Hate speech
Butler further elaborates her understanding of agency in Excitable Speech, her analysis of hate speech. She now accepts the distinction between two kinds of performative utterances: ¡[locutions and pcrlocutions. The former enact a deed as soon as they are uttered, as do the ceremonial statements ‘I now pronounce you man and wife’, and ‘guilty’ (Butler 1997, pp. 3, 17, 44). The latter aid in the performance of a deed, but do not enact the deed in and of themselves.
The speech act and its consequences arc ‘temporally distinct’ (ibid., p. 17). As a consequence, a gap opens up between word and event.
This distinction permits Butler to distance herself from those who advocate censorship because of the harmful impact of hate speech. One might have expected that her philosophy, emphasising the marginalising potential of linguistic labels, would have been sympathetic to censorship. However, Butler suggests that hate speech is an example of a pcrlocutionary utterance. Epithets do not immediately constitute the things they name. The political strategy of censoring hate speech assumes, mistakenly, the illocutionary approach, contending that slurs by definition take instant effect. Butler cautions against censorship laws while granting that slurs and epithets do harm.16 In the way that cultural norms of sex define our identities as men and women, we depend in general on being named (ibid., p. 26). Thus, when someone calls us by a slur, the word sticks and stings. Butler asks rhetorically: ‘Why should a merely linguistic address produce such a response of fear? Is it not, in part, because the contemporary address recalls and re-enacts the formative ones that gave and give existence?’ (ibid., p. 5). Our social existence is brought into being through language; thus, it is natural that words can hurt.
Yet Butler’s preferred strategy for challenging hate speech is to take the words that hurt us and mimic them in a slightly different context. As gender norms arc challenged through ironic parody, the best way to target hate speech is the ‘restaging and resignifying’ of hate words (ibid., p. 13). Similarly, pornography proliferating derogatory images of women is best challenged through the resignification of its content (ibid., p. 69). Women can make pornography that defies the meanings of most sexually explicit material. If we use hate speech or pornography in a positive or amusing ironic fashion, it can gradually lose most of its power to hurt. It will become so recontextualised that its original meaning will be almost (although not ever completely) lost (ibid., pp. 15, 150, 161). ‘[S]peech is ... vulnerable to failure’ (ibid., p. 12) Butler writes, because, ‘[it] is finally constrained neither by its specific speaker nor its originating context’ (ibid., p. 40). Were we to censor speech and pom - insisting that their meaning is fixed in the foundational intent of the speaker, or in a specific referent in the world - they would forever maintain their capacity to injure (ibid., pp. 15, 38).
Butler contends that much political action in the past several decades has been a form of this redefinition. Rosa Parks was reiterating the actions of whites when she sat at the front of the bus (ibid., p. 147). The attempts by various social movements to be granted equality, freedom, and justice arc performative and rearticulatory, as the definition of these words is challenged and rewritten (ibid., pp. 157-9). Butler periodically indicates that rcarticulation is the only form of struggle possible. While she does not explicitly rule out other types of political action, her statements about the potential of recontextualisation are quite dramatic. ‘[T]he logic of iterability ... governs the possibility of social transformation’, she insists (ibid., p. 147). She cites Pierre Bourdieu favourably when he writes, ‘The body believes in what it plays at: it weeps if it mimes grief’ (ibid., p. 154). If grief can be so constructed, the implication is that we can effectively change people if we alter what we say about them and, eventually, what they say about us. Butler closes Excitable Speech with the following declaration: ‘Insurrectionary speech becomes the necessary response to injurious language, a risk taken in response to being put at risk, a repetition in language that forces change’ (ibid., p. 163). The claims Butler makes here for performative politics arc thus substantial.
