P 86 - 6

Social Sciences and Political Projects: The Emergence and Demise of Reform Coalitions between Social Scientists and Policy-Makers in France, Italy, and West Germany

by Peter Wagner

For publication in: Stuart Blume, Joske Bunders, Loet Leydesdorff, Richard Whitley (eds.), The Social Direction of the Public Sciences. Sociology of the Sciences Yearbook XI, Dordrecht, Reidel, 1987. Earlier versions of this paper were presented at the conference on "The Social D ire c tio n o f the P ublic Sciences", Amsterdam, Novermber 1985, and at two workshops o f the Theory and Methodology Group o f the Swedish Collegium fo r Advanced Study in the Social Sciences, Amsterdam, December 1985, and Uppsala, April 1986. I would like to thank the participants in the discussions for valuable comments and criticism s.

Final Version, May 1986 Zusammenfassung

In diesem Beitrag wird die Entstehung einer Policy-Orientierung - einer Konzentration der Forschung auf Politikfelder oder auf politisch­ administrative Prozesse im Hinblick darauf, Voraussetzungen für "bes­ sere P olitik" zu schaffen - in den Sozial Wissenschaften in Frankreich, Italien und der Bundesrepublik Deutschland als Interaktionsprozeß von Wissenschaftlern und Nicht-Wissenschaftlern untersucht. In einem be­ stimmten gesellschaftlichen Kontext gingen Sozialwissenschaftler Koali­ tionen mit den Teilen der politischen und wirtschaftlichen Eliten ein, die die Notwendigkeit eines umfassenden g e se lls c h a ftlic h e n Modernisie­ rungsprozesses sahen, und definierten ihre eigene Aufgabe als die der Vorbereitung der erforderlichen politischen Innovationen auf der Grund­ lage sozial wissenschaftlicher Expertise. Nachfolgende soziale Entwick­ lungen führen zu einer Desillusionierung und zur Suche nach anderen gesellschaftlichen Akteuren zum Engagement in auf Forschung gegründeter politischer Aktion.

Der vorliegende Aufsatz versteht sich als Beitrag zu einer politischen Soziologie der SozialWissenschaften. Im Unterschied zu neueren Entwick­ lungen der Wissenschaftsforschung, die zu einer Betonung der mikroana­ lytischen Ansätze führten, wird die Auffassung vertreten, daß es mög­ lich is t, Einflüsse auf wissenschaftliche Entwicklungen makroskopischer und la n g fr is tig e r Natur auszumachen, wenn die Untersuchung des wissen­ schaftlichen Feldes und der Beziehungen verschiedener Forschergruppen in diesem Feld mit der Analyse der Beziehungen zwischen diesen Gruppen und nicht-wissenschaftlichen Akteuren in verschiedenen gesellschaft­ lichen Kontexten verbunden wird. Fälle intensiver Interaktion zwischen Wissenschaftlern und Nicht-Wissenschaftlern, die größere Gruppen von Forschern umfassen und über längere Z e it andauern, werden als g e e ig n e ­ ter Ausgangspunkt fü r die Entwicklung eines solchen Ansatzes angesehen. Summary

In th is paper the emergence o f a p o lic y o rie n ta tio n —a concentration o f research efforts on policy areas or on politico-administrative pro­ cesses with a view to "improving policy-making"—in the social sciences in France, Italy and West Germany is analysed in terms of an inter­ action between scientists and non-scientists« In a specific societal context social scientists joined reform coalitions with those parts of the po litica l and economic elites who saw the need for an incompassing societal modernization process, and defined their own task as designing the required political innovations on the basis of social science ex­ pertise. Later social developments led to a disillusionment with this approach and to the search for new social actors to engage in re­ search-based political action.

With such a perspective this paper is understood as a contribution to a political sociology of the social sciences. In contrast to recent dev­ elopments in science studies which led to an emphasis on micro-level analysis, it is argued, that the search for ordering principles of a macroscopic and long-term nature in scientific developments can be fru itfu lly pursued when linking the study of a scientific field and of the relations inside the field between different group researchers with the analysis of the relations between these groups and actors outside the field in a specific societal context. Cases of intense interation between scientists and non-scientists, which involve a larger number of researchers and last over some time, are seen as a promising starting- p o in t fo r e la b o ra tin g such an approach. - 1 -

Introduction

In the post-World War II history of the social sciences in France, Italy, and West Germany a distinct period can be detected in which a policy orientation was (re-)introduced into these disciplines. In this period major research efforts, either on specific policy areas or on the politico-administrative processes themselves, were undertaken with a view to improving policy-making by putting it on a "scientific" or "more rational" basis. These processes took place in the 1960's and early 1970's and thus followed previous developments in the USA where the term "policy sciences"^ was coined, but they differed significantly from the

US experience. The specific nature of the confrontation between the innovative approach and the established national science traditions and the interaction of the emerging policy researchers with actors in different politico-administrative systems had a significant influence on 2 the shape the process took.

