When Words Are Called For: A Defense of Ordinary Language Philosophy

In order to appreciate the radical thesis of this work and the case Baz defends -- that Ordinary Language Philosophy (OLP) has been fundamentally misunderstood and therefore unfairly put to rest in the analytic tradition when in fact it still constitutes a 'best-practice' for doing philosophy -- it is necessary from the outset to make clear that for OLP, a word's meaning remains 'in limbo' (for all intents and purposes) until it is determined by context. Until then, it better be looked upon as a variable and not as a given. What Baz reveals in his new book is the astonishing fact that, even in eulogizing OLP as it is being pronounced dead, many of its half-way sympathizers suffer precisely from continuing to hold on to a semanticist/pragmatist distinction in meanings when, for OLP, there is none.

The 'decline' of the interest in OLP as an approach is vouched for by the fact that, among other things, not a single book in English on the one person who first tried to articulate this approach, J. L. Austin, has been published in the last thirty years or so (the exception, this year, is a volume edited by Martin Gustafsson and Richard Sorli, published by Oxford University Press, to which Avner Baz contributed versions of his third chapter in the book under review). A biography of Austin -- the professional impact of whose sudden and premature death in 1960 at the age of 48 on OLP's founding early years and his colleagues cannot be underestimated -- is also now in the works, and will hopefully shed more light on a region of under-exploited intellectual wealth that has long been kept in the dark.

It may be well in this connection, and before turning to Baz, to say a few words on Austin's legacy as well as on OLP's inception. We turn our attention here to a buzzing philosophical activity in post-war England, and primarily among Oxford's young dons, animated by Austin, but including a number of older and already influential colleagues like Gilbert Ryle, editor of Mind. Here, Oxford seemed to be cutting a way for itself, leaving Russell and his Cambridge colleagues -- including their celebrated 'darling' Wittgenstein -- behind (and out).^[1] With Germany's defeat in WWII, an entire page in history was felt to have been turned. During the war, Austin had been recruited to set up, and ended up heading, the "order of battle" section of what became SHAEF (the Supreme Headquarters of the Allied Expeditionary Force) under Eisenhower. The section was responsible for collecting and analysing information from a variety of sources, including the top-secret Enigma at Bletchley Park, but also through the developing art of aerial reconnaissance (which later became satellite imaging) and human intelligence from the resistance across Europe, in support of the war effort generally and to prepare for the D-Day landing.

It is said that when the German army surrendered at Frankfurt, Austin was the only person amongst the Allies who knew where all of the German army was actually located.^[2] Returning to do philosophy at Oxford from this high-level Intelligence posting, it was natural for the young Austin to try applying this very special war experience in his resumed philosophical investigations. He set himself the task (again, as he preferred it, and had found more effective during the war, through team-work) of demystifying philosophical concepts in a somewhat parallel way, one imagines, to the manner he employed as scattered data (e.g., pictures) or separate pieces of information (e.g., a train movement) were painstakingly 'put to work' in order to interpret the data being gathered -- very much a bottom-up, piece-by-piece approach to finding out what these meant.

A full appreciation of this background may help clarify Austin's articulation of OLP.^[3] It is one thing to fit pieces of a jigsaw puzzle together, where each piece is already uniquely shaped to be inserted into just one place in the overall puzzle -- their meanings, one might say, already determined by the investigator to be part of one specific overall picture.  It is a fundamentally different thing to try to fit together intelligence pieces of information, where each piece could be fitted with another alongside it either in one way as part of one possible overall picture, or in some other way with other pieces alongside it as part of an altogether entirely different overall picture.  In the latter case, one and the same piece of information, such as a train movement, could either mean the entire army is about to move or that only a battalion is being relocated.  This case requires what we might call 'a pragmatist's approach' to determining meanings.

