The Beginnings of a Modern Debate in Epistemology and the Philosophy of Language Essay

The Beginnings of a Modern Debate in Epistemology and the Philosophy of Language Essay

Debates arose in the 1940s over the question of how much significance ought to be attached to the relation between ordinary uses of language and our understandings of philosophical problems. Arguments in support of the motion were championed by Austin, and arguments against the motion were (qualifiedly and reservedly) championed by Grice. These debates have impacted significantly on modern debates in epistemology, semantics and pragmatics. As such, we will do well to briefly outline one significant vein in the beginnings of this modern debate. I aim to (a) outline Austins commendation that we take ordinary language use seriously as playing a role in establishing the meaning of our concepts. (b) I consider objections raised by Grice, and developed in a specifically epistemological vein by Stroud, that aim to deflate the importance of appeal to ordinary language use as stressed by Austin. (c) I summarise some questions that these arguments have left requiring further answers to.

(a) Austin and the Ordinary Use of ‘Know’

Unconvinced, perhaps, with philosophers answers to epistemological questions, and dissatisfied with their methods of approach, Austin focused on considerations of how ordinary English sentences are used in order to understand ‘what sort of thing does actually happen when ordinary people are asked ‘How do you know?’’. A general example may serve to focus the issues. Consider a wholly usual instance where a sentence of the form ‘S knows that p’ is uttered, a statement involving a (supposed) particular, current, empirical fact; Austin’s favoured example was ‘I know that is a goldfinch’. On making such an assertion, one is immediately opened up to the question ‘How do you know?’. Austin specifies two such conditions through which responses to the question ‘How do you know?’1; are ordinarily judged as either adequate or inadequate,

(a) If you say ‘That’s not enough’, then you must have in mind some more or less definite lack … If there is no definite lack, which you are at least prepared to specify on being pressed, then it’s silly (outrageous) just to go on saying ‘That’s not enough’.
(b) Enough is enough: it doesn’t mean everything. Enough means enough to show that (within reason, and for present intents and purposes) it ‘can’t’ be anything else, there is no room for an alternative, competing, description of it. (1946/2007, p. 77)

Condition (a), the inadequacy condition, states that in order for a response to be deemed inadequate, one must have in mind some reason for judging the response as inadequate. For example, the proposed goldfinch might instead be a woodpecker, given that one’s claim to know is based only on seeing a red head, and the woodpecker also has a red head. Further, one cannot simply deem a response inadequate by saying “That’s not enough”; well, of course someone could adopt this strategy, however as long as one has offered a somewhat reasonable response and has heard no substantiated objection to the claim, anyone who raises an objection without any reason for doing so forgoes their right to be taken as a serious collaborator in a constructive conversation. Condition (b), the adequacy condition, states that in order for a response to be deemed adequate, one must offer enough grounds to allay, within reason, any cause for doubt. Sufficient grounds for allaying doubt do not entail disqualifying every conceivable possibility incompatible with one’s claim to know, such as that the goldfinch is stuffed, or a very cleverly disguised mule, one must only discount possibilities relevant to the conversational circumstances. Overall, Austin observes that, in ordinary discursive practices, conditions for applying or withholding that someone ‘knows’ are met by satisfying certain features deemed as relevant to the conversational situation.

What is of note is that Austin’s approach to the question ‘How do you know?’ is undertaken in stark contrast to many epistemologists often austere approaches to the question, where understanding ‘how we know’ is approached in terms of certainty, true justified belief or whatnot. It is however, a further, and problematic, question to ask what philosophical conclusions Austin thought ought to be drawn from these considerations. Austin readily emphasised that his studies of ordinary uses of language were often of interest in themselves, quite independently from broader philosophical considerations, and that appeal to ordinary use would not be the universus quod terminus of philosophical enquiry but only the beginning. On the other hand, Austin’s reflections on his preferred methodology are somewhat prescriptive. Acutely sensitive to philosophers propensities to muddle their way through language in order to tackle philosophical problems, Austin urged that as words are our tools ‘as a minimum, we should use clean tools’2;. Perhaps Austin’s most encompassing and useful methodological remark is that,

‘Ordinary language is not the last word: in principle it can everywhere be supplemented and improved upon and superseded. Only remember, it is thefirst word.’(1956/2007, p. 181)