The key actors were groups in the political and economic elites who saw the need for an encompassing societal modernization process, including the introduction of rational planning procedures and the formulation of reform policies which would enable all social groups to participate in 3 economic and societal progress. Social scientists sometimes joined outright reform coalitions with these groups, shared their basic political convictions and saw their own task in designing the required

1 Daniel Lerner, Harold Lasswell (eds.), The Policy Sciences, Stanford: Stanford University Press, 1951 2 For an attempt to analyse the long-term development of problem-oriented social science research see: Georg Thurn, Peter Wagner, Björn W ittro ck, Hellmut Wollmann, "The Development and Present State of Public Policy Research. Country Studies in Comparative Perspective", mimeo, Berlin 1984, an interim report from a research project presently pursued at the Wissenschaftszentrian Berlin and the Freie Universität Berlin in cooperation with the Swedish Collegium fo r Advanced Study in the Social Sciences. In the sections analysing the developments in Germany this paper partly draws on Hellmut Wollmann's analysis in this project. For further references on social science developments in Germany as well as in France and Italy see this report. 3 For an attempt to describe related processes in Austria in terms of a conflict between traditional and modernizing political forces, see Bernd Marin, Die Paritätische Kommission. Aufgeklärter Technokorporatismus in Österreich, Vienna: Internationale Publikationen, 1982, in particular part III. - 2 - political innovations on the basis of social science expertise. In Germany and Italy, the scientists' interlocutors were mainly members of groups in the major political parties, which had not. yet reached power positions, in France social scientists were more oriented to the process of "pianification", which as such took a more long-term perspective on social developments and seemed more open to heeding scientific advice and to in it ia t in g socia l re fo rm .T h e s e intense in te ra c tio n s between scientists and non-scientists, however proved to be of a short-term nature. Recognizing political and scientific deficiencies in their approach to a science-politics interaction, a learning process which was speeded up by social developments, researchers turned away from these coalitions. Some then started to search for new social actors to engage in research-based political action.

In the following an attempt w ill be made to analyse these developments by tracing the social processes which brought about the new orientation in the social sciences. To do this, an analysis at the macro-level—some concepts of the state of the discipline and of socio-economic change in­ fluencing disciplinary perspectives—w ill have to be tied to one at the micro-level dealing with the behaviour of particular actors and their strategies. In this regard, the three-country-comparison may serve to elucidate the relative importance of certain "macro-structural" determinants as compared w ith m icro -le ve l c h a ra c te ris tic s s p e c ific to the situation.

A comparative analysis starts out with im plicitly rejecting the idea of a world-wide sim ilarity of social science developments, an idea which could be based either on the assumption of universality of scientific

4 This difference may in part explain why in Germany political science was more involved in these processes, whereas in France sociologists played a major role. However, other factors enter into this, and I shall return to it below. The rationale of choosing different disciplines—sociology in France and Italy, political science in Germany—for the following analysis may be questioned, but this comparison, in my view, is the most appropriate here, as the interaction between actors in the science system and actors in the political system is to be stressed. Some unevenness in the argument, what concerns the comparison of disciplinary developments, is thus to be accepted. For sim ilar reasons the three countries were chosen for th is a n a lysis, as they were marked by short periods o f a p a rtic u la rly intense debate about modernization and concomitant social science- politics interaction. - 3 -

progress or on the diagnosis of an American intellectual hegemony which dominates developments in dependent countries. Whereas the firs t argument probably need not be discussed among sociologists of the sciences, the second one is worth mentioning. A dominance of social science concepts of American origin in the post-World War II period, which was also strengthened by a Europe-oriented social science policy 5 of some major U.S. foundations, can certainly not be denied here, but this doubtlessly strong impact does not lead to the same developments. As w ill be shown below, the actual debates, conflicts and coalitions are simultaneously shaped by deep-rooted intellectual traditions, structures and sizes of scientific fields and national political peculiarities.

With such a perspective this paper is understood as a contribution to a ß political sociology of science as it was advocated some time ago. For a researcher on scientific developments it is interesting to note that the programmatic t i t l e o f a book published in 1974, "Toward a P o litic a l Sociology of Science", reappears identically as a section title in a book published in 19867. Though a decade is not such a long tim e fo r a

science to develop, it is worthwhile to ask what has happened to the political sociology of science in between. Basically one might agree with Aant Elzinga's diagnosis that a "shift in the social paradigm of science" from structuralist-functionalist approaches in the study of the community of scientific investigators to actor-oriented approaches has o led to the loss of a macro-sociological perspective. This perspective, 5 European socia l s c ie n tis ts could compare th e ir view on the development of their disciplines with the one taken by Norbert Wiley, "The Current Interregnum in American Sociology", Social Research, 52. 1985. 179-207, who is careful enough to talk only about the ll.S.A.. His interesting attempt to relate social science developments to the "long waves", known from economic theory merits a closer consideration, which cannot be started here. 6 Stuart Blume, Toward a Political Sociology of Science, New York: Free Press, 1974 7 The third section in Björn Wittrock, Aant Elzinga (eds.), The University Research System: The Public Policies of the Home of Scientists, Stockholm: Almqvist and Wiksell, 1985, with contributions by S tua rt Blume, Tony Becher and Aant Elzinga Q Elzinga, "Research, Bureaucracy, and the D rift of Epistemic Criteria", in: Wittrock, Elzinga, op. c it., 1986. Note 7. 210 and 214ff. See also R. Whitley: "By insisting on jumping immediately to the micro-level of every-day scientific research, the ethno-scientists ignore the social processes by which some variations - 4 - according to his point of view, has to be reinstated now without losing the advances made by the micro-studies in the meantime.

