OCR’ed and uncorrected version
Man and World 17:1 75-195 (1984)
©Martinus Nijhoff Publishers. Printed in the Netherlands
AUSTIN'S FLYING ARROW: A MISSING METAPHYSICS OF LANGUAGE AND WORLD
DAVID WEINBERGER
Stockton State College
Introduction
If John Austin these days is considered to be a second-rate philosopher - a charming duck, but not to be taken too seriously - it is probably in large part due to the fact that he does what he does without telling much about how or why what he does works. He offers us a technique, demonstrates it in a small body of works, and only now and then wonders about the technique's philosophical underpinnings. When he does wonder, he does not wonder very long or all that well.
It seems that Austin has no positive philosophy to offer. Instead, he gives us a way of criticizing other philosophies. In this essay I want to use Austin's technique to criticize his own philosophy. First, we will examine his treatment of A.J. Ayer's philosophy of perception; we will see that Austin does more than he thinks he is doing. Then we shall look at some of the remarks Austin makes by way of explaining his own techniques. We shall apply to these passages the technique Austin uses against Ayer. We shall find that Austin's notion of the relation of world and language is inadequate. Finally, in proper Austinian spirit, I will not try to round out this negative criticism by presenting some positive doctrine of my own of the proper relation of world and language.
Given that I have just announced I intend to point out weaknesses in Austin's thought, I want here to state that I began this project because I find Austin's technique quite valuable. My conclusion will not be that the technique does not work, but that Austin has not adequately explained its success. I might also note that my fondness for the company of Austin's thought originally took me by surprise, for I am (what other would call) a Heideggerian. Heidegger, I think, can better account for the success of Austin's technique than Austin can, and this paper may be taken as preliminary to that argument.
176 I
s ae G o s i
Austin, of course, was reluctant to present a positive philosophy of his own straightforwardly. Perhaps we can infer one, but from his writin~ we get first-order applications of his method, and occasional second•
order hints Excuses") rough guidelines (admittedly incomplete, as in "A Plea for Excuses ) on how to use that method. Further, as is typical of much of 20th-century English philosophy, his method is one for understanding other philosophies. He aims at clearing up the errors and confusions of those professional philosophers who have preceded him, and at preventing future philosophers from making the old mistakes. Now, there is something positive about taking us off the wrong tracks as long as this involves (as it does for Austin) putting us on to the right tracks, but overall in his written work most of his effort is spent in the former pursuit and little on the latter.
Let us use as our example one of his most extended pieces of philoophical criticism: the witty and delightful Sense and Sensibilia,' j which Austin aims at dismantling the sense data theory, particularly
s expounded by A.J. Ayer in The Foundations of Empirical Knowldge (although he makes reference also to H.H. Price's Perception and .J. Warnock's Berkeley). Austin presumably feels that the sort of dificulties he finds in Ayer's work can be found in other presentations f the sense data theory.
Sense and Sensibilia proceeds mainly by showing that Ayer has misused important (although sometimes inobtrusive) words in important ways. Ayer has made the typical mistake of philosophers, namely, tretching the meaning of ordinary words until the words are virtually meaningless.2 Words only have meaning in context, so that it is meanmgless to ask "What does `Fire' mean?", for stripped of its context we cannot tell if it means there is a fire somewhere in the building, or if it s an instruction to a firing squad, or if it is the answer to the question: "Which island did you go to? Block Island or Fire Island?" Philosophers take perfectly good ordinary words, the meanings of which in context would give us no difficulty, and rip them out of their ordinary context. But words get their meanings from their ordinary contexts, so philosophers tend to use words meaninglessly.
If this is Ayer's problem, the possibility is open that Ayer might reformulate his theory in such a way that it might now be acceptable to Austin and yet be no different in substance from the original version. This seems to be an implication of part of "A Plea for Excuses" where Austin shows that an uneducated prisoner uses excuse words quite
277
precisely when describing an incident in which he scalded someone to death, whereas the judge and attorneys, in trying to be precise (or at least highfalutin), use excuse words quite sloppily. Here the judge and attorneys, if they paid attention to Austin, could reformulate their remarks in perfectly clear ordinary language with nothing of substance missing. Is that what Austin is saying about the sense data theory as well? In other words, is Austin merely turning back Ayer's paper with remarks such as "Poor word choice," "Awkward construction" and "Unclear" in the margins, with the expectation that Ayer will re-submit the same ideas in a clear idiom? If so, then Austin may be performing an important service all professors perform when they correct the spelling and grammar of their students' papers, but this service is not yet a fully philosophical one. In fact, it is a service necessary to get a paper into proper form so that it may for the first time be assessed philosophically. But that trivializes Austin's contribution. He becomes an English teacher -- perhaps the best English teacher this century has seen, but an English teacher nonetheless.
I do not think Austin's work is trivial. To defend him, I will try to show that Ayer could not reformulate his sense data theory in a sensible way. Although at times Austin plays English teacher, at least in sections of Sense and Sensibilia he does more: he is both English teacher and philosopher.
