Hayek at His Most Misesian

Steven Horwitz

Pete has often said that the best reading of Mises is a Hayekian one, and the best reading of Hayek is a Misesian one.  I completely agree, and I've spent energy over the years trying to convince people of the truth of the first half of that aphorism, mostly because it seems that self-proclaimed "Misesians" are more hostile to Hayek than Austrians more favorable to Hayek are to Mises.  But today I offer some evidence for the second half of the aphorism.

On Wednesday, my Austrian course covered Hayek's "The Facts of the Social Sciences" essay.  I think this is Hayek at his most Misesian, and it's worth noting that it was written 1942, well after the "Economics and Knowledge" essay that some argue was Hayek's split from Mises.  Consider the following passages from that essay and see if you think they aren't very Misesian.

Yet there seems to me to be no possible doubt that this [interpretation of the intentions of others] is exactly what we are doing, in ordinary life as well as in the social sciences, when we talk about other people's intelligible action, but that it is the only way in which we can ever "understand" what other people do;  and that, therefore, we must rely on this sort of reasoning whenever we discuss what we all know as specifically human or intelligible activities (62).

If that isn't Mises on "thymology," I don't know what is.  Compare this passage by Mises in Human Action (58):

Understanding is not a privilege of the historians.  It is everybody's business.  In observing the conditions of his environment everybody is a historian.  Everybody uses understanding in dealing with the uncertainty of future events to which he must adjust his own actions.

And is the following Mises or Hayek?'

The claim to which I have referred follows directly from this character of the first part of our task as a branch of applied logic.  But it sounds startling enough at first.  It is that we can derive from the knowledge of our own mind in an "a priori" or "deductive" or "analytic" fashion, an (at least in principle) exhaustive classification of all the possible forms of intelligible behavior.

It's Hayek from the same essay (pp. 67-8). Consider this too:

There can be no doubt that we all constantly act on the assumption that we can in this way interpret other people's actions on the analogy of our own mind and that in the great majority of instances this procedure works (64).

Compare to Mises, again from Human Action (p. 24)

But it is beyond doubt that the principle according to which an Ego deals with every human being as if the other were a thinking and acting being like himself has evidenced its usefulness both in mundane life and scientific research.  It cannot be denied that it works.

The footnote on the subsequent similar paragraph in Mises is to Alfred Schutz's The Phenomenology of the Social World and rightly so.

This essay, as much as any that Hayek wrote, shows the high degree of consistency between he and Mises when it came to the task of the social sciences and the appropriate methods for engaging in social scientific research. 

 

Comments

[Ryan Murphy]

I don't agree at all with the "Misesian" reading of this passage

"The claim to which I have referred follows directly from this character of the first part of our task as a branch of applied logic. But it sounds startling enough at first. It is that we can derive from the knowledge of our own mind in an "a priori" or "deductive" or "analytic" fashion, an (at least in principle) exhaustive classification of all the possible forms of intelligible behavior."

Hayek is foreshadowing his argument in The Sensory Order that one day science can hope to achieve a fully inclusive classification system of all possible phenomena. It is a purely "nominalist" statement that we can "a priori" have a classification system without looking into the world (I don't remember in which essay, but Hayek cites Popper in his belief in nominalism). For Hayek, these classifications hold no empirical content.

I've heard people make the argument that this is how Mises felt as well, but I don't see that in the text. Hayek made it clear on this issue (in one of those 1970s interviews, printed in Hayek on Hayek) that he never agreed with what he understood Mises's position to be on whether a priori belief formations hold empirical content. But it's pretty clear that he did not.

In a couple of the essays in Individualism and Economic Order, he mentions the pure theory of choice, and treats it as a useful tool that may not actually tell us anything about the real world. I think this passage works in connection with that as well.

Do you think that Mises believed that the pure theory of choice (max u) tells us nothing about the real world?

[Rafael Guthmann]

I think that Hayek build on Mises. The differences between the two represent the advances that Hayek did that Mises failed to absorb after.

The apriori point raised by Ryan, was that Hayek noticed that we need the empirical verification that people tend to learn what they need to know in order to formulate rational action plans. Without this tendency we cannot tell anything about the real world using a theory that assumes that decision makers have perfect knowledge.

But I think that this point was implicit in Mises. In part because his criticism of central planning is based on the impossibility of rational action given certain conditions. He assumed implicitly that people have limited leaning capabilities. He got this notion from empirical observation (trivial empirical observation).

[Greg Ransom]

Hayek moves (just a bit) away from this position after developing his notion that the mind something like a hierarchy of implicit rules -- work inspired by R. S. Peters and the later work of Wittgenstein.

Some dimension of these rules become a cultural achievement, and at bottom they are followed, but not articulated (see Hayek's "system of rules" paper, etc.).

Hayek nowhere makes sense of the division between these culturally inherited shared patterns of rule following, and the universal non-cultural, a priori structures of such things as mathematics and the logic of marginal valuation.

Hayek writes,

"The claim to which I have referred follows directly from this character of the first part of our task as a branch of applied logic. But it sounds startling enough at first. It is that we can derive from the knowledge of our own mind in an "a priori" or "deductive" or "analytic" fashion, an (at least in principle) exhaustive classification of all the possible forms of intelligible behavior."

[Greg Ransom]

Note that Hayek fails to understand that there are important differences between the explanation of action described in the language of meanings, intentions, etc. and the very different realm of open-ended learning and changes in understanding.

This is significant because Hayek locates the causal explanans in economics as _learning_, and he does not locate it in action explanations, as Mises does "officially" (but, significantly, not in fact in his actual explanatory efforts -- the causal explanation is the same as Hayek's, entrepreneurial learning).

[Greg Ransom]

Here's are two elements are suggested by this statement by Hayek, but which are in fact logically on a very different plane from belief / desire individual action explanations.

1. We act on the assumption that others are _learners_ on analogy with the way in which we ourselves are learners.

2. We act on the assumption that others pick up implicit pattern way of going on together just as we do -- implicit patterns of rule-governed behavior.

Note well that neither (1) or (2) can be captured in the logic of the "givens" found an action explanation -- common "beliefs" and "desires" and "intentions".

Hayek writes,

"There can be no doubt that we all constantly act on the assumption that we can in this way interpret other people's actions on the analogy of our own mind and that in the great majority of instances this procedure works."

[Greg Ransom]

Why aren't the social science causal / explanatory components of learning and imitated rule following on a plane with belief / desire action explanations?

Because learning and implicit rule following are open-ended, as emphasized by Hayek, Popper, Kuhn, and Wittgenstein -- they aren't capturable in a set of "given" and linked belief / desire atoms.

[Greg Ransom]

Rafael writes,

"Hayek noticed that we need the empirical verification that people tend to learn"

This isn't Hayek. He says the opposite of this.

The fact that relative global economic coordination exists is a given, a problematic explanadum that calls for solution.

Similarly, the fact that people are learners is also a given.

The empirical component of economics is linking the two together, order as an explanadum and entrepreneurial learning as an explanans.

Hayek in fact says we are not likely to advance a whole lot farther in understanding this learning processes than what we already know -- and there is really no _need_ to advance beyond what we already know, for the purposes of solving the problem at hand. (Hayek did not say, however, that we couldn't learn more about learning -- and Hayek was explicit in noting that those new things we might learn can't be fully anticipated in advance.)

[Greg Ransom]

I think I misunderstood Rafael.

I took him to mean "empirical verification" in the positivist sense.

But I think his sense is really the same as when he talks about "trivial empirical observation".

And this is trivial because we share common minds, the point of Hayek, Mises and Wittgenstein.

Rafael writes,

"I think that this point was implicit in Mises. In part because his criticism of central planning is based on the impossibility of rational action given certain conditions. He assumed implicitly that people have limited leaning capabilities. He got this notion from empirical observation (trivial empirical observation)."

[Troy Camplin]

"Hayek is foreshadowing his argument in The Sensory Order that one day science can hope to achieve a fully inclusive classification system of all possible phenomena."

Hayek explicitly denies that this is or could ever be possible in The Sensory Order. Further, Hayek also makes it pretty clear in The Sensory Order that his theory of how the brain classifies information supports Mises's aprioristic methodolgy.

[Nicolas Cachanosky]

There's an article by Hayek in "Studies" that follows the same line that Troy Camplin mentions. I don't exactly remember the title of the article and don't have the book at hand, but it's on rules and mind. Very good piece.
Good post! Looking forward for one on Mises at his most Hayekian now.
NC

[Per L Bylund]

@Nicolas:
It seems to me that if this is an attempt to "rehomogenize" Mises and Hayek, then the Boettkean statement of a Misesian reading of Hayek is false, just as there cannot be a Hayekian reading of Mises. Both of these "readings" suggest that Mises and Hayek WERE indeed different - why else would it make sense to use one as analytical framework for the other?

In other words, if Mises and Hayek are saying the same thing and stand for the same views of economic science, then it makes as much sense saying that one should adopt a "Misesian reading of Hayek" as saying that one should have a "Boettkean reading of Boettke."

[Nicolas Cachanosky]

I'm not sure what "rehomogenize" means, maybe because I'm not sure what the "dehomogenization" is supposed to mean. Sure there are differences between them, they were different persons, but I don't think that means all of their work is incompatible or all have to the same thing. So not sure what's the dehomogenization is about. I've read the article on dehomogenization more than once some time ago, but I still have problems to see the point.

As I see it, besides differences between them on some issues, I think Hayek's work can be seen as a continuation of the general approach Mises had on economics. In some issues Mises was more general and Hayek more specific. I guess that Hayek made explicit many things that were implicit in Mises. At least that's what I understand by a "Hayekian reading of Mises" and "Misesian reading of Hayek." Nothing more, nothing less.

[Mario Rizzo]

Hayek (in general) should not be read as a Misesian. I mean: why? Mises had so much methodological baggage that Hayek did his damnest to escape. Better to read early Hayek in the light of later Hayek if you can't just read Hayek.

A let me say quite bluntly: Hayek is so much more subtle a thinker than Mises. Mises was prone to excessive, and indefensible, generalizations about method, economic theory, and even economic history. Even where he is basically right he often overplays his hand.

Now a Hayekian reading of Mises might be a good idea if that means we qualify and balance some of the Misesian excesses.

[Adrian Ravier]

We have a rothbardian reading of Mises, and a hayekian reading of Mises. We have millions of readings of the same Mises. Even we have a machlupian reading of Mises´s methodology. Why? Is it because some authors try to show that Mises thinks like them? Or is it because Mises was not clear? I think that Mises was not completely clear on some topics. But, I need to say that my reading of Mises is more consistent with Hayek, than with Rothbard.

[Mathieu Bédard]

Is there any merit in the assertion that later Hayek and especially The Sensory Order is grounded in phenomenology and Schütz?

[Steve Horwitz]

@Mathieu: Roger Koppl and I find MUCH merit in that assertion. In fact, Pete's aphorism could be applied to Hayek and Schutz too: reading Hayek through Schutzian eyes is very productive, as is reading Schutz through Hayekian eyes.

I think TSO and Schutz are HIGHLY compatible.

[Jerry O'Driscoll]

Positions are very hardened on this topic, which doesn't make for useful discussion.

I do think one should read an author in terms of his own approach. But I also think one's interpreation of an author is deepended by understanding the influences on him. Mises clearly influenced Hayek, and both were influenced by common sources (Menger, the classical economists, etc.)

It would be odd, then, if Hayek and Mises didn't agree on many things. And since they are two different human beings, it would be unlikely that they didn't disagree on other issues. Why are these simple truths so much in dispute?

