Performatives

What is a "performative"? Here's some context:

All performatives imply propositions. There's no point in my operating a performative like, say, promising, or cursing, unless I have certain beliefs about the nature of reality: that there is indeed such an institution as promising, that I am able to perform it, and so on.

A reasonable test of a passage of argument is that the opposite makes sense or that someone might argue it. If it doesn't, or nobody would, the passage is likely to be meaningless in some way, perhaps tautological, perhaps hollow. Perhaps, and this is more often the case, tendentious. I am in favour of "fair" taxation, for example. Nobody would say the opposite, that they are in favour of "unfair" taxation. We're all in favour of fair taxation, we just don't agree what that means. So using the phrase, as some politicians do, is tendentious.

To take an example from the above passage, it talks of "certain beliefs about the nature of reality". Such a belief might be that God exists. Both that sentence and its opposite (God does not exist) make sense. Does it make sense to say that there is no such institution as promising? Would anyone say that? Would someone be frozen, unable to make a promise, because they do not believe there is such a thing as a promise, or that there is but they are unable - not morally unable but unable because of their "beliefs about the nature of reality" - to perform it? If that were the case with someone, it would tell us nothing about the nature of reality, but rather something about that individual's psychology. But nobody would, just as nobody argues for unfair taxation.

You'll probably have guessed we're in the hinterlands of postmodernism here. Wikipedia's page about performativity includes this:

Philosopher and feminist theorist Judith Butler has used the concept of performativity in her analysis of gender development, as well as in her analysis of political speech. Eve Kosofsky Sedgwick describes Queer Performativity as an ongoing project for transforming the way we may define - and break - boundaries to identity.

Peformativity is incredibly useful to a postmodernist, because it enables them to argue against "positivism" or a "representational idiom". What's a representational idiom? Here's where it gets sinister, as these things usually do. Also from Wikipedia:

Performativity is a concept that is related to speech act theory, to the pragmatics of language, and to the work of J. L. Austin. It accounts for situations where a proposition may constitute or instantiate the object to which it is meant to refer, as in so-called "performative utterances".

The concept of performativity has also been used in science and technology studies and in economic sociology. Andrew Pickering has proposed to shift from a "representational idiom" to a "performative idiom" in the study of science. Michel Callon has proposed to study the performative aspects of economics, i.e. the extent to which economic science plays an important role not only in describing markets and economies, but also in framing them.

In other words, it's being argued that in fields like science and economics, the "performance" of the scientist or economist "frames" the outcomes of their work - and not just that but also the very thing they are studying.

How is this useful to a postmodernist? It's a valuable tool if you're trying to argue something for which there is absolutely no evidence, or against which there is significant evidence. Which brings me to the full context of the first quote. A conversation with Terry Eagleton in the Monthly Review, as reproduced by Norm:

You say he [Dawkins] emphasizes a "propositional" account of religious faith above a "performative" one. But how far can one go believing in God performatively, through political acts, before it becomes a proposition?

TE: All performatives imply propositions. There's no point in my operating a performative like, say, promising, or cursing, unless I have certain beliefs about the nature of reality: that there is indeed such an institution as promising, that I am able to perform it, and so on. The performative and the propositional work into each other. But it is a typically positivist kind of mistake to begin with the propositional, just as it would be for someone trying to analyze a literary text, which is basically a performance. Somebody who didn't grasp that would be making a root-and-branch mistake about the kind of thing being confronted. These new atheists, and, indeed, the great majority of believers, have been conned rather falsely into a positivist or dogmatic theology, into believing that religion consists in signing on for a set of propositions.

Norm, rightly, remarks:

Terry is right to say, as he does further on in the interview, that 'It is a rationalist error to think that your opponents are simply stupid.' But he's wrong to deny 'that religion consists in signing on for a set of propositions'. That may not be all religion is about, but it is, centrally, about that.

Incidentally, thinking your opponents are stupid isn't just a rationalist error, but more on that later. First, I'd like to put in plain English what Eagleton is arguing - because one thing this language does is dress trivial or untenable arguments in fine clothes of verbiage. It is, as Richard Feynman used to say, like turning on a fog machine. Clouds of this fog billow out, and it's hard to see at all, let alone clearly, what the person responsible is saying.

Believing in God "performatively" to "begin with" means that acts, performances, behaviours can lead to specific beliefs. And that's certainly true. The problem with this is that enough ritual, incense, repetition - together with other aspects of religious behaviour, scourging, sexual abstention, sleep deprivation, fasting - can make anyone believe, eventually, in Archwaldo the Giant Turtle. Imagining that rationalists and atheists don't realise this is to accuse them of stupidity. Not only do they realise it, they think it's one of the problems with religious behaviour: it helps you believe in things that aren't true.

The fact that someone believes something has absolutely no bearing on whether or not it's true. The fact that someone can come to religious belief through, or helped by, performance - or that performance is a major part of their religious experience - has absolutely no bearing on whether or not the things they believe are true.

Chris Dillow wrote in support of Eagleton, in terms interesting for one so committed to evidence-based argument. He concluded:

One reason why writing about music is like dancing about architecture is that our language struggles to cross the barrier between practitioner and non-practitioner. Perhaps believer and non-believer will always be unable to understand each other. But then, why should all knowledge and beliefs be explicit rather than tacit and so amenable to “rational” debate?

People do, of course, write perfectly coherently about music, but they don't dance about architecture. I know what Chris means, writing about music is inadequate in some ways, the music has to be experienced. That's true, but it's a false analogy nevertheless. Equally false is the grouping together of knowledge and beliefs. These are not the same thing at all, in some ways they could be said to be opposites. We require belief for assertions for which we have insufficient knowledge, or that run contrary to knowledge. For me to think the world is flat would require belief and the suppression of knowledge, to think it is roughly spherical requires no more than knowledge. The shape of the world can only be determined through rational enquiry - and most importantly it can be determined like that. So can the question of whether, say, deficit spending is appropriate in a recession.

Ultimately, we might reach a stage of understanding of economics where we know - genuinely know - such things in sufficient detail to avoid recessions, or to understand that they are inevitable in the least bad forms of economies but that they can be mitigated according to certain rules. The more recessions we have and the more we study the reality of what happens, the closer we'll get to this stage. Belief, irrational, dogmatic belief, might help. It might produce hunches that can be tested and prove to be correct - as Polanyi argued (Chris's link on the word "tacit"). But the same could be said of throwing darts at a piece of paper with ideas written on it, or throwing dice to choose research avenues. Belief itself is of no value except to the individual holding it and that can be a problem for the rest of us. It's less often an asset.

The same is true of the question of whether or not there is a God. And that is the question, not whether or not some people believe in God, or even why they believe in God except where this is for rational reasons. Eagleton's argument isn't even wrong, it isn't even in the park. But that doesn't matter, it's not meant to be. It's designed to remove the subject from the sphere of the rational - that's what this means: "it is a typically positivist kind of mistake to begin with the propositional".

What looks like a clever dismissal of the inflexible Dawkins is actually a retreat from the debate. Eagleton cedes the ground of rational argument and says, instead, that religious belief is brought into existence through performance.

I think Dawkins might agree with that.