My take on Austrianism is almost precisely the same as yours, minus all the complicated math.

The more I study philosophy, the less impressed I become with a priori axioms. Although Cogito, Ergo Sum is on pretty solid ground, it still leaves open many unanswered (and perhaps unanswerable) questions. What is I? Is I a singular entity? Something appears to be thinking. But what is thinking? Is thinking necessarily a conscious process? Or can the illusion of thinking be replicated in a purely determinist world?

Truth, and knowledge of truth, exists.

I'm not sure what "truth" or "knowledge of truth" means in this context. Can knowledge of truth exist without a knower? What if that knower is always wrong about his knowledge of truth?

Conscious humans act.

This seems to imply that humans are intentional actors, capable of exercizing free will. But what if fatalism or hard determinism is true? Wouldn't those arguing for this axiom first have to reject fatalism or hard determinism before making such a claim?

Do you realize how incredibly dorky "I’ve been thinking a lot about Austrian economics lately" sounds? Awesome! ;)

In no subject is the ideal (or at least the formal) more frequently and deeply confused with the actual than in economics. Be that as it may, I think your battle may already be substantially won: two years ago Daniel Kahneman won a Nobel Prize in economics for using psychological studies to demonstrate, for example, that most people are more afraid of losing a certain amount of money than they are desirous to gain a equal amount of money; i.e. they are more attached to what they possess already than what they do not, and hence discriminate non-logically in the evaluation of utility on this basis, among others. So I think that rationlism may be slowly expunging itself from economics at the present, much as it did in philosophy after Bradley. Nevertheless, in defense of rationalism, it should perhaps be noted that Descartes' fundamental postulate, for example, is basically an empirical supposition, and Descartes himself severely reprimanded other philosophers who interpreted it as a logical deduction rather than as a datum. These categories, in other words, should not be taken too rigidly: I remember when Kuhn attended a conference about the merits of Kuhnian epistemology and stated: "I am not a Kuhnian." But as I firmly believe, the greatest thinkers have ever been the most convincing skeptics, and even if one doubts of the applicability of logic to human behavior (which could not possibly be other than partial in any case), one must still temper to some extent the impulse to induce from experience.

On a similar note, Keynes was not a Keynesian and Marx was not a Marxist.

even if one doubts of the applicability of logic to human behavior (which could not possibly be other than partial in any case), one must still temper to some extent the impulse to induce from experience.

Right. Although the post contains only critiques, I tried to include enough caveats to make it clear that I'm not completely trashing the Austrian approach. Pure empiricism/induction isn't the right way to do things, either.

"...However, it applies only in the case where there is a “supply of a good”, defined to be a homogeneous group of units “equally capable of rendering the same service to the actor”. Hence, in the example, the six horses would be a “supply” if they were all interchangeable. If one follows the deduction all the way through, though, one must conclude that there really is no such thing as a “supply of a good”. No two goods are completely interchangeable or homogeneous; to act or think as if they were is often useful, but does not follow logically...."

What you've overlooked is that the first sentence quoted above is effectively the definition of homogeneous, from the point of view of the actor. Absolute homogeneity is impossible, but also economically irrelevant.

If you want a bunch of marbles to serve as a game scoring system on the beach, positioning them as required, it makes no difference as to whether the marbles are different colors or different sizes, for THIS application. All of the marbles serve as a part of the supply and all are interchangeable. Every additional marble marginally increases the scope of the game that can be scored with the marbles at hand. At some point the cost of an additional marble is not justified by allowing an already large game to become larger.

Regards, Don

Don,

I agree with what you're saying, but my point was that the homogeneity of the marbles (and, hence, the applicability of the law of marginal utility to this scenario) stems from an intuitive understanding of how we think rather than from a deduction. What I'm saying is not that the law of marginal utility is invalid, but rather that one must be intuitive in order to be able to interpret it in any meaningful sense in the real world.

However, I admit that I'm not completely convinced that this objection is both relevant and non-trivial.

