Good Greif

by Henry Farrell on July 28, 2009

I’ve been thinking for the last ten minutes about how to start this post in a polite and reasoned way, and coming up with nothing very much. So, here goes with the bluntness. Charles Rowley’s “article”:http://www.springerlink.com/content/q6pq851k12738423/fulltext.pdf (behind paywall), ‘The curious citation practices of Avner Greif:
Janet Landa comes to grief’ in the new issue of _Public Choice_ is one of the sorriest hack-jobs that I’ve ever had the misfortune to read in an academic journal. The fundamental claim:

The commentary will establish that Avner Grief, by his curious citation practices, has failed to cite seminal papers by Janet Landa who pioneered in an important field of nonmarket decision-making—the economics of trust, which itself is part of the economics of identity—and that, by so doing, he now is inappropriately recognized among mainstream economists as the pioneer in the field of trust networks and informal enforcement of contracts.

is not supported, the insinuations are offensive, and the standards of argument are situated somewhere in that narrow region where the slipshod, the haphazard and the positively eccentric intersect.

Preliminary business before starting this post. I know Avner Greif to the extent that I have met him once, he read a bit of my dissertation when I was a not-especially-promising-sounding graduate student and said helpful things, and we swapped a couple of emails back and forth in the late 1990s/early 2000s. I haven’t been in contact with him since 2001 or so, when I may have cc-ed him on an email to one of his co-authors (I can’t honestly remember). I met Janet Landa at the first ISNIE conference in the late 1990s, and she seemed both nice and interesting. My sole acquaintance with Charles Rowley is that I have read his introduction to the “Edward Elgar Public Choice reader”:http://www.e-elgar.co.uk/Bookentry_DESCRIPTION.lasso?id=585, which, _inter alia_, describes public choice as a “program of scientific endeavor that exposed government failure coupled to a program of moral philosophy that supported constitutional reform designed to limit government,” and suggests that the program’s opponents are “scholars who had rendered themselves dependent on the subsidies of big government and whose lucrative careers in many instances were linked to advising … agents of the compound republic.” In contrast, my acquaintance with the literature on the economics of trust and the enforcement of informal contracts is reasonably extensive: I have a “book”:http://www.cambridge.org/uk/catalogue/catalogue.asp?isbn=9780521886499 on precisely that topic coming out shortly.

Rowley makes what seems to me to be a quite serious set of charges. He suggests that Avner Greif, a very well known economist, has systematically obscured the contribution of another scholar, Janet Landa, in ways that enhance his own reputation and damage her ‘intellectual property’ in the ideas that she put forward.

the informal scholarly rules concerning citations are grounded on a trust relationship between members of the community of scholars that they will assiduously
review the relevant scholarly literature for papers predating their own contributions,and that they will conscientiously cite all such prior contributions. This trust relationship
has been clearly severed in this case. Ironically, the intellectual property right that Avner Greif has expropriated from Janet Landa by his failure to cite her published works concerns the role played by trust relationships as an informal mechanism for facilitating trade in the absence of formal trading markets.

Janet Landa laid the groundwork … Her fieldwork and subsequent data analysis revealed that Chinese middlemen were not just a random collection of Chinese traders. Rather they were linked together in complex networks of personalized or particularistic exchange relations to form an ethnically homogeneous middleman group … first, and most important paper on the economic analysis of trust and identity … was published in _The Journal of Legal Studies_ (Landa 1981). … her theory of the EHMG is applicable to other kinship-related and ethnically homogeneous traders … Notably, the ethnically homogeneous Italian Mafia group is a low-cost club for the effective enforcement of contracts in an extra-legal environment. Similar arguments hold with respect to such homogeneous traders as: “the East Indians in East Africa, the Syrians in West Africa, the Lebanese in North Africa, _the Jews in Medieval Europe,_ (emphasis in original with snarky footnote saying ‘I have italicized the fourth example to indicate by just how far in time Janet Landa anticipated the supposedly path-breaking work of Avner Greif.’) and the Medici merchant-bankers in fifteenth-century Florence”

bq. … In 1994, Janet Landa integrated her previous writings on trust, ethnicity identity into a major book … The book develops a unifying theme: trust and identity matter for traders operating in an environment characterized by contract uncertainty, where the legal framework for the enforcement of contracts is not well developed. … One might have thought, by the late 1980s, that Janet Landa’s reputation as the pioneer of the economic analysis of trust was securely established … this has not come to pass. Instead, Avner Greif, by his curious citation practices, has attenuated Janet Landa’s intellectual property right.