Austin begins with the beginning of Ayer's book, and I propose we do the same with Sense and Seftsibilia. Let us take Chapter II as our example. In it, Austin dismantles the opening paragraphs of Ayer's book, which contrast the ordinary man's view of things with the philosopher's Philosophers, says Ayer,
are not, for the most part, prepared to admit that such objects as pens or cigarettes are ever directly perceived. What, in their
opinion, we directly perceive is always an object of a different kind from these; one to which it is now customary to give the
name of `sense-datum'.3
Austin quotes more of this passage, but I wish to direct our attention :o his important criticism of the lines I have quoted. Austin homes in >n what might strike us as the least contentious word in them; he does lot go after Ayer's use of `philosophers', `object' or `perceive'. He
ather chooses the word `directly', which Austin calls "one of the less
onspicuous snakes in the linguistic grass."4
Part of the joy of reading Austin is reveling in his ability to draw
178
significance from an apparently insignificant term such as 'directly' But he is not just showing off. It is fully part of his philosophic method to discuss the inconspicuous, for it is the ordinary word which the philosopher is likely to get wrong. Words with precise, stipulated definitions, such as `deontological' and `nomological', are much less likely to mislead. When ordinary words, however, are given technical, stipulated meanings, it will be much easier to go wrong, for it is easily forgotten that these words are not being used in their ordinary mean
ings.
Austin raises six points for us to ponder about `directly': (1) We only use the term `directly' (in ordinary language) in relation to perception when we have some reason to distinguish it from indirect per ception. If someone were to ask us, for example, if we were viewing the parade through a periscope, we might reply. "No, I'm viewing it directly." (2) Perceiving `directly' seems most at home when there is a change in the direction of sight, as when we view something in a mirror. (3) Indirect perception only makes sense when sight is the sort of perception we are talking about, and this is perhaps because of (2). (4) It is uncertain how far the idea of indirect perception can be extended, but there is a line somewhere to be drawn. (5) If X can be said to be perceived indirectly, it has to be the sort of thing which we (or suitable others) could perceive directly under other circumstances. Thus, if we see the path of a charged particle in a cloud-chamber, we are not seeing the particle indirectly, but are instead seeing a sign of the particle (or a trace of it, as we might say in the current American idiom). (6) In ordinary language, we will usually prefer to use the more explicit description of which `indirectly' is a vague metaphor. Instead of saying, "I can see the ships indirectly," we would much more likely say, "I can see the blips of the ships on the rader screen."
Now, some of these points are of interest without having much to do with Ayer's use of the term. Austin has, in other words, done what he thinks Ayer should have done before publishing his book: he has explained in some detail the ordinary language rules for the employment of the term `directly'. What does Austin make of this? He concludes that "it is quite plain that the philosophers' use of `directly perceive', whatever it may be, is not the ordinary, or any familiar, use; for in that use it is not only false but simply absurd to say that such
object as pens or cigarettes are never perceived directly."6 The philosophers break several of the rules governing the use of the term: they
say nothing about the special circumstances required before the word be used (looking at a parade through a periscope, etc.), they do not
.
179
note that it applies only to vision, and "moreover it seems that what we are to be said to perceive indirectly is never - is not the kind of thing which ever could be - perceived directly."
Ayer now seems to have two alternatives. He can find a more exact ordinary language term to replace `directly', or he can stipulate a meaning. But the stipulated definition ultimately will be in ordinary language terms (or else it will be mere gibberish), so in fact there is only one alternative: Ayer has to try to explain himself in ordinary language. And why can't he? Let's give it a shot. Ayer's original statement was:
P: ". ., philosophers who have recently concerned themselves with the subject of perception ... are not, for the most part,
prepared to admit that such objects as pens or cigarettes are ever directly perceived."
Will the following do as a translation?
P1: Philosophers who have recently concerned themselves with the subject of perception are not, for the most part, pre
pared to admit that such objects as pens or cigarettes are ever seen as what they are without the intervention of cer
tain mental steps or activities and/or phenomena which are not themselves pens or cigarettes.
If P1 is supposed to clarify P by putting it into ordinary language correctly used, P1 fails. In an attempt to clarify `directly' we find ourselves faced with the task of clarifying `intervention', `mental steps', `mental activities', and `phenomena'.
There is little point in looking for a P2 or P3 that might do the job, for we will in each case end up expanding `directly' into a more cumbersome phrase. Why isn't it possible to find the perfect English word to replace `directly', one that does what Ayer wants `directly' to do? We must ask what Ayer wants `directly' to do.
Austin is right when he says that we only use the term `directly' when we are contrasting it with some indirect way. In this case, of course, Ayer is implicitly contrasting our indirect perception of objects with our direct sensation of their sensory qualities. With this in mind, let us try to "cash out" Ayer's metaphorical use of `directly'. What he means, perhaps can be understood by seeing what he wants us to infer from his talk about not seeing pens and cigarettes directly: (1) From the ordinary use of `directly', we infer that what we see directly we see
180
more reliably than what we see indirectly, for there is less chance of an activity going wrong the simpler and more direct it is. (2) From the rule of use which says that we do not add a modifier unless there is some
reason to (more exactly, Austin calls this "No modification without aberration" in "A Plea for Excuses"),e we infer that besides indirect perception there is a sort of perception which is not indirect. (3) Ayer seems to want us to infer that when we indirectly see something, we do not really see it; rather, we see some phenomenon which stands in some sort of close relation (if perception is veridical in its normal sense) to the object of perception.