Despite the use of a different language, I am convinced that this analysis shows many signs of influence by twentieth-century behaviourism, Quinean and otherwise. Seyla Benhabib has similarly noted the connection (Benhabib 1995, p. 21), which Butler has explicitly and adamantly rejected (Butler 1993, p. 2). Certainly, Butler wants no part of the reductivist physicalism that inspired the behaviourist rejection of consciousness, instinct, and emotion as explanatory variables. Yet she redefines these variables in comparable fashion. For Butler, Quine, and behaviourism, there are no extra-linguistic or extra-cultural dimensions to anything, cither in the world or in our minds. The linguistic community mediates which aspects of the environment will be discriminated, and which aspects of the person will become the self. Behaviour or speech is either mimicry (albeit inexact) or reaction. Interestingly enough, the concept ‘performance’ that Butler champions (or her preferred variant, performativity, to note the ongoing nature of the process) was often used by behaviourists to describe the actions of their experimental subjects. ‘Performance’ supposedly captured the notion that subjects were behaving as scripted, rather than expressing internal states or capacities.17 Clearly this is a key principle for Butler as well.
There is, all the same, considerable truth to Butler’s thesis of ‘rearticulation’, as she calls it. Most serious thinkers acknowledge that the individual is not the purely self-constituting origin of his or her thoughts, utterances, and actions. Behaviourism surely explains some aspects of linguistic acquisition and transmission. Regardless, Noam Chomsky’s famous reply to Skinner years ago mounts important criticisms of Butler’s theory as well (Chomsky 1959, pp. 26-58). Consider these simple examples from Chomsky’s criticism. When individuals sec a painting and some respond with the word ‘tilted’, some with ‘beautiful’, and others with ‘clashes with the wallpaper’, Chomsky insists that Skinner can only conclude that each response is under the control of different physical stimuli (ibid., p. 31). Butler replaces the notion of physical stimuli with discursive stimuli. If individuals are heterosexual or homosexual, racist or egalitarian, this signifies only that each person is constituted within a slightly different discursive regime.
Chomsky locates two difficulties with this general approach. It cither makes implicit reference to internal qualities of the individual in question (why that person chose the response they did), or, failing these covert references, it turns to an empty notion of reinforcement. In the first case, there is agency, in that the individual somehow focuses on different stimuli or different discursive regimes. But how or why these differences occur is left unexplored. The notion that individuals could choose one alternative over another is apparently
Post-structuralist and constructivist feminism 73 dismissed. Chomsky argues that the behaviourist rejection of reasoned choice as a variable leads to the absurdity that torturing someone to say that the earth is motionless is the same as changing his or her opinion (Chomsky 1971, p. 21). Both are apparently conditioned responses. While we must continue to challenge the notion that the individual is fully self-constituting, the a priori alternative that people are simply determined or constituted into their social roles must be avoided. As Chomsky put it, to assume otherwise is to insinuate that ‘people have wills, impulses, feelings, purposes, and the like no more than rocks do’ (ibid.).
This leads to the second difficulty. Butler would surely deny that she is reducing people to rocks. Instead, she turns our attention away from inner factors to constitutive discourses, whereas Skinner emphasises reinforcing physical stimuli. Chomsky asserts that these forces arc so vague as to be completely non-explanatory:
Skinner’s claim that all verbal behavior is acquired and maintained ... through reinforcement is quite empty, because his notion of reinforcement has no clear content, functioning only as a cover term for any factor, detectable or not, related to acquisition or maintenance of verbal behaviour.
(Chomsky 1959, p. 38)
I suggest that this is a valid, and perhaps the most important, criticism of Butler’s notion of discursive constitution. It functions as a cover term for any and every fact related to the acquisition or maintenance of behaviour, verbal and otherwise. As Chomsky concludes, we know no more about the process of learning, or the role of agency, than we did before, when we were naive voluntarists (ibid.). In a way, the theory of discursive constitution reminds me of the rational actor model. Both arc totalising. If every action is rational, or discursively constituted, it becomes difficult to explain why people behave differently in similar circumstances. Butler’s rejoinder that these defectors arc constituted in reaction to the discursive norm may open up the explanatory door somewhat, but it is still extremely unhelpful in its generality.
