Mattei Dogan

An encounter between sociology of science and political science

Is scientific progress in political science built on the labor of thousands scholars, or, at the contrary, is it primarily generated by the work of a restricted elite?

In sociology of science, two theses confront each other. The first thesis is known in the literature as the "Ortega hypothesis": "The majority of scientists help the general advance of science while shut up in the narrow cell of their laboratory" (Ortega y Gasset, 1932, p. 84). According to this thesis, science is rarely advanced by "breakthough"; knowledge depends of the activity of many scholars. This interpretation is based on two assumptions: "that the ideas of the ancrage scientist are both visible and used by the outstanding scientist; that the minor work is necessary for the production of major contributions. In short, that the work of the ancrage scientist is indispensable if science is to advance" (Cole and Cole, 1972, p. 368).

The second thesis is known as the "Lotka-Price law": roughly 50 percent of all scientific papers are produced by approximately 10 percent of the scientists. "What remains problematic is the extent to which the 10 percent of the scientists who produce 50 percent of the research publications are dependent on the other 90 percent of research scientists and the 50 % of the total research they produce" (Cole and Cole 1972, p. 369). The reply to this question is of crucial importance for legitimizing the hierarchy of scientists and for the organization of scientific communities.

In two seminal books, Robert Merton has analysed with great pertinence the unequal distribution of scientists, and has formulated some of the findings in terms of the so called "the Matthew's effect" (Merton 1965 and 1973).

In a comprehensive inventory of the field of sociology of science drawn by Harriett Zuckermann (1988) there is no a single reference to the social sciences, dimension neglected also by Merton and by most sociologists of science. Notable exceptions are the book edited by T.C. Halliday and M. Janowitz (1992) and the book edited by K.W. Deutsch, A.S. Markovits and J. Platt (1986). Time is ripe for a bridge between sociology of science and political science.

Two recent books show clearly that the cleavages between specialties within the disciplines are deeper than the borders between disciplines. The Handbook of Sociology (1988) contains 20 specialties which are separated from each other like watertight compartments in large ships. The editor N. Smelser writes in his preface. "There appears to be no present evidence of an overarching effort at theoretical synthesis… and little reason to believe that such an effort is on the horizon", adding that "we have by now gone too far down the road of specialization and diversification" (p. 12).

R. Goodin and H.D. Klingemann in their introductory chapter to A New Handbook of Political Science, in search of what they call "integrators" (authors cited in at least five of the nine subdisciplinary parts), have found only 21 authors, among 1631 contributors, who stir over the borders of several specialties. Can these few authors "integrate" a discipline which counts thousands of professionals? The editors of this important book would have made a wiser choice by recognizing the predominant trend of specialization.

The thesaurus of this discipline consists in thousands of books and tens of thousands of articles. The Library of Congress in Washington contains today about 300.000 books considered by librarians as belonging to political science, about half of them published since the second World War. In this figure are not counted books relevant for political science but authored by sociologists, economists, historians or philosophers. Most books of great value published in other languages than English are not included in this account.

In the Social Science Citation Index were included each year, for all social sciences, during the period 1980-1994, between 1.300.000 and 1.900.000 citations of articles published in some 1500 journals (more than 90 percent in English). About 6 percent of these citations are referring to political scientists. Such a proportion transcribed in absolute figures gives between 8.000 and 10.000 citations per year. The number of citations had been smaller for the period 1966-1973 (about half a million each year) (S.S.C.I. Comparative Statistical Summary). It can estimated that political scientists had accumulated in the last third of century about 1.200.000 citations in periodicals published in English (assuming that 6 percent of all citations in social science concern directly political scientists).

A distinction is needed between scholars who confine themselves to teaching, most of whom undoubtedly are competent scholars, and the political scientists who have written books or articles that have attracted attention, as attested by the citations records during the last third of century. Considering exclusively the authors cited in the S.S.C.I., the number can fluctuate between 20.000 and 25.000, among whom a small fraction of Europeans, Asians and Latin Americans.