What would Austin say in response to this? (1) Insofar as Ayer proposes sense data as the objects we see directly, Austin is at pains in Chapter X to show that our knowledge of sense data is not incorrigible. (2) Austin has pointed out that Ayer believes there is no possibility of directly perceiving the pen and cigarette, and this seems to contradict Austin's fifth point, namely, that we can only say we see X indirectly if it is possible to see X directly. Might Ayer squeak by, by claiming that the contrast is between how we normally (and incorrectly) think about how we see a pen and how in fact we do see it? That is, we think we see the pen directly but in fact we do not; we see sense data directly and the pen indirectly. This would, however, make every case of perception one of indirect perception, which is not what we mean when we say we see something indirectly; some of what Ayer calls indirect perception - i.e., the normal case of viewing the parade without the periscope, as well as the rare case where we use the periscope - the rest of us in ordinary language call plain old (direct) perception. (3) Austin accepts that part of the conditions for saying that we see something indirectly is that there be "concurrent existence and concomitant variation as between what is perceived in the straightforward way ... and the candidate for what we might be prepared to describe as being perceived indirectly,"9 and sense data seem to fit this bill; as the pen moves, the sense data vary. But, if there are sense data, and if these sense data vary concurrently and concomitantly with the object of perception, this still does not mean that we never see the pen. Why not? All we need do is examine the meaning of the word `see'. If the sense data account is right, `to see' means `to have sense data closely and regularly related to the object of perception'. On such an account, we still see the pen. Do we see it directly? That depends on whether or not you are using a periscope.
But we seem to have left room for the sense data account, which Austin does not want to do. He wants to show that `directly' has been
l 1
h i
i t L
h
1 i a
i
3 i
181
`gradually stretched, without caution or definition or limit," until it ias become "first perhaps obscurely metaphysical, but ultimately neaningless."10 In what sense meaningless?
In fact, I believe Austin's account does not show Ayer's statement :o be meaningless. Rather, his analysis shows that it: (1) uses ordinary :erms in extraordinary ways, which (2) when cashed out yield a theory Zo different from what everyone believes. And this means that Ayer's iistinction between what ordinary people believe and what "philo;ophers who have recently concerned themselves with the subject of perception" say does not hold. As Moore says in "A Defence of Common Sense," it is not that some philosophers do not believe the tenets of common sense; it is that some philosophers hold in addition to those tenets certain philosophical ideas that contradict common sense.
After all, what does the sense data theory work out to, under Austin's analysis? It says that there is a difference between our perception of the pen and the pen itself. This is not meaningless. It is just so uncontroversial as to be philosophically trivial. What is truly distinctive about the sense data theory are additions to common sense which are neither meaningless nor trivial but false. (1) There is always a difference. In one sense, yes. We each see a different perspectival view of the pen, and no one view is the pen itself. This is trivial. In another sense, no. Sometimes we misperceive, but usually we perceive what is there." (2) Our sincere reports about our perceptions are incorrigible. This
Austin shows to be false,12 for we may be wrong when we say, "I seem to see a heliotrope patch"; it may not be heliotrope but mauve or
puce.
Austin thinks he is showing that what Ayer has written is meaningless. Ayer has taken a perfectly good ordinary word -`directly' - and has stretched its meaning so far that it has virutally no meaning left. Yet Austin should know that Ayer will take this as a criticism not of his philosophy but of his mode of expressing his philosophy. Ayer's next step (or ours, on his behalf) will be to reformulate his position. Upon doing so, however, that position turns out to contain only two sorts of statements: trivial ones, the hollowness of which was masked by the fancy philosophical usage of ordinary language, and false ones. Thus, Austin shows first that Ayer's actual way of expressing himself is at fault (and is at times meaningless) and that Ayer's position is cornPrised of the trivial and the false.
Showing that something is comprised of the trivial and the false is quite a bit different from showing that its mode of expression is meaningless. Austin seems to think his job is to expose the meaninglessness
of other philosophies, and does not pay much explicit attention to his hunting of the trivial and the false. But this is a most peculiar state of affairs, for by any normal standards, the second pursuit is more profound and important than the first. The first i,the exposing of the meaninglessness of the mode of expression of a philosophical position) is but a first step, for once the inadequacies of the mode of expression are exposed, the position may be reformulated. Once it has passed the
test of meaning, we will see whether it is true or false, and if truee whether it is trivial or significant_ Austin in his works engages in both pursuits, but seems to think he is only engaged in the first.
Why the emphasis on the first task. the exposure of the meaningless? For one thing. it is something Austin did better than anyone else, and it uses a technique which, if he did not invent, he refined to an unequalled purity. For another, the meaningless stretching of ordinary words does not only mask the triviality and falsity of positions; it also helps explain the falsity. It is because Ayer is imprecise in his use of thee words 'illusion' and 'delusion' that he makes false statements about the significance of the bent stick illusion.13
Austin's reluctance to talk about the truth or falsity of the position he is criticizing has important repercussions in his work. Because he thinks he is dealing primarily with meaning, and not with truth values, he deals inadequately with the world, and thus with that which enables his own philosophical program to go forward. To help see this, I wish briefly to contrast his program with one that is closely related to it: that of Martin Heidegger. Then, in Section II, we will pursue this matter more directly.
