Philosophy Essay 'austin's Performative Utterances'

1

Justin Beck 0401 0810595

Question 2: ‘Performative Utterances’ and an Analogy to Moore’s Paradox

The purpose of J.L Austin’s paper ‘Performative Utterances’ (1956) is to draw a distinction between two kinds of utterances. An historical understanding of language held that it was the business of every meaningful utterance to be either true or false. Against this historical backdrop, Austin notices a seemingly clear-cut class of utterances whose business is not to make any truth claim. Thus, he wanted to distinguish between different uses of language. However, in the same paper, under careful analysis, Austin’s distinction vanishes. In this essay, I will give a thorough exposition of Austin’s argument and conclusion, focusing specifically on the significance of an analogy Austin makes to Moore’s paradox. I will then consider a number of attempts to revive Austin’s distinction and conclude whether ‘Performative Utterances’ does or does not uncover distinct uses of language. Specifically, Austin makes the distinction between constative and performative utterances. Constatives are a class of ‘fact-stating’ utterances (Horn p159), that is utterances that ‘constate’ something true or false (Graham p54) 1. This includes reports, statements, descriptions, assertions, predictions etc., a simple example is ‘that yacht is white and blue’. Performatives however, although grammatically indiscernible from constatives have two distinctive properties: they do not constate something true or false, and a person making a performative utterance is doing something rather than saying something (Austin p235). Austin gives four straightforward examples of performatives (Austin p235): At a marriage ceremony someone says ‘I do’ (take this women to be my lawfully wedded wife). I tread on your toe and say ‘I apologise’. I have a bottle of champagne in my hand and say ‘I name this ship the Queen Elizabeth’. I say ‘I bet you sixpence it will rain tomorrow’. Notice here that it is nonsense, as Austin puts it, to regard these utterances as truth claims. We do not say anything true or false when we say ‘I apologise’. Nor are these utterances reports of what it is we are doing; we do not report that we are making a bet or christening the ship. Rather by making the utterance, the subject actually performs the action; when I say ‘I do’ I am not reporting a marriage I am indulging in it (Austin p235). A performative is not just saying something, it is doing something above and beyond making the statement (apologising, christening, betting, marrying). We have already said that performative utterances are not factual claims and are not therefore analysable as true or false, they are however analysable in terms of being felicitous or infelicitous. That is to say, a 1

#1: In Performative Utterances, Austin uses the term ‘statement’ to stand for the class of utterances used to make truth statements. This terminology becomes very confusing later on in Austin’s paper when he decides whether statements that state something are in fact statements. To avoid this we will use ‘constative’.

2

Justin Beck 0401 0810595

performative utterance is ‘happy’ – in which case it goes through, succeeds or is satisfactory; or the performative utterance is ‘unhappy’ – in which case it fails (Austin p237). A defining feature of a performance or action is that it is subject to felicity or infelicity. Austin outlines three rules for judging whether a performative is felicitous or infelicitous (Austin p237). Mentioning these rules and illustrating them with examples will help us gain a firmer grasp of the nature of Austin’s performative utterances: i) the convention of which the performative is of must exist. Put another way, there must be a valid way to execute the performative. For instance in order to marry someone by using the performative ‘I do’ there must first exist a convention which dictates that marrying is the process that the performative ‘I do’ accomplishes. ii) The circumstances in which we purport to issue a performative must be appropriate. If for instance, the priest suddenly suffers a stroke in the course of pronouncing a couple man and wife, the marrying would be considered to ‘misfire’. iii) There may be cases in which a performative is ‘abused’. Where the infelicity does not render the performative null and void, but something has certainly gone wrong. A prominent case is when the performative implies certain feelings, beliefs or intentions but those intentions are absent ie. I make a lying promise. This is by no means a comprehensive list of ways in which a performative may be felicitous or infelicitous, but does clearly illustrate the performative (felicitous/infelicitous) – constative (true/false) dichotomy. At this stage, Austin’s analysis shows that constatives and performatives have a mutually exclusive relationship. It will be helpful in what follows to diagrammatically express this initial distinction. Austin’s Initial Constative-Performative Distinction: Constatives

Performatives

Are factual statements that are either true or false.