The difference between the two approaches to determining what meanings words have is staggering, and far-reaching.  What Baz succeeds in showing is that its implication has not yet been fully digested even by contextualist practitioners of OLP, who continue to insist -- to use the intelligence-gathering metaphor -- that besides fitting the way it turns out to be actually fitted in the revealed enemy plan, that very same piece of intelligence information continues to retain its 'fittedness', so to speak, in one or another presumed (though disqualified) plan (or in any and all of them). Taking the intelligence metaphor one step further, one could also see how Austin and his colleagues might have come to deal with some pieces of intelligence information they suspected of being red-herrings, that is, as perlocutionary acts for which it was more important to assess what effect these were meant to produce than what truth-value their presumed information-content had.

Investigating classically challenging concepts like knowledge or truth in this totally new way-- from the language bottom-floor up, so to speak -- must have baffled professionals and non-professionals alike. But such had been the impact of this fresh and English-homegrown way of looking at the world, and of doing philosophy (freed of any Germanic or Continental 'baggage'), that it found itself by the end of the fifties suddenly in the eye of a public storm, generated by the publication, in 1959, of the highly polemical Words and Things by philosopher/anthropologist Ernst Gellner. This aspect of OLP's history is well-known. Gellner's work, as well as the circumstances of its publication, contributed to moving OLP out of academia and into the wider 'public intellectual' arena: Oxford's Gilbert Ryle, as editor of Mind, refused a request by Gellner's publisher to review the work in his journal, claiming that the book did not warrant a serious review. Provoked by Ryle's attitude, Cambridge's Bertrand Russell stepped in to question Ryle's motives through a letter he sent to The Times. Ryle duly responded in the same paper, signaling the start of an unprecedented historic duel in the press between the supporters of Oxford and Cambridge's preeminent philosophers.^[4]

Already at the academic level, such challenges as that of Oxford's young Strawson (1950) to Russell's already well-established and famous Theory of Descriptions (1905) -- published in none other but Mind itself -- reflected the underlying tug-of-war around the limits that should be set by natural languages over formal languages for specific philosophical purposes. But now OLP had become a public issue, warranting editorials in such leading newspapers and magazines as The Times and The Economist. Among other accusations Gellner made was the claim that Oxford's new philosophers were nothing but an elitist group of individuals belonging to a privileged class who, in line with Britain's forced withdrawal from its colonial role, also have decided to shrink to their island's size the traditionally global concerns of philosophy. Their focus on analyzing (the English) language was thus simply a reflection of the politically diminished and newly introverted Britain. By thus politicizing the issue, such accusations widened the scope of British public interest in what 'the privileged classes' in Oxford were up to. It should be remembered that, even discounting the class factor, only a tiny fraction (close to 6%) of British youth of age to attend university at the time in fact did. Colleges were the province of the elite par excellence. It didn't help, under those conditions, to accuse Oxford's theoretician elites of turning away from world problems, focusing their attention instead on the niceties of the English language -- however unfair (given the involvement of these professors in the war and in post-war efforts) this accusation was.

But, as Baz shows us, attacks also soon came from within the analytic tradition, ultimately forming the general impression that, whatever benefits it may have had, OLP had become something of the past. Taking the reader through a maze of criticisms and challenges, argument by argument and case by case, Baz dispels the notion that OLP has been discredited by any of the criticisms against it. Aside from the specific charges, there are according to Baz two fundamental faults common to most of these criticisms: attributing OLP with confusing meaning and use, and attributing it with the claim of possessing a systematic theory of meaning. Both attributions, Baz contends, are unfounded. Yet we can see them being implied or made explicitly in most of the major arguments from within the analytic tradition against OLP.

Concerning the first charge, Baz argues that it "presupposes one version or another of the very conception of (word) meaning that OLP, as I understand it, questions" (p. 13). Baz characterizes these 'versions' -- what he later calls 'the prevalent conception' -- as being based on what Wittgenstein described as a representational understanding of meaning. Major critics (Geach, Searle, Soames) presuppose such a conception without having first established it, and they proceed to call OLP to account for not portraying the meanings of words (and, eventually, the thoughts that are expressed by sentences which are made up of these words) in the representational sense mentioned, that is, as items that stand apart and are separable from the use to which these words and sentences are put. OLP is thus criticized for committing what Searle describes (p. 28) as the 'speech-act fallacy,' when this very fallacy presupposes the very sense of meanings that OLP questions.