bq. Avner Greif is very much a ‘Johnny-come-lately’ to the trust relationship literature. … completing his baccalaureate degree … when Janet Landa published her seminal 1981 paper. … not at all surprising, therefore, to learn that Greif’s 1989 and 1993 papers are now under critical scrutiny following a challenge from Edwards and Ogilvie … In his first (Greif 1989) paper, Avner Greif cites Janet Landa’s unpublished 1978 doctoral thesis as well as her unpublished 1988 Hoover Institution Working Paper … published papers are not cited by Greif. … terse comment on this 1988 unpublished paper … Again, he speaks tersely about her 1988 paper … Greif reviewed her book … review is not at all unfavorable; but it completely fails to acknowledge Janet Landa’s intellectual property right in the economic analysis of the ethnically homogeneous networks/groups. … notes that this explains why ‘she does not take advantage of game theory in analyzing situations that are inherently strategic’ … his own major book … Listed among the voluminous references to this 500-page tome are just the two unpublished works …

Greif was awarded a prestigious MacArthur Foundation Fellowship … Now famous, with a great number of citations … always somehow overlooking Janet Landa’s publications when compiling his citation lists. … Later-arriving mainstream scholars, perhaps themselves unwilling to expend time on researching the early literature, also bypassed Janet Landa’s scholarship and cited Avner Greif as the pioneer in the field. … In the July 2007 issue of Public Choice, Robert D. Tollison criticizes scholars of the new political economy research program—notably Persson and Tabellini (2003) and Acemoglu (2005)—for their failure to cite the preceding scholarship of the founding fathers of public choice …Tollison acknowledges the important contributions of these well-known Johnny-come-lately scholars.

This amounts to a quite unpleasant set of accusations. Words such as ‘curious citation practices,’ ‘expropriated,’ and ‘always somehow overlooking’ suggest that Greif has deliberately obscured his debt to Landa’s original arguments in order to burnish his own reputation, and that his failure to cite her has led others to overlook her work. The problem is, though, that the evidence that purportedly supports these claims is sorely inadequate.

Landa’s book is a smart and good one. But it is hardly the only book to consider this set of issues. The reason that Janet Landa was aware that mediaeval Jewish merchants had ethnically based trading networks is not that she had sieved through the original cartolaries herself. It’s that there was already a literature on this topic, as indeed there was on other trading networks. There is a thriving literature among anthropologists and sociologists on kinship and economic exchange, the origins of the gift relationship, and sundry other related topics.

Nor was Landa’s the only book that sought to apply basic economic arguments to the understanding of these relationships. Michael Taylor’s _Anarchy and Cooperation_ (1976) and _Community, Anarchy and Liberty_ (1982) use results from game theory to argue that non-state communities can support quite broadly based forms of trust and cooperation was published in 1976. Diego Gambetta’s original work on the Mafia and trust (see the relevant chapter “here”:http://www.nuffield.ox.ac.uk/users/gambetta/Trust_making%20and%20breaking%20cooperative%20relations.pdf ) was published in 1988 – neither Gambetta, nor any of the other contributors to this pathbreaking book on trust seems to have been aware of Landa’s work, despite frequent overlap in their arguments. As far as I can tell, Lisa Bernstein’s pathbreaking 1992 “article”:http://www.jstor.org/stable/724403 on the reputational basis for non-contractual relations among Jewish diamond traders doesn’t cite Landa either (bloody legal footnotes – but I both skimmed and did a wordsearch and Landa’s name popped up on none of these). The point is rather obviously _not_ that all of these people were deliberately trying to avoid citing each other so as to lay claim to the whole field, unless the academic conspiracy spread even further than Rowley seems to suggest. It is instead, that there was a happy coincidence between the development of new theoretical tools in the 1970s and 1980s and empirical information (much of which had already been considered by sociologists, anthropologists and historians) which seemed readily explicable using these tools. Hence, people like Gambetta (working on trust and the Sicilian Mafia), Landa (working on ethnic Chinese traders) and Bernstein (working on New York diamond merchants) could come to similar or related conclusions independently working with similar intellectual tools, and similar kinds of networked relations among individuals.

