September 2022

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

« Theory at MIT | Main | Methodological Pluralism and the Austrian School of Economics? »

Comments

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

Peter,

"...
a. there are no unpublished geniuses running around;

b. as a consequence there are no undiscovered geniuses;

c. brilliance is an description of work that is bestowed by the reader not the author...."

If the goal is the search for truth, this seems somewhat restrictive.

If an anthropologist (or an amateur on vacation)finds an important 'missing link' fossil while digging in Africa, and at least suspects that his find may be significant, or may be inconsistent with current beliefs, the fact that he cannot be considered a genius, or that he will likely never publish any work in the field, doesn't mean that his find isn't an important step on the path to truth.

If you assemble all the existing intellectual papers, and all the textbooks, it is highly doubtful that you have fenced in the complete truth. Or that they contain ALL of the factors that will lead to the next generation of papers and textbooks and an approximation of the truth.

Regards, Don


Heh. Boettke the title stealer gets hoisted by his own petard by being blocked from commenting in his own blog. There IS cosmic justice!

Don,

I am willing to be convinced on this, but can you give some examples since 1950 where someone completely outside of the academic community made a break-through in the social sciences with respect to "tracking truth"?

I guess I would use an example Jane Jacobs and her work on the life and death of cities.

As I said in my original post --- I am sure that my rules of thumb have their shortcomings, but I still hold that they are useful rules of thumb.

Pete

Peter,

"...I am willing to be convinced on this, but can you give some examples since 1950 where someone completely outside of the academic community made a break-through in the social sciences with respect to "tracking truth"?..."

Actually, my point is just the opposite. There is almost no possibility that an outsider can influence an academic discipline, independent of whether his insight is valid or not. From the outsider's point of view, academic publications first and foremost serve vocational ends. Authors have little or no incentive to actively seek out and correct errors in their previous publications unless other academic insiders complain. In turn, insiders typically won't complain unless doing so provides them with a topic of sufficient weight to justify a substantial publication of their own. Nitpicking someone else's paper is unlikely to be worth the chance of that someone returning the favor on your next paper. (Note : This is all supposition on my part, as I have no way of knowing whether or not what seem like relatively clear incentives are reflected in actuality.)

The very structure of the publication system would seem to tend to produce only large scale corrections, much delayed, if at all.

In the Spring 2004 QJAE, the following paper was published :

http://www.mises.org/journals/qjae/pdf/qjae7_1_4.pdf

ON THE OPTIMUM QUANTITY OF MONEY, by Barnett and Block.

To the best of my knowledge, no subsequent comments have ever been published in nearly four years. In my opinion, having spent several years reading and thinking about Austrian monetary theory, this paper contains several serious misconceptions that lead to erroneous conclusions. I find it hard to believe that most Austrian scholars would not agree with at least some of my opinions, but there is apparently no incentive to comment, or possibly to print any comments that may have been provided. Of course, the possibility exists that few read the paper in the first place.

Regards, Don


DGLesvic is an undiscovered genius. How can you deny that you know less economics and have less political savvy than him?

No undiscovered or unpublished geniuses, eh?

Then no opportunities for entreprenuerial discovery here... so you admit we are in equilibrium!

No undiscovered or unpublished geniuses, eh?

Then no opportunities for entreprenuerial discovery here... so you admit we are in equilibrium!

ArrowDebreu raises a good objection, but I believe it is only an apparent rejection. Remember I said these were my "rules of thumb" or working hypothesis, not settled positions. They are in other words first best approximations.

I would make the same claim about market phenomena in a relatively free market. In this sense, my first rule of thumb is that if a market is competitive (by which I mean relative ease of entry) and "thick" (by which I mean lots of participants on both sides of the market), then I would start with the Becker idea of the market is in equilibrium --- P = MC. Deviations from that ideal are not, however, evidence of "market failure", but that the market hasn't worked its way out yet; the entrepreneurial process is in the process of working itself out and thus the apparent inefficiencies will be eradicated or ameliorated in short order.

I think a lot of people get the Austrians wrong when they give NO role whatsoever to equilibrium properties. I think a lot of others get it wrong when they give nothing but equilibrium attention.

The equilibrium properties as laid out in the neoclassical system are not logically flawed, it is a question of the soundness (not validity).

Once we get to that point, we have questions of realism, we have questions of institutions, we have questions of dynamics, etc., etc.

Lachmann never questioned (a) maximizing [i.e., individual equilibrium], and (b) that markets exhibited systemic tendencies. What he objected to, was the claim that all entrepreneurship was by definition equilibrating. No, he countered, the data is impacted by the entrepreneurs move --- endogenous source of change. But if for the dog the rabbit is equilibrium and the rabbit moves precisely because the dog chases after him, we still can discuss the direction of the dog! A lot of readers miss this point in Lachmann --- again as an excuse to not study formal neoclassical general equilibrium theory. But we have to study formal theory in order to make sense (if we want to be academic economists which not everyone does) of the appreciative theory of the market we come to get from the work of Mises, Hayek, and Kirzner.

At least that is how I came to understand the importance of Kirzner after I went in a decidedly Lachmann/Shackle/Loasby direction for many years. And it was actually working through the calculation debate stuff and reading Kirzner much closer than I had before that turned me on to the issues.

Pete

The comments to this entry are closed.

Our Books