Austin – How to Do Things with Words [1955] (VIII & IX)

What happened thus far was this: in trying to search for a list of performative verbs, Austin stumbled over great many problems which gradually turned him away from the idea itself of a principled performative/constative distinction based on grammatical criterion (or even criteria). This eventually led him to distinguish the locutionary act, which was divided into the phonetic, phatic and rhetic act. When we perform a locutionary act, the ‘locution’ is (a) the phonetic act of making of certain noises, (b) the phatic act of uttering of certain words in linguistic constructions, (c) the rhetic act of the producing of meaning. In this lecture, Austin starts by making some further refinements, which prepare the introduction of the illocutionary and perlocutionary act. I believe it’s fair to say that this is the most important lecture so far.

If they’re still blurred, a good test for distinguishing the phatic act (b) from the rhetic act (c) is this: If one says ‘He said “The cat is on the mat”’, then one reports a phatic act – the uttering of certain words. If one says ‘He said that the cat is on the mat’, then one reports a rhetic act – the producing of meaning (of “sense and reference”, in a Fregean twist of phrase). Note that we can always report the phatic act by using quotation marks and the verb ‘said’, but we would want to report a, e.g., command or an exclamation, we would have to change the verb to “he told me to” or “he bade me welcome”. This stems from a deeper difference between the two acts, a difference which Austin spells out like this: “The pheme is a unit of language: its typical fault is to be nonsense – meaningless. But the rheme is a unit of speech: its typical fault is to be vague or void or obscure &c” (p. 98).
Phonetic Act (phone): series of bodily movements which result in the production of a certain sound.
Phatic Act (pheme): conforms to the lexical and grammatical conventions of English – that is, John has produced an English sentence. 
Rhetic Act (rheme): To use a pheme with meaning and reference.
Rhemes are a sub-class of phemes, which in turn are a sub-class of phones. One cannot perform a rheme without also performing a pheme and a phone. The performance of these three acts is the performance of a locution – it is the act of saying something.

As clear as the distinction between the pheme and the rheme may be, it is not in itself the main point; Austin is interested in a rather different sense in which to say something is to do something – the performative. So, we may have fully elucidated the locutionary act in ‘It is going to charge’ and still not be aware, as language users, if the speaker was performing the act of warning or not. To say something about that, we must distinguish the locutionary act from the illocutionary act: the illocutionary act is “in what way we are using the locution” (p. 98). Some examples of illocutionary acts are: ‘asking a question’, ‘announcing a verdict’, ‘making a criticism’, ‘giving a description or a definition’ etc. These different uses are called the forces with which the locutionary acts are used and this is the basis for the separation of force from meaning: We perform an act of saying something with a certain meaning, but we perform an act in saying something with a certain force. The noun ‘use’ here, Austin observes, is “hopelessly ambiguous” and can stand for both acts.

There is another way in which to say something is to do something: To say something is often (and, Austin notes, “even normally”), the producing of certain consequential effects, which can, but need not, be achieved intentionally. This is the perlocutionary act or simply perlocution. So, if the locution is the saying of ‘You can’t do that!’ and the illocution is the act of denying, or protesting against, or interdicting or whatever, then the perlocution is the act of persuading, or bringing to senses or stopping etc. Note that all three can be reported:

(A) Locution: He said to me, ‘You can’t do that!’

(B) Illocution: He protested against my doing that

(C) Perlocution: He brought me to my senses.

Austin makes a remark (p. 103) which is very important in the economy of these lectures: he says that, even if we can describe both B and C, that is, both the illocution and the perlocution, as ‘uses’ of language, the former may be said to be conventional, “in the sense that at least it could be made explicit by the performative formula”, while the latter cannot. The force is conventional, just as meaning partly is. So we can say ‘I (hereby) protest against you doing that’ but we cannot say ‘I (hereby) convince you that …’ Austin concludes: “the illocutionary act is a conventional one: an act done as conforming to a convention”.

The intricate relationships between locution, illocution and perlocution: For instance, perlocution is the consequential effect, but, one might wonder, the consequential effect of what? and how far can we co to extend the consequences? Deeper: what do we mean by consequence? We would certainly like to leave out the idea that the illocution is the consequence of the locution, but then we would equally like to say that there is a certain “regular point” at which the perlocution “breaks” from the illocution – the former being the consequence of the latter (p. 111).

The fact that we have names (i.e. verbs) for illocution should point to the idea that what we name when we name linguistic activity is the convention, not the consequence: “the conventions of certain illocutionary force as bearing on the special circumstances of the occasion of the issuing of the utterance” (p. 114). Still, one cannot simply overlook the fact that the illocution without a certain type of effect is infelicitous (‘unhappy’, ‘unsuccessful’). So, how to distinguish the consequence of persuading from the ‘certain type of effect’ of understanding an utterance? Even semantically, one could find oneself at pains distinguishing the two: what words to we use?

Since understanding (as opposed to any perlocutionary act) could be restated as ‘to take what has been said in a certain sense’ Austin settles this ‘certain type of effect’ as the uptake. We say that an illocutionary act, in order to be felicitous, must “secure an uptake” (p. 116). As a matter of fact, there are two more ways in which the illocution is ‘bound up’ with the effects – all of which can be separated from the perlocution itself.