October 2021

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
Blog powered by Typepad

« Your Hosts (minus one) | Main | We're Number 38!! »


Feed You can follow this conversation by subscribing to the comment feed for this post.

To reply to Roger Koppl's response to Paul Lewis: Roger quotes Don as if he *rejected* the actor's level of meaning and intersubjectivity. He's "puzzled by this."

No puzzle: Go back to Collingwood's detective. I was so motivated this morning that I read about 80 pages of Collingwood surrounding that example.
Collingwood -- which Don loved, lectured from C's *text*, and insisted we read further, makes NO denial of the historical actor's meaning or intentions. But as historians, both Don and Collingwood argue, WE MUST DO MORE. Not only try to reconstruct the *relevant* intentions and meaning of the actor, but to also bring our own "detective-like" understanding of the outcomes that followed *after* the actor was *formulating* and *beginning* to act upon in context. We know more about the consequences than the actor could have ever known (like us) ex ante. History is not copy-and-paste reconstruction of meanings and actions.

Don took this as pointing toward (not as fully consistent with) Gadamer's notion of fusing horizons. Collingwood might have rejected such an extension. Don -- being consistent with the "fusion of horizons" notion -- would not say that Collingwood would've agreed with this interpretation. Don was not creating a biographical reading of Collingwood, but instead was pointing to earlier developments in philosophy that might be found *meaningful* for our generation of scholars in light of hermeneutical and other developments.

I would put it this way to Roger: YES, we need the Schutzian context, which, YES, is not solipsistic because of his recognition of the importance of intersubjectivity. BUT, furthermore, the historian, the reconstructor, is also part of that intersubjectivity in the extended present. The historian can try to bracket large, irrelevant portions of his own intersubjectvity (of, say, his total "life horizon"), but he will use other portions of his own intersubjective understanding of the world -- which are contemporary and well beyond the actor under study -- to draw out a greater meaning of the actor's choices, decisions, and so on. That "intersubjectivity," by the way, might include (as a merely relatable example) our present understanding of monetary consequences that are much more "developed" than an agents of the 1800s. And that present (though restricted) horizon -- that intersubjectivity which could not have been part of the 1800 agent's intersubjective life world -- might be layed over or fused with or whatever metaphor is best onto our historical understanding of the agent.

Again, Collingwood didn't know Bernstein, are Weinsheimer, or Gadamer... not his lifeworld, but their claims might have enough value to not be bracketed for certain types of analysis. This is not, however, any call at all for *rejecting* the lifeworld of the agent (even Collingwood), even if large reaches of the agent's lifeworld and intersubjectivity might not only be ignored, but even in principle unknowable!

Don said this was similar to Mises's conception/understanding notion, but Don was troubled by both conception as pure praxeology (with historical adjustments over assumptions depending upon historical context of the actor), and also with the dichotomy between conception and understanding. And so, he pointed beyond Mises. He found comfort in that he emerging literature esp. in the philosophy of the *social* sciences was making similar arguments, which further said that one need not fall victim to a relativism. While, then, the harder-core Misesians were charging Don with relativism and nihilism, they did so because they thought that only pure praxeology can save us from the brink. The contemporary literature acknowledged that there is something like a brink, a horrendous relativism out there, but they suggested we need not fall victim to it. This was Don's argument, too.


I'm not sure I have an opinion on what might be "the best" argument, but I would like to push a Hayekian angle that may not be adequately appreciated even in our circles. The argument is completely parallel to Austrian arguments against socialism and interventionism. I'll start from the POV of my intellectual opponents and show that it doesn't really get them to the place they want to go.

Let us approach social science, therefore, from the most radically external and scientific perspective possible. For this purpose we adopt a totally mechanistic view of human action. At some fundamental level it's all "mechanistic" cause and effect. We observe, however, that similar stimuli sometimes produce dissimilar actions and dissimilar stimuli sometimes produce similar actions. To understand why we must examine the mechanisms linking stimulus to response, something we should do in any event. Those mechanisms turn out to be the workings of the central nervous system, which maps the environment and provides the agent an implicit model of its world. (Implicit because we are taking a scrupulously external perspective by argumentative strategy.)

