Human Reason and A Priori Economics
Jason Brennan has in posts at Bleeding Heart Libertarians rendered a great service to supporters of a priori economics in the style of Mises and Rothbard, though this is far from his aim. To the contrary, he endeavors to show that a priori economics is untenable.
He recounts a conversation with someone he calls “Austrian Dude.” Brennan asked Austrian Dude how he, as a supporter of a priori Austrian economics, would deal with irrational actions of the sort studied by behavioral economics. (Examples include “confirmation bias” and “framing effects.”) In response, the Dude said that these are examples of behavior and not action: true action is rational. Brennan rightly comments that the Dude has committed the No True Scotsman Fallacy. He purported to be talking about actions, as this notion is ordinarily understood. Confronted with a counterinstance to his claim, the Dude shifted to a question-begging claim about “true” action.
Brennan makes short work of Austrian Dude, but the Dude made a false claim about Austrian economics. Mises explicitly rejects the view that actors never make mistakes in reasoning. When he says that all action is rational, he intends something much less controversial. He has in mind only that an actor regards the means he selects as suitable to attain the end he wants. The actor may well be wrong in thinking that he has chosen a suitable means; but he thinks he has.
Mises says, “It is a fact that human reason is not infallible and that man very often errs in selecting and applying means. An action unsuited to the end falls short of expectation. It is contrary to purpose, but it is rational, i.e., the outcome of a reasonable — although faulty — deliberation and an attempt — although an ineffectual attempt — to attain a definite goal.”
Brennan would, if I am not mistaken, answer that I have missed his point. He intended the story about the Austrian Dude to be an example of a general thesis. Even if the example misses the mark, the general thesis remains good. If Mises makes a claim about action, the claim can be taken in two ways. If it is a claim about actions in the world, according to our ordinary concept of action, then it is an empirical claim. We can only find out if the claim is true by observation. We cannot know it to be true a priori. We can build into our concept of action that a particular claim about action is true, but then we will be committing the No True Scotsman Fallacy. We cannot find out the nature of the real world just by thinking about it.
Is this so? Brennan has given us no reason to think that it is. Mises intends to be talking, not about some contrived concept of “action,” constructed to make his assertions about action true by stipulation, but rather about actions in the world, as we ordinarily understand them. Precisely his thesis is that we can gain knowledge of actions, understood this way, by thinking. We have a priori knowledge that applies to the world.
What has Brennan to say against this? So far as I can see, he advances no argument. He merely states his own contrasting position. Mises says that we have a priori knowledge of action; and Brennan responds, in effect, “No, we don’t.”
Many philosophers would agree with Brennan about the a priori. Indeed, some philosophers, e.g., W.V.O. Quine and Michael Devitt, go further and reject the a priori altogether. But the status of the a priori, among those who accept its existence, is a controversial topic, and the view that Brennan has of it by no means is the only one in the field. Just to mention a few well-known thinkers, Tyler Burge, Christopher Peacocke, Laurence BonJour, Alvin Plantinga, and David Chalmers hold much more robust views of the scope of a priori knowledge than Brennan does.
Of course, to mention these philosophers does not begin to suffice to show that Brennan is wrong, and I do not intend here to defend a view of the a priori of my own. Just to give a very preliminary reason for thinking that it is not nonsensical to believe that we have a priori knowledge not confined to defining concepts, though, here are a few examples of statements that appear to be both a priori and to give us knowledge of the actual world: (1) Something exists; (2) I now exist [said by me now]; (3) I now know that 2+2 = 4 [said by me now].
Rather, my aim is a more limited one. It is to point out that if Brennan wishes to reject the a priori character of economics, he needs to do more than just state a view that denies that such knowledge is possible. He must argue that this view of the a priori ought to be adopted; and this, so far as I can see, he has not done.
I said at the outset that Brennan has rendered a great service to supporters of a priori economics, and the service I have in mind is this. He has suggested a very good way of evaluating a priori claims about action. If there are apparent counterinstances to such claims, and the person advancing the claim then responds by committing the No True Scotsman Fallacy, one has good reason to think that the claim has failed. This test is in no way dependent on accepting Brennan’s views about a priori knowledge.
Judged by this test, what can we say about the a priori status of Austrian economics? Do Austrian economists respond to counterexamples by resort to the No True Scotsman Fallacy? When Austrian economists say, e.g., that the parties to an exchange would not engage in it unless each expected to benefit from it, is it the case that “exchange” and “benefit” are being used in special senses, to immunize the statement from counterexamples? I do not think so. Rather, these words are being used in a perfectly ordinary way, and the Austrian claim is that, understood this way, the statement about exchange is known a priori to be true. What is supposed to be wrong with this Austrian claim? Again, when Mises says, “There is no ratiocinative operation which could lead from the valuation of a definite quantity or number of things to the determination of the value of a greater or smaller number,” he is making neither an empirical claim nor using words in special senses. He is stating an a priori truth about valuation, a process in the real world.
Brennan has given us no reason to think that Austrian economists commit the No True Scotsman Fallacy. Much less has he justified the broader thesis that pursuit of a priori knowledge of the actual world is a futile quest.
The most important of these posts is here: http://bleedingheartlibertarians.com/2013/10/extreme-austrian-apriorism-as-the-no-true-scotsman-fallacy/
 Human Action, Scholar’s Edition (Auburn, Ala.: Mises Institute, 1998), p. 20. Many neoclassical economists have stricter criteria for rationality than Mises does. The stricter criteria are sometimes used to show that consumers and producers in the free market act irrationally, and the state is called in to remedy these alleged problems.
Ibid., p. 122.