If one then tries to analyse in detail the play of different forces which shape scientific development, one has to come to the conclusion, as Pierre Bourdieu phrased it some years ago, "that it is pointless to distinguish between strictly scientific determinations and strictly social determinations of practices that are essentially ------overdeter-- mined." Such a finding does not, however, lead to the conclusion that the search for ordering principles of a macroscopic and long-term nature in s c ie n tific developments must be abandoned, as some seem to have concluded. It only means that things are more complex than they have previously been perceived to be. The study of a scientific field and of the relations inside the field between different schools and groups of researchers has to be linked w ith the study o f the re la tio n s between these groups and actors outside the fie ld .

have become dominant." ("From the sociology of scientific communities to the study of scientists' negotiation and beyond", Social Science In fo rm a tio n , 22. 1983. 710) q Pierre Bourdieu, "The Specificity of the Scientific Field and the Social Conditions of the Progress of Reason", Social Science Information, 14. 1975. 21 (original emphasis)

Similarly see Michael Poliak, "Organizational Diversification and Methods o f Financing as Influences o f the Development of Social Research in France", Elizabeth Crawford, Norman Perry (eds.), Demands fo r Social Knowledge, London: Sage, 1976, p. 131. Bourdieu in the above-mentioned article continues (p. 34f., emphasis added): "any s c ie n tific f ie ld . . . may be situ a te d between the two lim its represented at one extreme by the religious field ..., in which official truth is nothing other than the legitimate imposition (i.e. arbitrary imposition misrecognized as such) of cultural arbitrariness expressing the specific interest of the dominant - inside and outside the field - , and at the other extreme by a scientific field from which every element of social arbitrariness ... would be banished." Though this does not represent the core of Bourdieu's concept of science studies it is nevertheless astonishing that the programmatic implications also for a study of the macro-societal context of scientific developments seems to have escaped most of his readers. An obvious exception is Michael Poliak, see the above-cited article and "La planification des sciences social es", Actes de la recherche en sciences sociales 1976. 105-121 . Bourdieu applies this approach to the case of Martin Heidegger's philosophy in: "L'ontologie politique de Martin Heidegger", Actes de la recherche en sciences sociales 1975. - 5 -

Cases of intense interaction between scientists and non-scientists, which involve a larger number of researchers and last over some time, are a fru itfu l starting-point for elaborating such an approach. As in the case of university-industry relations, analysed by Stuart Blume in his c o n trib u tio n to th is volume, the emergence o f a p o lic y -o rie n ta tio n in the social sciences can be characterized as "the construction of a research programme in relation to a particular sort.of social network, which consciously embodies the goals, needs, interests or aspirations of actors within this network"^. The following analysis is devoted to such c o a litio n s between socia l s c ie n tis ts and p o litic a l actors, "common social projects", in Blume's terms, and to their demise and the related emergence o f new, p o lit ic a lly d iffe r e n tly orie nte d, c o a litio n s and research programmes. As an analysis of the social sciences it thus deals with struggles about the production and imposition of legitimate representations o f the social world and w ill have to pay p a rtic u la r attention to political actors and their positions in the different 12 societal contexts. The importance of the political context will already become clear from a short description of the situation of the social sciences before the p o lic y -o rie n ta tio n emerged.

The Social Sciences in an Early Phase of Institutionalisation

The processes under analysis here, by which a policy orientation emerged in the scientific fields of sociology in France and Italy and of political science in West Germany, took place mainly in the period between the la te 1950's and the e a rly 1970's. At the beginning o f th is period, the social sciences in these three countries were characterised by a low degree of academic institutionalisation, by strong links to philosophy and to normative thinking and by lively debates about consequences to be drawn from f i r s t experiences w ith modern, methodolog­ ically rigorous empirical research techniques.

S tu a rt S. Blume, in S tua rt Blume, Joske Bunders, Loet Leydesdorff, Richard Whitley (eds.), The Social Direction of the Public Sciences. Sociology of the Sciences Yearbook X l, Dordrecht, Reidel, 1987 12 Katrin Fridjonsdottir seems to take basically similar perspectives in her analysis of the development of sociology of work in Sweden, in Blume, Bunders, Leydesdorff, W hitley (e d s .), 1987. op. c i t . Note 11. ------6 -

In France, sociology formally received early recognition as an academic subject when Emile Dürkheim became the firs t regular professor in social science at the philosophy faculty of Bordeaux university in 1896 and when, more significantly, his later chair at the Sorbonne in Paris was renamed "education science and sociology", thus taking up the notion of sociology, in 1913. This success in terms of institutionalisation, however, must be ascribed mainly to Durkheim's personal endeavours. He had purposefully worked toward institutionally securing his scientific 13 approach. Though his school remained moderately influential in different fields in the inter-war time, no further steps towards full institutionalisation could be made. In the middle of the 1950's there were only four university chairs in sociology plus three chairs in 14 ethnology in the whole of France. The debate on the central importance of the social sciences in rebuilding French society in the immediate post war years had raised high expectations, which, however, material­ ised only to a small extent. The most important organisational innovations were the creation of the Centre for Sociological Studies (Centre d'etudes sociologiques, CES) in the National Centre for Scientific Research (Centre national de la recherche scientifique, CNRS) and o f the S ixth Section of the School fo r Advanced Studies (Ecole pratique des hautes etudes, EPHE, 1 ater Ecole des hautes etudes en sciences sociales, EHESS), ------which soon were to become the most important social science centres in France. These academic institutions remained outside the universities, which is a typical feature of the French research system, but which was also significant for the s till low academic re p u ta tio n o f the d is c ip lin e . A decisive change re su lte d only

