Michael Desch argues that American political science has become increasingly focused on technique and that the growing prevalence of mathematical and statistical methods has rendered the discipline irrelevant to policy makers. Desch’s essay is open to several criticisms. It does not engage with relevant sociological thought such as Max Weber’s and Karl Mannheim’s theorization of the relationship between social science and politics. It seems to narrowly equate political relevance with advising policy elites as opposed to critiquing these elites and the social order they dominate. The essay also contains some factual inaccuracies; the suggestion that a “majority consensus” (386) of political scientists opposed the Vietnam War is dubious at best. ^{Footnote 1}

Here I want to raise another issue. Desch implies that political scientists embrace quantitative techniques largely for “internal, disciplinary reasons” (378). He depicts technical specialization as an attempt by political scientists to increase their autonomy and protect their turf from encroachments by the government or by political interests. For Desch, the government is a force for “relevance” and whenever it “mobilizes” political science—which occurs mainly in times of war—the emphasis on technique recedes. But “once the war is over or the threat recedes, disciplinary dynamics once again” push the discipline toward technical specialization (382).

Without gainsaying the internal dynamics discussed by Desch (who rightly points out that trends of technical specialization are common to other academic fields), I want to suggest that he greatly underestimates the extent to which the valorization of statistical and mathematical methods in political science—rather than being emblematic of an estrangement between political science and the government—has actually been caused or at least reinforced by the government. The government has contributed significantly to the growth of technical political science, especially in the context of war and national security threats. Government funding played a key role in the development and diffusion of data analytical and mathematical techniques.

Consider the National Science Foundation, which was created in 1950 as an outgrowth of the activities of the Office of Naval Research and whose stated purposes included “secur(ing) the national defense.” ^{Footnote 2} In the late 1950s political science came to be included in the social science program of the NSF, and it is hard to overestimate the impact of that inclusion on the methodological orientation of the discipline. The American National Elections Studies—a massive data gathering and analysis effort that, more than any other project, symbolizes and strengthens the prestige of quantification in political science—depends almost exclusively on NSF funding. ^{Footnote 3}

Desch makes two passing references to recent Congressional restrictions on NSF funding of political research. He implies that these cuts are a bad thing but, if he is truly interested in curbing the technical specialization of political science, Desch should consider supporting the Congressional budget cutters, for few events have contributed to the privileging of quantitative and mathematical approaches to political science as much as the earlier Congressional decision to add the discipline to the mandate of the NSF.

The momentum toward quantification in political science received crucial support from other government agencies, especially national security agencies and allied foundations who hoped that rigorous social science techniques would aid the government in prosecuting World War II and then the Cold War. Let me offer several examples to establish the plausibility of my claim. The mobilization of social scientists during World War II produced not only qualitative analyses by the likes of Herbert Marcuse (which Desch discusses) but also cutting edge quantitative methods and methodological training. The Morale Division of the U.S. Strategic Bombing Survey (created by President Roosevelt in late 1944) made major technical advances in survey research. Gabriel Almond was a member of the division and the methodological tools he acquired while participating in a survey of German attitudes after the Nazi capitulation helped him in his subsequent quest to place the study of foreign policy and comparative politics on a more scientific, i.e., “behavioral,” footing. ^{Footnote 4} Almond’s mentor, Harold Lasswell, pioneered quantitative content analysis methods during World War II in his capacity as the director of the Experimental Division for the Study of War-Time Communication. Lasswell’s unit was part of the government’s intelligence apparatus and it served as a training center where young political scientists such as Karl Deutsch, Heinz Eulau, and Ithiel de Sola Pool acquired advanced methodological skills. ^{Footnote 5} Pool went on to become a key member of MIT’s government-supported Center for International Studies in the 1950s and 1960s, where he developed computer simulation techniques in the service of the government, including COMCOM, a model of communication in communist societies. ^{Footnote 6} COMCOM represented but one of many government investments in computer simulation methods. The scandal-ridden Project Camelot, which “drew extensively” on the work of political scientist Harry Eckstein, aimed to develop (in the words of its official title) computer simulation “Methods for Predicting and Influencing Social Change and Internal War Potential.” ^{Footnote 7} Finally, government support facilitated the creation of the infrastructure for the quantitative study of world politics. In the 1960s and early 1970s the Department of Defense and NSF made sizable investments (at least by the standards of social science) in the development of quantitative data sets such as the World Data Analysis Program at Yale University and the Dimensionality of Nations project at Northwestern University. ^{Footnote 8}

In the aftermath of the attacks of September 11, 2001, social science funding by national security agencies, which dried up in the 1980s and 1990s, has been resumed, culminating with the launching of the massive Minerva initiative by the Department of Defense in 2008. ^{Footnote 9} Desch appears to expect Minerva to mitigate, temporarily at least, the tendency of political science to equate rigor with technique. But if the past is any guide, the tightening relationship between political scientists and the government may accelerate rather than slow down the technical specialization of the discipline.

I do not doubt the sincerity of policymakers who conveyed to Michael Desch their frustration with the limited usefulness of political science research. But the fact of the matter is that the policymaking community is at least in part responsible for this state of affairs. If Desch wants to make political science less focused on technique, more accessible to the public, and more relevant to policy debates, he should probably seek to increase the distance between political science and policymakers rather than narrow the gap between them.