One last point I should mention in defense of rationalism: Clay, you criticize the "Austrians" (and I could surmise rationalists in general) for basing "logical" deductions from seemingly intuitive discrimination, but almost all of the rationalist philosophers that I have read are conscious (to varying states of alarm) of the fundamental disjuntion between intellectual concept and actual object or action. It should be noted, I think, that most of them, especially Descartes, considered the fact that their mental concepts were based on, and confirmed by, an intuitive correspondence with reality to be a strength of their way of thinking rather than a weakness. I think few of them would deny at least some difference in kind between the logical and the actual, or that all logical processes must begin with some sort of foundational assumption, intuitive or otherwise. However, this is often viewed as being far from fatal to the validitity of this system of thinking. Descartes, for example, actually thought that his clear intuition of God was a sufficient proof of God's existence, and considered God's existence to be a guarantee of the validity of his intutions (the Cartesian circle). This does not mean that such confidence is warranted, granted, but nonetheless I think that, although most students today are perhaps so grounded in an empirical mental framework that we cannot readily credit faith in intuition, this is not necessarily a self-evidently justified skepticism.

If I ever write on epistemology, I think you'll see that I'm actually a big fan of intuition.

shonk,

I wish you had broken up this post into multiple parts to make it easier to respond to. I will respond in parts.

For most of my life, I viewed economics as total bullshit. I saw people on TV talk about stuff that they themselves couldn't really explain or defend. They often argued with each other on fundamental issues. People who win the Nobel Prize in economics have 180 degree opposing views on the same topic. Further, I had seen what theories espoused by certain economists (Marx) had brought to the world. Most of the economics I had been exposed to (which wasn't much to be honest) was simply people making statements without having any 'deeper' justifications for them.

I like having formal systems of epistemology to study things. Intuition has its place, but for the most part, it's not very good at arriving at truths. Intuition tells most people today that if the economy goes bad, it's Bush fault. Or if kids started doing more drugs starting in 1992, it was Clinton's fault. Or if the economy booms in the mid-90's, it's Clinton's good work. But what if the policies of a given president have a lag time of say 4 years to their results? Or 12 years? Or what if the boom is actually not a good thing, but more like a cocaine high? Or what if the policies of a given president have nothing to do with the economy? How would you know, and how could you tell?

The formal system known as the scientific method was a great leap forward in epistemology. It's application to empirical knowledge is not perfect (i.e., can you really have a true control, how much statistical significance do you need, etc), but it is certainly better than anecdotal evidence. However, I realized that the scientific method cannot be applied to economics, due to the nature of man and due to the impossibility of doing controlled studies on a large scale. There is no way to have a controlled study to tell whether or not Clinton's policies resulted in the mid-90's boom. To do so, you would need two planet Earths which are alike in *every* way except in one, Clinton is president, and in the other, someone else is president, and you would need to be able to observe the results of both. Economics simply cannot be studied with the scientific method.

Yet, I knew there were other ways to obtain knowledge. Newton created differential calculus in just this way. He started with the axiom that the gravitational acceleration is constant (at a given location on the Earth), and used a formal system of mathematics to create equations which describe the motions of objects, thus giving rise to classical dynamics. There was no control needed. This was not the scientific method. Certainly, someone could have carried out a controlled experiment and dropped a boulder and a pebble to the ground from atop the leaning tower of Pisa to see that the velocity and acceleration attained was independent of mass, but with Calculus, there was no need to. Exact answers could be obtained on paper with a formal system which was not the scientific method. And for 200 years, this was the "truth", until the observed motions of very small objects and very fast objects didn't fit the equations.

My views on economics changed when I encountered a similar formal system on the internet on a message board (from Catallarchy's Don Lloyd). This system started with the fundamental axiom of human action and proceeded from there. It was the most rigorous defense of methodology in economics I had ever seen.

Just as Newton's classical dynamics equations were not perfect, i.e., they did not describe the motions of all objects, I don't think praxeology is perfect. In fact, no epistemology is perfect because it always depends on human reason, which is always imperfect. But different epistemologies work better for different subject matters, and I think the field that seeks to know how man makes choices toward his desired ends is best is best studied with praxeology.

I have said before that praxeology and neo-classical economics can sometimes be best thought of as different fields, although they overlap in topics of study.

But I am also one of those "Can't we all just get along?" people who is open to other ways of looking at things. If we put our heads together, we'll figure out more stuff.

Anyway, back to the axioms. My first complaint is that, although they are relatively basic propositions, none is really an “axiom” in the sense usually used by mathematicians and logicians. These aren’t mere starting points , they are the conclusions of deductions. For example, the first, which is just Descartes’ famous statement, is the conclusion of a chain of deduction, which itself must be based on earlier axioms. I make this point merely to point out that the use of the word “axiom” in this context is one that follows in the tradition of epistemology, but not in the tradition of formal logic.

