J.L. Austin’s How to Do Things with Words

2,296 words

John Langham Austin was an English philosopher of language, an Oxford Don, and (like Alan Turing) a Military Intelligence Officer, who worked on the D-Day landings. Born in 1911, Austin died relatively young, at the age of 48, and his three books were all published posthumously. Philosophical Papers was a collection covering a variety of topics, including epistemology and metaphysics, while Sense and Sensibilia (the title is a play on Austin’s own name and that of Jane Austen) dealt with the validity of perceptual problems. Perhaps his most famous work, How to Do Things with Words (HDW), was based on a series of lectures given at Harvard in 1955, and is a key text in English philosophy of language. It is this last book on which I will concentrate here.

But this will be a two-part essay. It is relatively rare that English and Continental philosophies cross-pollinate, but nature – as Darwin pointed out – will always make an exception. Austin seems the most quintessentially English of philosophers, with his briar pipe and thick-rimmed spectacles. Jacques Derrida, d’autre part, French father of deconstruction and boogeyman of the rational right, with his foppish showman suits and bouffant hair, seems a boulevardier by comparison. And that is just the sartorial aspect of the two men. Philosophically, Austin and Derrida could not have less in common, surely, but lack of certainty is what makes philosophy endlessly fascinating. Derrida’s 1972 essay Signature, Event, Context deals specifically with Austin, and the book written a decade earlier, HDW. Derrida’s essay was, like Austin’s book, based on a paper given to a conference. Speech and writing. Either thinker has much to say about both.

So, the first part of this essay will deal just with Austin’s HDW. I’ll isolate both the working concepts that Austin uses, and the curious change in English-language philosophy that he ushers in by his novel use of that methodology for which he (and others) became famous; speech-act theory. There is also the relevance of Austin’s ideas to our own time to be considered. Next week, the second part of the essay will deal with Derrida’s riposte to Austin in Signature, Event, Context. 

Austin deals with what are known as “speech acts”, and the book begins by drawing a dichotomy between two types of utterance or statement: the constative and the performative. A constative statement is the one we would more usually expect to see appearing in natural language philosophy, and is either capable or incapable of being verified. The statement, “a car is coming” can easily be verified or falsified by my looking down the street to see if a car is actually coming. But if I am marching into the street, looking in neither direction, and someone yells at me, “A car is coming!”, then I am not being invited to make use of any simple verificationist principle at my leisure, but rather to stay where I am on the pavement for my own good. The first use of the sentence is constative, the second performative. The first is aimed at description of the world, the second productive of action in the world.

This seems a simple division, but Austin shows that there is far more to performative language than a plain either/or. It is not essential that a performative utterance, speech act, or sign actually produce a performance, for example. If I see a sign in Trafalgar Square saying “Don’t feed the pigeons”, and nobody is doing so, the utterance on the sign is still a performative. A speech act (or equivalent) is not true or false, but the conditions to which it refers are, and Austin is explicit in this. If a man is asked at the altar if he takes this woman in marriage and he answers “I do”, we deem him to be married. If this takes place during a stage play, however, we do not. The raw, “phatic” statement does not change. The context does.

Austin continues to “the isolation of the performative”, this being a particular use of language which “is not to describe my doing of what I should be said in so uttering to be doing, or to state that I am doing it; it is to do it”. This covers such mundane statements as “I do” when uttered at the altar, or “I name this ship” while actually so doing. This is the performative, in which “to say something is to do something”.

Austin notes that performative statements are not self-contained, but rely for their validity on the circumstances in which they are made. To say that I am making a bet is not a performative if I announce my intention when the race has finished; answering “I do” when asked if I take this woman to be my lawful wedded wife is not a performative if I am already married, and so on. These anomalous circumstances produce what Austin terms “misfires”, “misinvocations”, and “misexecutions”. If none of these “infelicities” apply, as Austin also terms the class of dysfunctional performatives (referring, rather sweetly, to “happy” and “unhappy” performatives), then we have our performative statement. It has not described anything; it has performed an act.

There is also a further layer of context. Feelings, thoughts, and intentions all have their modal features and thus influence the performative. Entailment, implication, and presupposition all function in a marginally different way. Language, for Austin, is something of a wonderland of communicational nuances:

Much could be said about the connexions here with the phenomena of evincing, intimating, insinuation, innuendo, giving to understand, enabling to infer, conveying, ‘expressing’ (odious word) all of which are, however, essentially different, though they involve the employment of very often the same or similar verbal devices and circumlocutions.

It is possible to make sense of language (which is what we use to make sense of anything), but it is not always an easy task.

As with much English philosophy of language, Austin’s concern is to separate sense from nonsense. He credits Kant with the first real systematic review of this schism, but takes his own, unique philosophical position. If some statements can be shown to be nonsensical, then some contain sense (this follows logically). The nonsensical statements can be filed with the others, as it were, and philosophy and science can forge on with the useful stuff. It’s all very Vienna Circle. But Austin exercises some special pleading for some of these discarded and unwanted speech-acts:

Yet we, that is, even philosophers, set some limits to the amount of nonsense that we are prepared to admit we talk: so that it was natural to go on to ask, as a second stage, whether many apparent pseudo-statements really set out to be ‘statements’ at all.

