Literary Theory: A Very Short Introduction - Jonathan Culler 2000

Performative Language

In this chapter I pursue an instance of ’theory’ by following a concept that has flourished in literary and cultural theory and whose fortunes illustrate the way ideas change as they are drawn into the realm of ’theory’. The problem of ’performative’ language brings into focus important issues concerning meaning and effects of language and leads to questions about identity and the nature of the subject.

Austin's performatives

The concept of performative utterance was developed in the 1950s by the British philosopher J. L. Austin. He proposed a distinction between two sorts of utterances: Constative utterances, such as ’George promised to come,’ make a statement, describe a state of affairs, and are true or false. Performative utterances, or performatives, are not true or false and actually perform the action to which they refer. To say ’I promise to pay you’ is not to describe a state of affairs but to perform the act of promising; the utterance is itself the act. Austin writes that when, in a wedding ceremony, the priest or civil official asks, ’Do you take this woman to be your lawful wedded wife?’ and I respond ’I do,’ I do not describe anything, I do it; ’I am not reporting on a marriage: I am indulging in it.’ When I say ’I do,’ this performative utterance is neither true nor false. It may be appropriate or inappropriate, depending on the circumstances; it may be ’felicitous’ or ’infelicitous’ in Austin’s

terminology. If I say ’I do,' I may not succeed in marrying - if, for example, I am married already or if the person performing the ceremony is not authorized to perform weddings in this community. The utterance will ’misfire’, says Austin. The utterance will be unhappy - infelicitous - and so, no doubt, will the bride or groom, or perhaps both.

Performative utterances do not describe but perform the action they designate. It is in pronouncing these words that I promise, order, or marry. A simple test for the performative is the possibility of adding ’hereby’ in English before the verb, where hereby means ’by uttering these words’: ’I hereby promise’; ’We hereby declare our independence’; ’I hereby order you ...’; but not ’I hereby walk to town’. I can’t perform the act of walking by pronouncing certain words.

The distinction between performative and constative captures an important difference between types of utterances and has the great virtue of alerting us to the extent to which language performs actions rather than merely reporting on them. But as Austin pushes further in his account of the performative, he encounters some difficulties. You can draw up a list of ’performative verbs’ which in the first person of the present indicative (I promise, I order, I declare) perform the action they designate. But you can’t define the performative by listing the verbs that behave in this way, because in the right circumstances you can perform the act of ordering someone to stop by shouting ’Stop!’ rather than ’I hereby order you to stop.’ The apparently constative statement ’I will pay you tomorrow,’ which certainly looks as though it will become either true or false, depending on what happens tomorrow, can, under the right conditions, be a promise to pay you, rather than a description or prediction like ’he will pay you tomorrow’. But once you allow for the existence of such ’implicit performatives’, where there is no explicitly performative verb, you have to admit that any utterance can be an implicit performative. The sentence ’The cat is on the mat,’ your basic constative utterance, can be seen as the elliptical version of ’I hereby affirm that the cat is on the mat,’ a performative utterance that accomplishes the act of affirming to which it refers. Constative utterances also perform actions - actions of stating, affirming, describing, and so on. They are, it turns out, a type of performative. This becomes significant at a later stage.

Performatives and literature

Literary critics have embraced the notion of the performative as one that helps to characterize literary discourse. Theorists have long asserted that we must attend to what literary language does as much as to what it soys, and the concept of the performative provides a linguistic and philosophical justification for this idea: there is a class of utterances that above all do something. Like the performative, the literary utterance does not refer to a prior state of affairs and is not true or false. The literary utterance too creotes the state of affairs to which it refers, in several respects. First and most simply, it brings into being characters and their actions, for instance. The beginning of Joyce's Ulysses, ’Stately plump Buck Mulligan came from the stairhead bearing a bowl of lather on which a mirror and a razor lay crossed,' does not refer to some prior state of affairs but creates this character and this situation. Second, literary works bring into being ideas, concepts, which they deploy. La Rochefoucauld claims that no one would ever have thought of being in love if they hadn't read about it in books, and the notion of romantic love (and of its centrality to the lives of individuals) is arguably a massive literary creation. Certainly novels themselves, from Don Quixote to Modome Bovory, blame romantic ideas on other books.

