Introduction

Stein Rokkan would have been seventy-seven this year, if he were still alive, a veritable patriarch of European social science, looking back very likely on two more decades of his extraordinary contributions, innovations, and collaborations to the fields he enriched and wove together in the sixties and seventies. I am the lesser in years, and of course by far the lesser in stature, and never even had the privilege of collaborating with him directly, or rubbing shoulders with him at some of the great conferences of the early sixties whose attendance lists now read like a Who's Who of the great names in American social science.

But I was present when he arrived as a mature scholar like a whirlwind at Yale and on the American West Coast - he had been in the United States earlier as a junior scholar at Columbia and the University of Chicago.¹ Rokkan's whirlwind not only influenced such American luminaries as Robert Dahl, Juan Linz, Seymour Martin Lipset and others. It hit and inspired me as well, along with the many new trends in the social sciences, and transported me from a rather traditional graduate education in political philosophy and comparative constitutions and institutions at Berkeley to the brave new world of empirical and quantitative studies. This path from philosophy to empirical positivism, I gather, I shared with Rokkan and many another great name in the new social sciences of that period. This is a very important point, I believe, because a great deal of the original philosophicalthrust

---

* Peter Merkl, University of California, Santa Barbara. Santa Barbara, CA 93106. USA E-mail: merkl@sscf.ucsb.edu

sophicalthrustremains with you, the quest for universal principles and for the connections between all methods and goals of your research. No one can say of an ex-philosopher empiricist that he, or she, cannot see the forest for the trees, or is not aware of the laws of growth, of the development of the forest even when the actual research may be focused on the idiosyncratic peculiarities of just one tree. For my understanding of the translation of philosophical into empirical method, furthermore, the 'behavioral revolution'of late fifties played a big role. To use the words of one of its leading lights, Heinz Eulau, in the beginning it involved the operationalization'of of the 'big words' of political philosophy, such as power or legitimacy. If you could operationalize it in empirical terms, you could measure it, scale it, show its rise and fall; in a word, you could compare it (seeMerkl 1965, 58-86; 1969a, 141-51 (translation of Merkl 1965)).

By the time I first and briefly met Stein Rokkan at the Institute of Advanced Behavioral Studies on the hill at Stanford, in 1959 and again in 1967, of course, he had long been concentrating on the idiosyncratic social and political development of Norway. In fact, in the fifties he had developed the full panoply of insights and methods relevant to his later, universalized theories of national development from historical roots and coalitions via the advent of mass politics and the mobilization of the periphery. His work with Henry Valen and others on the single tree of Norwegian development became the core of the analytical explanation of tree development, or national development everywhere in Western Europe, perhaps in much of the developed world. In 1959, a freshly-minted PhD, I was about the same age, 27, as Rokkan in 1948 when he went to Columbia University and, working with Paul Lazarsfeld and people at the Michigan Survey Research Center, met up with the quickening of one of the mighty streams of social science research that soon came to dominate and popularize the social sciences. I am speaking, of course, of the dramatic evolution of opinion survey research, which, for better or for worse, has become an indispensable guide in every major election throughout the Western world and, increasingly, India, Japan, Russia, and ever new territories for an evergrowing of opinion subjects, from consumer preferences to religious beliefs. For Rokkan, too, it must have been a big step from the philosopher David Hume about whom he wrote his dissertation in 1948 to Lazarsfeld and the Michigan school of public opinion studies.

Now, I had still been exposed to survey research in the middle of my traditional doctoral studies at Berkeley in about 1956 or 1957 when the Michigan Center's Warren Miller was a visiting professor at the University of California. I took his seminar on public opinion research and will never forget how he made us do little research projects with the punch cards of the famous American national election study of 1952. I was cranking out possible relationships between the voters' political and religious attitudes