In other words, sense and nonsense depend on performative context, not just meaning. I may write “all triangles are green” and you would assume I had made a simple category mistake. But if it were part of a poem (although not one I would write or want to read), then it is admissible, for poetry does not recognize the category mistake. Austin does not dwell on the most obvious associative meaning English-speakers have for “performance”: the theatrical or cinematic, but his work is instructive in our own age of politics as theatrical performance.

Austin does not settle into a dogmatic position, and HDW at times seems to question itself (the best philosophy always does that). Its answer is often to come at the problem from another angle, doubtless good practice for a code-breaker. Austin suggests a tripartite division of language into what he will call the locutionary, the illocutionary, and the perlocutionary. Now we can get inside the performative and see what really makes it tick. What are the three stages of a speech act?

The locutionary act is any utterance, any manifestation of meaningful communication, the illocutionary stage sees the entrance of meaning, and the perlocutionary stage is the performative mode proper:

[We distinguish] the locutionary act… which has a meaning; the illocutionary act which has a certain force in saying something; the perlocutionary act which is the achieving of certain effects by saying something.

The latter pair, and their interconnection, are most important in their opposition:

[T]he performance of an ‘illocutionary’ act, i.e. performance of an act in saying something [is] opposed to performance of an act of saying something.

And it is the perlocutionary which is the true performative:

Saying something will often, or even normally, produce certain consequential effects upon the feelings, thoughts, or actions of the audience, or of the speaker, or of other persons: and it may be done with the design, intention, or purpose of producing them.

HDW does enter a thick forest of examples as Austin delegates certain performative functions to certain words in certain contexts. It can be a little dizzying, but the endless parsing of meaning is compulsive. An example – from many – of Austin’s simplicity in ordering his thoughts:

To congratulate may imply a verdict about value or character. Again, in one sense of ‘blame’ which is equivalent to ‘hold responsible’, to blame is a verdictive, but in another sense it is to adopt an attitude towards a person and is thus a behabitive.

There are other categories – exercitives, expositives, commissives – and Austin’s linguistic thoroughness convinces the reader that this was indeed exactly the sort of man British Military Intelligence needed. Philosophers helped win the war.

Austin’s book also has moments of dry humor, and one of them involves simplicity and philosophy:

And we must at all costs avoid over-simplification, which one might be tempted to call the occupational disease of philosophers if it were not their occupation.

And there is one very short, fascinating chapter which rather expands Austin’s remit by (apparently) bringing in time, and I will be anxious to discover whether or not Derrida addresses this (I am re-reading both works by Austin and Derrida for the first time in over 40 years). Most chapters of HDW are just called “Lecture I”, “Lecture 2”, and so on, like Kandinsky’s later paintings. This pivotal chapter is called Can Saying Make it So?, and, quite apart from being the best Country & Western song title ever, forms a perfect bridge in HDW between the idea of the performative and all its problematics, and the biggest problem of all – the promise.

In this short chapter, Austin opens by considering states of affairs which come into being without the need for speech acts, performative or not. A man need not say “I do” at the altar, but merely to cohabitate with his paramour and effect marriage that way. A bet is placed wordlessly by one who puts money in a fruit-machine. So, there are alternative ways for performatives to operate. But what of the seriousness of the performative? Austin elects to look at the promise, “one of the more awe-inspiring performatives”.

In a rare classical allusion, Austin quotes from Plato’s Hippolytus, in which Hippolytus himself describes his making of a promise: “My tongue swore to, but my heart (or mind or other backstage artiste) did not”.

This is the classic problem of the promise, and Boethius devotes some time to it in his exegesis of Aristotle because it is protentional, it looks to a future which is nowhere present outside of the promissory. Austin has already looked at the problem in the light of another of his divisions, that between the primary utterance and the explicit:

(1) primary utterance: ‘I shall be there’,

(2) explicit performative: ‘I promise that I shall be there’, and we said that the latter formula made explicit what action it is that is being performed in issuing the utterance: i.e. ‘I shall be there’.

But the subject of the promise must not become bound up with the performative nature of the promise merely because its intention has become skewed over the course of time:

Do we not actually, when such intention is absent, speak of a ‘false’ promise? Yet so to speak is not to say that the utterance ‘I promise that…’ is false, in the sense that though he states that he does, he doesn’t, or that though he describes he misdescribes—misreports. For he does promise: the promise here is not even void, though it is given in bad faith.

A broken promise (one which is to be broken at some point in the future) is still a performative speech act when it is made.

Austin’s central assertion, that language is not merely descriptive and representative, but also performative, is not confined to the dusty academic halls of 1950s Oxford. The banning of Donald Trump’s Twitter account in 2021, for example, shows that those who pushed for the ban were acutely aware that Trump’s statements, as well as describing situations or giving opinions, had an effect on the world. An utterance, in the Austinian sense, is not confined to the vocal. Any statement which has shared meaning is an utterance, be it a spoken sentence, a notice warning that trespassers will be prosecuted, a price tag on a supermarket item or, in this case, a Tweet. Language is all around us and it is not just saying things to us, it is doing things to us and making us do things in return. We just have to make sense of that, if we can.

Philosophy of language ought to be higher on the political right’s list of concerns because language is the battlefield on which we fight. Austin and other speech-act theorists show us the workings of language in its everyday use, not under the laboratory conditions usually associated with philosophical speculation. Their work represents a type of latter-day ars rhetorica to set with Horace or Aristotle, and is equally absorbing. As noted, next week will see what happened when John met Jacques, strange bedfellows as they may seem.