When is a performative contradiction not a contradiction? When the ‘performative contradiction’ objection is a tu quoque fallacy.

So what is a performative contradiction? We need look no further than Matt Stone and Trey Parker’s Team America: World Police to answer this question. Hollywood actor Gary Johnston (the Tom Cruise parody) is approached by Spottswoode, a United States Government agent, to join the Team America forces. Spottswoode proposes that Gary meet the rest of the team. Gary agrees; Spottswoode and Gary get into Spottswoode’s limousine to make their way to Team America headquarters. But to Gary’s surprise the limousine doesn’t drive, it flies. This is the dialogue that ensues:

Gary Johnston: OK, a limousine that can fly. Now I have seen everything.
Spottswoode: Really? Have you seen a man eat his own head?
Gary Johnston: No.
Spottswoode: So then you haven’t seen everything.

And Gary will never see a man eat his own head because to eat one’s head is impossible. A man simply cannot eat his own head for one good reason: the mouth is a part of the head. So to eat all of one’s head one would have to eat one’s mouth. But how can one eat one’s mouth with one’s mouth? You may get the lips down, but that’s about it. Once you get to the teeth and jaw, you’re done. At this point anyone who tries to do this will realise that he’s engaged in a performative contradiction. (No doubt dejection would follow.) Insofar as we maintain that eating involves chewing and to chew we need to have a mouth, including teeth and a jaw, then there comes a point when you cannot eat any more of your head because the very apparatus required for eating are part of the head.

Auto-cranial mastication clearly involves a performative contradiction, and anyone who is thinking about giving it a crack would be well advised to read Descartes’ Meditations for tips on avoiding performative contradictions. Descartes’ famous cogito ergo sum or ‘I think, therefore I am’ is his proof that he exists. It could not be the case that he could utter this statement and not exist. To say ‘I think and I am not’ would be to commit a performative contradiction. Accordingly, Descartes avoids the existential performative contradiction (as opposed to the auto-cranial mastication performative contradiction) by recognising that if one thinks then one must exist (at least as a thinking thing).

But does the performative contradiction always undermine the validity of the performance, or the claim being made? To cite that most reliable of sources, Wikipedia, a statement (and lets take statements as acts) is only a performative contradiction if the ‘propositional content of [the] statement contradicts the noncontingent presuppositions that make possible the performance of the speech act’ (Wikipedia 2010). To clarify, a statement is a performative contradiction if what is being claimed in the statement contradicts the condition that makes the statement possible. For example, Spottswoode knows good and well that it’s not possible for someone to eat his own head because having a head is the noncontingent condition of having a mouth. So eating your head presupposes that you have a mouth to eat that head with. But you cannot have a functioning mouth without a head. This is the head-eating impasse. Descartes also understood this impasse. He knew that his existence was a ‘noncontingent presupposition’ of thinking and, likewise, that thinking or cognition was the noncontingent presupposition of his uttering ‘I think, therefore I am’.

Given our definition of a performative contradiction, there are at least two conditions under which a statement would not contradict the noncontingent presuppositions that make possible the performance of the speech act: (i) when the statement contradicts the contingent presuppositions that make possible the performance of the speech act; and, (ii) when the statement refutes but does not contradict the noncontingent presuppositions that make possible the performance of the speech act. Let’s consider these conditions through a couple of examples. First, consider the postmodernist claim that reason is irrational.^[1] Defenders of modernity such as Jürgen Habermas argue that to use reason to prove reason’s unreasonableness is to commit a performative contradiction. You cannot possibly use reason to demonstrate that reason is unreasonable, or so it is claimed: ‘Performative contradiction occurs … when what I say is undermined by my saying it’. Further, ‘…an argument that attacks all rational argumentation, that reasons against reason as such … is in contradiction with its own performance’ (Matustik 1989, p. 146).

But is any use of reason to undermine reason necessarily a performative contradiction? Prima facie, yes. There seems to be a close resemblance between using reason to refute reason and the well-worn Liar’s Paradox, ‘This statement is false’. If the statement is true, then the statement is false; and if it is false, then it is true, in which case it is again false, in which case it is true again, and so on and so forth. Accordingly, the statement can never be true, or false. Such paradoxes can drive people to reject reason outright. But can we show that the use of reason to undermine reason is not a performative contradiction by some means other than just jettisoning the idea of a performative contradiction and embracing contradiction? Maybe.

How? Well, the charge by Habermas et. al. against self-refuting reason may be a tu quoque or ‘You do it too!’ fallacy. Tu quoque fallacies are committed by someone who objects to someone else’s argument (say, person A’s argument) in the hope of discrediting it. They take the form:

• (T1) Person A proposes X [The initial argument].
• (T2) Person A acts contrary to X.
• (TC) Therefore, proposal X is not correct [The objection].

The main problem with the tu quoque objection is that a person’s actions do not necessarily affect the soundness of her argument (vander Nat 2010, p. 297). What I say (my argument) counts; what I do has nothing to do with my argument. In argument, the adage ‘Do as I say, not as I do’ is sound.

To return to our main concern—whether reason can be self-refuting—does the use of reason preclude its use to prove the unreasonableness or irrationality of reason? Let’s see if the ‘performative contradiction’ objection fits the tu quoque form.

• (Tr1) Person A proposes X, that reason is irrational.
• (Tr2) But person A, in using reason to prove that reason is irrational, acts contrary to X.
• (TrC) Therefore, proposal X is not correct.