Is anyone capable to swim in this ocean? Most scholars are familiarized only with the literature of their own specialty, and are superficially informed about other specialties. Compendiums and treatises are structured in specialties, and leading authors are cited mostly by scholars working in their own domain. Citations from outside the specialty are rather rare, and they tend to be perfunctory. There are good reasons to manifest some scepticism concerning the accuracy of statistics on citations (Cf. Dogan and Pahre, 1990). It remains nevertheless the richest source of information for evaluating scientific productivity and professional reputation. Following the examples of its use in the natural sciences, citations patterns in political science should be conducted along the borders of specialties. A careful reading of the literature of sociology of science shows clearly that studies on scientific reputation and eminence are referring always to specialized domains and not to the entire disciplines. This is true for the research done by Lotka, Price, Cole, Kuhn, Merton, Zuckermann, Colhoun, Crane and the other sociologists of science.

In contemporary science, eminence is rooted in specialties. Geniuses, such as Copernicus or da Vinci are monuments of the past. The Nobel prizes are offered today to highly specialized scientists, working most of the time at the interstices between disciplines. In the social sciences, seminal theories are sectoral theories. Parsons was the last synthesiser of entire discipline, but he has lost the battle.

Given these constraints, the high flyers — about 2000 among 20.000 research political scientists — have to be selected by sub-disciplines and by periods. Even in the some sub-discipline the readership may vary greatly. Take, as example, two comparativists: Fred Riggs has been cited during the quinquenium 1981-85, 254 times (Klingeman, Grofman and Campagna, p. 263) and David Apter 261 times during the same period). The two lists of citing authors overlap very little because the two cited authors have dealt with different areas and advocate different theoretical frameworks. The fragmentation of the readerships by sub-disciplines is visible for methodological schools. Those who adhere to rational theory have little in common with the Marxist oriented scholars. Periodization is also needed. Gabriel Almond, Robert Dahl, Heinz Eulau, for example, belong to an older generation; Pierre Bourdieu, Margaret Levi and Norman Nie to another. They do not compete on the same territory.

To test the two theses ("Ortega hypothesis" and "Lotka Price law") we may select, among the 20.000 political scientists, the 10 percent outstanding scholars by criteria of number and quality of publications, innovative contributions, professional status and reputation as revealed by the number of citations. We could as well select only the top 5 percent or to enlarge the elitist category to 15 percent. The results would not be the same, but the threshold is of secondary importance. What matters is the configuration of the discipline, split into specialties.

In political science, as in all sciences, most scholars are specialists, not generalists, as evidenced by the fact that they are interested in a limited number of subjects and do not follow the scientific activity in other specialties. Sociometric studies show that they entertain relations with colleagues working in the same domain and that they are poorly informed about research in other domains. A specialist in American political system may ignore the literature on social stratification in Europe, the specialist of migrations may be a stranger in the domain of electoral systems. Because of this diversification, reputation can be established, eminence recognized, effective leadership observed at the level of domains, fields, research areas and schools. For this reason, attempts to rank the high flyers for the entire discipline are rather superficial and risk to fall, as in sports, into a kind of star-system.

The regular "political scientists and the high flyers coexist. Even, if empirical evidence does not validate the Lotka-Price hypothesis concerning the contribution of political scientists to knowledge, the role of the high flyers remains obvious.

Some tentative estimations can be suggested. The 2000 high flyers have been apparently cited each one during the period 1966-1990 in average 200 times. All together they have been cited 400.000 times.

The 18.000 regular scholars have seemingly been cited in average 40 times during the same period. According to such an estimation they would have accumulated all together 720.000 citations and have thus contributed almost twice that the high flyers, at least in terms of citations, which may not offer a sufficient base of evaluation. But the Ortega hypothesis and the Lotka-Price hypothesis are also based on citations patterns.

Why some scholars are more productive, more imaginative, more successful than others? This is a difficult question to which it is not yet possible to give a satisfactory reply. Nevertheless, three varieties of high flyers can be described here: the hybrid scholars, those educated in selected greenhouses, and the pioneers of dense fields.

Hybridization at the top. As already emphasised, there is little communication between specialties within the formal disciplines, but, paradoxically, there is a considerable interchange between cognate specialties across disciplines (Dogan, 1996, 1997, 1998, forthcoming 2001). This interchange can be formulated as a theorem: in the social sciences there are fundamental questions and issues of lesser importance; the more important a problem is, the more complex are the causes; when the causes are many, there is a greater need for an interdisciplinary approach.