Do not express anything true or false.

Are the saying of something.

Are performances; the utterance is the doing of something. Are assessed in terms of being felicitous or infelicitous.

Having described the properties of performative utterances Austin then proposes a grammatical criterion, which give us a means of 'picking out' performatives (Austin p241). Austin notices that his four primary examples all feature a verb that is in the first person singular present indicative active form. He further notices that while a verb in this form is performative, the same verb in other persons and other tenses is not performative. Promise for instance, when expressed in the first person singular present indicative active form as 'I promise [such-and-such]' is performative, however when used in an alternative person and tense: 'Willie promises (or promised) [such-and-such]' it yields a constative not a performative. 'Willie promises [such-and-such] is a report about the facts and does not constitute the performance of a promise. This first person singular present indicative active criterion however, is unsatisfactory. Not all performatives are in that grammatical form. Austin revises the criterion to distinguish between explicit performatives and primitive performatives. Take the example of utterances that are not in the first person singular present

3

Justin Beck 0401 0810595

indicative active form, but which are clearly performative: a sign that say's 'warning' or someone who says 'shut the door', neither utterance is making a truth claim, instead they are warning and ordering respectively. Austin says instead of the rigid grammatical criterion proposed above, an utterance could be considered performative if it is reducible or analysable into the form 'I .... [such-and-such]'. So 'shut the door' is performative because it is reducible to the equivalent utterance 'I order you to shut the door'; both do the same job, namely ordering. Austin calls the form 'I ...... [such-and-such] the explicit performative utterance while an utterance which is reducible to this form (such as 'shut the door') is the primitive performative. As long as we have correctly identified the primitive utterance as performative, this device makes it explicitly clear what precise act is being performed 2 (Austin p245). However this revised explicit grammatical criterion works against Austin's initial constative-performative distinction by showing (making it explicitly clear in fact) that all utterances, even classic constatives, are in fact speech-acts. He begins questioning his initial distinction by considering utterances which 'hover' in between constative and performative (Austin p247). 'I am sorry' appears to be a constative in that we are reporting a fact or feeling, yet it seems to act in the same way that 'I apologise' does, which performs an apology. So too when an umpire declares that Rahul Dravid is 'Out'. In this case the umpire is making an utterance that has a connection to the facts, specifically that Rahul Dravid was bowled middle stump, and thus has a 'duty' to be true or false. On the other hand however, the umpire’s declaration also performs the act of giving Dravid out. These cases all suggest that constatives are in some sense performative, but the case that is decisive to Austin treating all utterances as performing some act is the analysis of the classic constative 'I state [such-and-such]. Under the initial distinction 'I state [such-and-such]' is only making a true or false statement. But now, it does not look any different from an explicitly performative utterance. Indeed we are hard pressed to pick out any difference between 'I state [such-and-such]' and the archetypical explicit performative 'I promise [such-and-such]: the second utterance explicitly performs the act of promising, the first explicitly performs the act of stating. Furthermore other utterances previously assumed to be purely constative are actually instances of primitive performatives: 'that yacht is white and blue' is the primitive form of 'I state, notice, report, assert that that yacht is white and blue'. Thus Austin concludes that all utterances are a form of speech-act, that is, whenever we say something we are performing an action, we are at the very least speaking, stating, asserting etc. Austin's constative-performative distinction may now be represented as follows:

2

#2 An analogous case is where John bows in order to greet Mary. In order to make it explicitly clear however, that he is not stretching or inspecting his finely polished shoes, and is in fact greeting Mary, John also raises his hat (Austin p246).

4

Justin Beck 0401 0810595

Austin’s Diminished Constative-Performative Distinction: Constatives Are factual statements that are either true or false.

Performatives Do not express anything true or false.

Are performances; the utterance is the doing of something.

Are performances; the utterance is the doing of something. Are assessed in terms of being felicitous or infelicitous.