It is common to refer to The Sensory Order as late Hayek. It is in truth the earliest Hayek. The manuscript pre-dates all his economic work and reflects a conceptual framework pre-dating his economic thinking. It was published late, but conceived early in his career.

[Richard Ebeling]

In relation to Roger Koppl's and Steve Hotwitz's comment about reading Hayek in the context of Alfred Schutz.

Long ago (1975), when I was first becoming interested in Schutz's writings on intersubjective structures of meaning and ideal types, I asked Hayek what he thought of Schutz and if Schutz had any influence on him.

He told me that he never could make sense of what Schutz was saying in the discussions in Mises' private seminar in Vienna, and that he never understood Schutz's use of pheomenological philosophy in his book, "The Phenomenology of the Social World" (1932).

Hayek said that Schutz would have to be explained to him by Fritz Machlup and Felix Kaufmann.

But regardless of this, Mises' emphasize on methodological subjectivism (subjective meaning, the actor's point-of-view, the element of introspection to grasp certain essential concepts in human action, etc.); and Hayek's exposition and use of similar ideas in the article, 'The Facts of the Social Sciences' or the core chapters in "The Counter-Revolution of Science"; or Fritz Machlup's appreciation for the subjectivist foundations to economic analysis (see, for example, his insistence on the subjectivist perspective in the opening chapters of his "The Economics of Sellers' Competition" and his use of it to develop a theory of sellers' expectations in making decisions in the market place); all, are derived from Max Weber's theory of intentional action to which the actor assigns a meaning, and that "social action" refers to reciprocal subjective orientations in conduct. (And, indeed, is also very consciously the source and inspiration for much of Schutz's analysis of social action.)

In turn, the emphasis on the "introspective" glance as a source of useful knowledge for making intelligible how and why men act they way they do goes back to both Boehm-Bawerk and Wieser.

And, in fact, while some attempt to distinguish Mises and Hayek by arguing that Mises follows a Boehm-Bawerkian tradition in the Austrian School, while Hayek is following more closely in Wieser's train of thought, actually Wieser is the member of that older Austrian School who is most insistent upon the knowledge affirming usefulness of introspection for grasping the essential concepts and relationships in human conduct; and, thus, Mises is very "Wieserian" in his development of at least this aspect of the Austrian tradition.

Thus, if I may express it in this manner, the distinctions between Mises and Hayek in these and related matters are really "family squabbles" over the use and applications of aspects of the same methodological tradition of which they are both members.

Of course, Mises and Hayek are not "clones" of each other. They were of different generations and were influenced by different currents of ideas in their intellectually formative years. But they are often saying more in common from the same Austrian and Weberian roots than readers sometimes fully appreciate.

(And I would even say that Hayek did not always seem to appreciate how much he was closer to Mises on at least some aspects these methodological issues than he, himself, realized. Partly, I would argue, because he thought that Mises was saying things that Mises really was not. For instance, in later years Hayek accused Mises of being too much a child of Enlightenment Rationalism. And, yes, Mises was this, but for Mises that included the Scottish tradition of Adam Ferguson and Adam Smith on spontaneous order, just as much as for Hayek.)

Richard Ebeling

[Ryan Murphy]

Troy: Hayek makes it clear that we necessarily have apriori categorizations, meaning that there can never be any pure facts or empirical information. That isn't the same thing as saying our knowledge about the real world is a priori.

I don't completely "get" the need to read one economist from the point of view of another thinker. Read the words on the page, the footnotes, what else the economist said on the topic, and the conclusions of biographers. I feel like anything else is engaging in literary criticism.

[Greg Ransom]

Here's the Mises / Hayek overlap:

-- Both saw the pattern of social order as a problem.

-- Both saw entrepreneurial learning in the context of relative prices and property rules as the causal explanation of that order.

How do they differ?

Well, Mises makes a mess of fitting this explanatory strategy into the epistemological / philosophical tradition.

By contrast, Hayek over time makes significant advanced and frees himself from deep pathologies in the epistemological / philosophical tradition.

Examples:

-- Hayek provides deep grounding to our understanding of the social science of property rights -- adding to our understanding of traditional rules of conduct and the evolution of law, linking these up to human anthropology, biology, and brain science, etc.

-- Hayek breaks free of (a) the justificationalist tradition; and (b) the deductivist model (3) the formalist tradition which models language and the growth of knowledge in terms of a set of "givens" mapped in a formal construct.

-- Hayek makes plain the priority of problem raising patterns in our experience and the significance of open-ended causal mechanisms.

To coin a phrase, Hayek does a better job of being true to the insights of Menger than does Mises.

Hayek also differs from Mises in the fact that Hayek expands the forward looking logic of marginal valuation into the realm of production goods across time, while Mises reverts to backward looking Ricardian categories in Mises model of the "evenly rotating economy".

[Peter Boettke]

Ryan,

Two reasons to read differently. First, to contextualize an author. Most people have no clue about Hayek's context and thus forget where he came from and thus read his statements as if he was making a contribution to anglo-saxon philosophy of science. He is not. Second, to forge a creative intellectual synthesis of writers that often seem at odds but through your synthesis can go beyond what their limits were for a new theoretical framework.

It is not literary criticism, but it is about "reading" and "critical thinking".

I plan on taking up Nicolas challenge sometime this week after I get back from a trip.

As to Jerry and others --- yes people have hardened positions on this, even young scholars who have just started to adopt a position have hard and fast rules about who said what and why they said it. Isn't our goal as teachers to challenge peoples presumptions and make them feel uncomfortable about what it is they feel so comfortable with? I want to make Mario feel uncomfortable with his position on Mises because in the process perhaps he might see something there that opens up to a new way to view these issues and new directions to take the argument (he did that with his own efforts at reconstruction in the late 1970s).

Mises is a very rich thinker, as was Hayek, and both had more in common than any other set of economic thinkers with respect to their paradigm. We do a disservice to that intellectual heritage when we don't constantly make people feel uncomfortable about the easy reading of the tradition.

At least that is how I feel about it. Mises and Hayek should not be treated as fixed positions set in stone but instead as part of a live conversation that is part of our extended present in economics. Admittedly, I don't really care about what Mises said or what Hayek said, but where Misesianism and Hayekianism might be said to go as we develop economics further and apply it to new challenges.

[Richard Ebeling]

In reference to Ryan Murphy's comment, above, about whether a priori categories inform us about the "real world."

Certainly it would be absurd to attempt to, "a priori," deduce the workings of the laws of nature, or the chemical compositions of various physical objects.

But how do we understand "man"? Certainly, he is also a physical object that can be "scientifically" measured, weighed, dissected, and "observed."

However, if we accept that man is more than a "physical object," that he is a thinking, evaluating, and acting being as well, we must ask, how do we gain (at least some of the) knowledge about these characteristics?

How do I even "suspect" that man is thinking, evaluating, and acting? Well, I'm a human being, and I know that I'm a thinking, evaluating and acting person, because I do these things. And where do I do them (the thinking, evaluating, deciding)? In my mind.

And I am able to look "inside" and "observe" (that is, I introspectively reflect upon) what I am doing and the logic I follow in doing it.

This supplies very useful knowledge about man and his conduct (on the basis of one fairly reasonable "working assumption" -- other people's minds basically work like mine).

Now, in everyday life we take this very much for granted. We interact with people; we make and follow through with interpersonal plans and activities; we talk to each other; we attempt to anticipate others' actions (we attempt to "read" other people).

I never understand this "prejudice," that because we cannot understand "nature" through introspection and any claimed "a priori" knowledge, we therefore must ban introspection and the "a priori" knowledge it gives us about ourselves (and others).

Now, this is not to claim that such "a priori" knowledge tells us what products consumers will want to buy next week; or whether the events in Libya will set off similar activities in Saudi Arabia; or the direction of oil prices over the next six months.

And to be frank, Austrians like Mises HAVE NEVER CLAIMED that it could. Indeed, that is why Austrians such as Mises or Machlup drew attention to that Weberian tradition that Alfred Schutz extended for understanding how intersubjective structures of meaning and ideal types of individuals and groups of individuals (based on experience, interaction, or other similar sources of "empirical" knowledge) assist people in trying to anticipate the actions of others in the "real world."

Weber's, Schutz's and Mises' "ideal type" analysis is meant to show how individuals attempt to give "empirical" content to the concrete actions of men in the world, given the "a priori" categories in the context of which men think, evaluate, decide, and act.

When will we, finally, accept the wisdom of a phrase from the 1960s, when it comes to understanding an essential aspect of the human being -- "It's all in your head, man."

Richard Ebeling

[Greg Ransom]

What we want is a true understanding of the social world.

There is a lot of useful understanding in Mises, much of it embedded in a mistaken understanding of the nature of knowledge, explanation, and epistemology. As a result, Mises comes up with a confused and unhelpful explanatory strategy.

Reading Mises "thru" a Hayekian lens helps the useful parts of Mises pop out -- it allows us to filter out what is important in Mises, as it at the same time allows us to put the confusions and mistakes in the background.

Mario writes,

"a Hayekian reading of Mises might be a good idea if that means we qualify and balance some of the Misesian excesses."

[Mathieu Bédard]

Jerry O'Driscoll;

"It is common to refer to The Sensory Order as late Hayek. It is in truth the earliest Hayek. The manuscript pre-dates all his economic work and reflects a conceptual framework pre-dating his economic thinking. It was published late, but conceived early in his career."

D'Amico and Boettke in Making Sense Out of the Sensory Order (2010) offer compelling evidence that it's not really an early work, ie the original manuscript was only about 30 pages long and the only idea that can be said to have been already there is that memory precedes perception, which can be traced back to Kant. This interpretation has the benefit of being compatible with the idea that Hayek was influenced by Schütz. It would make The Sensory Order a book of its era.

[Nicolas Cachanosky]

I didn't mean the expression of "Mises at his most Hayekian" to be a challenge, or a work burden to you or Prof. Horwitz. I would enjoy reading a post like that; I think it can be written. I'm even sympathetic with Ebeling's mention that "Hayek did not always seem to appreciate how much he was closer to Mises on at least some aspects these methodological issues than he, himself, realized. Partly, I would argue, because he thought that Mises was saying things that Mises really was not."

On the methodological issue that came up in some comments, I'm inclined to think Machlup's "The Problem of Verification in Economics" is an interesting approach to bridge (or better understand) their similarities and differences. And also to see that Mises may not have been an "extreme apriorist" as sometimes he is presented, and that he acknowledged a specific role for "empirical assumptions." I think this piece Machlup may be an example of what Peter Boettke mentioned as "a creative intellectual synthesis of writers that often seem at odds but through your synthesis can go beyond what their limits were for a new theoretical framework." I'm curious on what others think about Machlup's paper.

I should add that I don't see these too much as a struggle of what "x" said about "y," but as an input to further research; after all (good) economics requires the study of history of economic thought.

[Ryan Murphy]

Professor Boettke,

I have no problem with contextualizing the author, but I think that's the purpose of biographies. I'm not trying to be sarcastic in saying that. But if you look at whom Hayek cites regarding methodology, it's not Mises. It's Popper. Yes, Hayek did study under Mises. He also said this:

"Let me get to the crucial point. What I see only now clearly is the problem of my relation to Mises, which began with my 1937 article on the economics of knowledge, which was an attempt to persuade Mises himself that when he asserted that market theory was a priori, he was wrong; that what was a priori was only the logic of individual action, but the moment that you passed from this to the interaction of many people, you entered into the empirical field." (p. 72 of Hayek on Hayek).

