Kant and His Categories Versus Mises and His Praxeology

Article of the Week from Fixing the Economists

by Philip Pilkington

You won’t see me dealing with the Austrian School of economics much on this blog. I wrote a long essay on them before and I think it says almost as much as needs to be said; namely, that the Austrians are just neoclassicals who substituted Faith for mathematics in their search for economic truth.

Insofar as practical differences between the Austrians and the neoclassicals it is simply a matter of ideology and extremity — not to mention a characteristic incoherence and general sense of confusion on the Austrian side. Take their business-cycle theory, for example. It is actually a Wicksellian business-cycle theory pushed to extremes. The Austrians claim that there is a natural rate of interest and that if the money-rate of interest (the central bank rate, basically) diverges from this inflation and deflation will result. Well, that’s basically the mainstream theory today used by most central banks who use a New Keynesian model with a Taylor Rule attached that essentially contains an implicit idea of a natural rate. The central banks then try to line their target rate of interest up with the so-called natural rate. (My criticisms of the New Keynesian position also apply to the Austrian position in this regard, by the way).

The Austrians are theoretically orthodox. The only reason they fall outside the mainstream is because they tend to be political extremists and — I’m going to dare to say it — they are usually not very thorough or fastidious thinkers. You can be a first-rate Austrian where otherwise you would be a second-rate neoclassical.

Today, however, I want to deal with a point regarding Austrian theory that Lord Keynes raised on his blog yesterday as it touches on a philosophical issue that I think is very important in economics more generally. This is the relationship between Mises theories of what he calls human action and Kant’s theory of a prioris and categories. First of all, let’s just let Kant himself get a word in on his a prioris and categories:

The a priori conditions of a possible experience in general are at the same time conditions of the possibility of objects of experience. Now I maintain that the categories, above cited, are nothing but the conditions of thought in a possible experience, just as space and time are the conditions of intuition for that same experience. They are fundamental concepts by which we think objects in general for appearances, and have therefore a priori objective validity.

A prioris and categories for Kant then are the conditions of the basis of our experience. A category in the Kantian system has the same status as our lived experience of time and space in that it is needed in order to comprehend reality (this is important, we will return to this in a moment). Now, what Mises is trying to do in his Human Action and his doctrines of so-called Praxeology is to derive something similar to Kant’s categories but which rather than form a basis for understanding the conditions that allow thought to be possible, they set out the conditions that allow action to be possible.

If that sounds a bit weird that shouldn’t surprise you. It should sound weird. Because it is weird. Deeply weird. Saying that the experience of time and space are necessary for our intuition is very different from saying that we can crystallise out conditions on which our actions rely. The former, as we shall see, is problematic. The latter is downright madness.

Kant’s was an epistemological project. The idea was to try to figure out the base-determinants of our thinking. Mises’ was more so a neurosis; an obsessive search for justifying the “correct” manner in which people should act. Actually, we can call such a project by its name: it was an attempt to form a rigid and codified doctrine of ethics and morals. I hesitate to call this moral philosophy because that’s not what it was at all. Moral philosophy tries to hit at Universal principles by which to evaluate Particular choices we must make. What Mises was doing was more so laying out a guide to life. Again, we can and should call this what it was: Mises was trying to write a Bible.

But back to Kant for a moment. What was fundamentally wrong with his project? Simple: it was too rigid. The question of time and space is interesting in this regard. Kant’s ideas about time and space, relying as they did on Newtonian physics, were basically overturned by Einstein and his relativity theory. Kant enthusiasts claim that it is unfair to criticise him on this; after all, they say, how could he have known? Well, actually the philosophical issues that Einstein’s theory raised were already there in George Berkeley’s De Motu. Naturally, Berkeley was (and is) usually ignored in these debates as parts of his theories were picked up and vulgarised by David Hume who then suppressed all the really interesting questions that these questions raised.

What does this tell us about the weakness of Kant’s approach? Again, it is all about its rigidity. Although we can break this down in a Berkeleyian manner and show that such rigid systems ultimately rely on what Berkeley called “occult qualities” — that is, overly abstract terms — it was, as I have shown before, Hamann that really nailed down the meta-problem with Kant and his ilk: it was their use of language that produced the rigidity. In trying to treat human language, which is the only real a priori that exists, as an overly precise tool it quickly becomes hard and fragile and loses all expressive power. It loses, in a very real way, its flexibility; and that is what is the key problem with Kant’s theory: it is in large part inflexible, trying always to subsume the New under its musty old categories and a prioris.

Now we can again return to Mises. If Kant’s project was doomed from the outset due to inflexibility what does this tell us about Mises’ project? Simple. It was complete and utter nonsense. Trying to stuff peoples’ actions and their reasons for doing various things into a tiny box of concepts is not only suffocating, it is borderline delusional. What Mises was really engaged in was the setting up of language traps that he then had people fall into. He tries to section off or limit the basis on which people can act and then purges everything else from his worldview. If an action doesn’t fit in with Mises schema it simply disappears.

Mises’ project was quite manifestly the product of a sick mind but it can tell us something about more mainstream theories. Neoclassical microeconomic theories, for example, seek to do something similar in a slightly less inflexible and dogmatic way, as I have pointed out elsewhere. And for this reason they too are doomed to fail. Human beings are infinitely mutable and changeable. There are very, very few constants under the sun when it comes to human behavior (the incest taboo being a notable exception) and the moment any constancy or regularity is found and articulated it is likely already on the way out.

Human beings are also extremely reflexive. Tell them that they act in one way and they’ll almost immediately start acting in the opposite way. This is not surprising given that the one doing the telling is just another human being and so the act of telling just an attempt to establish a relationship of authority — which is what Kant, Mises and the neoclassicals were and are ultimately trying to do. Human beings have a very ambivalent relationship toward authority and so the establishment of such relationships can only last so long before the system is shaken up. So our theories need to be flexible if they want to survive. And when I say “survive” I do not mean survive as a mosquito might in a glob of amber but rather survive in the sense that they can continue to grow and develop together with the events they seek to capture.