Gordon Tullock as a political scientist

Abstract

We consider Gordon Tullock’s impact in political science, focusing on his influence as a scholar and as an academic entrepreneur. It is common to think of Tullock as a “natural economist,” but his formal training at Chicago encompassed considerable coursework related to political science. We consider three sources of information to draw conclusions about Tullock’s contributions in political science: (1) Course syllabi; (2) Citations in academic political science journals; and (3) Impact on the careers of important political scientists, and shaping the intellectual agenda. Our conclusion is that, while Tullock’s work is clearly significant for central questions in political science, and has received some attention, his primary legacy lies in the impact he had on launching and shaping the careers of prominent political scientists, and thus the development of political science scholarship.

Appendix: Political scientists remember Gordon Tullock

1.1 Steve Brams

I always found Gordon a prickly but stimulating character. He and Jim Buchanan invited me to talk about a paper in Blacksburg in the early 1970s, which later appeared as an article with Bill Riker, “Models of Coalition Formation in Political Science.” This article was published in Mathematical Applications in Political Science, VI (1972), which was one of my early publications that we would have had difficulty publishing elsewhere. The same holds for other articles of mine that appeared in Public Choice, beginning in 1974.

Gordon had an opinion on all sorts of things, which he was quite willing to express. My coauthors and I had two exchanges in print with him in the APSR, once after my article with Riker, “The Paradox of Vote Trading,” appeared in 1974, and another time after my article with Peter Fishburn, “Approval Voting,” appeared in 1978. In the first case, Gordon was not happy that vote trading, which he and Buchanan had championed in The Calculus of Consent, could give a Pareto-nonoptimal outcome. In the second case, his animus seems a little more obscure–perhaps against the moderates that approval voting might elect, perhaps because it went against the grain of making unanimous consent a property of good decisions.

Ironically, Gordon’s dissent from (or may outrage with) our findings drew attention to our articles that they might not otherwise have received. Others later chimed in, both pro and con, so we probably benefited from the attention the controversies that Gordon provoked caused.

1.2 Bernie Grofman

I do indeed have very special memories of Gordon’s profound influence on my early career. My first submitted paper was my MA thesis (on which Duncan Black was an external reader, while he was visiting at the University of Chicago). While still a graduate student at the University of Chicago, I submitted that to Papers on Non-Market Decision-Making, the original title of Public Choice, and it was eventually accepted by Gordon after the third round of revision. As you know, in those days, Gordon was not just the editor he was the sole reviewer. Knowing that I was still a graduate student he was incredibly encouraging, giving me feedback after my first submission and then after resubmission, and encouraging me to try yet again. But Gordon’s influence on my early career does not end there. In 1970 Tullock had been asked by the President of APSA to write an essay analyzing the APSA Presidential and Council elections, and his (co-authored) analysis of the 1970 election was published in PS in 1971. He was asked to do it a second time, but opted out; instead he asked me whether I would like to write such an essay on the next APSA election. I delightedly said yes. I was still not yet a Ph.D., but I had already accepted a position as an Instructor at SUNY Stony Brook—a job that having an already—published article in Papers certainly helped me get. APSA provided me with a computer tape that had the images of the actual election ballots on it (with all voter identification stripped away) in a readable format. I believe my 1972 essay in PS on “The 1971 APSA Election” was one of the first computer analyses of raw ballot data at the individual level for a real world relatively large-scale election. I was able to do what were, for the time, some very innovative things, such as looking at the number of unique ballot patterns and looking at conditional vote probabilities across candidates and offices. Without Gordon’s having nominated me, there is no way that I, an unknown graduate student just starting his political science career, could possibly have been chosen to do the analysis of an APSA election. At the time he nominated me to do the APSA analysis, we had still not ever met. He knew me only from correspondence. In short, while I later observed Gordon to be often gruff and combative in person, and one who regularly emphasized the limited role that altruism played in the real world, I have direct evidence that he could also be one of the kindest people around.

1.3 Bill Keech

Gordon Tullock was an editor who felt that his judgment was better than that of anyone else. He may well have been right, and this stance worked in my favor. One wonders how he would have explained his practice in terms of the insights of public choice.

1.4 Mat McCubbins

I met Gordon Tullock at the Public Choice Meetings in 1983. He was outspoken, energetic, curious, and engaging. I felt honored to meet him as fresh-faced PhD. I used the Calculus of Consent and the Politics of Bureaucracy in my classes then, and I still do. We interacted at meetings, and I dealt with him often when he was editor of Public Choice. As editor, he encouraged the discipline to engage with the question of “why so much stability?” He was also a fine mentor, who helped me, and many others I knew, to bring out their best ideas in the pages of Public Choice. Few editors I have known have been so helpful and delightful to deal with. I had not spoken to him in years and I was sad to hear of his passing in 2014.

1.5 Gary Miller

Gordon had an enormous influence on the publication of my first political science paper–on bureaucracy and game theory. This was not exactly a “hot” topic back when Carter was president. Gordon’s encouragement meant a lot to me, and his professionalism as an editor gave me faith that the political science discipline was large enough and flexible enough to encompass my idiosyncratic interests.