We can interpret certain other passages all we want that make it look like Hayek was saying the same thing as Mises, but they don't stand up to scrutiny unless Hayek was constantly contradicting himself. There are alternative explanations of those passages that sound extremely "Misesian" but have crucially different meanings when considering everything else that Hayek argued (for example, the one I initially pointed to in my first post).

So either:

1) Hayek's conception of social science crucially differed from Mises, and Hayek didn't emphasize this until Mises died for the sake of politeness. Hayek believed that only the categories of thought (such as that people maximize utility in a tautological sense) were a priori, not anything about the interactions between individuals. He also believed that Mises believed otherwise.

2) Hayek never actually understood Mises's methodology. It would then be extremely curious to claim that the correct way to understand Mises was through Hayek.

I also agree that history of thought as primarily an instrument for generating new modern research, but if we are going to cite economists who died decades ago, we better be damn sure what they meant, not just what the most charitable reading of them means.

[Richard Ebeling]

I would argue, Mr. Ryan Murphy, that Hayek did not really understand Mises' views on the relationship between the theoretical (those insights of economic theory that might be deduced "a priori" through introspection about "human action," as such) and the "empirical" (that which is dependent upon knowledge of the content of peoples values, preferences, the institutional setting in which they act and interact, their expectations concerning others and potential future events, etc.).

It is true that except in passing in one or two of the essays that Mises published in the second half of the 1920s (and which were reprinted in his 1933 book, "Epistemological Problems of Economics"), Mises never attempts to explain in any detail the way that he considers Weber's ideal type construct to be a way that "acting man" attempts to gain the "empirical content" for anticipation and expectation in the market process.

But Hayek did review the German-language version of "Human Action" shortly after it appeared in 1940. And there Mises does try to explain in far more detail the method of "understanding" by which empirical content for market action is derived. (He attempted to expand on this further in "Theory and History" in 1957.)

So it has always seemed peculiar to me that Hayek kept in insisting that Mises believed that everything that was needed to know for market processes and coordination could be derived "a priori." This is especially strange, it has always seemed to me, since this subjectivist method for "understanding" is, as I said in a comment above, basically the same one that Hayek defends in "The Counter-Revolution of Science."

And Hayek's charge that Mises was an extreme "Rationalist," with all the negative connotations that Hayek liked to give to that term as times, is also a bit peculiar. For example, in Hayek's 1933 address on "The Trend in Economic Thinking," his gives his first clear formulation of the idea of society as,

" the spontaneous interplay of the actions of individuals [who] may produce something which is not the deliberate object of their actions but an organism in which every part performs a necessary function for the continuance of the whole, without any human mind having devised it."

At this point in the text Hayek, in a footnote, gives the reader only one reference to support this argument, and that is to Mises' discussion of this very point in "Socialism."(Hayek footnotes the same passage from Mises' "Socialism" when he discusses the Scottish "anti-rationalist" tradition in "The Constitution of Liberty.")

Thus, in "Socialism," Mises says at one point that advocates of natural law,

"erred in regarding this great change, which lifts man from the state of brutes into human society, as a conscious process; as an action, that is, in which man is completely aware of his motives, of his aims and how to pursue them. . . Because it led to present conditions, people regard the development of social life as absolutely purposeful and rational . . . Today we have other theories with which to explain the matter." (By which Mises means Menger's theory of the spontaneous order.)

And in one of the essays in Mises' "Critique of Interventionism" (1929), he says:

" . . . liberal social theory does not explain [the] formation of social ties and institutions as consciously aimed at human efforts toward the formation of societies, as the naive versions of the contract theory explain them. It views social organizations 'as the unintended result of specific individual efforts of the members of society' [Menger]."]

Hayek, it seems to me, never clarified in his own mind his own understanding of what Mises often was arguing. And as these quotes from Mises (and even Hayek's own references to some of them in contradiction to his charge that Mises was a version of a "constructive rationalist") suggest that Hayek cannot be considered a good judge of Mises' views on a variety of matters.

Richard Ebeling

[Steve Horwitz]

On the other hand, Richard, Hayek did point to Mises as the person most responsible for developing the subjectivist tradition in economics, which suggests at one level, he "got it."

[Steve Horwitz]

And let me say a word in defense of Pete's aphorism.

The point for me is more or less what Pete said. All readings are interpretations. We cannot simply "read the words on the page" as Ryan suggests. Words don't interpret themselves; humans interpret them. Pete's aphorism is the suggestion that because Mises and Hayek were, as several others have said, working from a common Mengerian tradition in which they agreed on far more than they disagreed, and in which they influenced each other, reading each through the other's framework is how to generate the most productive insights for *the furthering of that tradition*.

The idea of a "best" reading is not necessarily that which is most inline with the author's intentions, but rather which is most productive for moving forward the conversation in which that author is participating.

So perhaps I should have framed Pete's aphorism with "If one is interested in advancing the Mengerian/Austrian tradition in economics, then the most *productive* reading...." I'm far less interested, as Pete notes, in what Mises or Hayek "really" meant as I am in what their work has to say to us moving forward today. And that's what Pete's aphorism means to me.

[K Sralla]

"Admittedly, I don't really care about what Mises said or what Hayek said, but where Misesianism and Hayekianism might be said to go as we develop economics further and apply it to new challenges"

A progressive academic research program must move the ball forward or the school dies. The Austrian school hung by a thread in the 1930's and quite nearly went extinct. I would argue that Hayek single-handedly kept it alive in England. If left only to Mises, it would have died.

The reason for this is that Mises made a mess out of the order and priority of epistemology, metaphysics, and ethics. Hayek did not understand Mises's epistemology because it was pure Kant, and scarcely any human being can make sense of Kant on these issues. If Hayek could not understand some of Mises, it is very likely we can't either.

Hayek on the other hand raises many questions (sometimes more than answers) which show brilliant insight into the deepest problems of economics and science at large.

The unanswered questions are where the ball should be picked up. That is where the low hanging fruit hangs. I'm afraid that trying to harmonize Mises with Hayek is a dead end street. Hayek already did that work. He knew Mises better than any modern researcher, and found Mises arguments lacking, and this led to unsatisfactory grounding for his (Mises) most often correct conclusions.

Maybe it is because I am a natural scientist, but what I find most interesting is not the relationship between Hayek and Alfred Schutz, but rather Hayek and Michael Polanyi. There is an uncanny parallel between what these two men were working on throughout their carreers (not just in London), but later at the University of Chicago.

May I humbly suggest that if a researcher is looking to pick up another Nobel, this quite possibly may be where the ball should be picked up and advanced. These ideas both that Hayek and Polanyi worked on are some of the most facinating questions in the history of the world.

But if a researcher goes there, be forewarned to bring your A game, because this is very hard thinking, and will require a quite different formulation than the direction of much contemporary Austrian economics.

P.S. I think it is probably unfortunate that Hayek likely viewed Polanyi with some respect, but mostly as an amateur to economics and the social sciences.

Finally as an aside, I think it is important to ask why Hayek has continued to be read widely across a broad academic audience some 40-70 years after his major works, and Mises has largely been forgotten except by a few Austrian economists. It is no accident that Hayek won the Nobel and not Mises.

[Gene Callahan]

Ryan seems massively ignorant of how the process of interpretation works; we ALWAYS read texts in the context of other texts. There is no possibility of "just reading" the words on a page without a background context, as that would not be reading at all! Pete is trying to suggest some background we should highlight while reading Mises and Hayek. There is nothing outrageous about this suggestion at all.

[Greg Ransom]

Richard, let me suggest that to get a handle on the understanding gap between Hayek & Mises ones needs to appreciate the nature of a fundamental divide which exists between Mises and Hayek on how to do science, and how to think about advances in our knowledge and understanding.

Here is the divide.

The justificationalist model of knowledge and science demands various kinds of demonstrative justification for knowledge "claims".

Hayek rejects this model, Mises embraces it.

Hayek has abandoned the justificationalist tradition and replaced it with a conception where our understanding grows, and where scientific knowledge begins with problems in our experience and advances via mechanisms that explain those patterns and change our understanding of them at the same time. This is a view Hayek found in Menger and Popper, and he embraced it after finding reasons in brain science to reject the Mach/Hume/Russell/Carnap view of empirical knowledge and science as a demonstrative achievement built out of building blocks of phenomenological "givens".

Hayek doesn't look to ground scientific knowledge in any set of givens and he doesn't look to build up science from any set of givens known directly to the mind. He looks to problems first -- he says this again and again and in various books and articles, esp. those written after 1950.

Mises' effort is focused on defeating various kinds of skeptical and relativistic attacks on the scientific standing of economic science, and his strategy for vindicating economics is an appeal to a classic form of justificationalism -- the identification of a set of demonstratively true, a prior, apodicticly certain building blocks, from which to construct a larger structure of knowledge.

Why does this matter?

First, it leads them to do very different things with the pure logic of marginal valuation.

For example, Hayek uses it to help "see" the overall order of the market, which helps him to "see" the empirical order that demand explanation.

Second, it leads them to conceive of the explanatory role and logical status of entrepreneurial learning in a different way.

For example, Hayek identifies entrepreneurial learning as a _causal_ mechanism which provides the central causal explanans allowing us to understand the thing most responsible for giving us economic coordination.

Hayek may not have had a deeply well worked out conception of Mises epistemological / scientific project -- but he wasn't confused in understanding that the were on opposite sides of a great revolution in the history of science / philosophy.

[Ryan Murphy]

Professor Horwitz and Professor Callahan,

I think, ironically, you are taking me incredibly out of context. What I wrote was "Read the words on the page, the footnotes, what else the economist said on the topic, and the conclusions of biographers." Obviously I agree with contextualization.

Also, I agree with everything Greg Ransom said above. That is a better response to Professor Ebeling than I could have articulated.

[Richard Ebeling]

With all respect, Greg, from my reading, Hayek offers no theory of "entrepreneurial learning."

That is the great reversal that I see in part of Hayek's work. In "Economics and Knowledge," he constructs the argument in such a way that it should lead us to ask, precisely, how do people form expectations, how are their expectations modified and why (in terms of the direction of the modification) as they experience plan failure or success?

But he mostly drops this "research agenda" in "The Use of Knowledge in Society." Here individuals discover prices and use their various types of knowledge as seems appropriate by them, given the "price signals" they experience in their respective corners of the social system of division of labor. But there is little or no discussion of how individuals interpret the prices they find in the market to decide what it is the prices are telling them, so they can effective know how to use their individual bits of knowledge.

Thus, Hayek turns away from the "interpretive turn" (the process of interpreting the "signals" of the market to determine what they are "saying" in relation to the current and possible future actions of others to whom they must orient their own conduct). (I discussed this "problem" in Hayek along time ago in my essay, "Toward a Hermeneutical Economics: Expectations, Prices and the Role of Interpretation in a Theory of the Market Process in 1986.)

It is precisely in Mises and Schutz (building on Weber) that there emerged the beginnings of an Austrian theory of learning, interpretation, and expectations formation. This was (and is) meant to be basis for understanding the "empirical" content and the the meaning of the "empirical" in a subjectivist conception of man and social action.

With full admission that I am not a philosopher and cannot claim to have dove deep into the depths of the Kantian "mindset," I find the critical references to Kant in this Austrian context as bit of a "boggy man." As least as it relate to the needs of the economist to have some workable foundation for "justifying" how he attempts to ground the "truth-value" of what he trying to do in his sphere of social analysis.