13 See Robert L. Geiger, "Die Institutionalisierung soziologischer Paradigmen", W. Lepenies (ed.), Geschichte der Soziologie, Frankfurt/M : Suhrkamp, 1981, and Terry N. C lark, Prophets and Patrons. The French University and the Emergence of the Social Sciences, Cambridge, Mass.: Harvard University Press, 1973 14 Alain Drouard, "Reflexions sur une Chronologie: le developpement des sciences sociales en France de 1945 ä la fin de 1'annee 1960", Revue francaise de sociologie, 23. 1982. 66f. It may be argued that in the inter- and early post-war time sociological thinking in France had found its place in the disciplines of ethnology and history. In the latter case, the Anna!es school also succeeded in insitutionalising its approach early in the post-war years. 15 Michael Poliak, Gesellschaft und Soziologie in Frankreich, K onig stein/T s.: Hain, 1978. 3Ö - 7 - much later from the full recognition of sociology as a university subject, institutionally fixed by the introduction of a specific full degree, the " lic e n c e " , in 1958.

Compared to the already low level of development of sociology as an academic discipline in France, the situation in Italy before the 1950's was even less impressive. Neither the intense empirical research of early positivist sociologists such as Enrico Ferri or Cesare Lombroso nor the more theoretically-oriented works of Roberto Michels, Gaetano Mosca and V ilfre d o Pareto had any la s tin g impact on the academic organisation of the social sciences. Before World War II there was no university base for social science research except for some teaching in social philosophy and some social psychological research at the Catholic University of Milan (Universita Cattolica del Sacro Cuore di Milano)--if we exclude the Statistical Institute under the directorship of Gini and the faculties of political science, created at a number of universities during the fascist regime with the function of training future diplomats and high-level administrators in dealing with social facts and in e litis t ideology/6 In the early 1950's, there was just one university chair in sociology in Italy, the one in Florence, and not until the middle of the 1960's was sociology—after long disputes—fully recognised as a university subject leading to an own degree.

The academic tradition of sociology is much stronger in Germany with the early inquiries of the Association for Social Policy (Verein für Socialpoliti k) around the turn of the century, the foundation of a German Sociological Society (Deutsche Gesellschaft für Soziologie) in 1910, the important research institutes in Cologne and Frankfurt in the Weimar period, and about a dozen of chairs in sociology in 1932 as well as again in 1955 after the reestablishment of the discipline/?

See Orlando Lentini, L'analisi sociale durante il fascismo, Naples: Liguori, 1974; Franco Leonardi, "Italian Sociology within the Framework of Contemporary Sociology", Contemporary Sociology in Western Europe and America. Proceedings of tne First International Congress of Social Sciences of the Luigi Sturzo Institute, Rome 1967, Paolo M ie 'li, hI M a c h ia v e llin i", L'Espresso, 29. 1983. 66f. The strong ideological heritage of political science in Italy surely had its part in preventing an earlier reemergence of the discipline after World War II and in leaving sociology the firs t rank in the demand for societally relevant knowledge. 17 M. Rainer Lepsius, "Die Entwicklung der Soziologie nach dem 2. - 8 -

Political science had its predecessor in the Cameral and State Sciences which were to fu lfill analytical functions for the absolutist state. But the rise of the liberal ideology and of industrial capitalism went along with a decay of this quasi-interdisciplinary approach and political aspects lost their importance in favour of legal ones. It was only in the Weimar Republic that the idea of scientific analyses of political processes was revived and found one of its expressions in the creation of the German School for Politics {Deutsche Hochschule für Politik, DHfP) in Berlin in 1920. After World War II, American influence firs t led to a conception of political science as a science of democracy, which was to perform important reeducation functions. Chairs in a number of universities were created, but they were mainly intended for purposes of educating future secondary school teachers and students of all faculties as part of a Studium generale in democracy. During the 1950's, there was a prolonged debate about the appropriateness of a separate discipline of political science, which ended with the discipline's fu ll recognition at about the turn of the decade.

The lack of clear institutional profiles of sociology and political science until the 1950's went along with the absence of a consensual self-image of the disciplines, their concepts, methods and priority research tasks. In all three countries the debate about the social sciences was strongly shaped by dominant philosophical and ideological orientations.