Finally, there is a far more obvious explanation for Greif’s phenomenal citation rankings among economists vis-a-vis Landa (and me, and most other social scientists). Given the choice, economists _are far more likely to cite to formal than to non-formal work._ As the piece mentions, Landa doesn’t use formal models, even though she uses economic arguments extensively. Since the same is true on both counts of my own forthcoming book, I think there is absolutely nothing wrong with this. But I don’t expect economists to be citing my arguments in any very great number. Economists like formalism, and, if pushed, will defend their liking for it. “Paul Krugman”:http://web.mit.edu/krugman/www/formal.html, for example:

bq. Most of the topics on which economists hold views that are both different from “common sense” and unambiguously closer to the truth than popular beliefs involve some form of adding-up constraint, indirect chain of causation, feedback effect, etc.. Why can economists keep such things straight when even highly intelligent non-economists cannot? Because they have used mathematical models to help focus and form their intuition. … professional economists also have another task: to communicate with each other, and in so doing to help economics as a discipline progress. In this task it is important for your colleagues (and students) to understand how you arrived at your conclusions, partly so that they can look for weak points, partly so that they may find other uses for the technical tricks you used to think an issue through.

Like this or dislike this, it is a sociological fact. And one which (together with the genuinely substantial theoretical innovations that Greif is responsible for) is a far more plausible explanation than the one that Rowley advances.

I could go on. The Grandpa-Simpsonesque complaints about ‘Johnny Come-Latelys’ that veer into mutterings about how nobody pays attention to the Founders of Public Choice (among whom Rowley perhaps numbers himself) anymore. The gleeful citation to a working paper attacking Greif’s basic research and the failure to cite to Greif’s (as best as I can tell, as a non-expert) thoroughly convincing “response”:http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1153826 to that paper (isn’t this a bit rich for an article complaining about purportedly dodgy citation practices?). But you get the point. This article looks likely to create much embarrassment for a prominent political economist. But that economist isn’t Avner Greif.

Hat tip: “John Sides”:http://www.themonkeycage.org/2009/07/citation_bias_the_case_of_avne.html.

{ 13 comments }

1

Brad DeLong 07.29.09 at 12:17 am

It was amazing to see…

2

rea 07.29.09 at 1:52 am

somewhere in that narrow region where the slipshod, the haphazard and the positively eccentric intersect

If academics don’t work out for him, he could always write speeches for Sarah Palin.

3

temporarily anonymous 07.29.09 at 4:37 am

A true wonder in the annals of bizarre, ad hominem journal articles that make everyone wonder how they got through peer review is the article John French and Danny James published (pay-walled) a few years ago in Labor[1]. It was an extended (21 page) article attacking John Womack, questioning his sanity, mocking him for his low productivity recently, and–seemingly without conscious irony–complaining that he attacked his intellectual opponents personally, rather than purely intellectually. The content itself made it both bizarre and kind of a pleasure to read–I actually read bits of it aloud to my partner–but so too did their choice of venue (a journal overwhelmingly North Americanist in content and subscribers, and French, James, and Womack all study Latin America) and the fact that the journal then published a reply by Womack and then a counter-reply by French and James again.[2]

[1] John D. French and Daniel James, “The Travails of Doing Labor History: The Restless Wanderings of John Womack Jr.,” Labor: Studies in Working-Class History of the Americas 4 (summer 2007): 95-116.
[2] John Womack Jr., “On Labor History, Material Relations, Labor Movements, and Strategic Positions: A Reply to French and James (as Nice and Civil as I Can Make It),” Labor: Studies in Working-Class History of the Americas 5 (summer 2008): 117-123 (not currently pay-walled); John D. French and Daniel James, “Polemics and an ‘Army of One’: Responding to John Womack, Jr.” Labor: Studies in Working-Class History of the Americas 5 (summer 2008): 125-129. (Also seemingly available to all.)

4

Chris Dornan 07.29.09 at 10:36 am

How did that article get out–the reputation of Rowley? Horrible to behold, even if there were some truth in it.

5

John Quiggin 07.29.09 at 11:08 am

I’m very much an outsider as far as public choice is concerned. Still it seems to me that Rowley is not particularly notable as a contributor to the literature but rather as an academic entrepreneur, boundary rider and so on. So, this kind of exercise is less surprising than from someone with an academic reputation to defend.

6

monboddo 07.29.09 at 1:13 pm

But, but…this would imply that someone at GMU is a careless, vindictive hack…

7

DaveMB 07.29.09 at 2:29 pm

Rowley appears to be a co-editor-in-chief of the journal Public Choice, which may mean that this piece got “editing” analogous to that of Peretz’ pieces in The New Republic.

8

Dan Nexon 07.29.09 at 4:42 pm

The piece is so outrageous and craven that I think a more useful time would be spent discussing the general issues it raises about citation practices. I’m also curious what others make of the notion of someone having “intellectual property rights” to a subject. At one level, it doesn’t make much sense to me. No one really “owns” a subject.