Clearly, all men basically share a similar logical structure of thought, regardless of time and place. If this was not the case, I don't see how we can claim any reasonable confidence when we say we "understand" what ancient Greeks like Plato and Aristotle thought and said thousands of years ago, or what Confucius said in his parables also thousands of years ago in a civilization far removed from that in which have been born in "the west."

How does that logic manifest itself in "action"? And, thus, we come to the "a priori" categories and relationships of ends and means, costs and benefits, profits and losses, time judgments of "sooner" or "later." And it was why Boehm-Bawerk could say that centuries before economics formulated the logic of marginal decision-making, the human actor acted on the basis of it, since it is the way man's mind works.

If we do not assume this "a priori" basis of how we think (and on the basis of that thought, act), then I do not understand how we can be exchanging thoughts and (hopefully!) understanding each other -- both in agreement and disagreement.

Richard Ebeling

[Greg Ransom]

Richard, in his 1937 "Economics and Knowledge" essay Hayek says we don't need any special theory of learning -- and we can't really expect "science" to tell us much of great importance on the subject beyond what everyone already knows. He doesn't rule out advances in our understanding in this direction, he simply says it isn't important to putting together the explanandum and explanans of economics -- similarly, Darwin didn't need to fill in the underlying causal pathways of genetics (or even theorize about their logic) in order to put together a revolutionary explanandum, explanans which rocked all of science.

Richard writes,

"With all respect, Greg, from my reading, Hayek offers no theory of "entrepreneurial learning.

That is the great reversal that I see in part of Hayek's work. In "Economics and Knowledge," he constructs the argument in such a way that it should lead us to ask, precisely, how do people form expectations, how are their expectations modified and why (in terms of the direction of the modification) as they experience plan failure or success?

But he mostly drops this "research agenda" in "The Use of Knowledge in Society."

[Greg Ransom]

Richard, I don't see that I've used the name "Kant" anywhere in the above discussion.

Richard writes,

"With full admission that I am not a philosopher and cannot claim to have dove deep into the depths of the Kantian "mindset," I find the critical references to Kant in this Austrian context as bit of a "boggy man." "

[Greg Ransom]

Here are two different things, Richard, (1) learning; and (2) action. I.e. these two things are different:

1. This topic:

"in Mises and Schutz (building on Weber) that there emerged the beginnings of an Austrian theory of learning, interpretation, and expectations formation. This was (and is) meant to be basis for understanding the "empirical" content and the the meaning of the "empirical" in a subjectivist conception of man and social action.

2. This topic:

"How does that logic manifest itself in "action"?"

The belief / desire logic of action explanations is different in kind from what happens in learning.

Belief / desire explanations of individual actions assume "given" meaning entities as causes or carriers of interpretive significance.

Learning is open-ended and takes us into new realms of understanding, beyond any set of "given" meanings, e.g. the topics discussed in Kuhn, Wittgenstein and even Popper.

Hayek is getting clear about these distinctions, and showing how they linked very differently to the empirical problem of social economic order or coordination.

I've already suggested that there is much of great value in Mises work.

I really don't disagree with your account of Mises, and I certainly don't take the position that the elements you explicated above are not helpful or essentially sound.

I'm simply insisting that Hayek's work adds significant doses of clarity to what is going on, and what the logical status actually is of these components in Mises, and how the are interconnected in a sound explanatory frame.

Example -- Hayek clarifies the logical status of learning in economic explanation, and the link of that status to the empirical problem to be explained. Mises does NOT make this clear, in fact he muddles it and makes a mash of it, leaving Hayek baffled (among many others) if your account above is to be believed.

[Greg Ransom]

Here is a core difference between Hayek and Mises:

Hayek says that economics is an empirical science with the task of explaining the empirical pattern in our experience of global economic coordination, e.g. the pattern of order suggested by the repeated pattern in which prices repeated move toward costs.

Mises says that economics is a science of human action, that it explains human action, and that it isn't an empirical science.

But here is the muddle in Mises. You can dig out where Mises admits that there is a problem raising pattern out that that needs to be explained. And you can dig out where Mises acknowledges that the central explanatory element in economics is a causal mechanism called entrepreneurial learning.

So buried within Mises, you can find many of the core elements in Hayek. But these are really muddled and effectively denied by the explicitly stated account of Mises'scientific/explanatory project.

And I should add, when Hayek's explanandum / explanans nexus is laid out directly, a different understanding of the explanatory and conceptual roles played by the pure logic of marginal valuation is revealed (see, e.g. the first few chapters of Hayek's _The Pure Theory of Capital_).

The valid point that Horwitz and Boettke make is the suggestion that Hayek's work is enriched when the rich Misesian characterization of entrepreneurial learning and choice behavior are added to the bare bones of Hayek.

I see no where that either I or Hayek deny this, indeed the quote from Horwitz by Hayek on Mises and "subjectivism" suggests otherwise.

[Greg Ransom]

Richard, what I am trying to ward off here is a case of talking passed each other.

The point isn't about philosophy, the point is about science.

Science often has had to escape from aspects the philosophical tradition in order to advance -- the case of Darwinian biology is well explored by Ernst Mayr, and others have written on how physics over time escaped from various cul de sacs (e.g. the physics of Descartes).

[Troy Camplin]

The aprioristic logic of human action results in a spontaneous order catallaxy. I don't see what the problem is. If Hayek argued that Mises argued for a priorism in anything but understanding human action (such as in developing full economic theory), then from what I have read of Mises, I would have to argue that Hayek really didn't understand Mises. As I understand Mises, he was arguing that one uses apriorism to understand human action, and that one must understand that to understand human interactions, out of which results an economy that one can then understand emprically. But one must understand the source and nature of human action. The empricism comes about with Mises in his argument that economists should study what people do, and judge what those people's values are based on those actions. More, an understanding of economics allows one to recommend to people what they should do if they want to accomplish their goals. It seems to me that Hayek provides a model of how Mises's apriorism actually works, where it comes from. He demonstrates that it comes about because the brain is in fact structured similarly to the economy insofar as both are spontaneous orders and follow the same kinds of rules.

It also seems that very often Hayek thought he was refuting something by Mises, but then Mises would agree entirely with Hayek's work, and keep writing the kinds of things he was writing. This suggests that perhaps Hayek didn't quite get what Mises was saying all the time, and it also suggests that Mises saw their work as more complementary than anything. Hayek often covered things Mises did not, and vice versa. I see more overlap and complementarity than conflict -- but conflict is hardly a bad thing. Out of conflict comes new ideas and a healthy tradition.

[Roger Koppl]

Hayek never deviated from the "interpretive turn." In Bruce Caldwell's telling, TSO is the crucial moment when H is saved from hermeneutics by complexity. But the concluding philosophical chapter is ringing endorsement of "verstehende psychology" and apriorism. It's right there in plain words. In his 1964 essay "The theory of complex phenomena" (reprinted in the1967 Studies volume) H cites Empedocles’ “Knowledge is of like by like” and notes that he got the quote from Mises. That's very much in the Dilthey tradition. Way back in the 1920 essay that became TSO H says he using scientific methods to get the results very much like those "H. Bergson" got by another method. I don't know how Hayek could have been plainer or more direct. And yet, somehow, we think H somehow abandoned the "understanding" tradition of Dilthey and Max Weber.

I think might know the reason for this difficulty in seeing what is explicit in Hayek. Hayek did something weird. He took the epistemological tradition of lebensphilosophie (Bergson & Dilthey) and embedded it within a strictly scientific world view. He committed the phenomenological sin of "naturalizing consciousness" while vindicating the *old* classical hermeneutical tradition represented by Max Weber. (Hayek show no influence from Heidegger's ontological turn; on the contrary, that's out of whack with his system.) Hayek brought "science" and "humanism" together, but we can't see it because we are so convinced the two are oil and water.

[Richard Ebeling]

Greg,

To begin with your last point, I happen to consider Mises and Hayek to be "complements," not "substitutes." I not share the view that they need to be "de-homogenized" into two distinctly separate goods.

And, further, as I know you know, Hayek on more than one occasion over the many years said that the starting point of much of his own work -- on money and the business cycle, on socialist central planning, on the nature and workings of the market order -- was "inspired" and given initial direction from Mises' work. He often found Mises' conclusions persuasive, but found fault with some aspects of the chain of reasoning by which Mises reached that conclusion. And, thus, he attempted to develop his own explanation and chain of reasoning to more persuasively reach a similar conclusion as Mises'.

And there is no doubt that each in their (somewhat different) ways were both "children" of Menger's vision presented in both the "Principles" and in the "Investigations."

Let me suggest that both Mises and Hayek are attempting to sort out the learning of "patterns" that enable individuals to orient and coordinate their conduct in relation to others in the social and market orders.

This was precisely the use and application of the Weberian ideal type by, say, Mises and Schutz. And, in turn, the reasonable presumption of a common logical structure of to the human mind also serves for us to understand how men order and act upon their thoughts, which provides a basis upon which men can know and reason with each other.

This is, in my opinion, the Weber-Mises version of "pattern prediction," not in the quantitative sense of prediction, but like Hayek's variation on this theme, the quality and logic of how such social order and pattern emerges and serves to assist people in their orientations and patterned structures of social interdependency.

Neither this Weber-Mises version or Hayek's version provides a theory of actually how people learn in the sense of knowing how and why people draw certain conclusions from experiences rather than others. Do psychologists know, yet, why two individuals experiencing the same "experience" learn different lessons from it, in terms of what they think it means and how it makes them respond (to one degree or another) differently?

If they have, it would be interesting to know how it it that from the same (market) experience there are simultaneously "bulls" and "bears."

I have often wondered why Mises and Hayek -- coming from that same Mengerian tradition -- often discuss what has seemed to me, in essence, the same argument, yet in what appears to be significantly different ways.

The following is my (partial) conclusion. It has to do, to some extent, with the context in which they were led to think about these things in terms of the debates and opponents that they respectively faced.

Before the First World War and in the 1920s and 1930s, Mises' "opponents" are Historicists, Marxists, and National Socialists in the German-speaking world. They not only deny the logic and coordinating "harmony" of the market order, they question whether the "laws" of economics are even the same for all men or groups of men. These Marxists believes that social "classes" think in terms of different class-interest logics. The National Socialists insist that different races have different logics.

Mises is attempting to both challenge and refute these assertions. Mises is arguing there a common reality to the human condition, not only in the sense that the laws of the physical world are the same for all men and groups of men around the world (and in the past as well as the present). But, also, there is a common reality in the logical structure to the human mind -- whether it being in the thoughts that went through the mind of Plato in ancient times, or the "primitive" in a tribe in, say, New Guinea, or in the thinking and acting of a capitalist and his worker in industrial society, or an "Aryan" and a Jew in 1920s, 1930s, Germany.

And if there is a common reality of a logical structure of the human mind, then one can attempt to derive from this an understanding of the common reality and logic of how (all) men reason about choice and action. How, with a variety of subsidiary assumptions and "observations" about the institutional circumstances under which men act and interact, one can show the logic and order and patterns that emerge, "spontaneously," from the interactions of many men. And how such order and pattern can be disrupted by various forms of policies that interfere with the workings of a market order.

Hence, Mises' insistence on "reason," "rationality," and the "intentionality" of human action that results from that logical structure of the human mind -- but which also generates "order without design" when the individual intentionalities of many men interact in emerging social arenas of life. He is battling what he called in the title of one of his articles from this period, "The Cult of the Irrational."

Hayek, on the other hand, finds himself in the English-speaking world of the 1930s and 1940s. Here the intellectual climate is not one, like in the German-speaking world, of calls to the "blood" and race-think.