West German political scientists were influenced by the double experience of, on the one hand, the break-down of a parliamentary democracy and the establishment of a dictatorial regime and, on the other hand, witnessing in American exile the capacity of a different political system to deal with enormous economic, social and m ilitary problems without giving up the institutional order. Consequently, priority was placed, on the one hand, on the study of institutional

Weltkrieg 1945-1967", Günther Lüschen (ed.), Deutsche Soziologie seit 1945, special issue of the Kölner Zeitschrift für Soziologie und Sozial Psychologie 21/1979. 33. See for a historical sketch Horst Kern, Empirische Sozialforschung, Munich: Beck, 1982, and the report by Hellmut Wollmann in Thurn et a l, op. c i t . , 1984. Note 2. 18 See Hans Kastendiek, Die Entwicklung der westdeutschen P o litik­ wissenschaft , Frankfurt/M: Campus, 1977. 75ff. - 9 -

systems of government, in particular Western ones, with regard to the stability of formal democracy and with an explicit normative orientation towards them, and, on the other hand, on the study, of the history of political thought as a precondition for a debate on democratic values.. Philosophical reflection on political developments was seen as the specific task of the discipline which would distinguish it from the other social sciences and from law and would turn it into a somewhat 19 superior subject.

High political involvement and strong philosophical orientations were also characteristic of French sociology after 1945, but in a very different fashion. Not least due to its historical links to the philosophy faculties, sociology in France had a permanent point of reference in the ongoing philosophical discourse. From this perspective, the orientations of the intellectually dominant philosopher-sociologists of the inmediate post-war period, such as Raymond Aron, Maurice Merleau-Ponty or Jean-Paul Sartre, can be regarded as a reaction of revolt against the routinised and institutionalised sociological 20 approach of the Dürkheim school. H istoricity, class consciousness and commitment, the consecrated words of the philosophical semantics of the time, to use a phrase of Pierre Bourdieu and Jean-Claude Passeron, could not easily be linked either to the Durkheimian tradition of thought or to modern research techniques. In this intellectual climate, sociolog­ ical work proper had little chance of attracting the most enterprising minds, this philosophy "thereby helped to hold back the development of 21 the human sciences and especially the social sciences." The attitude of the "intellectuel engage", which doubtlessly had been furthered by the experience of the Resistance, rooted the intellectual debate firm ly in the political environment of the time and thus in some sense prepared the ground for a stronger practical orientation of the social sciences than in the inter-war period. Given the presence of a strong communist

ibid. 185ff. 20 On the origins of this attitude in the 1930's see: Johan Heilbron, "Les metamorphoses du durkheimisme, 1920-1940", Revue fra ncaise de sociologie, 26. 1985. 225ff. 21 Pierre Bourdieu, Jean-Claude Passeron, "Sociology and Philosophy in France since 1945: Death and Resurrection of a Philosophy w ithout Subject", Social Research, 34. 1967. 171ff., quotation from p. 179 - 10 - party as the major opposition force in society, it was almost an ob­ ligation for all intellectuals to clarify their position towards Marxism and Communist p o lit ic s .

In this latter respect, the intellectual atmosphere in post-war Italy resembled the French one. A strong Communist opposition which was soon almost totally excluded from political power and from the process of societal restoration attracted many intellectuals» In this environment, the rejection of modern sociological approaches was so strong that the word "sociology" was used with the pejorative connotation of an American imperialist instrument to secure bourgeois domination. This attitude found its counterpart in the still culturally dominant idealistic philosophy mainly represented by Benedetto Croce, whose influence had even increased not least through his anti-fascism. Throughout his life , Croce had opposed empirical social science research as unable to "understand" society. He believed this to be possible only through 22 historical-philosophical intuition. Sociology got its firs t foothold in post-war society in intellectual environments which were both non-Marxist and anti-idealistic: groups at and around the Catholic University of Milan and in the modernization-oriented industrial city of T u rin .

It was mainly in Turin that the firs t generation of sociologists found an opportunity to do empirical research supported and financed by industrialists such as O livetti. Industrial work organisation—and to a certain extent urban and regional planning—became the early thematic focus of sociological research related to topics such as modernization, rationalisation, and technological progress and the consequences for societal development. Similarly, French sociologists at about the same time got firs t experiences with empirical research on demand and with the support of non-academic actors. The research in industrial sociology pursued by Georges Friedmann, A lain Touraine and others were, though p a rtly based at CES and EPHE, financed by sources such as the European P ro d u c tiv ity Agency or the European Coal and Steel Community, and

22 For a short characterization of Croce's methodological writings see Franco Ferrarotti, Introduzione alia sociologia, Rome: Riuniti, 1981. 149ff.; for a description of the intellectual climate in the post-war period see Diana Pinto, "La sociologie dans l'Ita lie de 1'apres-guerre", Revue frangaise de sociologie, 21. 1980. 234ff. - n

favoured and made possible by industrial companies such as the Regie 23 Renault. In the same period, Jean Stoetzel, who was strongly influenced by Paul F. Lazarsfeld, tried to spread the use of quantita- 24 tive empirical methods at CES and in French sociology in general. In Germany, too, the debate on the relevance and necessity of sophisticated modern methods and, fo llo w in g th is , o f the re la tio n between theory and empirical facts was mainly led by sociologists who.had made more such experiences than political scientists. The normative orientations and the emphasis on the study of ideal-type workings of institutional systems of governments instead of the empirical distribution of power and influence made this a marginal issue until the 1960's in political science.