That is, Austin may be understood as claiming that analysis of ordinary language ought to be considered as an essential, and all too often overlooked, tool for setting the groundwork for philosophical analysis. Such analysis may then, in turn, prove to ‘supersede’ our reasonings arrived at through our ordinary uses of language. Important to note, however, is Austin’s stress on ordinary language being the ‘first word’. This emphasis could be representative of a fairly trivial reminder that we must, at some stage in our investigations, appeal to ordinary uses of words in order to check for whether theory and practice cohere. However, it is likely that Austin is here advising us to beginanalysis through appealing to ordinary uses of language. Austin’s painstakingly detailed accounts of ordinary uses of language indicate a commitment to this latter enterprise. This claim, however, would also be stronger than merely advising that theory must check in with ordinary practice at some stage. It seems one thing to say that a philosophical theory gains credence through its ability to cohere with and explain ordinary uses of language, but quite another to recommend that the theory must proceed directly from appeals to ordinary uses of language. At the least, supposing we ought to begin from there disguises an assumption that how words are used will be indicative of establishing what the meaning of those words will likely be. This is perhaps why Austin felt inclined to add the cautionary ‘in principle’ as to the possibility of philosophical analysis being capable of superseding what we understand by observing how words are used. Grice came to question the importance Austin attached to appeals to ordinary use, and it is to his argument, and Stroud’s adoption of it, to which I now turn.

(b) The Grice-Stroud Objection to the Adequacy Conditions

Grice was the first philosopher to present and analyse a distinction between what a speaker uses words for from the literal meaning of the words used. The former, Grice argued, can be understood in terms of pragmatic principles of conversation, and the latter in terms of fixed conventional standards3. For Grice, the primary goal of using words is to communicate one’s intentions to others4, however, he also notes that what is intended to be communicated by what is said, by the words that are used, often deviates from the standard literal meanings of the words themselves. I can say, for instance, “The grass is getting long”, the literal meaning of this sentence being a mere descriptive statement, but intend to communicate to you something quite different, for instance that it’s your turn to cut the grass. Grice claimed that we are able to bridge this deviational gap between the words used and the conventional meaning of the words through recognising implicit conversational implicatures, implicit suggestions in the conversation that are not explicitly stated by the words uttered.

What is important here is Grice’s claim that analysis of how words are used is a separable affair to understanding the meaning of such words. Grice initially used this distinction to defend a causal theory of perception against its ordinary language philosopher detractors (such as Austin). He argued that the common implication of someone using the ‘it looks to me as if…’ locution is that there is some doubt about the veracity of what one sees. A flouting of this implication would be to say ‘it looks to me as if…’ when one has no obvious reason to doubt; Grice remarks ‘there would be something at least prima facie odd about my saying “That looks to me” (not as a joke) when I am confronted by a British pillar-box in normal daylight at a range of a few feet’5. Nevertheless, whilst conversationally inappropriate and misleading, this need not detract from the literal truth of what has been uttered. Similarly, simply because one doesn’t say ‘I have the sense-datum as of a red pillar-box in front of me’, doesn’t entail that this is not the fact of the matter.

In broad agreement with Grice’s observations, Stroud argues that questions regarding the conversational appropriateness of the use of the word ‘know’, of the pragmatic principles that govern ordinary uses of the word, can be distinguished from questions regarding the truth of the knowledge claim. He gives the example of being at a party where the host asks “Do you know whether John, your good friend, is coming to the party?”6. On the basis that he has just spoken to John, who said he was coming, and that John is a reliable and trustworthy individual, who does not live far away, Stroud appropriately responds that he knows John will be arriving shortly. The host accepts that Stroud knows (Austin’s condition (b) is satisfied), and everybody continues mingling. However, suppose that, on his way to the party, John gets struck by a meteorite and so fails to turn up. As it turns out, then, Stroud didn’t in fact know John would be at the party. As such, whilst Stroud has appropriately claimed to ‘know’, the claim has turned out wholly false, and so Stroud didn’t in fact know that John would arrive, because John never arrived. Furthermore, even though it would be entirely inappropriate for the host to turn to Stroud and say “See, you didn’t know John was coming”, it would have to be conceded that the host was speaking the truth.

The conclusion Stroud draws from this is that whilst evaluations of uses of the word ‘know’ are formed according to standards as to what it is deemed conversationally appropriate to say, the conditions for knowing something can only be satisfied when what one claims to know is actually true. Furthermore, as the above example illustrates, there are often cases where the conversational appropriateness of what is spoken and the literal truth of what is spoken come apart. In short, there are conversationally appropriate uses of the word ‘know’ that are not cases of knowledge. Given this, Stroud concludes that ‘evidence from usage or from our practice will not establish a conclusion about the conditions of knowledge’7. The upshot of this analysis is clear, if how words are used do not, by themselves, determine the meaning of such words, then theories which stress the connection, such as Austin’s, may prove fundamentally misguided in virtue of disguising the multifarious ways through which use and meaning often diverge.

(c) Questioning the Sufficiency of the Grice-Stroud Objection

Stroud’s objections to Austin’s adequacy conditions are primarily founded on examples where the use of a word and its truth value are in opposition. This surely suffices to indicate that use and meaning are not co-extensive concepts. Nevertheless, Stroud pays scant attention to the thought that when the use of a word and the truth value successfully coincide there is nothing of value to be learnt here, in fact he devotes just one sentence to the thought: ‘even if John had actually arrived at the party—I do not think it was true when I hung up the telephone that I knew John would not be hit by a meteorite’8. We may accept this statement, and accept that Stroud, in the example, did not know that John wouldn’t be hit by a meteorite. However, it does not immediately follow that because one could not rule out this possibility, one cannot therefore know about John’s arrival. Why, we may ask, if one has ruled out a significant amount of possibilities in conjunction with its also being true that John arrives, should we not consider this as a case of knowledge? Why ought the possibility of a meteorite be a relevant consideration?

