distinct problems are now identified:

  1. The aim is to characterise (and possibly justify) a philosophical method. Grice perceives a problem in that the achievement of such a characterisation requires or presupposes the ability to apply the method, which in turm presupposes the ability to say what that method is.
  2. Some philosophers (Russell and Quine are mentioned) are hostile to the idea that it should be a primary concern of philosophy to study ordinary language. Grice associates this hostility with scientism which he glosses as the idea that science supercedes rather than depends upon ordinary language, and that philosophy should be a part of or an auxiliary to such science.
The second of these problems seems to be connected by Grice with the distinction between hoi poloi and hoi oligoi or of intellectual elitism, in Oxford and its connection with Athenian philosophy. In discussing this Grice compares Oxford philosophy with the methods described in Aristotle, and finds them not the same. On the other hand he notes Austin's preference for Moore over Wittgenstein, and finds Moore to be closer to the Aristotelian model, but has a difficulty in finding a convincing explanations for Austin's preference for Moore. The distinction at issue here appears to be that between studying ordinary language (Austin), and taking seriously common sense (Moore). However, even if this gap were bridged, there is still quite a gulf between defending common sense and Aristotelian dialectic.

Finally we get into fairy Godmothers.

By way of Critique


It seems to me curious how little credit is given to Wittgenstein.

I was rather pleased to see Grice's list, for it contains many examples which I have come across and which I have previously myself felt to be unsound. Pleased to see them, I suppose, because my familiarity with the literature has been insufficient for me to have seen these criticised before.

However, after the initial exitement at seeing the list, my eagerness to go on through Grice's further analysis, for example, of exactly what the attitude of these various philosophers is to the usage which they claim to be "inappropriate", soon begins to flag. Grice of course, unlike me, is interested in doing philosophy through the study of ordinary language, so he must carefully sort out this chaff so that he can continue the enterprise.

This I guess is how he gets to things like "conversational implicature", which begins now to seem like prophylactic philosophy applied to Oxonian elaborations of Wittgensteinian eccentricities.

Logic and Conversation

Further Notes on Logic and Conversation

The intro is a bit overelaborate so far as I can see. The explanation for "conversational implication" is in effect: "in systematic correspondence with" assumptions necessary in order to maintain the supposition that the cooperatice maxim is being maintained. But logically, this is surely: the things implied by supposition that the cooperative maxim is being maintained. (am I being picky?) But then he says: "non-trivially required".

Postwar Oxford Philosophy

Retrospective Epilogue

My personal feelings about the strands:

I'm pleased to see Grice defending causal theories of perception and the Analytic/Synthetic distinction. I appreciate some of the insights he has brought to the study of ordinary language. I don't concur with his attitude either to the value of ordinary language philosophy or to the ordinary man and common sense. I can't get exited about whether "the King of France" is genuinely referential, but if "genuinely referential" has an ordinary sense then I would be inclined to doubt that it is, in that sense.

Philosophical Method and Ordinary Language

Note how Grice in the opening sentences of this section tells us (implies or implicates) that philosophical analysis is subsumed by conceptual analysis.
In summary:

UP HOME © RBJ created 2003/5/2 modified 2010/5/12