The philosopher Robert Nozick raises some important criticisms of Austrian economics in his “On Austrian Methodology.” This paper came out in 1970 and it is conveniently available in Nozick’s SocraticPuzzles (Harvard, 1998). I’m going to talk about a couple of the criticisms here. You might think that I don’t have to do this. The libertarian economist and theorist Walter Block wrote a long and impressive paper, “On Robert Nozick’s ‘On Austrian Methodology,'” that responded to each of the many arguments in Nozick’s paper. Why then go over this material again? My answer is that my responses differ from Walter Block’s. It’s not that he is wrong—he is usually right, but I have something else to say as well. My readers should by all means study his paper, which is available here.
First, some good news. Many critics of the Austrian school dismiss out of hand the notion of a science of human action that consists of necessary truths. They tell us that this is an outdated idea that must be replaced. Not Nozick. He says:
The idea of elaborating what is contained in the essence of human action is certainly an interesting and challenging one. Such a project, involving as it does synthetic necessary truths, would have been ruled out by the logical positivists as impossible and empty. But the positivist position and arguments on synthetic necessary statements, as much else within the positivist position (for example, the verifiability criterion of meaning) has fallen upon hard times. Recently, there has been a resurgence of interest in such statements, and it is fair to say, I think, that there are no arguments generally acknowledged to be compelling against the possibility of such synthetic necessary truths.
By no means, though, does Nozick accept praxeology. He tells us that “Actions are a certain type of behavior, namely, purposive behavior. Not all behavior is purposive.” What if there were some way of explaining human behavior that, when applied to economics, turned out to be better than praxeology?
It turns out that there is such a science that might do the job: the psychologist B.F. Skinner’s operant conditioning. “There is this alternative, relatively elaborate conceptual scheme which does not appear to talk of or within the Misesian categories, yet whether it is true of people or not appears to be an empirical question, to be decided a posteriori…if so, even if human action theory is known a priori, it may be that it is not the best theory, for the Skinnerian theory actually may explain more, enable us to predict more, and so forth.”
In brief, operant conditioning deals with rewards and punishments as a way to modify behavior, in a way that need not involve the subject’s purposeful action. A funny example comes from Skinner’s autobiography A Matter of Consequences:
On a scrap of paper I wrote “Watch [Erich] Fromm’s left hand. I am going to shape a chopping motion” and passed it down the table to [Halleck Hoffman]. Fromm was sitting directly across from the table and speaking mainly to me. I turned my chair slightly so that I could see him out of the corner of my eye. He gesticulated a great deal as he talked, and whenever his left hand came up, I looked straight at him. If he brought the hand down, I nodded and smiled. Within five minutes he was chopping the air so vigorously that his wristwatch kept slipping out over his hand.
Of course, Skinner was here himself acting. He was consciously trying to modify Fromm’s behavior. But this doesn’t show that Nozick is wrong. He isn’t claiming that there is no such thing as action. He also doesn’t claim that operant conditioning shows that Austrian economics is false. He is just asking whether a theory of economics based on operant conditioning might turn out to be better than praxeology.
The answer to Nozick is simple. It isn’t a good argument against Austrian economics that someone might come up with a science that made better predictions. You have to show us the science, so that it can be compared with praxeology. Suffice it to say that this hasn’t been done. The praxeologist isn’t required to come up with an a priori argument refuting economics based on operant conditioning. It’s enough to say, “Put up or shut up!”
Let’s turn to another issue. Like many people, Nozick doesn’t like adding to praxeology the assumption that labor has disutility. He says:
Some consumption takes time, and cannot be done simultaneously with some other activities; for example, listening to Beethoven quartets and working a steam drill, lolling on the beach and teaching a philosophy class. Therefore, a person with multiple desires, some of which cannot be satisfied simultaneously with the particular labor he does, will want leisure time in which to satisfy these other desires. . .We do not need the additional assumption that labor has disutility. Even if labor is a good, we can still have tradeoffs of labor and leisure; that is, tradeoffs with the other goods that can be had only in that person’s laboring time.
There’s an interesting twist to Nozick’s argument, as there usually is with him. He isn’t claiming, as some people do, that it’s an a priori truth that labor has disutility, so we don’t need to add it as an extra assumption in praxeology (I discussed this issue in an earlier article). He is claiming rather that labor’s disutility can be derived from other non–a priori truths, i.e., the existence of desirable activities that can’t be done while laboring and hence the need for tradeoffs between labor and the leisure to engage in these activities.
Nozick’s argument doesn’t work. Nozick is right that satisfying some desires isn’t consistent with laboring at the same time. In that case, someone who has these desires can satisfy them only if he has leisure. But it doesn’t follow that this lets you get rid of the assumption that labor has disutility. We still need this assumption to rule out the case of someone who always preferred more labor to one of the other desires.
Nozick made a good point against those who rule out the notion of an a priori science of action. He should have quit while he was ahead.