Another remark of Stroud’s that is acceptable, but also disguising of further considerations, is: ‘once the question [‘Do you know that John won’t be struck by a meteorite?’] is asked, however inappropriately, can it be said that I do know that that possibility will not obtain? It seems to me that it cannot’9. It seems fairly obvious that one cannot rule out the possibility of the renegade meteorite, but again, suppose the meteorite doubt is never raised, suppose the only possibilities considered are ones overtly relevant to the situation (that John is reliable, trustworthy etc.), and again, suppose it is true that John arrives, why should we not consider this as a case of genuine knowledge?

These questions bring out a lacuna in Stroud’s argument; Stroud aims to show that considerations of usage, of the sort that Austin provides, do not establish a conclusion about the conditions for knowledge, and he proceeds to argue this by showing that one can appropriately use the word ‘know’, and yet fail to know. However, aside from the above brief remarks, Stroud offers no justification for denying the possibility that when one claims to know a certain matter, one has ruled out all of the conversationally relevant possibilities, and the matter is true, that one has not succeeded in possessing genuine knowledge.

These initial arguments may be described as establishing the following. As Grice brought out and Stroud applied to the use of ‘know’, how we use words does not, by itself, always determine the meaning of the words used; use and meaning often diverge. A moment’s reflection assures us that in order to know that something is true, that thing has to be true, and if the word ‘know’ is used positively on an occasion when, by fact, the claim is false, then no matter how socially appropriate or adequate one’s use is, it is not an instance of knowledge. Nevertheless, for all the above criticisms show, they show that (i) Austin’s appeal to use may not be sufficient for understanding the concept of knowledge, and that (ii) a claim to know must be true in order to count as knowledge. The question we are currently left with becomes: Can a theory based on considerations of use fill this lacuna and provide a response to Stroud?


J. L. Austin, ‘A Plea for Excuses’ (1956). Reprinted in his Philosophical Papers, Third Edition (2007); Oxford: Oxford University Press.

J. L. Austin, ‘How to Talk—Some Simple Ways’ (1952). Reprinted in his Philosophical Papers, Third Edition (2007); Oxford: Oxford University Press.

J. L. Austin, ‘Other Minds’ (1946). Reprinted in his Philosophical Papers, Third Edition (2007); Oxford: Oxford University Press.

J. L. Austin, ‘Truth’ (1950). Reprinted in his Philosophical Papers, Third Edition (2007); Oxford: Oxford University Press.

J. L. Austin, ‘Unfair to Facts’. Printed in his Philosophical Papers, Third Edition (2007); Oxford: Oxford University Press.

K. Graham, J. L. Austin: A Critique of Ordinary Language Philosophy, (1977); Hassocks: The Harvester Press.

H. P. Grice, ‘Logic and Conversation’, (1967a). Reprinted in his Studies in the Way of Words, (1989); Cambridge Mass.: Harvard University Press.

H. P. Grice, ‘Meaning’, (1957). Reprinted in his Studies in the Way of Words, (1989); Cambridge Mass.: Harvard University Press.

H. P. Grice, ‘The Causal Theory of Perception’, (1961). Reprinted in his Studies in the Way of Words, (1989); Cambridge Mass.: Harvard University Press.

H. P. Grice, ‘Utterer’s Meaning, Sentence-Meaning, and Word-Meaning’, (1967b). Reprinted in his Studies in the Way of Words, (1989); Cambridge Mass.: Harvard University Press.

O. Hanfling, Philosophy and Ordinary Language, (2000); London: Routledge.

A. Norman, ‘Epistemological Contextualism: Its Past, Present, and Prospects’, (1999), Philosophia 27.

B. Stroud, The Significance of Philosophical Scepticism, (1984); Oxford: Oxford University Press.

1J. L. Austin, (1946/2007), p. 77.
2J. L. Austin, (1956/2007), p. 181. It is of note that many of Austin�s methodological reflections referred to above occur in this article.
3It is of note that Grice�s own analysis of literal meaning in terms of the intentional states of the speaker is highly controversial (see H. P. Grice (1967)).
4See, for example, H. P. Grice (1957) for an influential early paper defending this view.
5H. P. Grice, (1961/1989), p. 227.
6See B. Stroud (1984), pp. 58-61.
7Ibid. p. 64.
8Ibid. p. 62.
9Ibid. p. 61. Emphasis added.

Related Essays

Leave a Reply

Your email address will not be published. Required fields are marked *