What sets Mises and Rothbard apart from nearly all other economists?

by Kwashiorkor last modified 2012-12-19T01:06:16+00:00
Why philosophy is important in the battle against bad economics.

(This is an exposition of section 1of Han-Herman Hoppe’s, Economic Science and the Austrian Method, available here.)

The Keynsian/liberal says, “Austrian economics is not worth considering, because it doesn’t use the scientific method, and doesn’t even bother to consider empirical data.”

The Austrian/libertarian says, “Complex equations, terabytes of data and computerized models are not needed to see that your system is on a track toward total economic collapse.”

What the opponents above are missing, and the reason they keep talking past one another, is that there’s something more fundamental about their differences than just theories of economics. There’s a philosophical difference, and if the libertarian is not aware of it, then he may be playing into the hand of his opponent.

Of course, the Keynsian/liberal is generally not aware of the philosophical aspect here.  He’s been raised to believe that there’s one way of determining what’s true (or at least the approach that people should use to determine a common course of action).  They’ll present it as such a common-sense approach that you’d have to be a nut to oppose it.  You just make a hypothesis, create a model to test it, and then evaluate the hypothesis based on the data that you collected.  That’s the empirical, scientific way.

But there’s more to knowledge than just using the scientific method (and I’m not talking about some kind of mystical means of transcending our consciousness, or innate knowledge, or intuition, etc.)

How do we know what we know? (This falls under the topic of epistemology, if you want to impress people.)

The philosopher Immanuel Kant had some important things to say about this. Unfortunately, he wrote in German using those multi-syllable words and long sentences that torture English readers. Kant said there was a two-fold way of looking at propositions.  (We may think of propositions as “units of truth” – declarative statements that are meaningful and can be judged as either true or false.)

The first way of looking at any proposition (according to Kant) was to consider if it could be shown to be true by the use of formal logic.  Many things can be proven by formal logic, but they tend to be simple things.  Even some apparently simple things require many long steps of formal logic to prove (like 1+1=2).

A positive example of this application of formal logic would be the sentence: “There are eight words in this English sentence.”  Formal logic is clearly sufficient to prove this to be true.  Kant called theseanalytic propositions.  Even if it’s a false proposition (e.g., “This sentence starts with the letter ‘X’.”), formal logic will still lead you to the correct answer (that the proposition was false).

A negative example of this would be propositions like:

  • “Paul Krugman is a smug asshole.”
  • “This statement cannot be proven true.”
  • “You would not regret coming to my place tonight.”


Formal logic would be useless for evaluating these statements. Kant called these synthetic propositions.  While the former analytic propositions were self-contained, the synthetic propositions tell us something new about the world.

Kant said that the second way of looking at propositions was to consider whether observations were necessary to establish their truth or at least confirm them.   Observations would help us with statements like, “Miranda Kerr has no scars on her body.” Observations would, however, NOT help us with statements like:

  • “A circle is made up of an infinite number of line segments.”
  • “Any conspiracy theory can be disproven by evidence.”
  • “She never really loved you.”


This is not to say that observations could not be applied to any proposition, just that the observations would not help us in determining its truth or falsehood.  We could count the number of sides on squares until doomsday, but all the statistics will be superfluous for “proving” the proposition that “All squares have four sides,” since that is the only number that a square could ever have, by definition.  We can have complete confidence that no square will ever be found with five or more sides.

Now, you’re free to imagine some bizarro, imaginary, fantastical universe where polygons have a variable number of sides, but I’ll leave you to that universe and the conclusions you may draw from it because:

  1. I’d be really surprised if it were free of any contradictions;
  2. It’s not a part of reality.


Getting back to Kant, he called these kinds of evaluations a posteriori (“from the latter,” meaning we can look back at them after performing our observations), or a priori (“from the earlier,” meaning no later observations are helpful).

To summarize, we have two ways of looking at propositions: whether formal logic is helpful or not, and whether observations are helpful or not.  Note that both considerations may be made of any proposition at the same time: any declarative statement may or may not be evaluated by formal logic, and that statement may or may not be proven or disproven by observations.

Students skip over all of this in today’s schools.  Somebody, somewhere, back when decided that formal logic was great if you wanted to study logic, but not really helpful to students in “the real world,” both due to its rigor and the nonsense about imaginary worlds where the rules don’t apply. All they’re left with is the use of observations, so they try to apply them to all cases.

Kant held that while certain simple statements of mathematics could be proven by formal logic, the vast majority of the propositions of mathematics are taken to be true without such formal proof. For example, the proposition from geometry that "a straight line is the shortest path between two points" is not subject to proof by formal logic, nor is it the result of a logical construction, yet it is still taken to be obviously true, and it is free from any internal contradiction.

Further, Kant held that mathematics and geometry are not aided by observations.  Finding all the other (infinite) paths a non-straight line will not help to prove that the straight line is the shortest path. He said that these propositions were held to be true, yet were not proven by formal logic, nor were they proven by observations.  They were synthetic and a priori.

(Note that this is unlike most laws of physics, which cannot be proven or disproven by logic, and rely entirely on observations.  For example, Coulomb’s law was determined entirely by observations, and it’s not even unreasonable to think that some future observation could result in a considerable change to the law, should extreme conditions be found elsewhere in the universe or one day be created in the lab. This not to imply that physics is any way unreliable, but rather to point out that the nature of knowledge about physics is of a different class.)

It was Mises who placed economics within this same class of propositions that could be true, and yet were not provable by formal logic or by observations. (This actually followed the trend of economic thought at the time, but it was Mises who first formalized and expounded on the thought.)

This is why most of those who are critical of Austrian economics think it’s irrational or insane to attempt to use anything outside of the “scientific method.”  They’re not even aware of acquisition of knowledge outside of analysis by observation.

It’s a philosophical difference, but they don’t know it.

Meanwhile, many libertarians are also unaware of this crucial difference.  Even many of those commonly referred to as Austrian economists are either unaware of this distinction or choose to downplay it so as to appear to be more reasonable. The result is that when we’re saying that Austrian economics is superior because it leads to economic advantages to more people, or that it is more just, or even that empirical data is inconclusive, or that forecasts are folly, we are arguing in non-Kantian turf. We’re basically agreeing with them knowledge is only gained by observation, but that our observations are somehow superior to theirs.

Mises and Rothbard are our icons who stand out from even other Austrian economists in their insistence that the legitimate realm of economics is both synthetic and a priori. Its propositions are not proven by nor constructed of formal logic, nor are they established by observations. They bring new knowledge about reality, and provided that they remain free of contradictions, may be relied on with a certainty no less than that given to mathematics.

If we miss the importance of this point, as many economists and libertarians have, we run the risk of being burying under the mountain of statistics that Keynesians love. The data may not actually prove their points, but the spectators don’t really know how to analyze the data anyway, they just let their judgment favor the side that appears to have the greater weight. The best way we have for discrediting this great mass is to undermine its validity from the beginning.  If economics does not fall into the class of things that can be proven by observations, then the mass of observations actually has no weight at all.

So, the first step in bridging this gap with our opponents is to get them to admit that there can be a kind of knowledge that doesn’t need to be confirmed by observations.  The next step is to show them that the propositions of Austrian economics are of that class.