In short, the performative brings to centre stage a use of language previously considered marginal - an active, world-making use of language, which resembles literary language - and helps us to conceive of literature as act or event. The notion of literature as performative contributes to a defence of literature: literature is not frivolous pseudostatements but takes its place among the acts of language that transform the world, bringing into being the things that they name.

The performative is linked with literature in a second way. In principle at least, the performative breaks the link between meaning and the intention of the speaker, for what act I perform with my words is not determined by my intention but by social and linguistic conventions. The utterance, Austin insists, should not be considered as the outward sign of some inward act which it represents truly or falsely. If I say ’I promise' under appropriate conditions, I have promised, have performed the act of promising, whatever intention I may have had in my head at the time. Since literary utterances are also events where the intention of the author is not thought to be what determines the meaning, the model of the performative seems highly pertinent.

But if literary language is performative and if a performative utterance is not true or false but felicitous or infelicitous, what does it mean for a literary utterance to be felicitous or infelicitous? This turns out to be a complicated matter. On the one hand, felicity may be just another name for what critics generally are interested in. Confronted with the opening of Shakespeare's sonnet ’My mistress' eyes are nothing like the sun', we ask not whether this utterance is true or false, but what it does, how it fits in with the rest of the poem, and whether it works happily (felicitously) with the other lines. That might be one conception of felicity. But the model of the performative also directs our attention to the conventions that enable an utterance to be a promise or a poem - the conventions of the sonnet, say. The felicitousness of a literary utterance might thus involve its relation to the conventions of a genre. Does it comply and thus succeed in being a sonnet, rather than a misfire? But more than that, one might imagine, a literary composition is felicitous only when it fully becomes literature by being published, read, and accepted as a literary work, just as a bet becomes a bet only when it is accepted. In short, the notion of literature as performative enjoins us to reflect on the complex problem of what it is for a literary sequence to work.

Derrida's performatives

The next key moment in the fortunes of the performative comes when Jacques Derrida takes up Austin's notion. Austin had distinguished between serious performatives which accomplish something, like promising or marrying, and ’non-serious' utterances. His analysis, he says, applies to words spoken seriously: ’I must not be joking, for example, or writing a poem. Our performative utterances, felicitous or not, are to be understood as issued in ordinary circumstances.' But Derrida argues that what Austin sets aside in appealing to ’ordinary circumstances' are the numerous ways in which bits of language can be repeated - ’non-seriously' but also seriously, as an example or a quotation, for instance. This possibility of being repeated in new circumstances is essential to the nature of language; anything that couldn't be repeated in a ’non-serious' fashion wouldn't be language but some mark inextricably tied to a physical situation. The possibility of repetition is basic to language, and performatives in particular can only work if they are recognized as versions of or quotations of regular formulas, such as ’I do,' or ’I promise.' (If the groom said ’OK' rather than ’I do,' he might not succeed in marrying.) ’Could a performative utterance succeed', asks Derrida, ’if its formulation did not repeat a “codified” or iterable [repeatable] form, in other words if the formula that I utter to open a meeting, christen a boat, or undertake marriage were not identifiable as conforming to an iterable model, if it were not thus identifiable as a kind of citation?' Austin sets aside as anomalous, non-serious, or exceptional particular instances of what Derrida calls a ’general iterability' that should be considered a law of language. ’General' and fundamental, because, for something to be a sign, it must be able to be cited and repeated in all sorts of circumstances, including ’non-serious' ones. Language is performative in the sense that it doesn't just transmit information but performs acts by its repetition of established discursive practices or ways of doing things. This will be important to the later fortunes of the performative.