We can't make much progress, then, in our mechanistic account of human action until we account for these cognitive mechanisms. Our attempt to do so, however, quickly leads us to an appreciation of their great complexity relative to the classificatory apparatus with which we are attempting to understand them, namely, the human mind. Because the mind is not more complex than itself, it is a complex object of inquiry. It is complex in a rather specific sense: A simple Cantorian "diagonal argument" (http://en.wikipedia.org/wiki/Cantor's_diagonal_argument) shows that we cannot correctly predict a human's responses to the stimuli it receives if stimulus and response are both described in physical language. We therefor lose information when we attempt to describe human action in purely "scientific" physical terms. We must rely on "understanding" in precisely the German-language hermeneutic sense of Dilthey and others. Our radically empiricist exercise leads us to an unexpected *defense* of "Verstehen" and hermeneutics.

That's pure TSO, of course, but I think the structure of Hayek's argument is not well appreciated. If we took Hayek seriously, we would look to complexity theory, computable economics, and such as forms of argument that help us to make the case for "humanistic" economics. In this sense I completely favor the *integration* of "scientific" and "humanistic" economics. Unfortunately, most economists like "science" and dislike "humanism" or vice versa.

This view can today no longer be sustained in that way. At some point in history those who used certain kinds of mathematical tools were also the interventionists, statists etc. - so some people started thinking there is an essential or necessary connection between the two - but that need no longer be true. We should separate the math character of methods from ideological content. A good example is provided by developments such as agent-based computational economics, which is very mathematical but very Hayekian, evolutionary etc. at the same time...

Ideology has absolutely NOTHING to do with it, it is a methodological point about the appropriateness or inappropriateness of the methods chosen to the task.

The Austrian objection to both historicism and scientism are epistemological objections, not ideological ones.


It certainly could be true that I've misread Don on this point, but then I would feel justified in complaining that he sure made it hard for even a sympathetic reader to see that we was not somehow diminishing the idea of meant meaning. I think he tended to stretch the meaning of "meaning" too far, which made it hard to see which distinctions he was happy with and which he was effacing.

Personally, I parse it out like so: Individual actions have meant meanings and may be "interpreted." Unintended consequences are meaningless precisely because they were unintended. The catallaxy has no meaning. Thus "Verstehen" applies at the one level, and "mechanistic" causality applies at the other level.

I certainly agree with you that ideology _should_ nothing have to do with it; that´s exactly my point. But I believe the objections, if there are any, should be of an economic-scientific character. The epistemological objections you are referring to leave that matter largely undecided. They are at best of the order of preliminary generalities, preliminary phenomenology etc. Science is largely pragmatic and there is no problem with that; scientists should be allowed to be eclectic in the choice of their methods. In fact many Austrians themselves illustrate this fact. In their "epistemological" pronouncements they are very much against this, against that. But as soon as they start doing some more serious work, they cannot help using some game theory, some statistics etc.

From the trenches_Austrian economics and its Methodological Dualism in particular have worked just fine analyzing our modern political economic travail.

Anyone applying Austrian theory to the Greenspan era and beyond has no problem seeing that the US is the road to perdition.

This is an area of Austrian economics of which I have little knowledge --- Mises' books on epistemology and verstehen never quite did it for me, in part because there is not much of audience for these ideas outside a select group of scholars.

But here is a paradox that I don't believe has been addressed in the literature: the existence of verstehen (interpretive understanding of purposive human action) and the subjectivist conception of cost (opportunities forgone). These two ideas cannot simultaneously be advocated by the subjectivist because no observer can access the sacrificed alternatives of an individual decision-maker. So if the purpose of verstehen is to make human action intelligible, but we as analysts are prevented from *apprehending* the costs of those actions, then an understanding of human action and the plans that guide it would seem to be impossible. James Buchanan is explicit about this: we can NEVER know what alternatives the decision-maker considered before following through with an action. Consequently, an assessment of the performance (ex post) of any action does one no good in trying to make sense of the plans that undergird all human action.

I wrote a paper on this subject in a course I took with Lawrence H. White and would be happy to email it to anyone interested in reading it. These ideas occurred to me when reading Lachmann's work because he simultaneously maintained the validity of verstehen and subjective (opportunity) cost --- this is logically incosistent.

To DPryhcitko: How is this for a sweeping statement ---- A true subjectivist cannot be a Misesian!

As far as I can see, the US are great and will stay so for quite a while...


I think your position results from a confusion between the knowledge of the agents (subjectivism and opportunity costs) and the knowledge of the scientist (verstehen). There is no contradiction.

One could also discuss this in terms of first level and second level verstehen (and draw on the work of Alfred Schutz in making this distinction). But perhaps your argument demonstrates that these distinctions fade.

Please send me your paper. What was Prof. White's reaction?



As I told my 101 (!) class yesterday:

(1) I cannot know what your opportunity cost is unless I ask. I.e., I see you're here, but I don't yet know what you would've done had you not come to class today.