In spite of all differences in the national situations of these social sciences in the 1950's, there emerges a common picture of the discip­ lines as hardly secure of their legitimate roots in academic institu­ tions; strongly linked to philosophical discourse, not least through a lack of consensus on basic conceptual matters; disputing imported methodologies with which they were as yet unable to cope; and firm ly tied to developments in the political environment as a point of reference for their normative intellectual self-image.

Social Transformations

After a short period of a seemingly open political situation, in which new stable political majorities had not been yet formed and various paths of societal development seemed possible, around the year 1950 a constellation emerged, favouring reconstruction on the basis of re s to ra tio n o f the economic system and the re tu rn to power o f the political elites of the pre-war time. At the end of the 1950's the

23 See Claude Durand, "Les ouvriers et le progres technique: Mont-Saint-Martin vingt ans apres", Sociologie de Travail, 22. 1980. 5, and Alain Drouard (ed.), Le developpement des sciences sociales en France au tournant des annees soixante, Paris: CNRS, 1983. 89f. and 1 2 6 ff. 24 See his one-sided account: Jean Stoetzel, "Sociology in France—An Empiricist View", Howard Becker, Alwin Boskoff (eds.), Modern Sociological Theory in Continuity and Change, New York: Holt, Rinehart and Winston, 1957 - 12 - in itia l room for economic expansion—which had resulted from reconstruc­ tion needs, influx of labour and changed consumption patterns—was f i l l e d . Comparatively high rates o f employment were reached, in te rn a l migration or immigration increased and led to social tensions, inflation rates rose rapidly, growth rates though s till high showed tendencies to decline. Not least due to shortage of labour, the influence of the trade unions and—in Italy and West Germany—of the socialist and social- democratic party rose. The response in the political and economic system to these signs of crisis was to strengthen those forces which sought modernization of both the economy and the politico-administrative apparatus, which should become capable of supporting economic restruc­ turing as well as of planning and implementing reform projects to prevent or mitigate social tensions.

One of the early topics of this emerging modernization debate was the need for a science and technology policy. Public funds for research, which had traditionally been considered as a consumption expenditure, were now increasingly seen as a national investment, returns from which would come from improved international economic competitiveness based on technological progress and increased productivity. The creation of science policy institutions, such as the establishment of science policy ministries, research councils and commissions, in all three countries 25 dates from the mid- and late 1950's. The social sciences, however, were rarely the subject of science policy debates in these early years. A rather sim ilar debate concerned the need for improving and expanding the education system to increase the qualification level of the work force. However, this issue had a second, equally important, political focus the discussion on social reform, on equality of educational opportunities.

This "reform mood", common to all three countries in these years, found different expressions in changes in political majorities. In France, the crisis of the Fourth Republic led to the return of Charles de Gaulle in 1958, whose p ro je c t o f a m odernization-oriented conservatism emphasised rapid technological innovation accompanied by social reforms. It was to

25 For a comparison of French science policy institutions with those in the Netherlands and the U.S.A., see Ronald Brickman, Arie Rip, "Science Policy Advisory Councils in France, the Netherlands and the United States", Social Studies of Science, 9. 1979. 167-198 - 13 -

be designed and directed by a reformist elite in the planning apparatus and in high administration and bore heavy technocratic tra its. In Italy, the Christian democrats (DC) who had been the majority party in the 1950's, had to rely on the political support of the socialists (PSI) as of 1962. This "opening to the left", the formation of the so-called Centre-left governments, had since the 1950's been sought by a current of left catholic reformists in the DC, mainly from the industrialized North, who saw the need for a nationwide modernization policy which had to embrace the workers' organisations and would have to put an end to the particularistic and diente!ist practices common in government policies until then. In this situation, the reformist intentions matched the desire of the socialists to end the power monopoly of the DC and to introduce reform policies in favour of the workers. The German trade union federation and the social democrats (SPD) had been moving from a more far-reaching to a moderately reformist position during the 1950's and early 1960's. On the programmatic basis of a comprehensive set of social reform policies without major societal restructuring, the SPD had steadily gained influence on the local and Länder level, until it joined the Federal government after a crisis in the Christian democratic m a jo rity p a rty in 1966 and became major government pa rty in 1969. In all three cases, reform policies were based on the concept of sustained economic growth, harmonised and regulated by government in te rv e n tio n , linked to a set of social reforms by which the increase in welfare would be more evenly distributed in society and external costs of growth would be diminished.