Instead, Hayek finds an intellectual environment in many believe so much in the power of reason and rationality, and to such an extent, that they believe that the human mind can know, comprehend, and integrate all that is needed to know to systematically design and re-order society according to "plan."

Rather than like Mises, who must insist upon the primacy and importance of common reason and rationality in a cultural and political climate of calls to "polylogism," Hayek, instead, must focus his "attack" on demonstrating the limits of reason and "rationality" in a climate of opinion in which too many intellectuals believe that they can know "everything" and rationally construct the good and just socialist society.

Thus, Mises, in the German-speaking world of that time, must emphasize the common reality and universal logic of human thinking, choosing and acting in a setting in which this commonality is being denied. In other words, the arguments that run through most of Menger's "Principles."

Hayek, in the English-speaking world of that time, must emphasize the common reality and limits to men's ability to think, choose, act and try to reshape the social order. In other words, the arguments that run through part of Menger's "Investigations."

But, of course, Mises knows and fully appreciates Menger's theory of spontaneous social processes. And Hayek, of course, knows and fully appreciates the common logic in human thought, from which the economist can derive the individual's "logic of choice."

Thus, they are playing from the same music score -- Menger's subjectivist conception of man and society -- but they are underscoring different notes on the music sheet given the different "audiences" for whom they are playing the music.

Richard Ebeling

[Peter Boettke]

well said Richard, very well said!!!!

[Steve Horwitz]

A few quick comments:

1. The next time Pete complains about the quality of comments on economics blogs, I'm going to send him this discussion. This really has been like being at one of our FEE workshops or around the table at GMU or NYU. Regardless of the disagreements, this is some really good stuff.

2. I too think Richard nails it above and I think Pete's point about historical context is correct as well.

3. Having just re-read Menger for my Austrian class, there is NO doubt that both Mises and Hayek are faithful Mengerians, stressing different sides of the argument, perhaps for exactly the reasons Pete suggests.

4. On the learning point: I have argued for many years that although we don't have the empirical explanation of learning that is being discussed above, the best way of understanding modern Austrian economics is through Kirzner, in the following sense: what Israel understood is that the "solution" to the observed pattern in the world (the explanandum to use Greg's term) that motivated Hayek is the entrepreneur as found in Mises.

Hayek does not refer to the entrepreneur very often (though re-read the 1940 piece on socialism carefully), but when he does, he gets it. Mises focused in on the intentional, entrepreneurial element of action, downplaying though not ignoring the explanandum (though read his discussion of the methodological implications of Menger's theory of the origin of money on p. 405 in HA). What Kirzner does is to bring these two together and use Mises's explanada to attack Hayek's explanandum.

As I pointed out in my SDAE presidential address a few years back, this is the way we should be doing business as Austrians. I have made the same argument in the Journal of Private Enterprise's recent forum on Dan Klein's critique of Kirzner.

Here's my SDAE presidential address: http://www.gmu.edu/rae/archives/VOL17_4_2004/1-Horowitz.pdf

The JPE paper is not online that I'm aware of, but if you email me, I can provide a copy.

[Richard Ebeling]

If I may just add, Steve, I was going to mention your point about Mises' conception of the entrepreneur "filling in" the essential element in Hayek's story of markets, prices, and knowledge in terms of tendency for coordination.

But since my comment was already so long . . .

It has always seemed peculiar to me that clearly, as you suggest, Hayek understands the central role and function of the entrepreneur in the market process, but he never emphasizes or high-lights the entrepreneur, or gives him an integrative place, in his writings.

Maybe he took too much for granted when thinking about these in terms of what he learned from Wieser (see his, "The Law of Power,") or Mises through many of his writings, or even Schumpeter (who was influenced by Wieser during his student days on the "innovative," leadership role of the entrepreneur.) And, therefore, failed to think it necessary to make this more explicit in some of his writings.

But it is certainly the case that if we integrate the "Misesian" entrepreneur into Hayek's analysis of knowledge, expectations, and prices, one comes far closer to a more "complete" understanding of the market process of expectations, adjustment and coordination.

Richard Ebeling

[Steve Horwitz]

My own suspicion Richard is that there are a number of places where Hayek sort of takes for granted elements of Mises's thought (for example, I think all of *Road to Serfdom* assumes, but rarely if ever articulates, the Misesian critique of planning). Whether he is just assuming his readers know it, which they might not, given his English context and Mises's German context (e.g. Keynes), or what, I'm not sure. But you can find him arguing as if what Mises has said is right without ever explicitly saying so.

[Julian F. Mueller]

Very compelling conclusion Mr. Ebeling.

This would also give an answer to the question, why Hayek (afaik) has never given an own account of the logical status of the very core assumptions of economics. If I understand you right, Mr. Ebeling, your account would be, that this is because Hayek has basically the same view as Mises on this basic economic - ceteris-paribus-shaped - laws?

Or would you side with Mr. Ransom on his account that Hayek "reject[ed] the Mach/Hume/Russell/Carnap view of empirical knowledge and science as a demonstrative achievement built out of building blocks of phenomenological "givens""?

Julian F. Mueller

[Richard Ebeling]

Steve,

It is a peculiar aspect to "The Road to Serfdom" that nowhere in the book does Hayek present even a general summary of the types of criticisms that Mises and he had been making of the economic "impossibility" of successful socialist central planning.

I have always taken this to be the case because Hayek wanted to present the strongest "political" arguments against socialism, and collectivism in general, in terms of its dangers to political and personal liberty.

Or, as you are suggesting, he may have assumed that the really interested reader would have read summaries of Mises' and his own critique of socialist planning in other writings. For instance, the very readable and clear summary in Walter Lippmann's "The Good Society" (1937).

I wonder if Bruce Caldwell has any thoughts (or knowledge) about this.

Richard Ebeling

[Greg Ransom]

Richard, I would tell this history embedded story a bit differently.

The key background for Hayek is the fact that Hayek was soaking from the very beginning in the new understanding of logic and science that rocked the intellectual world -- the work of Wittgenstein, Mach, Russell, Carnap, and the rest. This revolution offered something of an empiricist alternative to the various strands of neo-Kantian which dominated the Continent -- a neo-Kantianism which (for many) stood in shatters in the wake of Frege's revolution in logic, Einstein's revolution in science, and Hilbert's revolution in math.

Hayek read Wittgenstein's _Tractatus_ in 1921, he studied Mach and brain cells in the year before.

When Hayek turned to economics, the leading modern account of economic science put Machian epistemology right at the center of the modern view of economics as a science -- Schumpeter's 1908 book.

You can see Hayek using the language of formal logic repeatedly in the 1920s and 1930s describing the logical properties of the elements of the explanatory strategy of economics. In answer to Lowe (etc.) Hayek is looking for a strictly _deductive_ account integrating money into the "barter" equilibrium construct, attempting to vindicate the monetary explanation of business cycles found in Wicksell and Mises.

So Hayek's table we set for him in Vienna.

Hayek from the beginning was an empiricist -- and the empiricism he soaked up was that of Mach and Wieser, who viewed the "psychological" aspects of economics as empirical phenomena.

But Hayek had alredy identified the internal illogic of the Machian system, and he was on a collision course with the Wieser strategy, which viewed economics as at once as both deductively closed and as grounded in empirical psychology.

Here is Hayek's problem situation.

You start with Wieser's "god's eye view" dictator model of the "economy as a whole". And with a commitment to some sort of empiricism about science and economics.

As Hayek pushed down on the deductive conception of the equilibrium construct and the logic of marginal valuation, an empirical element popped to Hayek's attention -- learning was empirical, learning was guided by profit and loss signals, the equations of equilibrium math did not coordinate the system, discovery and changes in understanding within local conditions and the relative price system drove the coordination process.

And money and the coordination of production processes were the "loose joints" in the system where learning and profit and loss were most vital, and it was just there were other economists like Lange and Lerner and Kaldor were fooled by various equilibrium constructs.

Hayek saw that a commitment to the deductive idea didn't work as he has anticipated it in 1929 -- he had to rethink the use of equilibrium constructs, and he suddenly perceived that the world of learning -- including profit and loss signals -- laid outside of the "givens" of formal constructs built upon the pure logic of marginal valuation.

Hayek was already much of the way down this path in his 1929 book, but it wasn't until the collectivist economic planning debate -- and his re-reading of Menger (while editing Menger's collected works), that Hayek put it all together.

And reading Popper's book the winter of 1936-1937 gave Hayek a great sense of confirmation -- empirical problems do come first, the Carnap/Mach/Tractatus division between deduction and induction was a bust as a picture of science, etc.

Well, enough for one comment.

[Greg Ransom]

Richard, in one sense -- in comparison with Mises -- this is true. But in another sense, this isn't true at all.

Hayek's case against Lerner & Lange, etc. on socialist economic planning come down to nothing less than an insistence on the unreplaceable role of the entrepreneur in the coordination process.

As early as 1929 Hayek identifies profit and loss signals as the core of the market coordination system. Who uses profit and loss signals? The entrepreneurial learner.

When Hayek identifies learning as causal explanatory solution to the problem of overall economic coordination, he's talking about entrepreneurial learning.

When Hayek talks about competition as a discovery process, again, he's talking about the entrepreneurial learner.

We don't' want to fall into a dispute over semantics.

And as I've said, I agree that Mises gives us a rich account of entrepreneurial judgment and learning, while Hayek doesn't focus his research energy in this direction.

But, none the less, the entrepreneur as learner definitely does have an integrative place in Hayek's work.

Richard writes,

"Hayek understands the central role and function of the entrepreneur in the market process, but he never emphasizes or high-lights the entrepreneur, or gives him an integrative place, in his writings."

[Julian F. Mueller]

That's quite the cliffhanger. Dear Mr. Ransom, please continue!! This seems to me one of the rare occassions to hear some thoughts on Hayek and Mises from someone with a profound knowledge in philosophy of science.

[Steve Horwitz]

Greg,

You and Richard don't disagree. Richard's point, and mine, was not that the entrepreneur is absent in Hayek's work, but that it is rarely explicit even though it lingers in the background. The interesting question is why it's never as explicit as it seems it should be.

[Greg Ransom]

Larry Wright ("the most important writer on teleology since Aristotle - A. Rosenberg) does something similar with teleological explanation -- Wright "naturalizes" function and teleological kinds which are DIRECTLY perceived by human beings, we can directly perceive the purposive behavior of the rabbit or the functional character of things in nature, because we automatically perceive such things in such a way. BUT we can also provide a causal etiology behind these teleological entities, e.g. the way Darwin does with adaptations -- physically multiply realized teleological entities with a causal etiology, but which can be directly perceived by humans because our brains allow us to do such things.

Wright naturalizes functional talk and teleological talk while vindicating both teleological perception and a naturalistic causal understanding of the underlying mechanisms of nature.

"He took the epistemological tradition of lebensphilosophie (Bergson & Dilthey) and embedded it within a strictly scientific world view. He committed the phenomenological sin of "naturalizing consciousness" while vindicating the *old* classical hermeneutical tradition represented by Max Weber."

[Greg Ransom]

A copy of Wright's _Teleological Explanations_ will cost you $331:

http://www.amazon.com/Teleological-Explanations-Larry-Wright/dp/0520030869

[Greg Ransom]

Wright's widely cited paper "Explanation and Teleology" is here:

http://www.jstor.org/pss/186722

[Greg Ransom]

Here's something to consider.

Our language for talking about the universal element of mind (e.g. the logic of choice, math, etc.) is largely inherited and acquired -- and evolves and develops over time, along with our inderstanding.