When Mises talks about axioms, he talks about phenomena that cannot be reduced futher to their own origin:

Since time immemorial men have been eager to know the prime mover, the cause of all being and of all change, the ultimate substance from which everything stems and which is the cause of itself. Science is more modest. It is aware of the limits of the human mind and of the human search for knowledge. It aims at tracing back every phenomenon to its cause. But it realizes that these endeavors must necessarily strike against insurmountable walls. There are phenomena which cannot be analyzed and traced back to other phenomena. They are the ultimate given. The progress of scientific research may succeed in demonstrating that something previously considered as an ultimate given can be reduced to components. But there will always be some irreducible and unanalyzable phenomena, some ultimate given.

Re: circularity of arguments. A lot of the arguments do appear circular, but that is because of the nature of the epistemology. Irrefutable logic is like that. The statement 1+1=2 is irrefutable because any empirical observation that one plus one is not true implies an error in observation. One could object, "But that depends on what 1 is. You cannot say that 1+1=2 if you don't know what one is, because that is a circular argument" but that gets back to the basic axiom of unity. Given that axiom, the statement 1+1 is irrefutable.

Re: intuition vs. deduction. Human Action is much more rigorous in the steps of deduction taken from the action axiom. MES is the book I recommend anyone serious about Austrian Econ to read (after reading _Economics for Real People_ by Callahan) because it is an 'intermediate' level text that is much more fluid and accesible to the average joe, but if you want a rigorous defense of the methodology, and a better philosophical underpinning, HA is the better book.

Similarly, the statement that “there is never any possibility of measuring increases or decreases in happiness or satisfaction” (pg. 15) seems similarly intuitive. So far as I can tell, this does not follow from the Axiom of Action (or any of the other listed axioms); rather it seems to be a pragmatic, almost empirical observation. Rothbard’s treatment of scales of values is nice (though we’ll get to this), but the incomparability of values and happiness does not flow from this. Instead, this appears to be an ad hoc observation; from our experience with people, we know that it isn’t realistic to think that the utility that I derive from, say, a good meal is comparable in any meaningful way to the utility that a friend derives from drinking a good stout.

You're right, in a sense that does seem very intuitive rather than deductional. However, Mises does offer a defense of this as rooted in methodological individualism. A person's own value scale is unknowable to an outside observer except as action undertaken to pursue an end, i.e., action reveals preference. In any meaningful sense, the only thing we can know is that they preferred one action to another because they actually did it, and thus, ordinal rankings are all that are relevant in the context of what an outside observer can know about a person's preferences. I submit that this is more deduction from the axiom that an individual takes action to pursue his mostly highly desired state of satisfaction than it is intuition.

No two goods are completely interchangeable or homogeneous; to act or think as if they were is often useful, but does not follow logically.

They are interchangeable if viewed in the praxeological context that means are valued by what ends they help pursue.

Especially since the valuations of any particular good are subjective, it would actually seem to fly in the face of the logical framework to suppose that any two goods should even have the same label, let alone be considered homogeneous or interchangeable. The fact that we think of, say, horses as being relatively homogeneous is the result of a sort of intuitive shorthand, one that is usually useful and rarely leads to difficulties, but not one necessitated, or even condoned, by pure logic.

I believe you are missing one of the key points of utility, as given by praxeology. No object has utility on its own. It only has utility because it helps an individual attain a specific end-state. Thus, any two objects that help achieve the same end are effectively interchangable units. If my desired end is to quench my thirst, then a coke and a water are effectively interchangable units of the same good - a 'thirst quencher'. If my desired end is to drink in calories, a coke is a different good from water. These 'lumpings' and 'splittings' of different objects into the same and different categories of goods is very much a real world phenomenon.

This is one of the main objections Bryan Caplan has to Austrian methodology - his support of indifferent states - and I think he's completely wrong.

Another problem I have with the Law of Marginal Utility is that, as presented, it seems useful, but that’s really only because the examples are so easy. Obviously, in the case of six interchangeable horses doing six different jobs, applying the law is easy. But what if those six horses are doing twenty different jobs? Then each is doing multiple jobs, perhaps in pieces. When the man has to give up one horse, it isn’t necessarily just a matter of cutting the three lowest-rated jobs; instead, the man must rate the combinations of jobs and choose to cut the lowest-rated combination that can be done by five horses.