Barring exceptions, it is not possible to inquire into the major phenomena of civilization within a strictly monodisciplinary framework. Only by taking up position at the cross-roads of many branches of knowledge can one try to explain the collapse of democracy in the Weimar Republic, the implosion of the Soviet Union, the proliferation of giant cities in the Third World, the decline of the United Kingdom in the last 50 years, the phenomenal economic growth of Japan, the fall of the Roman Empire, the absence of a socialist party in the United States or how a child learns to speak.

Whenever a question of such magnitude is raised, one finds oneself at the intersection of numerous disciplines and specialities. In a library catalogue a- book can be included in several sections at the same time, but the actual book can be placed only on one shelf. Where should librarians place Karl Wittfogel's book on Oriental Despotism, or Gunnar Myrdal's The American Dilemma, Louis Dumont's Homo Aequalis. Alfred Kamark's The Tropics and Economic Development, or Joseph Schumpeter's Capitalism, Socialism and Democracy? Under economics, history, sociology, political science, geography or agronomy? In these major works, numerous subdisciplines, or rather numerous specialities, join hands. The analytical index to Paul Bairoch's De Jericho à Mexico, villes et économies dans l'histoire or to Barrington Moore's Social Origins of Dictatorship, and Democracy may show 15 to 20 specialities. Many great books, past and present, could be referred to here.

In the cumulative index to the seven volumes of the Handbook of Political Science, published under the direction of F. I. Greenstein and N. W. Polsby (1975), more than 3500 authors are listed. Among those who are cited at least 12 times, about half can be considered to be scholars working in hybrid fields. Needless to say, the degree of hybridization varies greatly.

Among the hundred or so major innovations listed by Karl Deutsch and his colleagues in their Advances in the Social Sciences, two-thirds lie at the intersection of various disciplines or specialties.

The higher one goes up the ladder of innovations, the greater are the chances of attracting the attention of other scholars, to be cited by them, and consequently to become high flyers.

Greenhouses and networks. A significant relationship has been established between publication record, frequency of citations, professional reputation and undergraduate or graduate education in the most prestigious colleges and universities. This relationship appears clearly in some countries such as Britain, France, Japan and the United States. The privileged educational institutions are greenhouses for elite recruitment in all sectors of the society . An important literature is devoted to them. In a study focusing on political science it has been shown that a high proportion of American political scientists have graduated from the most prestigious universities (Klingemann, Grofman, Compagna, 1985). This recruitment has occurred early in life. Professional achievement came much later. Have they been successful in their career because they were well selected from the beginning, or because they have benefited from a privileged intellectual environment during their studies, and also because they had the opportunity to build a network of personal relations in the high strata of the academic circles? Retrospectively, the first hypothesis (good initial selection) seems to be validated, even if we take into consideration the relatively high proportion of intellectual failures among graduate students in the most selective universities (these who have abandoned their studies and never finished their dissertation).

The second hypothesis (privileged intellectual environment) suggests the notion of elitist network. The success is explained, at least in part, but the stimulating academic entourage. The two explanations are not incompatible, they may have a cumulative effect. To the question: who are the high flyers the reply is clear enough: they come in large number from the high places of learning, at least until the recent decade.

Leaders in dense fields. Increasing the amount of effort applied to any given subfield should in principle increase the amount of research results. In reality, after some time of high creativity, "research in a given area is ultimately subject to the law of diminishing marginal returns" (Dogan and Pahre, 1990, p. 29). As a result, "a major proportion of what participants see as innovative work is completed before the field has begun to acquire a significant proportion of its eventual membership" (Lemaine et al., 1976, p. 5). The addition of more scholars may increase the amount of innovative research done, but it does so at a decreasing rate. The tendency of densely populated subfields to produce less innovation, notwithstanding the greater effort applied has been called "the paradox of density" (Dogan and Pahre, 1990, p. 32). Good examples of overcrowded subfields are American presidency, American Congress, electoral behavior, centre-periphery in Norway, regional diversity in France, and more recently, comparative survey research on values. The scholars who lead these highly dense subfields, and who may have been their founders or may have played a pioneering role, have all the chances to be often cited in a perfunctory style by their followers. A minor research, like the improvement of a correlation coefficient, may be considered by densely packed scholars worth citing. Such citations count in rating scholars as high flyers.