So the "apriori" is intertwined with evolved and acquired patterns of language and
"going on together".

This is a point found in both Hayek and Wittgenstein.

What Hayek and Wittgenstein want to do is show how purely logical formal systems misrepresent social phenomena such as the social phenomena of language and economc coordination -- these "bird's eye view" constructs leave out the empirical and sociallly achieved coordination that stands behind these formal systems and symbols -- and gives them signficance.

Mises tries to address this issue by focusing on individual action explanations in a socal context -- but he fails to grasp the larger problem of making sense of the relation of the pure logic equilibrium construct to the overall perception and explanation of market order. Because economiics fundamentally is interested in the problem of explainng a social pattern, and we intuitively understand human behavior, and we don't need to know the formal logic of choice to order our own affairs accordng to it, or perceive that ordering in the doings of others.

[Richard Ebeling]

Greg,

I'm not sure what Hayek was "soaking up" in some Vienna bathhouse, but the fact is as late as 1935, in the introductory chapter to "Collectivist Economic Planning," Hayek presented a conception of economic science and its grounding that was completely in line with Wieser and Mises on the logical character of action and choice that that inner, "empirical" reflection provides us with.

In the 1920s and part of the 1930s, Hayek was still much taken with Walrasian-Paretian general equilibrium theory and its logical coherence. (Recall Hayek's story that he suggested to J.R. Hicks that he look into Pareto's indifference curve analysis. -- Boy, did that not end up influencing the direction of microeconomic theory in the 20th century, if true!!)

(Hayek was also David Rockefeller's thesis adviser at the LSE after the war -- what went wrong, there!!)

Hayek already appreciated the "guiding role" of prices in 1929, in "Monetary Theory and the Trade Cycle":

". . . in the exchange economy, production is governed by prices, independently of any knowledge of the whole process on the part of individual producers." (pp. 84-85)

But it seems fairly clear that it was the challenge of explaining the "problem" of coordination and mis-coordination in the business cycle in replying to his critics such as Gunner Myrdal (who took Hayek to task for not integrating a theory of expectations into his monetary analysis) and the attempt by socialists such as Oskar Lange and Dickenson in the mid and late 1930s to formulate a "market socialism" with the tools of Walrasian theory, that made Hayek see the limits and inherent problems in the general equilibrium and perfect competition frameworks.

This "breaking away" from Walrasian-Paretian theory shows itself in all those articles between 1937 and 1946 -- "Economics and Knowledge," his article on the false "Competitive Solution" to the problem of socialist planning, his analysis of plans and plan coordination in the early chapters of "The Pure Theory of Capital," and then "The Use of Knowledge in Society" and "The Meaning of Competition."

Over this ten year period, Hayek is attempting to formulate a different way of thinking about and understanding the nature and workings of markets, and what men can know and how they act in this world of change and divided knowledge.

Indeed, I would argue that in rethinking the nature of markets and their workings, Hayek was moving more in the direction of the conception of the market process formulated by Mises in "Nationalokonomie" (1940) and "Human Action" (1949), but which were already presented in its aspects in Mises' earlier works on socialism and the market order in reply to to the interventionists of that time.

And this separate from any "misgivings" Hayek may have had about Mises' "grounding" of economics.

Yes, Gottfried Haberler recommended Hayek to read Popper's "Logic of Scientific Discovery" shortly after it appeared in print in 1934. And all that played on his mind -- as well as the influence of those philosophical disputes that you mention that were "soaking" in Vienna in the 1920s.

But, it seems to me, that it was not these philosophical matters that made him "rethink" various aspects of economics and its foundations. Rather, it was these challenges within economics that made him rethink how we had to reconsider the properties and assumptions underlying any more realistic and relevant theory of market processes and coordinating institutions. And, that made him ask questions about the "scientific" method and its uses and applications in a field such as economics.

But however much Hayek may have been rethinking what he considered to be more satisfactory ways of thinking about knowledge, learning and "knowing" in the social sciences, the fact remains he published "The Counter-Revolution of Science" in 1955. And that book is and remains an excellent example of that methodological subjectivist tradition that one finds in Weber and Mises. (How else do we understand the meanings and logic in people's action if not from "inner" interpretations and analysis of how the actors think and see and give meaning to the world and their relationships to others in it?)

At the most, what Hayek kept grappling with was the search for what to him would be conceptions and foundations that would more persuasively ground the same Mengerian (and Misesian) ideas about intentional man whose interactions often result in unintended patterns of social coordination.

Richard Ebeling

[Steve Horwitz]

Hayek's dive (back) into the philosophical and methodological issues in the 1940s was, as Richard says, the result of him having been perceived as losing those two key debates and trying to figure out why both the Keynesians and the market socialists didn't see economics and markets the way he did. He concluded, in my view, that the problems were bigger and broader, hence the turn toward the issues in *Counter-Revolution* and elsewhere.

And that is why the quotes in "The Facts of the Social Sciences" are so interesting to me: Hayek was being VERY Misesian AFTER those debates, in the very process of trying to figure out what went wrong. This is also consistent with Kirzner's point about the near simultaneous publications of *Human Action* and *Individualism and Economic Order* as representing the point of conception for the modern Austrian school.

Richard is right, I think, that Hayek in a number of ways moved closer to Mises after those debates because he perceived, correctly in my view, that methodological issues about the nature of economics were key to understanding why people were talking past each other and where economics, in Hayek's view, had gone wrong.

[Greg Ransom]

Richard, we aren't that far apart. I simply note again that Hayek was identifying the causal coordinating role of profit and loss signals and (implicitly) learning already in his 1929 book -- and he was explicitly saying the the Walras-Pareto construct didn't explain this adjustment process.

And Hayek already has his problem situation -- justifying the explanatory and scientific adequacy of monetary explanations of the trade cycle against Lowe's demand for a deductively implied result.

Most of this background is German economics -- although Hayek's discussion of the coordinative function of proft and loss signals outside of the equations of math economics was addressed to a claim by an American economist.

So, contrary to Richard, this breaking away was already happening in 1929, and he was already articulating a market process view in that 1929 work, in just the part of Hayek's 1929 book from which he quotes.

"This "breaking away" from Walrasian-Paretian theory shows itself in all those articles between 1937 and 1946"

[Greg Ransom]

What I sense Richard is that you seem repeatedly to miss Hayek's focus on the central role of the problem of overall global economic order in thinking about economics -- and the way that Hayek rethinks the equilibrium construct and the role of entrepreneurial learning in relation to this empirical problem.

When Hayek talks of pattern perception and pattern prediction and explanation of the principle, he's talking not abou the behavior of an individual, he's talking about the social pattern in overall economic coordination of all individuals, and Hayek's parallel case is Darwinian biology, which explains overall patterns but does not explain individuals as individuals.

[Greg Ransom]

What Hayek develops post 1935 is a combination of intertwined insights.

-- He builds upon his 1929 insight that empirical patterns give rise to the problems of trade cycle economics to be soved by "pure theory. Hayek comes to see that the coordination problem of the overall economic order is an empirical problem to be explained

-- He builds upon his 1929 insight that the math of "pure theory" does not provide an explanation of the coordination process, that this coordiation process is outside of "pure theoyr" in the realm of indviduals acting on profit and loss signals.

-- He gains clarity on the realm of pure tautologies -- e.g. the equilibrium construct and the pure logic of choice. And he gains clarity about what lies outside of that realm of the tautological and purely logical -- e.g. the empirical problem order to be explained, and the realm of price signals and learning which causally explains that empirical problem. And Hayek comes up with an account of the different roles the purely tautological logic of marginal valuation can play in understanding the causal realm which stands outside of the merely tautogical.

-- The background for Hayek's development here is both the modern literature on the nature of formal systems and the character of empirical scieces -- and also the communication gulf which Hayek experiences in his engagement with the economics profession. And I note that the communication gulf included "Austrian" educated economists, such as Schumpeter. Again, Hayek's engagement with Lowe and the deductive ideal makes clear that the problems involved here are on Hayek's table already before the move to Britain.

[Julian F Mueller]

If we really want to understand how close Mises's and Hayeks conceptions of economics are, I think it could proof to be a fruitful endeavour to narrow down the dicussion from general discussion abouth Hayeks intellectual background to the basic propositions of economics.

Let's just consider this text-book proposition

[A]Mandating a price floor above the equilibrium wage should cause unemployment.

Mises would negate (!)this proposition. Because in Mises view economics can just make ceteris paribus laws. But [A] clearly is just a (bold) proposition about the laws of society.

Mises would say:

[B]Mandating a price floor above the equilibrium wage ceteris paribus causes unemployment.

My intuition is that [A] and [B] though the difference seems subtle, actually refer to whole different paradigms of science and understandings of causation.

My question would be, what would be Hayeks handle on that proposition?

I think this question is actually not easy to answer, because of this: If Hayek really changed his mind regarding the foundation of basic economic laws, he never wrote about his own approach on the logical status of the basic economic laws.

I think it is deeply troubling that even the late Hayek postulated Laws in Economics ans Society (not regularities, but laws (!)) that allow us to predict patterns. Furthermore, in his paper about complexity he also his highly critical about quantitative methods (as he was most of his life).

So Hayek is critical about both a) a prior economics and b) quantitative methods. But that leaves me puzzled - what (again) is than Hayeks view on the basic economic axioms?

[stickman]

Some thoughts from an outsider (as it were)...

This has been an interesting - if rather lengthy and, at times, difficult - thread to follow. I, for one, will admit to having already separated out Mises and Hayek in mind; largely because of my distinct problems with the epistemological/ praxeological method of Mises versus the more Popperian approach of Hayek.

Indeed, it is interesting to read about some of Mises's more nuanced qualifications regarding the use and limitations of the a priori method, because I certainly don't get much sense of that when I visit the Mises.org site.[*] IMHO, this "fundamentalist" view (if I can call it that) will do more harm to Austrian economics than good in the long-run, simply because it acts as an immediate deterrent to the multitude of people who hold that theory and empirical observation can (should) be mutually reinforcing. Once that initial mental-barrier is in place, it's really hard to overcome.

[*] For what it's worth, this is one reason that I find reading Coordination Problem a far more rewarding experience than sifting through the Mises Echochamber... er, Institute.

[Greg Ransom]

Here's the internal dynamic already in Hayek in 1929.

Hayek is grappling with making sense of deviations between the monetary economy and the barter economy -- and trying to fit both within the deductive ideal of necessity prescribed by Lowe for a scientific / theoretical economics.

And he's already identified the causal coordative process -- e.g. profit and loss signals -- as lying outside of the deductive frame.

And this problem situation sits for Hayek in a rich background understanding of the logical status of formal systems, and the logical demands of scientific explanation -- informed by the modern post-Fegean / post-Machian literature as much as by the older neo-Kantian and verstehen literature. In other words, as much informed by early Wittgenstein, Carnap and Schumpeter as by Mises.

The 1929 Hayek/Lowe deductive ideal failed _internally_ in Hayek's work -- and he resolved these problems by coming up with a revolutionary change in his understanding of the uses of equilibrium constructs and pure logic of choice.

Hayek's answers here are not the same as Mises, although the causal element of entrepreurial learning is shared. This shared causal component, however, is embedded in different conceptions of the explanatory task and strategy of economics.

[Richard Ebeling]

Greg,

This discussion could go on, obviously, for a long time. And both of us, in terms of the writing we have both been churning out in reply to each other and others is, well, quite a bit!