on a counter-sorter machine. Miller was a wonderful teacher, but I have long forgotten the result of my first little attempt at empirical research. It was still another ten years before I got involved in organizing major opinion surveys myself, in 1967/68, on social and political attitudes of Social Democratic and Christian Democratic party members in Italy and West Germany (see Merkl 1971; 1976). Anyway, the memory of cranking the counter-sorter machine reminds me of another mighty and accelerating stream of developments that had an enormous impact on us empirical social scientists everywhere, the development of advanced computers and other methods of numbers-crunching. I do not know how exactly Stein Rokkan came to this revolutionizing new technology, but I am certain that it must have been a major spur for his penchant for quantitative analysis, because it made possible the ever-faster processing of ever-larger bodies of numerical data. He had already relied heavily on quantitative data of the development of Norway, region by region, local electoral behavior, and other social and political indicators, but may well have had to make his calculations at first the old-fashioned way, with pen and paper and an ofnce-style calculating machine (see, e.g., Rokkan 1966b, 70-115; Rokkan & Valen 1974, 315-70; Rokkan 1966a, 241-66).

The new possibilities of quantitative processing of large amounts of data were highlighted further by the revival of the ecological approach that had lain dormant since the work on the 'electoral geography' of France by Andre Siegfried, long an attraction and model for Rokkan's studies of Norway. Revived by an 1950 article of James Robinson and imaginatively applied by Mattei Dogan, the statistical analysis of aggregate data for geographic areas could now be explored at a symposium organized by Rokkan and Dogan in Evian, France. The results were edited by the two scholars in a book, Quantitative Ecological Analysis in the Social Sciences (Dogan & Rokkan 1969) which went beyond the logic of political ecology and ecological inference. Rokkan in particular demonstrated his predilection exploring historical, geographic, and contextual dimensions along several, mutually reinforcing dimensions rather than only on the single track of ecological inference, and possible fallacy.² The new means of accelerated data processing also gave enormous impetus to the growth of data archives and the establishment of international data handbooks such as the Cross-Polity Survey by Arthur Banks and Robert Textor (Banks & Textor 1963) or the World Handbook of Social and Political Indicators by a team that still publishes new editions every few years.³

I shared some of Rokkan's aversion to one-sided or unidimensional quantitative analysis and felt at the time that there were too manyquantitativeresearchers seemed to follow the maxim 'have computer, will travel' without plumbing the depths of their respective subjects. But the advent of the quantitative approach did affect me in two major ways that

redirected my professional development in the sixties: One was my selfconsciousgrappling the rich methodological innovations in a series of lengthy review articles for the German Politische Vierteljahresschrift, and later the American journal World Politics and, once again, the German Zeitschrift fur Politik (see Merkl 1964; 1969b; 1977). Yes, I also got to review some of Stein Rokkan's work. The other new departure for me was the opportunity to quantify political attitudes and motivations in a wellestablishedhistorical I correlated and crosstabulated biographical and attitudinal data I had drawn from nearly 600 autobiographical statementsof pre-1933 members of the German Nazi party and its storm troopers which had been collected in 1934 by a Columbia sociologist, Theodore Abel. I had always felt a keen curiosity about the factors that motivated the early Nazis and discovered the collection of autobiographies in 1967, the year of Stein Rokkan's second stay at Stanford where my data were stored at the Hoover Library. Back in 1967, no one to my knowledge had ever undertaken quantitative research on Hfe course data, a field that now can be found in many piaces, though rarely with my focus on extremist motivations. It turned out to be a very labor-intensive, long drawn-out project which was finally published in the late seventies (see Merkl 1975; 1980a). This work also introduced me to the new historiographical subfields of social history, quanto-history and, since my research emphasized politicalsocialization motivation, psycho-history, as well as what the Social Science Research Council at that time called 'humanistic applications of computer research.' Above all, I had learned the value of combining historical study, in this case of the Weimar Republic, with social science data. Again, Stein Rokkan had taught us ahistorical social scientists the benefits of proceeding along several methodological lines at once. I had also learned from him to appreciate the importance of timing and duration, for example in the process by which a Nazi recruit gradually became fully activated as a militant engaged in street violence or proselytizing for the movement.

