Reading Andrew Kliman’s comments on Samuelson, I was put in mind of the latter’s interventions in the debate about Friedman’s “as if” methodology, and in particular of the following passage (Samuelson 1965, page 1166):
"To help my understanding of these issues, Professor Massey kindly provided me with minimal references to related writings that should be of interest to economists, namely the Hempel and Nagel references of my bibliography. I benefited from this course of study, which I believe deepened my understanding of my own position."
One imagines that Massey, contra Samuelson, thought that the immediate purpose of reading other people’s work was to deepen one’s understanding of *their* position, rather than one’s own.
Samuelson’s conclusion is perfectly reasonable in itself, but given the emphasis he places on it in this passage it is difficult to believe that a need for accuracy about others’ work was uppermost in his mind.
Julian Wells
Reference:
Samuelson, Paul (1965) ‘Professor Samuelson on Theory and Realism: Reply’, The American Economic Review, Vol. 55, No. 5, Part 1 (Dec., 1965), pp. 1164-1172.
Stable URL: http://www.jstor.org/stable/1809235 .
Accessed: 23/08/2011 03:14
Dr Julian Wells
Director of Studies
School of Economics
staff web-page: http://fass.kingston.ac.uk/staff/cv.php?staffnum=287
personal web-site: http://staffnet.kingston.ac.uk/~ku32530
Principal lecturer
School of Economics
Faculty of Arts and Social Sciences
Kingston University
Penrhyn Road
Kingston-upon-Thames
KT1 2EE
United Kingdom
+44 (0)20 8417 2341
________________________________________
From: Societies for the History of Economics [[log in to unmask]] On Behalf Of Kliman, Prof. Andrew J. [[log in to unmask]]
Sent: 23 August 2011 01:25
To: [log in to unmask]
Subject: Re: [SHOE] Is textbook HET systematically false?
Roger, I think you may misunderstand what I claim to be inaccurate (about Samuelson's paper) and what I think reflects a lack of concern with accuracy.
In my view, it is NOT inaccurate to:
1. "build a mathematical model of the views held by a group of economists who held different views, did not talk in terms of models, and did not express their ideas using the same type of mathematics as the modeler," or
2. "abstract and make assumptions that they did not explicitly make," or
3. "construct rational reconstructions."
In my view, it IS inaccurate to:
4. attribute the result of all this to the classical economists, as Samuelson did by by calling his paper “The Canonical Classical Model of Political Economy.”
“A New Model of Mine that Captures what I Believe to be the Salient Features of Classical Political Economy" would have been an accurate title (assuming that he did believe this).
To put the matter in terms of interpretation vs. basic facts:
a. Whether Samuelson's model captures the salient features of classical political economy is a matter of interpretation.
b. That Samuelson's model is not the canonical model of classical political economy is a basic fact (since it not only "abstract[s] from details of what individual economists said," but also makes up stuff that they did *not* say).
c. That Samuelson was aware that he made up stuff that they did not say is a basic fact.
d. That Samuelson selected a title for the paper that nonetheless attributes the model--including the stuff he made up and conclusions he would not have obtained without making stuff up--to the classical economists is a basic fact.
Andrew Kliman
________________________________
From: Societies for the History of Economics [[log in to unmask]] on behalf of Roger Backhouse [[log in to unmask]]
Sent: Monday, August 22, 2011 4:05 PM
To: [log in to unmask]
Subject: Re: [SHOE] Is textbook HET systematically false?
There is a sense in which Bruce's response answers this, though not directly, so I will add something brief.
If anyone builds a mathematical model of the views held by a group of economists who held different views, did not talk in terms of models, and did not express their ideas using the same type of mathematics as the modeler, it is clear that the modeler has to abstract and make assumptions that they did not explicitly make. Samuelson is perfectly clear that this is what he is doing. One may or may not like this method, and one may or may not agree with the particular rational reconstruction that results, but though there is a trivial sense in which any model is not a completely accurate description of what is being modeled, I do not find it helpful to see this as a lack of concern with accuracy. The presumption must be that those who construct rational reconstructions believe that they capture the salient features of the ideas being reconstructed. Whether or not they do this is a question of historical interpretation. Samuelson argued that his model did capture the salient features of classical economics, and Hollander responded that it did not. This is all very different from simply getting basic facts wrong in the way Bruce talks about.