Derrida also relates the performative to the general problem of acts that originate or inaugurate, acts that create something new, in the political as well as literary sphere. What is the relationship between a political act, like a declaration of independence, that creates a new situation, and literary utterances, that try to invent something new, in acts that are not constative statements but are performative, like promises? Both the political and the literary act depend on a complex, paradoxical combination of the performative and constative, where in order to succeed, the act must convince by referring to states of affairs but where success consists of bringing into being the condition to which it refers. Literary works claim to tell us about the world, but if they succeed they do so by bringing into being the characters and events they relate. Something similar is at work in inaugural acts in the political sphere. In the ’Declaration of Independence' of the United States, for example, the key sentence runs: ’We therefore ... do solemnly publish and declare that these United colonies are and of right ought to be free and independent states.' The declaration that these are independent states is a performative that is supposed to create the new reality to which it refers, but to support this claim is joined the constative assertion that they ought to be independent states.

Performative-constative relations

The tension between the performative and constative emerges clearly also in literature, where the difficulty Austin encounters of separating performative and constative can be seen as a crucial feature of the functioning of language. If every utterance is both performative and constative, including at least an implicit assertion of a state of affairs and a linguistic act, the relation between what an utterance says and what it does is not necessarily harmonious or cooperative. To see what is involved in the literary sphere, let us come back to Robert Frost's poem ’The Secret Sits':

We dance round in a ring and suppose,

But the Secret sits in the middle and knows.

This poem depends on the opposition between supposing and knowing. To explore what attitude the poem takes to this opposition, what values it attaches to its opposing terms, we might ask whether the poem itself is in the mode of supposing or of knowing. Does the poem suppose, like ’we’ who dance round, or does it know, like the secret? We might imagine that, as a product of the human imagination, the poem would be an example of supposing, a case of dancing around, but its gnomic, proverbial character, and its confident declaration that the secret ’knows’, makes it seem very knowing indeed. So we can’t be sure. But what does the poem show us about knowing? Well, the secret, which is something that one knows or does not know - thus, an object of knowing - here becomes by metonymy or contiguity the subject of knowing, what knows rather than what is or is not known. By capitalizing and personifying the entity, the Secret, the poem performs a rhetorical operation that promotes the object of knowledge to the position of subject. It thus shows us that a rhetorical supposition can produce the knower, can make the secret into a subject, a character in this little drama. The secret who knows is produced by an act of supposing, which moves the secret from the place of the object (Someone knows a secret) to the place of the subject (The Secret knows). The poem thus shows that its constative assertion, that the secret knows, depends on a performative supposing: the supposing that makes the secret into the subject supposed to know. The sentence says that the Secret knows but it shows that this is a supposition.

At this stage in the history of the performative, the contrast between constative and performative has been redefined: the constative is language claiming to represent things as they are, to name things that are already there, and the performative is the rhetorical operations, the acts of language, that undermine this claim by imposing linguistic categories, bringing things into being, organizing the world rather than

simply representing what is. We can identify here what is called an ’aporia’ between performative and constative language. An ’aporia’ is the ’impasse’ of an undecidable oscillation, as when the chicken depends upon the egg but the egg depends on the chicken. The only way to claim that language functions performatively to shape the world is through a constative utterance, such as ’Language shapes the world’; but contrariwise, there is no way to claim the constative transparency of language except by a speech act. The propositions which perform the act of stating necessarily claim to do nothing but merely display things as they are; yet if you want to show the contrary - that claims to represent things as they are in fact impose their categories on the world - you have no way to do this except through claims about what is or is not the case. The argument that the act of stating or describing is in fact performative must take the form of constative statements.