The three types of high flyers may overlap at margins.

Calhoun, Craig, 1992 "Sociology, Other Disciplines and the Project of a General Understanding of Social Life", in Halliday and Janowitz, op. cit. 137-195.

Cole, Jonathan R., and Stephen Cole, 1972, "The Ortega Hypothesis", Science, 178, Oct. 368-178.

Crane, Diana and Henry Small, 1992, "American Sociology since the Seventies: the Emerging Identity Crisis in the Discipline", in Halliday and Janowitz, op. cit. 197-234.

Deutsch, K.W., A.S. Markovits, J. Platt, eds. 1986, Advances in the Social Sciences 1900-1980, Lantham, Md., University Press of America.

Dogan, M. and R. Pahre, 1990, Creative Marginality Innovation at the Intersections of Social Sciences, Westview Co.

Dogan, Mattei: "The New Social Sciences: Cracks in the Disciplinary Walls", International Journal of Social Sciences, 1997, 153, pp. 429-443); "Fragmentation of the Social sciences" International Journal of Social Sciences", 1994, 139, pp. 27-42; "The Hybridization of Social Science Knowledge", Library Trends, 1996, pp. 296-314; "Political Science and the other Social Sciences" in R.E. Goodin and H.D. Kligemann, eds., A New Handbook of Political Science, Oxford University Press, 1996, pp. 97-130; "La Thèse de l'Interdisciplinarité dans les Sciences Sociales", CNRS, Sciences de l'Homme et de la Société, 1998, pp. 22-27; "The Moving Frontiers of the Social Sciences", in S. Quah and A. Sales, eds. International Handbook of Sociology, Sage Publications, 2000; "Sociology among the Social Sciences" in E. Borgatta, ed. Encyclopedia of Sociology, New York, MacMillan 2000; "Specialization and Recombination of Specialties" Forthcoming 2001 : International Encyclopedia of Social and Behavioral Sciences, London, Pergamon-Elsevier Science, 2001; "Are there Paradigms in the Social Sciences" (idem) Forthcoming 2001.

Goodin, R. and H.D. Klingemann, 1996, "Political Science: The Discipline", A New Handbook of Political Science, Oxford, Oxford University Press,
pp. 3-49.

Greenstein F.I. and N.W. Polsby, eds., 1975, "Cumulative Index", Handbook of Political Science, Reading Mass. Addison-Wesley Publishing Co. pp. 1-91.

Halliday, T.C. and M. Janowitz, 1992, Sociology and its Publics: The Forms and Fates of Disciplinary Organization, University of Chicago Press.

Klingemann, H.D. B. Grofman, J. Compagna, 1989, The Political Science 400: citations by Ph. D. Cohorts and by Ph.D. Granting Institution", Political Science and Politics XXI, 2, pp. 258-269.

Kuhn, Thomas, S. 1962 (1973) The Structure of Scientific Revolutions, University of Chicago Press.

Lemaine, G., R. MacLeod, M. Mulkay, P. Weigast, (eds) 1976, Perspectives on the Emergence of Scientific Disciplines, The Hague, Mouton.

Lotka, Alfred J., 1926, The Frequency Distribution of Scientific Productivity", Journal of the Washington Academy of Sciences, 16, 317-23.

Merton, Robert K., 1965 (1985) On the Shoulders of Giants, New York, Harcourt Bruce Janovich.

Merton, Robert K., 1973, The Sociology of Science, Theoretical and Empirical Investigations, Edited by N.W. Storer, Chicago University Press.

Merton, Robert, K., 1988, "The Matthew Effect in Science, Cumulative Advantage and the Symbolism of Intellectual Property", Isis 79, 607-23.

Ortega y Gasset, José, 1932, The Revolt of the Masses, New York, Norton.

Price, Derek de Solla, 1963, Little Science, Big Science, Columbia University Press.

Smelser, Neil, J., 1988, Handbook of Sociology Newbury Park, Sage Publications.

Social Science Citation Index, 1994, "Comparative Statistical Summary" 1966-1979, and 1980-1989 and 1990-1994, pp. 61-63.