Heidegger dispenses with sense data theories rather quickly and brusquely in The Basic Problems uf Phejtvmeuolog.i. written in 1927. shortly after Being and Tijne. He mentions no philosophers by name in the two relevant paragraphs. He says not aword about meaninglessness. Here is his critique:
Let us take a natural perception without any theory. without a preconceived opinion about the relationship of subject to object
and other such matters, and let us interrogate this concrete perception in which we live, say, the perception of the window. ...
In everyday behavior, say, in moving around in this room, taking a look around my environment. I perceive the wall and the win
dow. To what am I directed in this perception'? To sensations Or, when I avoid what is perceived, am I turning aside from
representational images and taking care not to fall out of these representational images and sensations into the courtyard of the
university building`.'
183
To say that I am in the first place oriented toward sensations is all just pure theory. In conformity with its sense of direction, percep
tion is directed toward a being that is extant. It intends this precisely as extant and knows nothing at all about sensations that it
is apprehending.14
~or Heidegger, it is not the thoughtless use of ordinary words out of heir ordinary contexts that lets the false position slip by as one that is easonable, if not obviously true. Rather, it is theory that blinds us. ~or Heidegger, to understand the origin of the error (that is, to exlain the error) will require understanding the origin of the theory, nd this task is nothing other than uncovering the history of philosohy (and, in Heidegger's words, the history of Being, Seinsgeschichte). bus, Heidegger begins his lifelong exploration of the history of Westrn philosophy.
But one does not need to know the origin of the error in order to °e that it is an error. To see that it is an error, Heidegger would have s ask ourselves, "Does this theory accurately describe my experience?". Husserl, Heidegger's teacher, would have us adopt a special sort of ~flection which is free of theory; Heidegger thinks ordinary expeence is accessible without Husserl's "reductions.") If the sense data ieory says that when I kiss my wife, I am actually kissing the sense ata of my wife, I must ask if that theory accurately describes my exerience this morning of kissing my wife. If it says that we receive uniterpreted sense data which we then construe, has it accurately de:ribed how I perceived the window when I went to open it? Heidegger xpects that we will all answer no. (If a voice in the back of the class nswers "Yes" in all seriousness, Heidegger will not be able to conince the person otherwise.) Heidegger, in short, has us compare direct
the theory and the way in which we live in the world. He has us ask, The theory sounds good, but is it true? That is, is it true to my exerience of the world?". Notice that Heidegger does not have us cornare the theory with the theoretical explanations we normally give ithout special tutoring. Heidegger does not have us compare one ieory with another, one explanation with another. Rather, he has us )mpare a theory with our experience of the world. Because of this, Heidegger has to spend a long time telling us about le world and our relation to it. (Considering the results of Heidegger's 'quhY, this is a misleading way of putting it, for Heidegger finds that ~ be a human is to be in the world. Humans and the world are not two ~stinct sorts of entities which are accidentally related.) Austin, because
184
he mistakenly believes he is simply showing that other philosophies are meaningless, does not feel he must talk about the world very much. He thinks he is showing how philosophical explanations extend to the point of meaninglessness terms we use in our ordinary explanaa tions; because he pays scant attention to the role of falsity in his own critical philosophy, Austin pays scant attention to the role of the world. This in turn means that he fails to understand what makes his own philosophical method possible.
We now turn to Austin's attempts to explain why his method works.
II
If we now try to find Austin's own positive philosophy, we find very little. By combing through his published works, we can find passing references, even whole paragraphs, devoted to explaining, for example, why his method works, but we find no systematic treatment. Instead, Austin provides three sorts of explanatory comments: (1) General reasons why philosophers (and other hightfalutin types) go wrong; (2) a sort of theory of the origins of language; (3) metaphorical statements of our general relation to the world and to language.
(1) The general reason philosophers tend to go wrong is, of course, that philosophers too often take an ordinary word and use it out of context. Austin mentions two other general sorts of mistakes. First, he says we sometimes use simple models for complicated cases,ls or, we think that what holds of ordinary words must hold of extraordinary words as well.lb Second, philosophers have tended to look for some special form of words to meet certain requirements - the example he discusses is that of philosophers looking for a special form of words in which we make no assumptions - instead of realizing that words get their meanings from the circumstances in which they are uttered, not from their form."
If the overall basis of Austin's philosophical criticism is that words receive their meaning from ordinary usage, we may wonder how Austin ~ can at times say that ordinary usage has got things wrong. The answer is not that ordinary usage conflicts with states of affairs, but rather - that in some cases philosophical prejudices have managed to infect ordinary usage.18 The term `universal', for example, started out as a~
7
philosopher's term but crept into ordinary parlance, and has been mis• ` leading philosophers and ordinary people ever since.19 (2) Austin's explicit remarks about language at best constitute a
185
sketch. Language is a too1.20 There is a stock of symbols (which may be words, but could be semaphore signals, dots and dashes, etc.) available to the linguistic community,21 and which stock embodies all the necessary distinctions.22 Why say that the stock of words embodies all necessary distinctions? Austin holds to a form of Darwinian theory of the origin of language. Words, as tools, were invented to make the necessary distinctions; those distinctions that remained necessary tended to keep the appropriate words in circulation. Words that do not embody necessary distinctions tend to drop out of use and disappear.23 That is why Austin consistently looks to the etymology of words for hints about their meaning.24
Here is one point where one wishes Austin had pushed a bit harder. His Darwinian theory indicates that ordinary usage is not a matter of whim (". .. the distinctions embodied in our vast and, for the most part, relatively ancient stock of ordinary words are neither few nor always very obvious, and almost never just arbitrary...."25 ). The necessary distinctions are maintained, but they are necessary for what? He says our words "have stood up to the long test of survival
of the fittest ...,"26 but surely this is a case of using a simple model
to cover a complicated case. For Darwin, the traits that survive are those which enable an organism to leave behind more of its offspring. Yet relatively few of our words have much to do with enabling us to survive long enough to have lots of children. What, in the case of language, could count as an arbitrary distinction?