The heart of Austin's initial constative-performative distinction has essentially been disproved; constatives are performances every bit as much as performative utterances. However Austin's initial distinction also held that performatives were subject to felicity or infelicity. To entirely assimilate constatives into the class of performatives Austin uses an analogy of Moore's paradox to show that constatives are also subject to felicity and infelicity3. Although the utterance 'The cat is on the mat but I don't believe it is' is truth assessable, we do not want to say that it is false. Likewise 'All John's children are bald, but John has not children'. But there is no doubt that these two utterances are 'unhappy', even if they aren't straightforward contradictions (Austin p248). Austin says that the above constatives are subject to infelicity in the same way performative's are, in fact now we realise constatives and performative were always really the same thing (Austin p248). In this case, the infelicity occurs because each utterance is conjunction of two propositions that reject each other. The two propositions are not contradictory however because one is a statement of fact while the other is in the first person and is only a ‘belief’. This is akin to the third kind of infelicity outlined earlier. But now we can imagine that truth claims are subject to all sorts of infelicities, 'The King of France is bald' for instance is not straightforwardly false, but is unhappy because it fails to refer. So in answer to the question what is the significance of the analogies to Moore's paradox in Performative Utterances we may say the following: i) although the analogies to Moore's paradox could be used to show that all utterances are in fact speech-acts, Austin shows this in his preceding analysis of 'I state [such-and-such], and is therefore not significant in this respect. ii) The analogies are instead used to show that constatives that make truth claims and are speech-acts are also analysable as felicitous or infelicitous. iii) This completely breaks down the constative-performative distinction, all utterances are performative. Austin’s conclusion can be shown below:

3

This is of course expected considering constatives are speech-acts. Being subject to felicity and infelicity is a defining feature of a performance or action.

5

Justin Beck 0401 0810595

Austin’s non-existent constative-performative distinction: Constatives

Performatives

Are factual statements that are either true or false.

Do not express anything true or false.

Are performances; the utterance is the doing of something.

Are performances; the utterance is the doing of something.

Are assessed in terms of being felicitous or infelicitous.

Are assessed in terms of being felicitous or infelicitous.

Before evaluating Austin's argument, and his conclusion, we must say something about an evident flaw in the exposition just given. Constatives may still be distinguished from performatives because constatives make truth claims while performatives do not. On this Austin says that loosely speaking true and false is just a relation to the facts. Similarly, felicity and infelicity is just a relation to the facts. The distinction is not significant and depends only on how stringently we define true and false (Austin p250). A lot of the literature on Austin does not dispute his classification of all utterances as speech acts, however, many authors attempt to salvage the constative-performative distinction. I will deal with just one such attempt as put forward by Jacobsen. I then wish to deal with Black’s argument against Austin’s conclusion, although I will argue against it, and ultimately reject it, Black’s argument highlights the general reservations Austin’s critics have with the notion that constatives just are performatives. Jacobsen argues that Austin overlooks an essential quality that can be ascribed to constative utterances but not to performative utterances, and it is this distinction that allows us to maintain the constativeperformative dichotomy. This quality is the property of ‘inter-personal transference’, whereby utterances such as ‘The cat is on the mat’ may be transferred from one person to another without any agreement between the persons (Jacobsen p358). I am as free to use the utterance ‘the cat is on the mat’ as you are. Contrary to this, performatives cannot be freely transferable to other persons. The intrinsic quality of performatives is that an act performed explicitly by a person is intimately connected to that person. If I perform the act of promising using the performative ‘I promise [such-and-such]’, that act cannot be taken over by another person without some agreement ie. you must agree to fulfil my duties and I agree to let you fulfil my duties (Jacobsen p359). Presumably Jacobsen’s notion of free inter-personal transference rests on the idea that one can freely transfer a constative without damaging the truth or felicity of the utterance, but one cannot transfer a performative without damaging the felicity of that performative. We can reject Jacobsen’s distinction outright for the simple reason that constatives cannot always be freely transferred between people. Like performative’s the speech-act performed by the constative is intimately linked with the utterer. We cannot for instance transfer the utterance ‘that yacht is white and blue’ from a particularly attentive person to, say, a blind person without in some sense rendering the utterance infelicitous. Once again, constatives are liable to infelicity every bit as much as their performative counterparts4. 4

For an alternative attempt to salvage the constative-performative distinction that fails see Appendix 1: Olsen’s attempt.