Butler's performatives

The latest moment of this little history of the performative is the emergence of a ’performative theory of gender and sexuality’ in feminist theory and in gay and lesbian studies. The key figure here is the American philosopher Judith Butler, whose books Gender Trouble: Feminism and the Subversion of Identity (1990), Bodies that Matter (1993), and Excitable Speech: A Politics of the Speech Act (1997), have had great influence in the field of literary and cultural studies, particularly in feminist theory, and in the emerging field of gay and lesbian studies. The name ’Queer Theory’ has recently been adopted by the avant-garde of gay studies whose work in cultural theory is linked with political movements for gay liberation. It takes as its own name and throws back at society the most common insult that homosexuals encounter, the epithet ’Queer!’ The gamble is that flaunting this name can change its meaning and make it a badge of honour rather than an insult. Here a theoretical project is emulating the tactics of the most visible activist organizations involved in the fight against AIDS - the group ACT-UP, for instance, which in their demonstrations use such slogans as ’We're here, we're queer, get used to it!'

Butler's Gender Trouble takes issue with the notion, common in American feminist writing, that a feminist politics requires a notion of feminine identity, of essential features which women share as women and which give them common interests and goals. For Butler, on the contrary, the fundamental categories of identity are cultural and social productions, more likely to be the result of political cooperation than its condition of possibility. They create the effect of the natural (remember

Aretha Franklin’s ’You make me feel like a natural woman') and by imposing norms (definitions of what it is to be a woman) they threaten to exclude those who don’t conform. In Gender Trouble Butler proposes that we consider gender as performative, in the sense that it is not what one is but what one does. A man is not what one is but something one does, a condition one enacts. Your gender is created by your acts, in the way that a promise is created by the act of promising. You become a man or a woman by repeated acts, which, like Austin’s performatives, depend on social conventions, habitual ways of doing something in a culture. Just as there are regular, socially established ways of promising, making a bet, giving orders, and getting married, so there are socially established ways of being a man or being a woman.

This does not mean that gender is a choice, a role you put on, as you choose clothes to put on in the morning. That would suggest that there is an ungendered subject prior to gender who chooses, whereas in fact to be a subject at all is to be gendered: you can’t, in this regime of gender, be a person without being male or female. ’Subjected to gender but subjectivated [made a subject] by gender,’ writes Butler in Bodies that Matter, ’the “I” neither precedes nor follows the process of this gendering but emerges only within and as the matrix of gender relations themselves.’ Nor should the performativity of gender be thought of as a singular act, something accomplished by one particular act; rather, it is ’the reiterative and citational practice’, the compulsory repetition of gender norms that animate and constrain the gendered subject but which are also the resources from which resistance, subversions, and displacement are forged.

From this viewpoint, the utterance ’It’s a girl!’ or ’It’s a boy!’ by which a baby is, traditionally, welcomed into the world, is less a constative utterance (true or false, according to the situation) than the first in a long series of performatives that create the subject whose arrival they announce. The naming of the girl initiates a continuous process of ’girling’, the making of a girl, through an ’assignment’ of compulsory repetition of gender norms, ’the forcible citation of a norm'. To be a subject at all is to be given this assignment of repetition, but - and this is important for Butler - an assignment which we never quite carry out according to expectation, so that we never quite inhabit the gender norms or ideals we are compelled to approximate. In that gap, in the different ways of carrying out the gender's ’assignment', lie possibilities for resistance and change.

Stress falls here on the way the performative force of language comes from the repetition of prior norms, prior acts. So, the force of the insult ’Queer!' comes not from the intention or authority of the speaker, who is most likely some fool quite unknown to the victim, but from the fact that the shout ’Queer' repeats shouted insults of the past, interpellations or acts of address which produce the homosexual subject through reiterated shaming or abjection ( abjection involves treating something as beyond the pale: ’anything but that!'). Butler writes, ’Queer' derives its force precisely through the repeated ... invocation by which a social bond among homophobic communities is formed through time. The interpellation echoes past interpellations, and binds the speakers, as if they spoke in unison across time. In this sense it is always an imaginary chorus that taunts ’queer!'