Austin's reply probably would be that that is not important for his project. He is offering us a technique. He has found, empirically, that checking etymologies often gives hints. Why that is so is a matter of speculation.
(3) In "A Plea for Excuses," Austin gives us a bit of "metaphysical speculation," in metaphor, about the relation of world and language. 1-et us look at that and see if we can treat it the same way Austin treats Ayer's writings on perception.
Austin in this essay briefly defends ordinary language philosophy, which examines "what we should say when."27 "Perhaps this method, at least as one philosophical method, scarcely requires justification at present - too evidently, there is gold in them thar hills ...."2a (But how can Austin tell gold from pyrite?) Reluctantly, however, he wili justify his method.
First, he says, "words are our tools, and, as a minimum, we should
use clean tools ...."29 Second, he writes:
186
. . words are not (except in their own little corner) facts or things: we need therefore to prise them off the world, to hold them apart from and against it, so that we can realize their in. adequacies and arbitrariness, and can look at the world without blinkers .°
The third justification is that "our common stock of words embodies
all the distinctions men have found worth drawing," and presents his theory of the "survival of the fittest." It is on the second justification
that I wish to focus.
Austin closes this small section by attempting to counter a misunder
standing:
When we examine what we should say when, what words we would use in what situations, we are looking again not merely at words (or `meanings', whatever they may be) but also at the realities we use the words to talk about: we are using a sharpened awareness of words to sharpen our perception of, though not as the final arbiter of, the phenomena.31
Here Austin explicitly tells us he is looking beyond meanings to truth: This is, for our purposes, an important text, especially taken in conjunction with the second justification. Unfortunately, we are left wondering exactly what Austin means by three words which seem to be related, if not synonymous: `world', `realities', and `phenomena'.
Of the three terms, `phenomena' and `realities' most clearly are philosophical and not ordinary. But `world' does have ordinary usages, and if we hope to find the meaning of one of these three terms, `world' is most likely to turn out not to be meaningless.
We meet the term `world' in this passage as something distinguishable from words, although something which ordinarily is not distinguished from words; hence the need to prise words off the world. Further, the plain implication is that facts and things do not need to be prised off the world. I suppose that Austin here is considering the world to be composed of facts and things (as well as anything else?). What, then, is the relation of words and world?
The passage gives us three hints: (1) Words do not comprise the world, unless they are taken to be (in their own little corner) facts or things. (2) There is a certain adhesion of word to world which Austin's method can loosen. (3) If we can loosen the adhesion, we will be able to see the world without blinkers.
It is not like Austin to mix metaphors, yet he does, and we can ask
!
187
why. Why does he switch from the prising metaphor to the blinkers metaphor? Consider what the expected continuation of the prising metaphor would be: We prise words off the world, hold the words apart from and against the world, compare word and world, and now can see the world for the first time. (This assumes that words have entirely and opaquely covered the world. But Austin may mean that words do not black out the world, but only obscure it, in which case we can see, after prising off the words, the world for the first time clearly or - Sacre Bleu! - directly.) The metaphor of prising leads one to think one will see what is underneath for the first time, or for the first time clearly. The metaphor of blinkers, however, is quite different. The horse's blinkers keep it from seeing very much of the world, but what it does see it sees clearly and directly. Removing the blinkers enables the horse to see more, but not better.
The prising metaphor is attractive to Austin because it helps make clear the difficulty of his task and why it is that no one before him noticed the task's necessity. Words stick to the world so firmly that one might not even notice that there is anything underneath them. To be philosophical, however, one can begin by prising off the words and then comparing the two realms, rather as one might compare a map with the landscape. We should note, however, that this takes language to be some sort of copy of the world. Yet the point of talking about words as tools surely is to emphasize that words are not copies.
That is why he has to switch metaphors. When we look out at the world, we see the world, not a copy of it. Thus, he cannot keep with the copy metaphor, and instead uses the blinkers metaphor. Words do not copy the world. Rather, they direct (and that means limit) our awareness of the world the way blinkers do.
It is extraordinarily difficult to see how either of these two metaphors helps us to understand how and why Austin's method works. If we take an example of a philosophical word that has given us difficulty (`real,', for example), after Austin has cleared up the problem for us, it does not seem that he has prised anything off the world, and compared it with the world. He says that `real' is used only to distinguish something from something else which is not real in any of the wide variety of ordinary senses: real vs. imitation, vs. artificial, vs. toy, etc. He has thus dissolved the idea that "The Real" is some entity or class of entities. This shows the fly the way out of the bottle, but it does not prise anything off the world. Or, if we take an example of a word that has not gone wrong - his famous discussion of the differences between shooting a donkey by accident or by mistake32 -
188
where is the prising, and what is the world apart from the words which is revealed?