6

Justin Beck 0401 0810595

Max Black also attempts to salvage the constative-performative distinction. He argues that performatives should be understood primarily as doing something above and beyond saying something true or false. Constatives, although a form of speech act, do not do anything beyond saying, asserting, stating, reporting, describing something true or false. Black proposes the following definition to make this clear: An utterance is said to be performative, when used in a specified circumstance, if and only if its being so used counts as a case of the speaker’s doing something other than, or something more than, saying something true or false (Black p403). The idea of Black’s distinction (and which is common to many arguments against Austin), is that the performative ‘I promise [such-and-such] is not merely a speech-act in virtue of something being uttered, but does something beyond this, it performs a promise. But the constative ‘[such-and-such]’ is only a speech act in virtue of something being uttered. Likewise ‘I state [such-and-such] is constative because it is a speech-act in virtue of something being stated, but does not perform any action beyond this. Forguson rejects this distinction because Black’s constatives may be used to perform actions beyond the act of the utterance. Consider ‘the bull is in the field’, according to Black this is a constative because it performs no action beyond what is performed in virtue of it being uttered. However as Forguson shows, that utterance can be used to perform a multitude of other actions: ‘the bull is in the field’ may be used to warn, inform, to express surprise, to express anger etc. Usually it is not necessary to make the extra force of the utterance verbally explicit, we can ascertain whether the utterance is warning, informing, blaming etc. by the context. However using Austin’s ‘I X [such-and-such]’ grammatical form we can make the force of the utterance verbally explicit, eg. ‘I warn you that there is a bull in the field’, ‘I am surprised that there is a bull in the field’, ‘I confess that there is a bull in the field’ (Forguson p418). Austin has clearly shown, using his explicit performative criterion ‘I state [such-and-such] that all uses of language are performative. It naturally follows from this that all uses of language will be subject to felicity or infelicity. If an utterance is a performance, then it can go wrong, and Austin shows just one case of this using an analogy to Moore’s paradox. The feature that most of Austin’s critics find unsettling about Austin’s conclusion, that constatives are performatives, is that ‘I promise [such-and-such]’ just feels like it is doing more than ‘I state [such-and-such]’ or ‘[such-and-such]. This however is symptomatic of social conventions that frame our use of language. ‘I promise [such-and-such]’ does more performing because the social convention stipulates that the utterance ‘I promise [such-and-such]’ performs the act of binding one to his word. However this is not an intrinsic property of the utterance and cannot be considered a distinct use of language. Any utterance could be given the extra work that ‘I promise’ gets, it is just a matter of which utterances our social conventions pick out.

7

Justin Beck 0401 0810595

REFERENCES: Austin, J. L., ‘Performative Utterances’, 1956. Black, M., ‘Austin on Performatives’, in Symposium on J. L. Austin ed. K. T Fann, Routledge, London, 1969. Forguson, L. W., ‘In Pursuit of Performatives’, in Symposium on J. L. Austin ed. K. T Fann, Routledge, London, 1969. Graham, K., J. L. Austin A Critique of Ordinary Language Philosophy, The Harverster Press, Sussex, 1977. Horn, J., ‘Linguistic Acts’, in The Oxford Companion to Philosophy ed. Honderlich, T., Oxford University Press, New York, 1995. Jacobsen, K., ‘How to Make the Distinction Between Constative and Performative Utterances’, in The Philosophy Quarterly, Vol. 21, No.85, The Philosophy Quarterly, 1985. Olsen, C., ‘Austin’s Worries about ‘I state that…’, in Mind, New Series, Vol.76, No.301, Oxford University Press, 1967.