But we do often expect certain scholars to be cited on specific subjects, which strikes me as pretty much the same thing as saying they have an “intellectual property right” to the topic. And surely, sometimes, those scholars are cited for reasons unrelated to the actual “merit” of their contribution in the subject area. Individual citation choices may have ripple effects on the status of a piece of scholarship down the road. Scholarly careers can be made or broken on whether their work becomes a part of the “citation cannon” for a particular domain of research.

So what should we do–if anything–to make sure that scholars get their “just desserts” for their work? Once we move beyond the domain of plagiarism, what kinds of rules of justice and fairness apply to citation practices?

9

Henry 07.29.09 at 4:59 pm

Dan (about to head to Ireland and have been meaning to phone you for weeks about lunch – will you still be around next academic year in yr new responsibilities?) – this is a more interesting question for discussion. There is a fair amount of work on the sociology of citation, I think, and there are obviously norms. But a norm of attribution does not necessarily equate to a property right in having your work attributed. I may feel highly annoyed when someone doesn’t cite my pathbreaking work on _XXXX_, but I don’t feel (or, at least, _I_ don’t feel) that my property rights are being abrogated. Rowley’s theory is perhaps based in a kind of crackpot Lockeanism, perhaps not.

Monboddo – since I don’t otherwise know Rowley or his work, I am not going to comment on ’em. Could be that he is otherwise an intellectually meticulous supergenius whose hobbies comprise rescuing fluffy kittens from burning skyscrapers and discovering the cure for cancer. But this article is a real piece of work.

10

monboddo 07.29.09 at 8:49 pm

OK, maybe a little unfair to Rowley, but too tempting a line to pass up.

11

LFC 07.30.09 at 3:12 am

D. Nexon: “But we do often expect certain scholars to be cited on specific subjects, which strikes me as pretty much the same thing as saying they have an ‘intellectual property right’ to the topic.”

No, I do not think this is “pretty much the same thing.”

D. Nexon asks: What rules of justice and fairness apply to citation practices?
Since it is virtually impossible to cite all the relevant literature on any topic of any consequence that has been around for any length of time, it seems to me that there must be a large element of judgment and discretion in citation practices. The notion that, as a broad proposition, “rules of justice and fairness” apply to citation practices seems very dubious to me. Rules of accuracy and attribution obviously apply, and I think we all know what those are. But these are not the same as rules of justice and fairness. If, for example, I write something and I take an idea directly (or indirectly) from Nexon, then I damn well better cite Nexon for that idea. But if I write about a topic that Nexon has written about and I don’t cite Nexon (say in a string citation of “see XYZ on [the general topic]”), I don’t agree that I am being “unfair” or “unjust” to Nexon. In that case I may (or may not) be revealing myself as a shoddy, careless, lazy, and/or inferior scholar, but that’s my problem. I don’t think fairness or justice much enters into it. That, at least, is my off-the-cuff reaction.

12

Dan Nexon 07.30.09 at 5:44 am

“Property rights” is obviously a deeply strained way of looking at it, but the underlying notion seems to be that X has some connection with research program Y such that any mention of Y requires a mention of X. And we see claims to that effect all the time in academia–particularly in peer reviews. I’m not saying that’s a good practice. One of “Nexon’s rules for peer reviewing” is that criticisms for “failing to invoke a work” should be coupled with reasons why invoking that work might change or impact the argument.

But there’s also a sense in which one has a responsibility to find out what’s been published in an area of work and to avoid replication without citation. That’s been a semi-regular topic of discussion here (think Kieran’s complaints about physicists and social network analysis, or the “Eurovision” incident). In my subfield (International Relations), I do think there’s certain laziness about citing articles from certain journals (*cough* IO *cough*) when articles in other journals are actually more on point.

And there’s also the larger question of how citation practices impact the careers of scholars; certainly there are cases in which scholars routinely fail to cite work they’re familiar with, and even if those reasons are benign, that failure helps to marginalize the people they’re failing to cite.

Yes, of course, like Henry I’ve read articles or books and thought “WTF” when I don’t see a reference to my work, but I’ve learned to get over it. What I do see (particularly when reviewing) are scholars who aren’t adventurous enough to stray beyond one or two journals–or “must read” articles–and thus miss stuff that they’re replicating or really ought to deal with in order to improve their argument. On the other hand, I’ve had reviewers criticize me for not including a reference to some article that makes the same point as a number of other articles in a string citation, and at least twice that’s because I needed to reduce my word count.

Which raises another issue: how the decreasing maximum length of journal articles encourages us to skimp on citations…..

13

Dan Nexon 07.30.09 at 5:46 am

Wow, that was a babbling comment. Must to bed.

Comments on this entry are closed.