Roger Backhouse
On 22 August 2011 04:06, Kliman, Prof. Andrew J. <[log in to unmask]<mailto:[log in to unmask]>> wrote:
In response to Roger:
Yes, the words "Canonical Model" signal that Samuelson was "abstracting from details of what individual economists said," as you note. But they don't signal that he was making up stuff that they did *not* say--"supply[ing] the equations missing for their additional unknowns"--and attributing this stuff, and the results to which it leads, to the classical economists. If one supplied different equations, or no equations, the results would differ from those that Samuelson christened as "canonical classical." His procedure is the mathematical equivalent of putting words in people's mouths. So when he called his model “The Canonical Classical Model of Political Economy” (rather than “A Revision and a Reworking of Some Elements of Classical Political Economy into a New Model”) he wasn't being accurate. And since Samuelson was aware that he wasn't just "abstracting from details of what individual economists said," I submit that he lacked concern with accuracy.
I agree that there's "a fundamental difference between someone like Samuelson who read widely in the history of the subject, but applied an interpretive method that many historians do not accept and those economists who claimed that Keynesians held a naive view of the Phillips curve" on the basis of their reading of secondary sources. But I think they're fundamentally different forms of the same core problem, lack of concern with accuracy.
You write that "I don't consider the Samuelson example a lack of concern with accuracy. It reflects something very different, namely a view on the appropriate way to study history." This strikes me as a distinction without a difference (or maybe a difference without a difference), since I think that accuracy isn't a central concern of the whig history that Samuelson celebrated and practiced. Yes, whig history is an interpretive method, but I don't see that lack of concern with accuracy that's a feature of an interpretive method is morally superior to lack of concern with accuracy that's not.
Andrew Kliman
________________________________
From: Societies for the History of Economics [[log in to unmask]<mailto:[log in to unmask]>] on behalf of Roger Backhouse [[log in to unmask]<mailto:[log in to unmask]>]
Sent: Sunday, August 21, 2011 10:42 AM
To: [log in to unmask]<mailto:[log in to unmask]>
Subject: Re: [SHOE] Is textbook HET systematically false?
Sorry, but I don't consider the Samuelson example a lack of concern with accuracy. It reflects something very different, namely a view on the appropriate way to study history. You say the title is misleading, yet there are strong clues in the title about Samuelson's intentions. "Canonical model" suggest that he is abstracting from details of what individual economists said, for they did not talk in terms of models. There is a fundamental difference between someone like Samuelson who read widely in the history of the subject, but applied an interpretive method that many historians do not accept and those economists who claimed that Keynesians held a naive view of the Phillips curve because, having read Friedman (1968), they deduced, without reading the Keynesian literature, that these economists must have been naive.
Roger Backhouse
On 20 August 2011 23:49, Kliman, Prof. Andrew J. <[log in to unmask]<mailto:[log in to unmask]>> wrote:
In response to Roger:
I think the common disparaging attitude to questions of “what X really meant” is clear evidence of lack of concern for accuracy.
And I think the following is “reason to doubt that economists making statements about the history of their subject believe them to be true.” In a JEL (16:4, Dec. 1978) paper in which he developed what he called “The Canonical Classical Model of Political Economy,” Samuelson wrote (p. 1423):
“it would be ahistorical to read into the classicists a full-fledged post-Clarkian model of neoclassical type. Nonetheless, if we wish to flesh out the torsos of their logically incomplete models, we must supply the equations missing for their additional unknowns. And, once we commit ourselves to ... free-entry and widely-shared knowledge, ... constant-returns-to-scale technology, and ... smooth variability of the (LtKt) components of the Vt dose, ruthless competition will enforce the neoclassical marginal productivity relations in the canonical model whether or not the classicist is yet aware of those relations and is able to apprehend them.”
Nonetheless, he called the paper “The Canonical Classical Model ...,” not “A Revision and a Reworking of Some Elements of Classical Political Economy into a New Model.”
Andrew Kliman
________________________________
From: Societies for the History of Economics [[log in to unmask]<mailto:[log in to unmask]>] on behalf of Roger Backhouse [[log in to unmask]<mailto:[log in to unmask]>]
Sent: Saturday, August 20, 2011 4:54 PM
To: [log in to unmask]<mailto:[log in to unmask]>
Subject: Re: [SHOE] Is textbook HET systematically false?
I deliberately did not speculate on the reasons for this phenomenon, though I am skeptical about whether relativism or instrumentalism play a significant role. I have no reason to doubt that economists making statements about the history of their subject believe them to be true.