But, if I may, I like to point out that Mises, too, considered the "starting point" the realization that there were and are these regularities and seemingly patterned relationships of coordination in the social sphere that needed "explanation" for their observed ("empirical"?) existence.

Mises often pointed out that economics began, as a serious field of study, when some of the thinkers of the 18th century (the French Physiocrats, the Scottish philosophers) discovered and began to systematically attempt to analyze and explain them.
(See, as just one example, because it is on my desk in front of me, the opening pages of Mises' 1942 article on 'Social Science and Natural Science,' reprinted in "Money, Method and the Market Process: Essays by Ludwig von Mises" [1990]:

"The foundations of the modern social sciences were laid in the eighteenth century . . . The founders of Political Economy discovered regularity in the operation of the market. They discovered that to every state of the market a certain state of prices corresponded and that a tendency to restore that state made itself manifest whenever anything tried to alter it.

"This insight opened a new chapter in science. People came to realize with astonishment that human actions were open to investigation from other points of view than that of moral judgment. They were compelled to recognize regularity which they compared to that which they are already familiar in the field of the natural sciences." (pp. 3-4)

In this sense, Mises, too, viewed himself as a part of an "empirical" research agenda: how does one demonstrate and show the origin, emergence, and evolutionary institutional basis of a social order that manifests such patterned regularity?

His insistence on that introspective "a priori" starting point with that "logical structure of the human mind," is based on his argument that nothing happens in the social and market arenas that does not originate in the choices and actions of individuals. How, then, do we study how the individual "works."

He rejected Bahaviorism and Positivism for reasons that I think are familiar. Then, from whence do we derive our "knowledge" or "empirical" information about what makes the human "mechanism" work the way it does, and which may generate those observed regularities when individuals interact under certain specified conditions?

We look "inside" the human mechanism, which is made possible for the social scientist (the economist) because he, like his object of study, is also a man. The social scientist is both "object" and "observer." He can perform "experiments" on himself. He can look into his own mind, and ask his "object": how does your (my) mind work, what logic does it (I) follow in thinking, evaluating, deciding, and doing (initiating and following through with intentional actions)? And so on.

This also helps tell the social scientist what his object of study (himself) can claim (with reasonable confidence) to be those things that it (he) can possibly know by looking within himself and what requires "external" knowledge if it (he) is to know other things.

Thus, reflecting in this introspective way can also assist the social scientist in determining the limits of such "a priori" knowledge.

All this are stepping stones to understanding and analyzing the interactive results of social association with their various intended and unintended outcomes. All of which are the elements of constructing a successful theory of the processes that generate the "empirically" observed regularities crying out for explanation.

This is Menger's method and approach; it is Mises' method and approach; and it is Hayek's method and approach in "The Counter-Revolution of Science," when he speaks of the "composite method" for understanding the workings of the social order and its "unintended" properties.

Richard Ebeling

[Steve Horwitz]

I think I'm just going to start cutting and pasting "What Richard said."

[Greg Ransom]

Well, we agree with all that Richard -- I've indicated above that I'm well aware that all of that is in Mises.

This makes it all the more clear that Mises produced a misleading muddle when he characterized economics as an apriori / apodictic science of human action. It's subject matter isn't individual action -- as Mises indicates its task is to explain global economic coordination. And the causal explanans isn't human action, it is entrepreneurial learning.

We know we can square all of these muddles and give a charitable reading of Mises which puts everything in a frame which makes economics an empirical science focused on causal processes and mechanism of learning -- but this is essentially Boettke & Horwitz's recommendation to give a Hayekian reading to Mises.

And it still doesn't successfully get us to the picture of the uses of equilibrium constructs and the pure logic of valuation that Hayek identifies in his 1937 essay "Economics and Knowledge" and in his 1941 book _The Pure Theory of Capital_.

There is clarity in Hayek about the explanatory significance that comes in linking up the empirical problem with the causal solution (and the advance in understanding which this linkage creates), which isn't achieved in Mises. Similarly, there is clarity in Hayek about the empirical / causal nature of entrepreneurial learning in relation to the empirical problem of overall economic order that is obscured and muddled in Mises.

Again, these considerations recommend a Hayekian reading of Mises.

So we can agree on all of the things that are contained in Mises, and agree with all of the overlap between Hayek and Mises, and still see reason for a Hayekian reading of Mises, and still see important elements in Hayek that are not to be found in Mises.

[Steve Horwitz]

I agree with your last paragraph Greg. Would you agree with the reverse? I think there are important elements in Mises that are not to be found in Hayek, thus a Misesian reading of Hayek is also valuable.

[Greg Ransom]

Yes, I agree. I'm a Mises fan boy ...

"Would you agree with the reverse? I think there are important elements in Mises that are not to be found in Hayek, thus a Misesian reading of Hayek is also valuable."

[Steve Horwitz]

As I'm teaching "Economics and Knowledge" and "Use of Knowledge" tomorrow, I just re-read the former. Maybe it's because I have all this on my mind, but I can't help but think about how Schutzian the whole argument is, especially in the idea of knowledge about others' plans. That's exactly Schutz's definition of social action - action that requires taking into account the actions of others. Hayek even mentions, favorably, verstehende social science in fn 18 on p. 52 in the context of the essential subjectivism of social science.

More interesting though is this sentence "It is only because of [the fact that the relevant data is what is believed by the actor to exist] that the propositions we deduce are necessarily a priori valid and that we preserve the consistency of the argument." What's interesting is the footnote that follows cites Mises' *Epistemological Problems*.

Later, when Hayek introduces the idea of "the division of knowledge" as the key to understanding empirically why/how equilibrium will come about (and uses language that foreshadows the 1945 essay), his immediate citation on that phrase is to Mises' *Socialism*.

So Hayek cites Mises on the nature of the a priori and on the idea that markets are about the division and use of knowledge. Now maybe this is just a Misesian reading of Hayek, but it sure seems to suggest that Hayek thought Mises and he were largely on the same track.

[Greg Ransom]

I should also say I'm a big fan of Richard Ebeling, and much like his book on Mises.

[Greg Ransom]

Hayek sometimes suggests he was trying to gently persuade Mises with his 1937 paper. Not very plausible, given what else we know about the paper, but interesting never the less.

[K Sralla]

Hayek contends in “Economics and Knowledge” that both a priori and empirical types of propositions are common to all economists. A priori tautologies (the pure logic of choice) isolate the individual components of the social order in a kind of reductionist manner, while empirical learning relates pure theory to the big problem of catallaxy (later spontaneous order).

But Hayek’s distinct yet common sense construction was of a sort that Kant said was impossible, namely *analytic a posteriori.* This is the Kantian concept that trips up Mises, and I think mystifies Hayek. Hayek, in Economics and Knowledge is trying to do something that Mises largely contends in Human Action is impossible, namely marry the analytic and empirical, and find means by which empirical propositions may be verified (falsified). This is a significant turn in Hayek away from Mises (I think).

In the end however, I somewhat disagree with Greg who says this is more about the philosophy of science. I would contend that the main difference between Mises and Hayek is EPISTEMOLOGY. The big question that both seek to answer is “HOW DO WE KNOW?” Mises says the answer is found within us all due to the logic patterns that have evolved in the Human brain, but also asserts dogmatically that this can never be subject to any empirical claims.

Hayek knew there were all sorts of problems that could be raised with Mises’s formulation, tracing back even to the methodologies of Menger and Bohm-Bawerk.

I suspect Hayek thinks Mises muddles the use of the very concept of “a priori” knowledge, and as a result his usage is incomprehensible. Hayek agrees that under pristine isolation, there are indeed purely priori axioms that are also important economic propositions, but also asserts that many economists failed to differentiate them properly from empirical knowledge, and that the latter is the stuff of interesting real world questions.

[K Sralla]

"The second point is that I do of course not want to suggest that the sorts of problems I have been discussing were foreign to the arguments of economists of the older generations. The only objection that can be made against them is that they have so mixed up the two sorts of propositions, the a priori and the emperical, of which every realistic economist makes constant use, that it is frequently quite impossible to see what sort of validity they claimed for a particular statement. More recent work has been free from this fault, but only at the price of leaving more and more obscure what sort of relevance their arguments had to the phenomena of the real world."

(Economics and Knowledge p. 56)

[K Sralla]

"The second point is that I do of course not want to suggest that the sorts of problems I have been discussing were foreign to the arguments of economists of the older generations. The only objection that can be made against them is that they have so mixed up the two sorts of propositions, the a priori and the emperical, of which every realistic economist makes constant use, that it is frequently quite impossible to see what sort of validity they claimed for a particular statement. More recent work has been free from this fault, but only at the price of leaving more and more obscure what sort of relevance their arguments had to the phenomena of the real world."

(Economics and Knowledge p. 56)

[K Sralla]

Let me try to briefly unpack a few more problems:

Proposition: We make sense of human action by understanding our own actions due to the fact we are human. This allows for the construction of axioms of human action along the same lines as mathematics or pure logic.

This is said by Mises to compose a kind of "a priori" knowledge.

*However*, "a priori" literally means independent of conscious experience. But if we carefully examine ourselves, we find that without experience of many situations, we can have only limited learnings of how we ourselves act. Without such past *experience*, it is hard to make sense of the actions of others, especially in more than the instinctual processes of life. Therefore in some form, we concede the role of empirical knowledge in the formation of economic axioms. Now to most of us, this would not be controversial. Of course we are more or less rational beings, but also beings who learn from experience.

However, Mises says we are born with an inherited (via evolution) purely logical conception of human action (kind of), although it somehow springs forth (I guess sort of like instinct) after we grow from cell stage to embryo, to conscious person.

Now if we aren’t really born with it, but it suddenly springs forth as we become conscious, it is fairly hard to say categorically that these are not learned patterns of thought in one degree or another.

This leads to the observation that if the pure logic behind praxeology is inherited in our genetic code, then it is largely instinctive, and destroys the free purposeful action of human beings. Mises even concedes this to some extent in admitting that there is a kind of hard determinism that is guiding human action, and that the concept of “free will” is largely weakened.

A counter argument to the Misesian construction of purely logical economic axioms holds that the pure logic of choice is not as pure as the type of a priori knowledge which whispers to our conscious mind that when four apples fall from a tree, four apples will lay on the ground, and four less will be hanging from the tree. We may have never seen an apple tree, but we intuitively know that the statement is true a priori.

In the sense of catallaxy, the a priori and empirical are quite hard to separate. The fact is that the pure logic of choice may not be "pure" logic at all, but polluted by the empirical knowledge of experience. There are two categories, as Hayek argues, both a priori and empirical, and both of these operate within the economic endeavor

[Troy Camplin]

Mises does not have to be correct about the nature of a priori knowledge to be correct about its proper use. Thus, Hayek can provide an emprical explanation of how apriorism comes about that is at variance with Mises, yet affirms Mises's use of apriorism as valid.

All in all, This has been a fantastic discussion. Many of my senses of how Mises and Hayek are related has been given clear form by Richard, with Steve's observations on Kirzner's role wrapping it up.

I am surprised in all the discussion by Greg on Hayek's other influences that Bertalanffy has not been brought up. Surely he and his work on General Systems Theory had a prfound effect on Hayek.

[Greg Ransom]

Hayek in "Economics and Knowledge" uses the terms "tautology" and "verification" in working to clarify the a priori and empirical elements in economics.

The modern notion of "tautology" was introduced by Wittgenstein in the _Tractatus_ and "verification" was Carnap's standard for establishing significance, derived from his reading of the _Tractatus_.