1.6 Nicholas Miller

By 1972, when I first (unsuccessfully) submitted a paper to Public Choice, the journal had a secondary editor (first Peter Ordeshook and later Ken Shepsle) who handled more “mathematical” (essentially social choice related) submissions. Peter or Ken handled all of my various submissions in the 1970s, so I actually never dealt directly with Tullock as PC editor. Incidentally, Peter and Ken used a regular referee process, whereas of course Tullock typically made decisions on his own, and he usually did so within no more than 48 h. I recall that PCS folks would observe that Tullock didn’t necessarily disagree with the principle that journal editors should get reviews from the world’s foremost experts on the topic of a submission, but that 95 % of the time he determined that expert was himself.

I do remember a few other interactions with Tullock, however. First, when I visited VPI (as it was then generally known) as a job candidate (in the Political Science department) in March 1971, I asked to visit the Public Choice Center. Gordon welcomed me, showed me around, and introduced me to Duncan Black (who was visiting at the time), with whom I was able to talk for 20 min or so. I also remember that Tullock was recruited to drive me from Blacksburg to the Roanoke airport. (I was heading from there to my job interview at UMBC.)

Second, in December 1973, I sent Tullock a copy of “Logrolling and the Arrow Paradox” (which would become my first publication) that I had just submitted to Public Choice through Ordeshook, as I thought it might interest him. I didn’t necessarily expect a reply, but I received quite long and detailed letter in reply, which he must have prepared (much like his editorial decisions) more or less instantaneously. (My letter was dated December 19 and his reply December 19.) As I recall, I sent him several other things (including a copy of my dissertation) as well in those early years and that he always responded promptly and in an encouraging way.

A final remembrance is more recent. In 2002, I sent written comments to Bob Erikson on a paper on “Voting on Many Issues, One at a Time” that he had presented at the Public Choice Meetings. I wrote: “It struck me that Gordon Tullock made an argument along the same lines many, many years ago. Yesterday, I checked some of his stuff and confirmed my recollection. In Chapter 2 of his book Toward a Mathematics of Politics, based on what he calls “pencil [and compass] exercises”, Tullock concludes that it is “obvious” that “if the number of dimensions is [equal to or] less than the number of individuals minus one,” the Pareto region will assume one form, which will change as the N-1 threshold is crossed. This chapter is vintage Tullock. He reaches insightful conclusions based on sheer intelligence undisciplined by formal analysis or (despite the title of this book) mathematics, doing this years or decades before the Ordeshooks, McKelveys, or Eriksons, etc., of the world pin them down rigorously (and show that Tullock got things more or less right). This is pattern is particularly exemplified by Tullock’s next chapter “The General Irrelevance of the General Impossibility Theorem,” which argues that majority cycling and a two-dimensional space is unlikely to be much of a problem with many voters. And in passing, Tullock essentially defined the “yolk” years before formal theorists did.

1.7 Ken Shepsle

From 1975 to 1980 I served as the “mathematical political science” editor of Public Choice. Gordon was the overall editor. This meant that I was responsible for one of the four issues published each year. Peter Ordeshook has preceded me in this capacity.

This was, I believe, an act of generosity on Gordon’s part (but not entirely selfless). Gordon very much wanted this part of the public choice community to prosper, even though it really wasn’t his part. He was in the business, along with the other founders, of assembling an intellectual coalition, and the modelers in political science were an important part of that. He gave a young pup like me complete discretion over editorial matters for that one issue. (And, I might brag, nearly every year a paper from my issue won the Duncan Black Prize, the most notable of which was Romer and Rosenthal’s famous paper.)

1.8 Barry Weingast

The early 1980s were a very productive time for me, with many on-going strands of research. My 1981 Public Choice paper with Shepsle represented one of these strands.

Tullock had published his paper, “Why so much stability?” that became a touchstone for the literature. He posed exactly the right question, and Shepsle and I wrote one of several papers that Tullock published on this topic. We titled our paper, “Structure-induced equilibrium and legislative choice.” I believe this paper won the prize for the best paper in Public Choice that year.

This paper, along with Shepsle’s 1979 paper, “Institutional arrangements and equilibrium in multidimensional voting models,” helped validate the idea of structure-induced equilibrium as a means of understanding how institutions solved the problem of cycling. This approach served as the basis for a great many works, including Krehbiels 1988 Pivotal politics (although Keith does not mention the concept). Cycles in preferences still existed, but institutions prevented a legislature from continuing to cycle. Our paper also highlighted the central importance of agenda institutions. Institutions that control the agenda play a major role in shaping legislative outcomes.

So, we can say that the series of papers that Tullock published on “why so much stability?” had a significant impact on how political scientists studied legislatures.

For Shepsle and me, this line of research culminated in the publication of our 1987 APSR paper, “The Institutional Foundations of Committee Power.” Two other papers in the series were our 1984 AJPS paper on the “uncovered set” and a 1984 JOP paper on, “When do Rules of Procedure Matter?”