I put in the remark about Debreu's Theory of Value because I anticipated that if I did not, people would respond that it was not a typical textbook, or not what they understood by a textbook, and I did not want to get into that debate. If the example makes my point, fine. If it does not, then delete the sentence.
Roger Backhouse
.
On 20 August 2011 20:11, Kliman, Prof. Andrew J. <[log in to unmask]<mailto:[log in to unmask]>> wrote:
Why is Debreu's _Theory of Value_ a bad example? Because it's not a textbook?
In any case, I think that the core problem is lack of concern with accuracy or truth. This problem is pervasive throughout the whole culture. In some disciplines, a turn toward relativism has also contributed to it, while in economics, instrumentalist attitudes toward ideas have contributed to it.
Uncovering what was actually said or took place isn't sufficient when there's lack of concern with accuracy and truth. The lack of concern needs to be challenged, too. I think Harry Frankfurt's _On Bullshit_, which identifies lack of concern for the truth as the distinguishing feature of bullshit, and argues that it is morally worse than lying, is a good place to start. Susan Haack has also written some good things about the moral implications.
Andrew Kliman
________________________________
From: Societies for the History of Economics [[log in to unmask]<mailto:[log in to unmask]>] on behalf of Roger Backhouse [[log in to unmask]<mailto:[log in to unmask]>]
Sent: Saturday, August 20, 2011 1:35 PM
To: [log in to unmask]<mailto:[log in to unmask]>
Subject: Re: [SHOE] Is textbook HET systematically false?
I suspect that the reason why we have more examples in macroeconomics textbooks is simply that, for a variety of reasons, microeconomists feel less obligation to present a historical perspective and remain silent on the history. Of course, they use labels that incorrectly imply that they are doing the same as some past economist (Marshallian theory of the firm/market/..., Coase theorem, etc), but I suspect they do not so so routinely tell historical stories. I know this is a bad example (no need to point this out) but I conjecture that Debreu's Theory of Value contains few historical errors.
There was a HOPE mini-symposium, which Roy put together, in 2005 (Vol 37 issue 2) exposing a few very bad examples. My contribution related to macroeconomics, on "Misunderstanding the history of the business cycle" but there were several others on other fields. However, whilst some cases are easy to identify, there are many incorrect claims that cannot be identified without undertaking serious historical work first, and though some people are doing this, such work is remains very limited. The volume of work being done on recent economics has increased dramatically in the past thirty years, but it is still a much smaller fraction of work on the history of economics than, in my view, it should be.
Roger Backhouse
On 20 August 2011 04:06, Bruce Littleboy <[log in to unmask]<mailto:[log in to unmask]>> wrote:
May I ask a wider question? Apparently the history of ideas is also written by the winners. “Our” team’s contributions are packaged as insightful and novel corrections to “their” prior errors and foolish misjudgements.
More is involved here than the Whig view of HET. I refer to the systematic re-touching of the evidence. I mean hatchet jobs.
JB Say and other classicists were presented in a way that made Keynesians look brilliant; the textbook Keynes is not the true Keynes; the instability of the Phillips curve was not Friedman’s discovery; Keynesians in the 1960s were not as they are routinely depicted etc.
Is macroeconomics afflicted by this more than other disciplines? Important figures write textbooks that distil what we’ve learned from tumultuous times and debates, and the profession generally trusts them.
Is there a scholarly book that exposes this apparently widespread mal-description of what happened in the history of macroeconomics? More than inevitable simplification for undergraduates is involved (i.e., it is fair if a textbook simplifies, because everybody hedges and qualifies their claims, but they may basically have meant what the summary attributes to them). Do we all apply the spin required to fit the ideas of others into our own (or our School’s) pet narrative? Or are these examples of error merely rare blemishes?
Bruce Littleboy
From: Societies for the History of Economics [mailto:[log in to unmask]<mailto:[log in to unmask]>] On Behalf Of Brad Bateman
Sent: Friday, 19 August 2011 3:20 AM
To: [log in to unmask]<mailto:[log in to unmask]>
Subject: Re: [SHOE] functional finance
The history is also told (well) in Michael Bernstein's A Perilous Progress (2001).
Brad
Bradley W. Bateman
Office of the Provost
Denison University
Granville, Ohio 43023
Office: 740 587 6243<tel:740%20587%206243>
Fax: 740 587 5790<tel:740%20587%205790>
This email has been scanned for all viruses by the MessageLabs Email
Security System.
This email has been scanned for all viruses by the MessageLabs Email
Security System.
|