What gives the insult its performative force is not the repetition itself but the fact that it is recognized as conforming to a model, a norm, and is linked with a history of exclusion. The utterance implies that the speaker is the spokesman for what is ’normal' and works to constitute the addressee as deviant, beyond the pale. It is the repetition, the citation of a formula which is linked to norms sustaining a history of oppression, that gives a special force and viciousness to otherwise banal insults such as ’nigger' or ’kike'. They accumulate the force of authority through the repetition or citation of a prior, authoritative set of practices, speaking as if with the voice of all the taunts of the past.

But the performative’s link with the past implies the possibility of deflecting or redirecting the weight of the past, by attempting to capture and redirect the terms that carry an oppressive signification, as in the adoption of ’Queer’ by homosexuals themselves. It’s not that you become autonomous by choosing your name: names always carry historical weight and are subject to the uses others will make of them in the future. You can’t control the terms that you choose to name yourselves. But the historical character of the performative process creates the possibility of a political struggle.

Stakes and implications

Now it is obvious that the distance between the beginning and the (provisional) end of this story is very great. For Austin, the concept of the performative helps us to think about a particular aspect of language neglected by prior philosophers; for Butler, it is a model for thinking about crucial social processes where a number of matters are at stake: (1) the nature of identity and how it is produced; (2) the functioning of social norms; (3) the fundamental problem of what today we call ’agency’ in English: how far and under what conditions can I be a responsible subject who chooses my acts; and (4) the relationship between the individual and social change.

There is, thus, a big difference between what is at stake for Austin and for Butler. And they seem to have principally in view different sorts of acts. Austin is interested in how the repetition of a formula on a single occasion makes something happen (you made a promise). For Butler this is a special case of the massive and obligatory repetition that produces historical and social realities (you become a woman).

This difference, in fact, brings us back to the problem about the nature of the literary event, where there are also two ways of thinking of it as performative. We can say that the literary work accomplishes a singular, specific act. It creates that reality which is the work, and its sentences accomplish something in particular in that work. For each work, one can try to specify what it and its parts accomplish, just as one can try to spell out what is promised in a particular act of promising. This, one might say, is the Austinian version of the literary event.

But on the other hand, we could also say that a work succeeds, becomes an event, by a massive repetition that takes up norms and, possibly, changes things. If a novel happens, it does so because, in its singularity, it inspires a passion that gives life to these forms, in acts of reading and recollection, repeating its inflection of the conventions of the novel and, perhaps, effecting an alteration in the norms or the forms through which readers go on to confront the world. A poem may very well disappear without a trace, but it may also trace itself in memories and give rise to acts of repetition. Its performativity isn't a singular act accomplished once and for all but a repetition that gives life to forms it repeats.

The concept of the performative, in the history I have outlined, brings together a series of issues that are crucial to ’theory'. Let me just list them:

First, how to think about the shaping role of language: do we try to limit it to certain specific acts, where we think we can say with confidence what it does, or do we try to gauge the broader effects of language, as it organizes our encounters with the world?

Second, how should we conceive of the relation between social conventions and individual acts? It is tempting, but too simple, to imagine that social conventions are like the scenery or background against which we decide how to act. Theories of the performative offer better accounts of the entanglement of norm and action, whether presenting conventions as the condition of possibility of events, as in Austin, or else, as in Butler, seeing action as obligatory repetition, which may nevertheless deviate from the norms. Literature, which is supposed to ’make it new' in a space of convention, calls for a performative account of norm and event.

Third, how should one conceive of the relation between what language does and what it says? This is the basic problem of the performative: can there be a harmonious fusion of doing and saying or is there an unavoidable tension here that governs and complicates all textual activity?

Finally, how, in this postmodern age, should we think of the event? It has become commonplace in the United States, for instance, in this age of mass media, to say that what happens on television ’happens period', is a real event. Whether the image corresponds to a reality or not, the mediatic event is a genuine event to be reckoned with. The model of the performative offers a more sophisticated account of issues that are often crudely stated as a blurring of the boundaries between fact and fiction. And the problem of literary event, of literature as act, can offer a model for thinking about cultural events generally.