Stein Rokkan demonstrated magnificently the importance of timing, for example in his study of the evolution of historically rooted cleavages in Norway and other countries, and with the advent of mass participation in democratic politics. With the step-by-step process of recognition of oppositionand extension of voting rights to the point of universal suffrage, the cleavages became 'frozen' into a particular party system, as his phrase put it, because the widening of mass participation at that point in history could not go much further; every adult now could vote and, given the privilege, most of them indeed did. In the same fashion, peripheral resistancemight mobilized and incorporated in parties of resistance, and not simply absorbed into the alignments of national parties. The book Party Systems and Voter Alignments (Lipset & Rokkan 1967), or rather its famous

introduction by Rokkan and Seymour Martin Lipset is still the most widely known and cited work of Rokkan's in the United States and, quite possibly, in Europe as well, at least among specialists on parties and party systems. Rokkan was involved in many co-authored publications and, as you know, such co-authorship often raises questions as to who did what in a book. Well, in this one Rokkan is known to have been responsible for most of the theoretical sections while the substantive, historical ones reflected more of a give and take between him and his famous co-editor, Lipset. I know that I for one and many other American scholars were deeply influenced by it and by other writings of Rokkan comparing the development in different Western countries of mass participation and parties. His book, Citizens, Elections, Parties (Rokkan 1970) which brought together his previous studieson complex of subjects, really showcases his rich contributions to this discipline, to the sociology of parties, and to a long and very distinguishedinternational of sociologists of parties.

The emphatically comparative perspective of most of these studies, of course, was another lifelong preoccupation of Rokkan's social science ever since he began, in 1948, to concern himself with democratic government and its problems in industrial societies. His close association with UNESCO, the International Sociological Association, and the International Social Science Council, moreover, permitted him to give his comparative research interests the special twist that has distinguished most of his publications. He believed in well-coordinated international collaboration across the countries and cultures of the Western world which brings to bear the advantages of the most seasoned expert knowledge, and lessens the burden on any one person of having to know many complex systems inside and out (see also Rokkan 1969). The vast majority of direct political comparisons around the world has been co-authored in this fashion, including some of my own work on European party politics and party systems (see, e.g., Merkl 1980b; Merkl & Lawson 1988). I have been a comparativist all my professional life, beginning with my studies in the mid-fifties at Berkeley under the guidance of the late Eric Bellquist whose inspiration lives on through his many graduate students in comparative politics. Bellquist taught us to compare whole systems and institutions, including parties,⁴ but this was long before the burgeoning of the comparison of processes of national development which received its greatest boost with the explosion of interest in the politics of developing areas. Many former American specialists on European politics switched to Third World studies in the late fifties and the sixties.⁵ The new development theories at first were largely derived from what presumably had been learned from the European historical literature. It took a few more years for the developmentalists to find their theorizing wanting and to feel the need to 'return to Europe' for another look at European history. That famous 'return to Europe', among other notable books, produced Charles Tilly's The

Formation of National States in Western Europe (Tilly 1975) and Raymond Grew's Crises of Political Development in Europe and the United States (Grew 1978), the last volumes of the ambitious Princeton series on political development.Stein had an essay on his new macro-model of European development in the Tilly volume. The 'return to Europe' also helped Rokkan's work on Norway to receive the broader attention it deserved, beyond American specialists on Scandinavia.

So far, I have described Rokkan's work and its interaction with contemporary social scientists mostly in terms of a system of mountain streams and rivers that, in the fifties and sixties, began to increase mightily in strength at the same time that they came together in that gigantic flow of the social sciences of today: electoral and public opinion research, behaviorism in political science and sociology, accelerated quantitative of data, the ecological approach, the establishment and use of data archives and handbooks, and the comparative study of national development and of cleavage structures and party systems. Stein Rokkan appears in all this as the immensely skilful riverboat captain who instinctively the varying strength of the currents and who could successfully maneuver through the whirls and eddies of the raging waters that would have caused others to spin and get swamped with water. Or perhaps we should look at him as the crafty hydro-engineer who organized the international teamwork needed to tame and channel the newly emerging hydrodynamics of Western social science as we know it.