26 For an attempt to assess the importance o f changes in p o litic a l majorities for the demand for policy research see: Peter Wagner, Hellmut Wollmann, "Fluctuations in the Development of Evaluation Research. Do 'Regime Shifts' Matter?", International Social Science Journal No. 108 (June) 1986 27 Such an ultra-brief description can obviously not account for the complexity of specific historical situations in three countries and—for the purpose of this paper—almost unavoidably tends to neglect im portant d iffe ren ces between the co u n trie s. For more detailed analyses see for example: Joseph LaPalombara, Ita ly. The Politics of Planning, Syracuse, N.Y., Syracuse University Press, 1966; S.S. Cohen, Modern C a p ita lis t Planning. The French Model, Cambridge, Mass., Harvard University Press, 1969; Elmar Altvater, Jurgen Hoffmann, W illi Semler, Vom Wirtschaftswunder zur Wirtschaftskrise, Berlin, Olle und Wolter, 1979; Jack Hayward, Michael Watson (eds.), Planning, Politics and Public Policy. The B r itis h , French and Ita lia n Experience, Cambridge (UK), Cambridge University Press, 19^5 - 14 -

The government apparatus also had to be adapted to a new and more demanding s ty le o f policy-m aking. M in is te ria l re o rg a n isa tio n , the introduction of improved techniques for monitoring administrative activities and efficiency control, and the creation of new institutions were measures intended to allow fo r p u rpo sefu lly and e ffe c tiv e ly intervening into economic and social developments. The establishment of a planning unit in the Chancellor's office in West Germany, the creation of an interministerial committee for economic programing in Italy or the introduction of an instrument called "rationalisation of budgetary choices" into French policy-making illustrate the debate on "meta policy-making".

It is against this background of a pronounced "reform mood" in the political system that the development of the social sciences w ill be sketched in the following section. We shall see that the interactions between so cia l s c ie n tis ts and actors in the p o lit ic a l system were not part of a "top down" social science policy: this did not yet exist. Moreover, and though the non-scientific actors involved in these interactions doubtlessly can be considered part of a larger societal elite, most of them were in political opposition and strong advocates for change.

Policy-Oriented Social Sciences: The Rush for Relevance

Having remained almost completely outside the academic institutional framework, Italian sociology from the 1950's on developed a strong 28 social position in close links with the emerging Centre-left politics. T h e o re tic a lly , the American fu n c tio n a lis t approach provided a comprehen­ sive and coherent framework under which social phenomena could be subsumed and allowed the steering capacity of specific political actors to be made the central point of reference in theoretical reasoning. It can be plausibly argued that the absolute reliance on private funding sources with specific policy interests, namely private and public

28 Diana Pinto uses the phrase "cultural centrality" in: "Sociology, Politics and Society in Post-War Italy", Theory and Society, 10. 1981. 676 - 15 -

enterprises, was crucial in bringing about the predominance of the func­ tionalist approach in Italian sociology which was stronger and imported 29 more uncritically than in other European countries. .

Modernization was the thematic and conceptual focus of almost all socio­ logical research. The basis of the modernization process was seen in technological innovation; sociology's major task was to analyse the impact of technological change on work organisation and productivity, on social mobility and the class structure, on urban and regional developments and, more generally, to focus on the possibility of harmonising and controlling the social processes set in motion by technological development. This orientation can be illustrated by some examples of important contemporary sociological work.

Alessandro Pizzorno studied the re la tio n s between a Lombardian commune and the expanding textile industry which increasingly came to dominate 30 the social and political life in the village. Luciano Gallino analysed the impact of technical change on industrial organisation in the 31 O liv e tti company. A number o f s o c io lo g is ts or economists w ith sociological interests studied Southern development with a view to the 32 need for agrarian reform or to get hold of the migration issue. Regional disparities and their social significance were also the main subject of the firs t congress organised by the newly-founded Italian Association of the Social Sciences in Bologna in 1958; again in 1961 a conference was held in Turin and Saint-Vincent on "Regional Disequilib- ria and Public Intervention" now making the policy-orientation visible in the conference title . For the Fourth World Congress of the Interna­ tional Sociological Association which was held in Stresa in 1959, the Italian sociologists chose for their collective contribution the title

A strong argument in th is d ire c tio n is made by Carlo Guido Rossetti in: "Sur la sociologie italienne vue par Diana Pinto", Revue fran^aise de sociologie, 23. 1982. 284f. 30 Alessandro Pizzorno, Comunita e razionalizzazione, Turin, 1960 31 Luciano Gallino, Proqresso tecnologico e evoluzione organizzativa negli stabilimenti O livetti, Milan, 1960 32 This research seems to be undervalued in Diana Pinto's accounts, see e.g. Salvatore Cafiero, "La contribution des sociologues italiens ä 1'etude des problemes du developpement national", Social Science Information, 3. 1964. 7ff. - 16 -

" .ta lia n Economic Development and it s Social Problems". In 1960, an international conference was held in Stresa again on the subject of "Technological Progress in Italian Society". One year before the long sought Centre-left coalition would shape politics at the national level, in 1961, the Italian Association of the Social Sciences chose as a topic for their meeting in Ancona the problem which was of central concern for most sociologists looking to become socially relevant: "Sociologists and the Centres of Power". The main thrust of the Ancona debates can be summarized in saying th a t most o f the social s c ie n tis ts f e l t th a t th e ir field had now been consolidated, that they had analytical tools for interpreting social developments and were now at the point where they desired to translate their findings into practice. For Franco Ferrarot- ti, for instance, an essential function for sociology consisted in clarifying the technical significance of socio-political choices that had been made and in proposing alternatives. For De Rita civil commitment of sociologists included not only scientific tasks, but also the willingness to modernize the sphere of public life , and Luciano Gallino saw the sociologist as expert in the technique of managing human 33 relations, and thus not excluded from social responsibility.