(2) Even if you inform me that you would have slept, as your next-best good, I could not know the "value" or "utility" or "well-being" that that would have provided you -- subjective.

(3) And even more: you, the student, could not know the "value" or "utility" or "well-being" that that would actually provide you because that choice, that trading-off of sleep, was based on your expectation, and so the expected benefit of sleep can never be compared against what actually would have occured. [No ex post, only ex ante]

Matt, yes, Buchanan is really good here. I teach this even to my EC 101 students. Vaughn (so little mentioned on this blog) also has much to say. Indeed, both Buchanan and Vaughn's discussions have much to say about the nature of opp. costs and subjectivism in a world of general equilibrium and a world of processes of incessant change.

I'm not sure a Misesian would disagree with your claim that one cannot be Misesian and (this kind of?) subjectivist. Just wait for them to respond.

I must add, however, that I really don't have time to care anymore.


DPrychitko has done a good job at summarizing the main points of Vaughn and Buchanan on opportunity cost (Vaughn "Does it matter that costs are subjective"; Buchanan "Cost and Choice") --- and I agree with Prychitko in that I would like to see Vaughn's work discussed more on this blog. Every new student of Austrian economics should be directed to read her 1994 book and pay close attention to her criticisms of Mises, Rothbard and Kirzner.

The most important implication of *this kind* of opportunity cost for verstehen is that costs are fundamentally *subjective*. Here is Buchanan:

"[costs exist] in the mind of the decision-maker and nowhere else. ... [They] cannot be measured by someone other than the decision-maker because there is no way that subjective experience can be directly oberseved."

That is the basic argument I put forward against the proponents of the verstehen doctrine --- Opportunity cost prevents at the very outset the theoretical validity of interpretive understanding.

Now to Dr. Boettke. I am unfamiliar with Shutz's work, but I would argue that it makes no sense to make the distinction between agent and scientist unless you can prove that the scientist can access the opportunities forgone by individual agents. I don't think he can (and I got Vaughn and Buchanan to support me).

Now Professor White's reaction, as I recall, was that we can just simply ask the agent what he sacrificed in performing his action and determine cost that way. It was here that I responded in a similar way to DPrychitko's #3 above. Alternatives are based on expectations, and once a decision is made, no cost comparison is possible. Buchanan is great on that too.

--- Prychitko, I am sorry to hear that you are too tired to care about this stuff anymore because I think you are one of the most insightful Austrians working today. (I had my library order your collection of essays and am reading through them now.) I hope you stick around.

Oh, and for anyone interested, here is a link to a paper I delivered at the ASC last March on this topic. I am not sure if the link still works, but I hope it does. I am the last speaker.



I'm completely stumped by this claim that verstehen is bogus if we can't "know" opportunity costs. "Verstehen" is the human understanding of human meanings. You wave your hand in front of me and I have to interpret the meaning. Are you greeting me? Are you threatening me? Are you greeting me by pretending, in a jovial way, to threaten me? Are you swatting away gnats? Are you even in control of those movements? There is an indefinitely long list of possible interpretations of your hand's movement. If an angel swooped down from heaven with the magically complete list of every different thing you might have meant by the gesture and I picked one at random, the probability of picking the right one is infinitesimal. And yet, in reality we get it right all the time. Errors are waaaay less frequent than probabilities would imply. *That*, my friend, is "Verstehen." Without Verstehen, social cooperation would be truly impossible. We are good at guessing the meanings in one another's words and deeds.

Independently of that, there is an academic theory of action that says we minimize opportunity cost. Strictly speaking, such costs are purely subjective. Yep. But I can still say things like "Raising the tax rate tends to lower the tax base." If we had zero good guesses about what opportunity costs look like, we wouldn't get that much useable knowledge our of economics. Some, I suppose, but not much. Thanks in part to "Verstehen" our guesses about other people's opportunity costs are good enough to make economics quite useful in understanding the world.

Where is the contradiction between Verstehen and subjective opportunity cost?


But you don't have Buchanan and Vaughn to support your position. They actually support the position I am laying out. You can have an objective science of subjective phenomena. As Buchanan has said repeatedly, Yes, Virginia there is a science of economics.

I think you should read Roger Koppl's comments VERY carefully here, he pretty much hits the nail on the head.

BTW, the point about the distinction between the knowledge of the actors in the economy and the knowledge of the scientist studying the economy was originally made by Hayek in the 1920s and 1930s. Verstehen is not a theory of control, if it was, the sort of criticisms you are raising WOULD be devastating (that is the point of Buchanan and Vaughn's critique of cost/benefit analysis and planning and regulation). Verstehen is an approach to understanding of the human condition.