But actually his story continued into the exploration and organization of yet another field of developmental insights that blossomed in the last years of his life and has been carried on since by Derek Urwin and others. The foundation for these insights was laid early in his career when Rokkan was mapping out the national development of Norway, its center and periphery in history and geography, and then inspired others to do the same for the other European nation-states. Couldn't such a map of centers and peripheries in the history of individual European countries be developed for the whole of Europe over the centuries? Strong encouragement in this direction also came from a 1965 conference of the International Social Science Council which recommended that the Council establish closer links with historians and, between them and the generalizing social scientists, launch seminars and conferences on comparative national development, state-formation, and nation-building. This program eventually resulted in the two volumes on Building States and Nations, one on models, theories, and data resources, and the other on development patterns of different regions of Europe.⁶ The first regional workshop, in 1968 near Gothenburg, focused on European development, particularly in the Low Countries and Scandinavia, but also on the formation of territorial centers, economic and cultural links, and local and regional boundaries. Comparisons with the

non-European world brought out the distinctiveness of Western state- and
nation-building.7

In one chapter of this work, Stein Rokkan set forth his 'conceptual map of Europe' in which I saw, for the first time (1973), his telltale trident of economic, cultural, and military penetration from one center to its periphery, three directions of differentiation in large-scale territorial systems. Roman empire in its heyday was the original great center of European development, and even its military collapse (in the fifth century) still left much of its economic and cultural (church) networks intact to form the trading belt of European cities in the middle of Europe, including Bergen. New military and bureaucratic centers then arose along the Western Eastern rims of this city belt, such as the crusading empires of Iberia and Hapsburg and the monarchic states of France, Britain, Sweden, and eventually Prussia. By the nineteenth century, these new rim states and empires had succeeded in taking over much of the trading city belt and forced the rest of it, the German and Italian states, to try and form their own nation-states. The end result was and is a polycephalic Europe.⁸ The history of European development, moreover, was accompanied by evidence of the by now familiar survival of distinct cultural identities, their politicization, the evolution of new games of organized peripheral resistance and central responses.⁹ Rokkan's 'conceptual map of Europe' had ushered in a new, geopolitical kind of history of Europe and its constituent states.

I was fascinated when I first saw this conceptual map, but must confess that at first I did not appreciate the more elaborated versions of this revival of geopolitical analysis of European development. This occasion has made me reread it all in greater detail, and now it dawns on me that I have been working in that vineyard all along, if without the guidance of Stein Rokkan and his disciples. After all, my rather voluminous doctoral dissertation of 1958 not only dealt with West German and historical German federalism which by definition involves the various ways in which a national center tries to accommodate the organized resistance of the peripheries. Rather to the surprise of my doctoral committee, I also mapped out the boundaries and geographic patterns of West German interest group and party organization, religions, and of economic geography in an effort to demonstrate the 'federal societal patterns' underlying the governmental patterns of federalism. I tried to understand other federal systems in the same manner.

And in the late seventies and since the eighties, I have been working on a study of the rural and microcommune periphery of the already somewhat peripheral state of Bavaria, where I was born and grew up, a periphery within a periphery. These center-periphery patterns are also governed by laws of centrality that separate small, medium and large centers from my microcommunes and minitowns. Actually, 1 got involved in this VolkswagenFoundation-supported because I had discovered how the

Bavarian state government in the seventies was trying to eliminate all communes below a certain size by forcing them to join other communes or to give up their autonomy and become members of so-called local administrativeunions As a result of this 1978 reform, the number of Bavarian communes shrank from over 7000 in the sixties to about 2000 today, with the planning-minded state bureaucracy nearly taking over what used to be such a prominent part of the rural periphery in Bavaria. One of my indicators of changing peripheral status was not only the size of the surviving local community but, with it, the rather obvious changes in local party systems and electoral patterns: In local units larger now than 3000 residents, the national parties - Christian, Social or Free Democrats, perhaps also the Greens - tended to be the dominant presence. In units under 3000 and, until the seventies, in nearly six out of seven of the original 7000 communes, local parties and independentvoters' tended to predominate in the elections of First Mayors, county, and communal councils. Does this sound familiar from Rokkan and Valen's study of Norway? I also looked into instances of organized local resistance and of state responses to the local protesters against the reform to gauge the extent of the 'mobilization of the periphery' against the centralizing state bureaucracy.¹⁰