I thus question whether Skinnerian conditioning or Butlerian resignification adds sufficiently to our knowledge of human motivation and behaviour. In the example of Rosa Parks, Butler’s argument that she was ‘rearticulating’ the concepts of freedom and equality tells us nothing more about Rosa Parks, racism, or the fight against it. Why was it Rosa Parks that was able to engage in her brave act? Why did she refuse to give up her scat in 1955, and not 1950? Butler can only reply that Parks was somehow able to tap into a relationship between the discourse of racism and another discursive network. The reference to the individual is thus thrown back to discourse, in an apparent loop with no origin other than the whole of language, culture, and the individuals inhabiting their domains. If the regress is ever to be terminated, behaviourism and now post-structuralism seem to adhere to the view that human behaviour is random, and only fixed as it is selected and reinforced by the community. Such a solution to the very difficultproblem of explaining human behaviour appears to be a covert retreat into cither metaphysics or, perhaps, even the mathematical concept of chance.
There arc certainly other ways in which Butler’s project is remote from Quine’s and Skinner’s behaviourism. Yet she has primarily defended herself against the charge of behaviourism by arguing that her ‘subject’ of discourse is never final or fixed. As I have summarised, the capacity for intention or agency is located in cither the individual’s inexact mimicry of norms, or the reaction made possible by relationships between two or more norms. Butler writes:
To be constituted by language is to be produced within a given network of powcr/discoursc which is open to resignification, redeployment, subversive citation from within, and interruption and inadvertent convergences with other such networks. ‘Agency’ is to be found precisely at such junctures where discourse is renewed.
(Butler 1995, p. 135)
Every single woman will perform a slightly different version of womanhood. A lesbian woman’s actions arc constituted in relation to the sanctioned norms of heterosexuality. A constituted identity is never fully determined, and thus the thesis of rcarticulation is allegedly not behaviourist (ibid., and 1993, p. 2).
Yet it was no part of Skinner’s doctrine to argue that human behaviour was inalterable, or that identities were fixed. For both Butler and Skinner, as we arc always already conditioned, social change is the result of reconditioning. Butler has argued that hate speech can be rccontcxtualiscd and hence redefined because it has no fixed links to either the hateful speaker’s intent or to a condition in the world. The only thing cementing the meaning of a word - and this is a relative cementing - is its repeated association with a particular context. Individuals have thus been conditioned into these word-context linkages. Because context varies, it becomes possible to orchestrate the variation to the advantage of individuals injured by hate speech. When minority members accept an epithet, but insist that its ‘meaning’ is now altered, individuals who originally utter the word with hateful intentions will be reconditioned. The target of the hate speech apparently accepts the designation ironically, thwarting the original speaker’s goal. The original speaker has, in Skinner’s language, been punished. The best chance for a more equal society, for both Skinner and Butler, depends on getting people to associate old words with new contexts; in effect, reconditioning people in the expectation that this will change behaviour.
I do not think Butler’s refutation gets to the heart of the allegation: it is the rejection of instincts, natural capacities, and the like, that is the hallmark of behaviourism. Butler has consistently maintained one of two things. Either we have no means of knowing whether there is any innate sex or sexuality because discourse mediates all of our knowledge, or there are no such inner forces because these arc either overridden by or intertwined with cultural forces. She has now articulated a more general theory of agency in which reference to spontaneous instincts or a capacity to reason existing prior to culture or language is similarly forbidden. Agency, as I just summarised, is a cultural ‘rearticulation’, not an expression of innate tendencies, urges, potentials, or reason. Agency is cither inexact mimicry or reaction. Butler’s overall philosophy is extremely close to Quine’s and Goodman’s in its relativism and nominalism; her political analysis is not as far from Skinnerian and Watsonian behaviourism as she has proclaimed.