[Roger Koppl]

Steve,

Can you give some examples of "important elements in Mises that are not to be found in Hayek"? I think it can sometimes be hard to get what Hayek's talking about if you haven't also read Mises, but I'm so sure Hayek just omitted anything important and valuable.

FWIW, I tend to think of Hayek as the one who brought Mises into the scientific world view. Mises is the master who shaped Hayek's program. It is hard to find a structural element in Hayek's work not already there in Mises. The one exception is the theory of mind. BUT, that theory of mind is precisely the thing that let Hayek put Misesian economics on modern scientific foundations. If I am looking at Hayek and Mises as an historian of economic thought, Mises is the great pioneer and Hayek the disciple. As a work-a-day economist, however, it is Hayek who gives me sustenance. Hayek is the more modern, scientific figure who helps me work through new problems.

[Roger Koppl]

I'm *not* so sure Hayek omitted anything . . .

[Greg Ransom]

Hayek was also -- and independently -- a great pioneer.

Examples.

Hayek pioneered the notion of intertemporal equilibrium.

Hayek pioneered growth theory -- inspiring Harrod.

Hayek pioneered the extension of the logic of marginal valuation to production goods through time.

And Hayek was as much a "disciple" of Wieser, Menger, Wicksell, and Bohm-Bawerk as he was a "disciple" of Mises (I have this language).

Indeed, many of the things we credit to Mises are to be found earlier in Wieser, Menger, Wicksell, and Bohm-Bawerk -- and Hayek picked up and developed or transformed elements from these folks that Mises missed.

Roger writes.

"If I am looking at Hayek and Mises as an historian of economic thought, Mises is the great pioneer and Hayek the disciple."

[Greg Ransom]

Sorry. Make that:

I _hate_ this language.

In reference to the language of "disciple".

[Jerry O'Driscoll]

As this facsinating discussion winds to an end, I find that Roger Koppl and Mario Rizzo are in agreement on substance. Roger is more positive in tone. But Mises' "baggage" (Mario's terminology) makes it more difficult to join his work with that of contemporrary thinkers than for Hayek.

I didn't like Roger's use of "disciple." I would say "influenced," and then agree with Greg Ransom's widening the sources of influence. Again, I would add the classical political economists.

Recall how Hayek developed the model of Prices and Production by linking it to the classical debates over money and central banking. Mises insisted his monetary theory of the trade cycle was the working out of the currency principle, also harking back to 19th century controversies (which keep repeating because economists no longer read the earlier work).

[Greg Ransom]

Jerry, agreed.

For brevity's sake I skipped reference to the classicals.

[Richard Ebeling]

Jerry's observation, above, should not be lost before this discussion comes to an end.

Both Mises and Hayek considered themselves building upon the entire ediface of the best of economic theory since the time of the Scottish philosophers in the 18th and the Classical Economists in the 19th century.

Both Mises and Hayek were widely read in the history of economics, and many related disciplines.

Mises, more than once, emphasized that while it was flattering for many to refer to the theory of money and the business cycle developed by him and Hayek as the "Austrian" theory, he told readers that it was all a continuation and refinement of the works of the Currency School and Wicksell.

The same applied to the general theory of the market system, and its applications, from the classical contribution to it being based on a "sounder" foundation with the subjectivist approach of the thinkers of the late 19th century.

For all of the seeming "dogmatism" that Mises is often viewed as expressing, I doubt very much whether it would have considered it very fruitful to argue over who is a "real" and "true" Austrian. He would have asked, does it advance our knowledge and understanding of our theory of human action, and its most "highly developed" subfield, economics?

Richard Ebeling

[Jerry O'Driscoll]

Amen to Richard's last observation: let us focus on the quest to advance the theory of human action.

[Roger Koppl]

I guess the word "disciple" makes Hayek sound unoriginal. Happy to ditch the word and that connotation, though Mises was more than just one influence among many as all here would likely agree.

[Greg Ransom]

We are back to square one.

Hayek is after a science of global economic order, the pure logic of marginal valuation is used to help understanding that causal domain.

The science of human doings isn't limited to the pure logic of marginal valuation -- in encompassed brain theory, the imitation of patterns of behavior, etc. And economists aren't especially good in these larger domains.

Jerry writes,

"let us focus on the quest to advance the theory of human action"

Richard writes,

"He would have asked, does it advance our knowledge and understanding of our theory of human action"

[Steve Horwitz]

From Bruce Caldwell:

"Hopelessly late as ever: Just a line or two on a question raised by Richard and Steve yesterday about the Road to Serfdom and why Hayek did not repeat or refer to the arguments Mises made about socialism and that Hayek had reiterated in Collectivist Economic Planning. Hard to know why but here are 3 possibilities.

First, his audience was not economists. As he said in its preface, it was a political book, and he was aiming at people like the Webbs but also others (recall the "To the Socialists of All Parties" dedication) who thought socialism and democracy had marched hand in hand throughout the 19th and into the 20th century. His goal was not to show that calculation under socialism was impossible, but that socialist planning was incompatible with democracy, and indeed would kill it off. He was worried about what the postwar political situation was going to be, and the argument that planning equals freedom was a key theme of his opponents.

Second, the book was part of a larger project in intellectual history from which it had become detached. Part of the point of that larger history was - and this became increasingly important to point out in 1944 - to argue that all that had happened was not due to the viciousness of the German people but to the bad ideas that they had embraced since Bismarck's time. This was an important point to make in 1944 because who knew whether the allies were going to put another round of reparations on the German state after the war, something some people (e.g., the French historian Etienne Mantoux) would have endorsed (the beginning of the Cold War altered all that).

Finally, as some have said, he could have presupposed that everyone who would have been interested in the economics already knew about the earlier arguments. "

[K Sralla]

Hayek is the big thinker. Mises is a very very smart economist, but largely a man of his time IMHO.

Some of Hayek's ideas (or those he develops in a much more insightful sense than earlier Austrians) portend later developments in science such as complexity that are still in their infancy. These go way beyond economics proper.

Let me suggest that Edward Lorenz's classic paper on meterology is not unrelated to the phenomena Hayek was wrestling with at the University of Chicago, and really throughout much of his career. Such an assertion would have called down the anathema of Mises.

Where most of pre-late 20th century science is focused on order, Hayek dared to look at disorder and ask how order might emerge. Mises researches the tidy. Hayek researches the untidy. Mises sees the rational, Hayek considers the non-rational. Mises sees linearity. Hayek sees non-linearity.

Mises sees that his formulation of economics as praxeology was the big idea. Hayek sensed a big idea in his research that was larger, and that extended beyond only economics.

This is not to turn this into a contest on who’s the smartest, just to let economists know which was onto ideas which more thoroughly pervade fields outside strictly economics.

[Greg Ransom]

Let me re-emphasize the divide here between Hayek's view of the problem of science of interest to economists and the Misesian view expressed by Jerry and Richard.

Hayek identifies domains of science in terms of empirical problems -- problem raising patterns in our experience that demand explanation.

The topics of science are defined by their empirical problems. Problems come first. All sorts of things help us identify the problem and see the problem, but the problem defines the domain of inquiry.

The domain of inquiry in economics is defined by the thing to be explained -- systematic patterns of global economic order and disorder. The empirical problems are at the global level, they are not at the individual level.

Economists are developing a science that identifies empirical problems and provides causal explanations which allow us to understand those patterns.

Mises gets us off track by this notion that the topic of economics is "human action", and that economics is a "science of human action".

Jerry writes,

"let us focus on the quest to advance the theory of human action"

Richard writes,

"He would have asked, does it advance our knowledge and understanding of our theory of human action"

[Greg Ransom]

Say we reject Hayek's view of science and knowledge, and we go back to Mises' view that what we are after is a theory of human action, and we go back to the pre-Hayek/Popper view that knowledge is justified true believe and scientific knowledge is knowledge which is demonstrable.

So how do we go about developing a science of human action, or economics as a science of individual behavior or choice? Well, we either look for necessities in our minds, or we look for "verified" or "tested" empirical patterns. The first doesn't take you into the empirical world of the contingent and causal, and the second leaves out much of what is important, and doesn't produce regularities that are universal.

Hayek takes us out of this morass. He re-sets the explanatory strategy and logical status of economic science in a way that is true to the phenomena and succeeds as an explanatory enterprise.

[McKinney]

I'm not as well-read in Mises or Hayek as the other posters, but what I have read is more in line with Ebeling's postings.

Principles of hermeneutics are important here. Sound interpretation requires understanding the historical context, the intended audience, the purpose of the piece, etc.

When these principles are considered, it seems to me that Hayek and Mises were saying the same things but in different ways to different audiences for different purposes.

Clearly, Hayek wasn't going to devote his career to parroting Mises. Hayek wasn't the type to plow the same field that someone else had already done a good job at. Hayek built upon Mises without going over the same ground.

While Hayek didn't emphasize the a priori nature of economics, it seems he took it for granted and worked more on the empirical side. And while Mises emphasized the a priori side, he never neglected the empirical.

[Greg Ransom]

Imagine a Darwinist explaining the alternative Darwinian explanatory project to a 19th century German biologist focused on characterizing Aristotelian biological essences.

The Darwinist points out that there is an important problem raising empirical pattern in our experience, constituted by the problematically interconnected patterns of adaptive features and the evolution of species by descent.

Imagine if the 19th century biologist said he accepted all that, then added:

"Let's not argue over what is "real" and true biology. Rather, lets ask, does it advance our knowledge and understanding of our theory of Aristotelian essences, and its most "highly developed" subfield, anatomy?""

[K Sralla]

"Hayek takes us out of this morass. He re-sets the explanatory strategy and logical status of economic science in a way that is true to the phenomena and succeeds as an explanatory enterprise."

Bam. That's it. Greg nails it down.

[Troy Camplin]

Except that, according to Hayek in TSO, one can understand human action, but not the economy as a whole, because it is more complex than those trying to study it. The economy emerges out of the interactions of humans acting. We can understand one level completely, but not the other. This insight is one of Hayek's most important.

[Richard Ebeling]

This limit on the power of the mind to fully understand itself, and what the mind creates (in conjunction with other minds through social and market interaction) is a theme that Hayek discussed in a 1969 piece on "The Primacy of the Abstract," which to my knowledge has not been reprinted after its original appearance in an conference proceedings volume.

To fully understand the working of his own mind, Hayek argues, man would have to be able to step outside of himself, and have a perspective greater than what the limits of his own mind permits.

Thus, man can never fully know himself and his potentials. Even as man learns more, including about himself, he can never know if he has learned everything that there is to know.

Richard Ebeling

Richard Ebeling

[Steve Horwitz]

Richard,

"Primacy of the Abstract" is in *New Studies*.

[Roger Koppl]

Richard,

Hayek's argument about the mind understanding itself was also developed at the end of TSO and in some other places. He explicitly links it to the mathematics of Cantor and Goedel. Markose picked up on that link in her 2005 EJ paper, as I did in my Decemberr 2010 JEBO paper.

[Greg Ransom]

Hayek had invited both Popper and Polanyi to his "Analogies" seminar at Alpach, at which he presented and further developed his "The Primacy of the Abstract" paper.

Neither Polanyi nor Popper accepted the invitation.

Instead, John Watson and Paul Feyerabend took the place of Popper and Polanyi.

A number of folks working in jurisprudence and other fields also attended.

In 1962 Hayek sent a note to Popper recommending that he read Thomas Kuhn's _The Structure of Scientific Revolutions_.

Kuhn includes Hayek many-many problem of perception which block reduction in one of his papers (without citation to any source).