One last thing --- this distinction is important in the work of Gadamer, who contrast the sciences of problem/solution, with those of questions/answers.

To the readers of this blog --- I am still waiting to debate the issue of the UNIQUE Austrian methodological position in the sciences of man (that they must avoid historicism and scientism simultaneously) because to succumb to either is to in fact undermine understanding of the human condition.

One of Lavoie's best essays was titled Between Institutionalism and Formalism to capture this idea.


Let me just second Pete's endorsement of Roger's comment here. THAT is what verstehen is.


Just to clarify, pretend I'm dense for a moment and please give your one to two sentence definitions or examples of scientism and historicism.

The more I think about this thread the more I start to see a similar structure in the history of thought surrounding criminal justice and criminological theory (though you were hoping for econ, or more broad examples of human sciences?).

I'm working on a paper for Advances where I plan to hijack your "science of control" language. I think the Beckerian models of crime and punishment fell easily into cj policy makers' quests for a science of control, while the psychological perspective of criminology and its accompanying rehabilitation practices loosely fit the bill as historicism. Look you call those people rational they cant be.

Ideally I want to present the work begun in my dissertation as constructing a new alternative framework - an Austrian political economy of criminal justice. By recognizing markets as a social epistemic process, we can understand a system of individual cooperation whereby individuals express preferences for punishment qualities and quantities by buying and selling security, legal and punishment services. Rather than constructing a rationale of punishment through philosophical deduction, social preferences for justice and fairness emerge from the bottom up.

The historicism psychology line might be a stretch??? But I am curious to hear your clarification.



First before definitions, let me say that I think you are (a) correct, and (b) that a good place to start is Buchanan's essay in the Rowley edited volume on the Origins of Law and Economics, where is lays out his objections to cost-benefit analysis in law and economics. I would also look at his discussion in Good Economics, Bad Law in Freedom in Constitutional Contract.

Second on the definitions --- historicism --- meaning that all we can hope for in the cultural sciences is historical description of particulars, not time invariant principles are in operation, and scientism --- the belief that there is only 1 true method of science and what is done in physics and the applied branches of physics must be copied in the sciences of man if they are to mature to be real sciences as opposed to pseudo sciences.

I do think that the idea of economics as social physics with the applied doctrine of social engineering is what is behind the idea of the economics of control. This an animal that is unrecognizable as the discipline of political economy for those who practice economics as a philosophical science.

Does that make sense (it is late)?

I doubt whether the unqualified scientism critique à la Hayek still applies to today´s entire mainstream economics. Of course mainstream economists still have that stance of superiority typical of the "École Polytechnique" mentality which Hayek described so well in The Counter-Revolution of Science; but note that this is largely justified: mainstream ecnomists ARE far superior qua technical sophistication. In the meantime Austrians go on criticizing the mainstream as if we hadn´t witnessed disequilibrium economics, asymmetric information economics, agent-based modelling etc.


Not sure I understand the reference to Austrians ... every Austrian that is working seriously in economics certainly knows about disequilibrium economics, asymmetric information, and agent-based modeling. Those of us, like me, who has acknowledged these developments, but have also argued that they fail to capture the essential point of the Austrian argument do not make the claim out of ignorance, but instead out of a profound sense of discomfort about the ability of those methods to capture the essential points.

What I am curious about is precisely who within the Austrian camp you are pointing to who supposedly doesn't know that economists have tried to build disequilibrium models, asymmetric information models, or agent-based models?

Dr. Boettke is right about the Austrian acknowledgment, but not acceptance, of these recent developments in economic theory. But what is not so clear are the essential points that these models are failing to capture or address? It seems to me that Austrians can respond in two ways:

(1) Failing to capture the role of the entrepreneur in coping with these problems, and thus ultimately eliminating them; or

(2) Failing to capture the essence of uncertainty which makes the problem more serious than one of simple asymmetric information and bounded rationality. The problem now is one of fundamental uncertainty in which Reason is absent.

I think Austrians should be emphasizing the second point because to invoke the entrepreneur is to dismiss these problems as unimportant. Yes, there is a process whereby these problems are gradually eliminated, and perhaps never eliminated, but instead recurrent; but to emphasize the importance of uncertainty is to take the analysis further than one of optimization in response to asymmetric information. Boundedly rational agents are admittedly imperfect, but they are perfect in responding to these imperfections. But is this good enough?

The comments to this entry are closed.

Our Books