The second metaphor also does not immediately help us understand what Austin accomplishes. By taking off blinkers, we are given a fuller view of the world. But what Austin does is give us a fuller view of how we use words. To take the same examples: (1) His discussion of `real' eliminates a philosophical entity and awakens us to the variety of ways we use the word `real'; (2) his discussion of shooting the donkeys awakens us to the sophistication of ordinary language, a language we necessarily ordinarily take for granted.
There is, however, one important way in which the prising metaphor works_ It applies (not especially aptly) to how Austin actually treats Ayer, as opposed to how he claims to treat Ayer. He shows that Ayer's way of expressing himself is meaningless, but also that once his manner of speaking is straightened out, what Ayer has to say is either trivially true or just plain false. To try to put it into Austin's metaphor, if one prises Ayer's words off (off of what? the world? no, it would be more accurate to say that they are to be prised off of what Ayer is trying
, to say), one can see that those words either state trivial truths or describe the world inaccurately.
The mixing of the metaphors is due, in part, to Austin's failure to keep these two fundamentally different projects distinct in his own mind. In his treatment of Ayer, Austin first removes the blinkers: he shows that Ayer has been using words which belong to a broad, everyday context, in a narrow, restricted and `blinkered' way. Having removed the blinkers, we now can compare what Ayer really has to say with the way the world is, and we see that Ayer has in fact inaccurately described the world.
The difficulty is, of course, that `world' as Austin uses it is not within the domain of ordinary discourse. He is using it as a philosophical term, and what he seems to be saying (or unintentionally implying) about it is most un-Austinian. The world, for Austin, is that from which words are to be distinguished. In the essay "Truth," Austin says that for there to be "communication of the sort that we achieve by language" there must be words and "There must also be something other than the words, which the words are used to communicate about: this may be called the `world'."33 (He notes here that the world may also include words, which explains what he means when he says that words are not facts or things "except in their own little corner"). But this distinction is one only a philosopher would make. Austin's brilliant analysis of `real' works because he is able to show us the many
189
I
different phenomena we mean to distinguish by use of the word `real'. But in ordinary language we rarely (never?) use the word `world' in distinction from words. And this is for two reasons: first, the word `world functions in a different way than Austin implies; second, the distinction Austin is trying to get at is rarely made at all.
A brief look at the ordinary use of `world' makes clear the first point. The word `world' is used in ordinary language to indicate millieu or sub-world ("She comes from a different world; they'll never get along", "The world of a numbers runner"), or as an adjective to indicate international scope. We sometimes talk about "Making one's way in the world", and undergraduates refer to "the real world," referring (loosely) to either the sub-world of business and economics, or to any dog-eat-dog world as opposed to the sheltered world of the campus. The distinction Austin offers us between the verbal and the worldly just is not to be found in ordinary language; it is a philosopher's distinction. (And who would say, except a philosopher, that the world is what words are used to communicate about? I use words to communicate about - the awkward locution is Austin's, and indicates his extraordinary use of the word `world' here - this feeling, that situation, this circumstance, that fact. Very very rarely do I use words to communicate about the world. Is the world the set of all which words communicate about? Even if this were a plausible interpretation, clearly by now we are over our heads in philosophical stretching of the ordinary word `world'.)
Austin seems to be setting a distinction between the signifier and the signified, but it is striking how infrequently we actually need to make that distinction, or at least need to talk about it. The distinction between signifier and signified - or, more prosaically, word and thing - is so obvious that we rarely need to distinguish them explicitly. It is an extremely nervous person who does distinguish them explicitly: "Please pass me the salt, and not the word `salt', but the thing which is salt." Such a person clearly needs a vacation from philosophy.
So, we are now in a position to help Austin attempt to escape from our charge that he is using `world' in a philosophical - and thus Possibly meaningless - way, just as Ayer uses `directly' in a philosophical way. Ayer's first recourse will be, as we saw, to claim that Austin ' just quibbling over words, is playing the pedantic English teacher, etc., and that Ayer can make himself perfectly well understood by substituting ordinary words for the offending ones. In the same way, Austin may want to say that he was lax in using the word `world', but he can explain the distinction easily in other, inoffensive terms.
190
In response to this, we are claiming that the distinction is, in any words really a philosopher's distinction, not one we make in ordinary dis~ course.
Yes, but what does that mean? True, we do not ordinarily make the distinction, but surely it is there to be made whenever we care to be philosophical. To this Austin has given us the material of a reply:
But
When it is stated that X did A, there is a temptation to suppose that given some, indeed perhaps any, expression modifying the verb we shall be entitled to insert either it or its opposite or negation in our statement . . ..3a
The natural economy of language dictates that for the standard case covered by any normal verb ... no modifying expression is required or even permissible.3s
It is on those grounds that Austin can claim that if I enter a room and sit in a chair, I did not sit in that chair intentionally, for the modifier `intentionally' is only permitted in special circumstances (such as my seeing the place-card there for someone else but deciding such social conventions should be flouted). Now, the distinction between intentional and unintentional action (if those two are truly contradictories and can be applied equally well to the same verbs, as `voluntarily' and `involuntarily' cannot36 ) is always there to be drawn, but Austin's point is that we only draw the distinction in certain circumstances. In just the same way, the distinction between word and world may always be there to be drawn, but we only draw the distinction in very peculiar circumstances. Austin, as a philosopher, has drawn the distinction without the justifying circumstances. This is just the sort of situation Austin thinks leads to philosophical nonsense.