So is the objection, (Tr2), a tu quoque fallacy? That depends on what is meant by the claim that reason is irrational. Is the claim that all reason is irrational, in which case the objection is sound because the argument used to show that reason is irrational is thereby self-refuting; or is the claim more limited than that? Is the claim that although any one piece of reasoning may be valid, ‘reasoning’ as a project eventually leads to its own refutation, in which case the specific piece of reasoning claiming that reason is irrational may be valid although reasoning in toto is self-refuting? At minimum we need an exhaustive account of rationality to know whether or not rationally-established irrationality is precluded by reason.

But in the meantime, what can we say about the postmodernist’s use of reason to prove reason’s irrationality in light of the potential tu quoque fallacy of those who charge pomos with committing a performative contradiction. Let’s consider a second example. You are a child and your father tells you not to smoke, but he himself smokes. The tu quoque looks like this:

• (Ts1) Dad proposes that I do not smoke.
• (Ts2) But dad smokes.
• (TsC) Therefore, dad’s proposal is not correct.

Now, you commit a tu quoque fallacy if you reject the validity of your father’s argument that you should not smoke based on the fact that he himself smokes. Just because your dad smokes doesn’t mean that you can smoke. Dad’s ‘Do as I say, not as I do’ attitude may reveal his hypocrisy, but, remember, hypocrisy is not a legitimate objection to an argument. There may be very good reasons why you should not smoke which your father may or may not invoke. Nevertheless, your dad’s hypocrisy only illustrates that he’s a hypocrite. It doesn’t refute his argument (Ts1). The same may apply for Habermas’s charge against the postmodern auto-refutation of reason.

The fact that postmodernist’s (Tr2) engage reason to (Tr1) disprove or refute reason may make them hypocrites without locking them into a performative contradiction, much as (Ts2) the father’s smoking makes him a hypocrite without locking him into a performative contradiction. Like the child to the father, Habermas may have shown that pomos are hypocrites—they do one thing (use reason) while recommending another (the turn away from reason)—but that may be all. The arguer’s hypocrisy doesn’t disprove his or her position. The performative contradiction would thus be reduced to an act of hypocrisy, but hypocrisy doesn’t undermine one’s arguments.

The case of the hypocritical father as an analogue of the postmodernist’s self-refuting reason stands or falls based on whether or not your father, by smoking while recommending that you don’t smoke (like Habermas charges of the postmodern), contradicts the noncontingent presuppositions of an anti-smoking argument. That is, is it necessary for an anti-smoking argument that the advocate not smoke? In this example the answer is no because we can assess the validity of the father’s argument independent of his smoking. We can separate the argument and the act. And here may be the weakness with Habermas’s argument.

The son in our example conflates the argumentative act or performance and the content of the argument (the father’s claim). By making the argument an act the son traps the father into a performance. Thus, when the act of arguing against smoking is made by someone who also smokes, a performative contradiction arises. But as we’ve seen, the son commits the tu quoque fallacy unless he separates the father’s actions and the content of the argument. When the act and content are separated, our strongest charge against the father is that he’s a hypocrite. Likewise, Habermas’s charge of performative contradiction against postmodernists may be reducible to a charge of hypocrisy. The postmodernist then has two options: accept Habermas’s charge; or try to show that the argument can be removed from the act of making the argument—that the content of the argument is separable from the utterance of that argument. After all, a performative contradiction can only arise where there are two performances to contradict each other. When there is only one performance you cannot have a performative contradiction.

But Habermas has a card up his sleeve. He will argue that while auto-cranial mastication is quite obviously a performative contradiction because it is physically impossible, whether or not the auto-refutation of reason is a performative contradiction is a more complex matter. The auto-refutation of reason is physically possible, no doubt. Many people engage in it. But the question that concerns Habermas is whether or not it is a legitimate action. Whereas auto-cranial mastication is de facto illegitimate because impossible, the legitimacy of the auto-refutation of reason is a question of whether or not it is a de jure action. And insofar as the auto-refutation of reason is a question of the action’s legitimacy, the sheer possibility of postmodernists making the auto-refutation of reason claim is insufficient justification for its legitimacy. So will this move give Habermas the devastating critique of the postmodern refutation of reason he so needs?

[I'll have to fill this section out later, once I've checked the nuances of Habermas's argument. Nevertheless,]

At this point let’s concede that Habermas’s claim holds and reason cannot prove its own irrationality—that the auto-refutation of reason by reason is illegitimate. What remains of reason? A great deal too much. If one doesn’t permit that reason could demonstrate its own irrationality then doesn’t one reduce reason to dogmatism? Reason without the possibility of legitimate, i.e. valid and sound, auto-refutation lacks a theory of error. If one holds that the ‘performative contradiction’ objection holds in all cases of reasoned auto-critique of reason then one could never use reason to legitimately argue for its irrationality. One would thus preclude the possibility of reason’s irrationality. And approaches that preclude the possibility of their own refutation reduce themselves to dogmatism. On this model, all reason is in the service of reason.

But this may simply be a piece of hyperbole. Validity and soundness are safeguards built into reason itself. And on any day I’d prefer to be subject to a dogmatic (pseudo-communicative) yet rational power than a theological, political or any other form of power that rejects its own refutability and demands submission without also possessing an internal system of checks and balances.

---

Notes

[1] Maybe incorrectly, in this post I take reason and rationality as synonymous, and unreason and irrationality as the converse of each and, therefore, also synonymous. [Back to text]

---