I did not set out consciously to learn this approach from Stein Rokkan, although I must have seen it in his study of Norwegian national development. truth is that I too am a 'son of the periphery,' of the Bavarian periphery of Germany and that I was investigating with great sympathy the fate of a periphery within that periphery, especially of the rural and smalltown microcommunes in depressed West Middle Franconia where even some ancient medieval towns with their original walls, turrets, and gates still intact seemed unable to keep out of the grasp of the Bavarian state. Of course Germany itself, while hardly a periphery of Europe, still shows signs of being a periphery of the United States which has penetrated it militarily, economically, and with its popular culture for the last fifty-odd years. As an American citizen now, living and teaching in California for the last 40 years and intensely interested in many countries, I also consider myself a 'cosmopolitan' son of the periphery, much like the great man we are commemorating today.

---

NOTES

1. For biographical background, see Allardt & Valen (1981,11 -38). Also Daalder (1979).

2. See esp. the introduction by the two editors, 'Quantitative Ecological Analysis: Contexts, Trends, Task' which also appeared in Social Science Information (Rokkan & Dogan 1967).

3. See also the remarks of Karl Deutsch, Richard Merritt, Bruce Russett, and Erwin Scheuch in Merritt & Rokkan (1966, chapters 2, 4, and 5).

---

4. A typical expression of the nature of my graduate training, namely the mixture of political philosophy and comparative government, was my 1967 textbook, Political Continuily and Change (Merkl 1967).

5. I revived the comparative parts of my earlier text, Political Continuity and Change in rewritten form in Modern Comparative Politics (Merkl 1970; 1977), and this time incorporated the new developmental literature of the sixties. There was a developmental of sorts in my earlier text, but it was along the lines of a history of liberalism in the West.

6. This development was to be basedon the earlier data resources and archives discussed above, such as Merritt & Rokkan (1966), Rokkan (1966c), and Dogan & Rokkan (1969), as well as Rokkan (1968). See also the preface to Eisenstadt & Rokkan (1973, 5-8, 13-14).

7. See Rokkan (1973, 13-25). See also the tables and explanations on European nations (ibid., 240-51).

8. Ibid., 73-97. The European parts of Rokkan's essay,' Dimensions of State Formation and Nation-Building: A Possible Paradigm for Research on Variations Within Europe' in Tilly (1975, 562-600), and Derek Urwin's posthumously edited Rokkan essays (Rokkan & Urwin 1983) make largely the same points except for further details and refinements. Rokkan also accounted for the Völkerwanderung of the 3rd through 6th century A.D. and the establishment of the United Netherlands and the Swiss confederacy for the diffedrent kinds of peripheries 'colonized' in the course of the formation of new centers: External and subject peripheries, 'interface peripheries' between competing centers, enclave peripheries in mountainous backwaters like Friuli and the Grisons, and 'failed center peripheries' like Bavaria. See esp. Rokkan & Urwin (1983, ch. land 2).

9. Ibid., chapters three to five.

10. For details, see Gibson (1985, 3-28) and Merkl (1985, 29-122). My collaborator, P. Gibson, and I also found that the autonomous communal life of this old rural periphery had largely shrunk and shrivelled up years ago as a consequence of the automotive revolution, educational reforms, and the transformation of Bavarian agriculture since 1945.

REFERENCES

Allardt, E. & Valen, H. 1981. 'Stein Rokkan: An Intellectual Profile,' in Torsvik, P., ed..
Mobilization, Center-Periphery Structures, and Nation-Building (Festschrift for Stein
Rokkan). Oslo: Norwegian University Presses.

Banks, A. S. & Textor, R. B. 1963. A Cross-Polity Survey. Cambridge, MA: The MIT Press.