Since we do not ordinarily draw the distinction, and when we do the circumstances are most pecuilar, and when we do we do not do so in terms of word vs. world, Austin's use of the word `world' so far is unintelligible in the same way Ayer's use of the word `directly' is. Can he put it into other words to make it intelligible?
I think not, and for reasons Austin should have appreciated. The world, we have seen, for Austin seems to consist of facts and things. In his essay "Facts," Austin defends his use of the term `fact' against Strawson's criticisms. While his defense and analysis of the word 'fact is complex, at least part of his point is that a fact is "something in the
191
world";37 he argues against the position that "Facts are what statements state, not what they are about'."38 If a cat has mange, "The condition of the cat is a fact, and is something in the world."39 Presumably, anything that is a fact is something in the world. This makes the world rather extensive, with which I have no quarrel. What is left out, however? From what can the world be distinguished?
Clearly, the world can be distinguished from any sub-world, and since words "in their own little corner" can be included as `things' in the world, we can distinguish word and world as subset from set. But that is not the distinction Austin is getting at when he talks about prising words from the world and removing the verbal blinkers.
I wish to maintain, crudely, that Austin is a realist of some sort. It is true that Austin himself disavows realism. In Sense and Sensibilia he writes that realism, "the doctrine that we ... perceive material things (or objects)" as opposed to perceived sense data40 is "no less scholastic and erroneous than its antithesis,"a' for the antithesis of sense data vs. material things is spurious. "There is no one kind of thing that we `perceive' but many different kinds, the number being reducible if at all by scientific investigation and not by philosophy."a2 Still, I think it fair to take Austin as holding that: there is a world; we are in some way aware of the world; the world is independent of our awareness of it. But what is the relationship of our awareness to the world, and what might that relationship be if we pursue philosophy? (Obviously Austin thinks his sort of philosophy will make some sort of change, for he tells us it will enable us to take our blinkers off.)
What is distinctive of Austin is, of course, that he does not talk in such metaphysical terms. He does not talk of awareness or consciousness, but rather of language, or (to be precise) "what we should say when". If you want to know what you think, see what you can say. Or perhaps, at least begin by seeing what you can say: "Certainly, then, ordinary language is not the last word .... Only remember it is the first word."a3
What is the relationship? Let us look at another curious metaphor. Austin is talking about "adjustor words", of which `like' is the great example. `Like' is "the main flexibility device by whose aid, in spite of the limited scope of our vocabulary, we can always avoid being left completely speechless."aa And here is the metaphor:
If we think of words as being shot like arrows at the world, the function of these adjustor-words is to free us from the disability
of being able to shoot only straight ahead; by their use on occa-
192
sion, such words as `pig' can be, so to speak, brought into con, nexion with targets lying off the simple straightforward line on
which they are ordinarily aimed.4s
Now, let us try to think of words as being shot like arrows at the world. What does this metaphor mean? On the face of it, it is perfectly clear, but then so was Ayer's idea that we do not see the broken stick
directly. The more one contemplates the metaphor, the less helpful it becomes. What Austin means to say (judging from the context) is that situations sometimes arise for which we have no exact word, so the adjustor words enable us to talk about the new in terms of the old. What does the arrow metaphor add? Consider it in contrast to a metaphor Austin did not choose: why are words like arrows shot at the world rather than like bullets shot at the world? Of course here we are speculating freely, yet it seems probable to me that Austin chose the arrow metaphor as opposed to a bullet metaphor because: arrows travel much more slowly than bullets, and so we have a clearer image of the arrow in mid flight; arrows protrude from their targets, whereas bullets vanish in or behind their targets; arrows can be either weapons or directional pointers, whereas bullets are only the former. Thus, the metaphor brings to mind three possible relationships between words and world: words are not attached to things but are on their way towards that attachment; words are attached to things and we recognize the things by the words attached to them (= the arrows sticking out from them); words are not attached, and never will be, but point to things.
Did Austin have any of these in mind when he composed the metaphor? This of course cannot be known for sure. Yet it is legitimate to try to flesh out a metaphor for an author, especially if the metaphor ; seems not to work. My contention - for which I can offer no proofis that the arrow metaphor presented itself to Austin because it is ambiguous in its depiction of the relationship of words and world. He would never say, "If we think of words as being copies of things" of "If we think of words as being labels attached to things," for he is certain that those metaphors are inadequate. The arrow metaphor seemed better to him, I contend, because its inadequacy was hidden behind its ambiguity. Austin prefers the ambiguous because he is so uncertain about the nature of the relation of word and world.