Baz makes this very fundamental accusation, in one form or another, against all the different groups of critics of OLP mentioned throughout the book: traditionalists, intuitionists and even contextualists.  According to Baz, contextualists, while recognizing the organic link between meaning and use, still work under the assumption that, through working with the different meanings-in-use, it is still possible to identify or form ultimate meanings associated with the various words or expressions being investigated. Such meanings or contents are assumed to remain associated with these words or expressions even as these are admitted to have a primarily illocutionary function in some speech-context. A contextualist holding this view could be seen as having taken an important step towards adopting the OLP approach, Baz points out, but as not having gone the full length (of 'disowning' such independent contents in the first place).

The second general accusation Baz makes against OLP's critics is their assumption that OLP claims to have (but fails to defend) a systematic theory of meanings -- a charge which he says is not true since OLP does not make this claim in the first place. Citing Strawson ('Truth') and Austin ('Other Minds'), Baz argues that regardless of anything else that might be said on behalf of or against what these two leading figures of OLP claim in those papers about the subject-matters they address (truth and knowledge, respectively), their critics miss the point entirely when they assume -- and target in their criticism -- a final word in the form of a systematic theory or a full analysis of the meanings of knowledge or truth being presented by the authors. Quite the contrary, Baz points out: what both Strawson and Austin try to do in those papers is simply to challenge the philosophical tradition's perspective at its most fundamental level by bringing meanings down to their earthly, hum-drum size, so to speak, and, in so doing, to show it is precisely through our attending to these hum-drum uses, primarily, that we can give an account of the meanings we seek to explain.

Against the backdrop of these two general criticisms, Baz turns to particular cases (arguments as well as examples) that have been or can be viewed as constituting challenges to OLP, in each case showing how these challenges do not stand up to scrutiny. For example, he argues that Searle's critique (that a comprehensive analysis of the meaning of a word must account for its use not only in indicative but also in other types of sentences, such as interrogatives and conditionals, etc.) rests on two unwarranted assumptions: that OLP presumes to be providing such an analysis (which OLP does not claim), and that there is one thing that a word means or can mean (which is a thesis that is questioned by OLP, and is presupposed but not established by the criticism). Baz adds, at first seemingly not in synchrony with his main argument, that if OLP's claim were that only a particular set of things that it describes as being what may illocutionarily be done with simple declarative sentences that contain a philosophically troublesome word were all we needed to know about the meaning of that word, then indeed OLP could be charged with committing what Searle called the 'speech-act fallacy' (p. 55). Surely, here one might respond, Baz already told us that this supposed fallacy is itself fallacious. This leaves the reader with the conclusion that what Baz wishes to clarify here is that OLP does not pretend to be in possession of all the potential meanings of a specific word, its potential uses by speakers of the language still being undetermined. Indeed, he does go on to state that OLP simply seeks to dispel some of the philosophically troubling features of certain words, without pretending to exhaust the entire spectrum of uses to which these words may be put.

Another criticism Baz addresses is one made by Geach about the propositional (or descriptive) function of sentences, where straightforward valid inferences of the modus ponens form containing such sentences risk being seen as losing their inferential force if those sentences were to be understood in their illocutionary rather than their descriptive roles. Geach constructs an imagined inference in this context using Austin's (much maligned) example from 'Other Minds' where he draws a comparison between 'I know . . . ' and 'I promise . . . ', arguing that the former should also be read as an illocutionary act (in this case, of giving warrant) rather than as an indicative sentence expressing a truth-value bearing relationship between subject and predicate. (Baz, by the way, points out, both here and elsewhere in the book, that Austin was wrong to suggest this was the only illocutionary use of 'I know . . . ', and he points to other uses Austin does not mention in that article).^[5]

Baz's main critique of the Geach example (and of others he considers later) is that it is too contrived: that it imposes a preconceived and very narrow meaning on words (such as 'know') that in normal situations speakers of the language are not wont to have in mind or to be using. Indeed, he points out here (and makes much use of a similar argument in later chapters) that Geach's constructed inference is too unrealistic to be helpful. Furthermore, and given the traditional semanticist's requirement that a strict separation be made between a knowledge-claim by someone and what entitles that person to make that claim, Baz draws attention (p. 68) to the glaring mystery of how 'generations of readers of Geach's imagined argument' have missed seeing that it is not at all an example of a valid inference in the first place: one is not entitled to the conclusion that one knows a painting to be a clumsy forgery from the claim that one knows -- as a non-expert -- it is a forgery.