Again, all goods are value based on the ends they help achieve. The fact that a horse must be given up means the owner has to rank the end-states of his own satisfaction that result, whether they each perform one job or ten. This objection is, I think, your weakest in the War and Peace post.

To consider a simple example, perhaps there are three ends that I value as follows: I value A over B and B over C. There are cases where this may not be transitive (i.e. I may actually prefer C over A), but we need not get into that case. Instead, simply consider the case where I must cut some of my factors of production, meaning I can no longer accomplish all three ends. Suppose also that, although, individually, I rate A over both B and C, I prefer the combination of B and C to A alone. Hence, if I can still accomplish both B and C with my remaining goods, it will turn out that A is actually the relevant marginal utility of the supply, even though it occurs at the top of my scale of preferences.

Again, the relevant question is not, "What do I prefer - A, B, or C?" but rather, "What combination of A, B, and C do I prefer?"

In other words, for any given scale of preferences, I must have a meta-scale of preferences, rating each of the combinations of preferences. This, then, is the relevant scale for consideration in the Law of Marginal Utility. Of course, logically, there is no reason to stop at this level. I would also have a meta-meta-scale, and a meta-meta-meta-scale and so on. Since at each level I am strictly increasing the cardinality of the list, this presents a serious problem. In the basic example just explained, I have three things, A, B and C, on the original scale of preferences, but 6 combinations on the meta-scale, and then 720 meta-combinations on the meta-meta-scale.

Your logic is correct, but as above, this does not refute the Law of Marginal Utility, because the 'meta' part is already intrinsic to end-states.

This, ultimately, is the point I hope the reader draws from this little exposition: Austrian deduction depends fundamentally on intuitive, inductive and even empirical reasoning in order to arrive at its conclusions. It only provides meaningful results because it is not strictly deductive. As I’ve tried to show, even simple results have this dependence. This is easy to overlook, as our own thoughts are shaped so strongly by intuition, but it is, to my mind, an almost devastating critique of the formal deduction of the Austrian methodology.

You're right that perhaps intuition cannot be completely quashed from a purely deductive epistemology. However, I submit that an empirical methodology is much more susceptible to intuitional fallacies than praxeology.

You’re right that perhaps intuition cannot be completely quashed from a purely deductive epistemology. However, I submit that an empirical methodology is much more susceptible to intuitional fallacies than praxeology.

Yes, and this is a point I wish I had made more clear. My goal in writing this thing up was twofold: to provide an accurate account of objections that were raised in my mind (regardless of how pertinent or serious they are) and to provide evidence to support the contention that Austrian methodology isn't purely deductive. It was not and is not my intention to say that any other system is necessarily better or more reasonable.

Let me put it this way: I'm not saying that the conclusions of the Austrians are wrong; all I'm saying is that the Austrian _conceit_ is wrong. That is to say, I think it's a fallacy to say that Austrian economics (or praxeology) is _purely deductive_. Which, again, I want to emphasize is not necessarily a bad thing -- when the intuitive leaps taken are reasonable (as they seem to be), you conclusions arrived at thereby remain valid (at least if one accepts intuition as a means of arriving at truth, which I do).

The last sentence above should read "the conclusions...", not "you conclusions...".

My final thought on the subject: I don't think Clay's criticism is only methodological, or at least it need not be. In other words, it is not simply that deductive or rationalistic economists arrive at valid, substantially correct conclusions by means other than the logical principles they claim to be using, but rather that a faulty methodology often leads them into error. If economic predictions were always correct, then the methodology behind them would be rather irrelevant (at least in my opinion). But they are not, and it is to investigate the source of error that I believe this whole discussion began. So while Jonathan may be correct in contending that it is better to have certain deductive, logical principles than to proceed wholly by induction, like Diderot's blind man groping in the dark, it may still be true that the flaws in the deductive economic analytical system under consideration derive from an insufficient sensitivity to the non-logically-derived bases of many of the fundamental assumptions of this system. For just as many economic decisions are compound rather than pure either/or propositions, so too are the modes by which we think about them probably inevitably compound notions composed of different elements, rather than pure methodologies.