I believe it is Austin's realism that kept him from understanding his own method. Because he held on to the belief that language somehow is between human and world (the fyling arrow), and is in sotne
1
193
; sense a sort of label, he thought he could deal with the labels other philosophers had supplied, and discard those labels as meaningless (they either label nothing or do not label what the philosopher thinks
' they label). Yet he also had a keen awareness (as well as, probably, a rather dull, thudding throbbing in the back of his mind) that language is not a system of labels, for he is the discoverer of performatives. Performatives are words that do not mediate between humans and
; world, and are neither labels nor copies of the world. Rather, they are actions in the world, and are fully part of the world.
r This also helps to explain Austin's seeming lack of ambition. In Sense and Sensibilia, when talking about `seeing as', he writes that
', "there will sometimes be no one right way of saying what is seen, for
i the ... additional reason that there may be no one right way of seeing it."46 The distinctions embodied in language are drawn for utilitarian
;(Darwinian) reasons, and thus are neither right nor wrong. They may, however, be meaningless. Thus, while as a realist Austin thinks we have access to the real world, that world yields many different linguistic
I schemes which are to be judged in terms of utility. Thus, it is not up to philosophers to declare statements true or false. Rather, philosophers
; are to declare statements meaningful or meaningless. Since meaning is granted by ordinary usage, only extraordinary statements need assessing. Therefore, it is the philosopher's job to assess the meaningfulness of extraordinary statements, particularly those of other philosophers. Therefore, although Austin believes we have access to the world. the philosopher's job is not to discuss the world, but rather other philosophies. Why bother doing this? Either because it is entertaining, or because the other philosophies are confusing people. Because philosophies tend to mislead only philosophers, the lack of urgency in Austin's writings is understandable.
But Austin, as we have seen, did not merely search out the meaningless in other philosophies. He also searched out the false. Asking "What we should say when" is a good way of finding out what terms mean. But when trying to evaluate some other philosophy, we can also ask, "Does this other philosophy - now that I've understood its terms - accurately describe what it is talking about?" If the other philosophy ~ Ayer's talk of perception, we can ask, "Has Ayer accurately described what happens when I go to open a window, drive my car, or see the stick in the water?" To answer this question we examine the experience. That is, we turn from Austin's "linguistic phenomenology" to phenomenology proper.47
Now we can ask if Austin's picture of language accurately describes
194 ;
,
our experience of language. Has he accurately described what it is to talk, what it is to say what when? I think it is clear that he has not, ` That is, his realist picture is false (except in very unusual circumstances, ' and Austin has warned us that what holds of the extraordinary may not hold of the ordinary.) We do not ordinarily face a world in which distinctions may be drawn in various sorts of ways: if we can manage to see the sky as the foreground and the row of trees as the background, we will have accomplished something most extraordinary, , The duck-rabbits of the world are few and far between. Indeed, we do not apply words to situations at all, except in unusual circumstances, ` Much more ordinarily, when language does its job, the words present ' themselves within situations. It is only because of this that we can rely on Austin's method of seeing what we would say when. If words = are tools, they are not under our control (at least not ordinarily). 'I'he `' world thrusts itself upon us in words; we do not shape or even tinker at the world with words. '
One might add that in any case words are not ordinarily our tools, even when we do things with them. After all, we ordinarily do things I with the earth - we walk upon it, farm it, etc. - but to call the earth £ a`tool' is to so stretch the meaning of the word `tool" as to render it virtually meaningless. The same is true of calling language a tool. ' Rarely, perhaps, it is, as when I test a microphone by saing, "Testing, ' one, two, three." But the disanalogies between language and tools far = outweigh the analogies. A tool is a thing in the world used for altering ` other things in the world; words are not things "except in their own 3 little corner", as Austin knows. Tools are used by humans and mediate ; between humans and the rest of the world; language, I wish to suggest; ; is not something mediate between us and our world. Tools are shaped ' according to the projects humans wish to accomplish; language, I wish to suggest, shapes the projects humans desire. Tools depend on us in a. way that language does not. But to support these assertions of mine would require my breaking my promise not to end by offering a positive philosophy.
My point is simply that Austin's meta-philosophy (which is really to 'n say, his philosophy) founders because it itself offers no consistent, well- ~ thought understanding of the relation of word and world. Such an understanding would, of course, also have to illuminate the relation of humans and world. But if I am right about the difficulties and confu- ; sions Austin encounters, my prescription works out really to saying ; that what Austin's work lacks is a good dose of old fashioned meta• physics. '
195
NOTES
1, John Austin, Sense and Sensibilia (Oxford University Press: London) 1962. Henceforward, referred to as SS.
2• SS, 15.
3• SS, 6--7. 4. SS, 15.
5. SS, 15-19. 6. SS, 19. 7. SS, 19.
8. In Philosophical Papers (Oxford University Press, Oxford), 1979, p. 189.
Henceforward, referred to as PP.
9. SS, 17. 10. SS, 15.
11. SS, see Chapt. V. 12. SS, 42-3.
13. SS, Chapters III-VI.
14. Martin Heidegger, The Basic Problems of Phenomenology, translated by Albert
Hofstadter (Indiana University Press: Bloomington) 1982, p.63.
15. PP, 15.
16. PP, 127-8. 17. SS, 137.
18• PP, 39, 182. 19. PP, 39. 20. PP, 181. 21. PP, 121. 22. PP, 182.
23, PP, 274, 281.
24. SS, 63; PP, 17, 72, 163, 164, 201, 202, 211, 260, 274, 283;How to Do Things
with Words. 25. SS, 63. 26. PP, 182. 27. PP, 181. 28. PP, 181. 29. PP, 181. 30. PP, 182. 31. PP, 182. 32. PP, 185• 33. PP, 121. 34. PP, 189. 35. PP, 190.
36• PP, 190-1. 37• PP, 158.
38• PP, 158. 43. PP, 185.
39. Pp 158. 44. SS, 74. 40, SS, 3• 45. SS, 74. 41. SS, 4. 46. SS, 101. 42. SS, 4. 47. SS, 182.