Dogan, M. & Rokkan, S. 1967. 'Quantitative Ecologica! Analysis: Contexts, Trends, Task.'
Social Science Information, VI, 35^7.

Dogan, M. & Rokkan, S. 1969. Quantitative Ecological Analysis in the Social Sciences.
Cambridge, MA: MIT Press.

Eisenstadt, S. N. & Rokkan, S., eds. 1973. Building States and Nations. Beverly Hills: Sage.

Grew, R., ed. 1978. Crises of Political Development in Europe and the United States. Princeton,
NJ: Princeton University Press.

Lipset, S. M. & Rokkan, S. 1967 Party Systems and Voter Alignments. New York: Free Press.

Merkl, P. 1965. 'Behavioristische Tendenzen in die amerikanischen politischen Wissenschaft.'
Politische Vierteljahresschrift, VI, 58-86.

Merkl, P. 1967. Political Continuity and Change. New York: Harper & Row

Merkl, P. 1969a. 'Behavioristic Tendencies in American Political Science," in Heinz Eulau,
ed., Behavioralism in Political Science. New York: Atherton Press.

Merkl, P. 1969b. 'Political Cleavages and Party Systems,' World Politics, XXI, 469-85.

Merkl, P. 1975. Political Violence Under the Swastika: 581 Early Nazis. Princeton, NJ: Princeton
Press.

Merkl, P. 1976. 'Party Members and Society in West Germany and Italy,' in Wildenmann,
R., ed., Form und Erfahrung fur die Demokratie, Festschrift fur F. A. Hermens. Berlin:
Duncker & Humblot.

Merkl, P. 1977. 'Wohin steuert die amerikanische politisene Wissenschaft?,' Zeitschrift fur
Politik, 163-94.

Merkl, P. 1980a. The Making of a Stormtrooper. Princeton, NJ: Princeton University Press
(reprinted by Westview Encore editions 1987).

Merkl, P. 1980b. Western European Party Systems. New York: Free Press.

Merkl, P. 1985. 'Territorial Reform and Bavarian Local Politics: Patterns of Resisitance,' in
Merkl, P., ed., New Local Centers in Centralized States. Lanham: University Press of
America.

Merkl, P. & Lawson, K.., eds. 1988. When Parties Fail. Princeton, NJ: Princeton University
Press.

Merritt, R. & Rokkan, S., eds. 1966. Comparing Nations: The Use of Quantitative Data in
Crossnational Research. New Haven: Yale University Press.

Rokkan, S. 1966a. 'Electoral Mobilization, Party Competition, and National Integration,' in
LaPalombara, J. & Wiener, M., eds., Political Parties and Political Development. Princeton
NJ: Princeton University Press.

Rokkan, S. 1966b. 'Norway: Numerical Democracy and Corporate Pluralism,' in Dahl, R.,
ed., Political Oppositions in Western Democracies. New Haven: Yale University Press.

Rokkan, S., ed. 1966c. Data Archivesfor the Social Sciences. Paris: Mouton.

Rokkan, S., ed. 1968. Comparative Research Across Cultures and Nations. Paris and The
Hague: Mouton.

Rokkan, S. 1969. 'International Cooperation in Political Sociology,' in Allardt, E. & Rokkan,
S., eds., Mass Politics. New York: Free Press.

Rokkan, S. 1970. Citizens, Elections. Parties. Oslo: Universitetsforlaget and New York: David
McKay.

Rokkan, S. 1973. 'Centre Formation, Nation-Building, and Cultural Diversity: Report on a
UNESCO Programme,' in Eisenstadt, S. N. & Rokkan, S., eds., Building States and Nations.
Beverly Hills: Sage.

Rokkan, S. & Urwin, D. 1983. Economy, Territory, Identity: Politics of Western European
Peripheries. London: Sage.

Rokkan, S. & Valen, H. 1974. 'Conflict Structure and Mass Politics in a European Periphery,'
in Rose, R., ed., Electoral Behavior: A Comparative Handbook. New York: Free Press.

Tilly, C, ed. 1975. The Formation of National States in Western Europe. Princeton, NJ:
Princeton University Press.