Turning to Soames' criticism of OLP -- that a necessary condition for speakers of a language to understand new sentences they come across is that they be familiar with the meanings of their component words (apart from and prior to their use in those sentences) -- Baz argues that this empirical claim, like all empirical theories, is 'underdetermined by [the] data' (p. 78). Like phlogiston, meaning-posits may provide an explanation for a phenomenon but prove nonetheless to be a fictitious entity. In any case, the Soames criticism shares with its predecessors the assumptions that such fixed and use-independent meanings exist and that OLP's approach to philosophically troublesome words must therefore be irrelevant to dissolving philosophical difficulties -- neither of which assumptions is established by Soames.

In Chapter 3 Baz takes on reliance on intuition as an alternative approach to doing philosophy, pointing out (p. 88) that objections to intuition as an investigative philosophical method have compelled philosophers to focus once again on what is to count as a viable method of inquiry. Baz calls the two main objections to intuition 'the cognitive diversity' (a line of criticism originally broached by Stitch) and the 'calibration' (associated with Cummins) objections. The former adopts an empirical -- more recently, an experimentalist -- approach, on the basis of which, basically, it is shown that our intuitions (about how to apply words to concepts) diverge quite significantly. The so-called 'calibration' objection, on the other hand, is more to do with an absence of an (independent) assurance criterion that our intuitions in fact track those matters (e.g., what knowledge is) they are supposed to be tracking. Baz thinks that neither of these objections quite goes to the heart of what is wrong with what he calls 'the prevailing research program' -- that is, tracking how ordinary speakers of the language might respond or react to targeted key words (e.g., 'know') when these are introduced by the theorists into hypothetical or imagined contexts/conversations. His own misgivings about these objections have to do with the contrived nature of these imagined contexts -- that real people in real situations on the whole would not respond to or understand the introduced words through the meanings forced on them by the theorist. As we set out to imagine a stretch of ordinary discourse we must not only bring to it the world of which we speak, but also the world in which we speak and think (p. 128).

But Baz spends the better part of this chapter critiquing Williamson's projection of the intuitive approach as simply being a natural extension of our ordinary capacity to judge the application of words to concepts from ordinary into professional speech (2007). On Williamson's interpretation, the cognitive objection dissolves into the natural explanation that some of us are more skilled than others, and different responses (e.g., about what 'knows' means in imagined examples) simply reflect that some of these responses are right and others are wrong. From Baz's point of view, this directly places Williamson in the traditional (representational) camp and makes the presumed skills his subjects have simply those of being able to assign words we come across in such new speech contexts to their pre-established concepts. But this attempted mapping of hypothetical speech contexts to ordinary speech contexts is misconceived in the first place for assuming that the hypothetical speech-contexts are sufficient unto themselves -- that they truly reflect how speakers of the language would linguistically behave or respond -- when the exchanges that are imagined to take place can be totally different in ordinary life, with the words used being dealt with (used and understood) in ways that have little if anything to do with how the theorist conceives them in an imagined 'pure state'.

Having argued that neither the (philosophers') intuitions approach nor its Williamson version seems to be a viable method for dissolving philosophical difficulties, Baz asks why not, then, in the face of a philosophical difficulty, revert to OLP, given that it relies on our shared knowledge of what competently employing these words in different situations would normally require, without any preconceived notions about what specific meanings different speech-contexts may reveal to us, and without, also, pretending to seek or possess an entire theory of language (pp. 126-7). In concluding the chapter Baz sets out some possible objections to his proposal: Isn't it, then, just another empiricist method? Doesn't it rely on intuition, after all? Hasn't it been shown that OLP confuses meaning and use? Isn't OLP only concerned with language as opposed to the things in themselves?  He then responds to these objections, saying with respect to empiricism that philosophers need not become experimentalists but can remain seated in their armchairs as they consider the different ways words are and can be used, and that

Far from focusing on mere words and leaving the world aside, OLP's procedures are precisely designed to bring a world lost through philosophical theorizing back into view -- but again, not just the world we speak and think of, but also the world we speak and think in. (p. 133)

Admitting the charge of intuition in OLP's procedures, Baz nonetheless points out, first, that on this view, as opposed to the traditional view, the intuition involved just serves as a preliminary investigation into how a key word is or might be used in ordinary contexts, without the claim that there will always be unique answers, or the expectation that such an investigation is guaranteed to produce all the required answers.

In the final two chapters Baz turns to addressing the debate between the traditionalist (invariantist) and contextualist schools of thought, finally bringing into focus that fundamental question underlying what might be OLP's major insight (and controversial claim) in this context, namely, whether its truth-condition is the inseparable partner to a sentence-in-use. Here, in spite of the tendency to see Austin as a forerunner of the contextualist school, Baz insists -- referring to the works of Keith DeRose and Charles Travis -- that, insofar as this school still insists on a semanticist/pragmatist distinction, and hence on a semanticist role for the truth-conditions of sentences beyond their illocutionary functions, it stops short of taking the extra step needed of, in certain contexts, doing away altogether with the notion that indicative sentences always have to answer to the question whether they are true or false (p. 143).

Baz returns to this point in Chapter Five, challenging what he considers to be a forced (theory-laden) 'reading' of how 'know that . . . ' and its cognates are put to use in three imagined speech contexts by representatives of the two schools (DeRose's Bank example, Stewart Cohen's Airport example and Jason Stanley's Bank example). In the final analysis, the underlying point of tension between the various schools -- one that is brought to the forefront by invoking the question of truth-conditions -- is precisely how to address skepticism. Rather than anchoring its response to skepticism in a representational model, OLP's approach

suggests that the felicitous adoption and expression of any cognitive attitude or stand, including that of wondering about the existence of the 'external world', presupposes, in the background, the existence of the world. It presupposes, in other words, a rationally assessable relation to the world that is not one of representing it in either speech or thought. (p. 158)

Baz concludes his book by invoking Kant, arguing that there are interesting points of similarity between him and OLP (as well as lingering dissimilarities, especially on the question of representation). Kant speaks of mental presentations or pictures we form of words and the world that spellbind us in ways that distract us from seeing clearly, in a somewhat parallel fashion to the way Wittgenstein speaks of the captivating effect the pictures we form have on us, preventing us from seeing our way around the world clearly. And just as Kant, referring to the 'pointless' (because unverifiable) disagreements between rationalists and empiricists, describes them as suffering from a 'transcendental illusion' that mocks them 'with a reality where none is to be found' (p. 202), Baz similarly refers to the barren contextualist/anti-contextualist debate with respect to knowledge as also being in a state of stalemate, 'bespeaking a crisis of philosophical paradigm' (p.198), a crisis that Baz shows his readers can only be overcome by returning to the investigative, bottom-up and piece-by-piece approach to determining meanings. But in successfully pushing aside from philosophy's center of attention what could well be seen as mere tinkering with marginalia, Baz also pushes the doors wide open for another fresh start in philosophical investigations.

Cited Works

Gustafsson, M. & Sorli, R., The Philosophy of J. L. Austin, Oxford University Press, 2011.

Russell, B. 'On Denoting', Mind, Vol. 14, No. 56 (Oct., 1905), pp. 479-493.

Strawson, P. F. 'On Referring', Mind, Vol. 59, No. 235 (Jul., 1950), pp. 320-344.

Williamson, T. The Philosophy of Philosophy. Oxford:Blackwell Publishing, 2007.