In fact, the period of the Centre-left dominance brought for sociolog­ ists an increased importance in defining reform policies. In disciplin­ ary terms, these developments even increased the dependence o f sociology on a policy model, since academic institutionalisation proved to be a much slower process, beginning in 1962 with the--also politically motivated—establishment of the Institute for Advanced Studies in Sociology at the University of Trento, but bringing fu ll recognition of the subject only in 1968.

The non-academically constituted Italian sociology was thus influenced by a politico-scientific constellation to such an extent that one may speak of an absolute predominance of an approach functionalist in scientific terms and modernization- and reform-oriented in political terms. By contrast, in French sociology, with its moderate but more secure in s titu tio n a l basis, changes in the socio-economic environment

As summarized by Renato Treves, S ociologi e Centri del Potere, B a ri: Laterza, 1962, p. 24f., quoted from Guido Martinotti, "Il condizionamento della ricerca", Pietro Rossi (ed.) Ricerca sociologica e ruolo del sociologo, Bologna: 11 Mulino, 1972, p. 146 - 17 -

led to the emergence o f divergent approaches in the socio lo g ica l community, rather than a complete and unanimous orientation towards reformist policy-making.

This latter position was in France clearly taken by Michel Crozier, who, in a very e lu c id a tin g a r tic le , bases his view on a profound change of the role of the intellectuals from the immediate post-war time to the 34 early 1960's. Based on Raymond Aron's version of the "end of ideology" theorem, which emphasises steady economic growth, internal sta b ility and cumulative progress in the sciences and in technology as factors allowing for a new form of rationality and for knowledge eliminating the necessity of force, Crozier argues that intellectuals have to move closer to action, their thought having to be much more pertinent and usable in a direct way. In political and intellectual developments which started with the late Fourth Republic Mendes-France government, he sees the emergence of a new elite which brings the whole sphere of political and social action and the strategy of reform into the intellectual domain. For the social sciences, this will lead to an increasing influence of those groups who "form a sort of link ... between the intellectual tradition seeking to reform itse lf and the world of action which is trying to renew itself through a process of more scientific • ,.35 reasoning" .

Seeing himself as a representative of this new type of social scientist, Michel Crozier at the time of writing had already contributed to this approach by elaborating an analytical model for the study of bureauc- racies. His book on "the bureaucratic phenomenon" has been at the root of a French school in the sociology of organisations centered at the Centre de Sociologie des Organisations (set up in 1966) under Crozier's directorship and has been innovative in opening up private and public administrations for empirical research. Reluctant to give policy advice in this book, Crozier in his later publications came closer to the new

34 Michel Crozier, "The Cultural Revolution: Notes on the Changes of the Intellectual Climate in France", Daedalus, Winter 1964. 514-542 35 ib id . 537

36 Lephenomene bureaucratique, Paris: Le Seuil, 1963 - 18 - role of the intellectual as he conceived it, giving more direct policy recommendations. During the post-1968 government he took up the role of inform al p o lic y advisor to Prime M in is te r Chaban-Delmas,

Basically sharing Crozier's view on the desirable future position of the sociologists as policy intellectuals, Raymond Boudon voiced a different 37 perspective on a strategy for the development of the social sciences. In his opinion, the way simultaneously to achieve greater u tility of sociology for societal practice and greater scientificity lay in methodological advance and, concomitantly, the extension of the research infrastructure which this would necessitate. In spite of profound differences, Crozier and Boudon can be grouped together with some other sociologists under the heading of a politics-science interaction model, in which the social sciences are oriented towards providing knowledge for the socio-technical steering of societal development by government and administration.

In contrast, the sociological approach of Alain Touraine, though it too seeks to break with the traditional philosophical discourse and to be more empirical, tries to maintain a more distanced relation to politics. Touraine acknowledges the legitimacy of a call for societal relevance and recognises the pressures to produce knowledge instrumental to politico-administrative needs. But his reaction is a refusal to make the analysis of the functioning of social systems his primary research objective. Instead, his "sociology of action" will analyse the commitment and engagement o f in d iv id u a ls and c o lle c tiv e a cto rs, by which 38 they create situations and establish meanings . Consequently, he sees a possible application or u tility of sociological knowledge not necessar­ ily and not primarily in the politico-administrative system, but by other socia l a cto rs, such as social movements. This he was to elaborate in more detail later.

Different again, the group of sociologists around Pierre Bourdieu were not centrally concerned by the potential practical use of their sociology. Linking up to the Durkheimian tradition and to structural

37 See e.g. his contributions to the debate, as reprinted in the collection: La crise de la sociologie, Paris: Droz, 1971 38 Alain Touraine, La sociologie de 1'action, Paris: Le Seuil, 1965 - 19 - analyses of society, their work on the education system and on in­ tellectual milieus tried to relate micro-level analysis to macro- sociologically conceived developments. Refusing a "voluntarist orientation towards knowledge application" (Michael Poliak), their analysis of the reproduction of social structures in and through the 39 education system in fact played a part in the gradual disillusionment with the possibility of social reform through education policy.

Compared to developments in Italy, a rather similar socio-economic environment in France did not lead to an identical reaction among sociologists. Strong stimuli for producing politico-administratively relevant knowledge accompanied the emergence o f p o lic y -o rie n te d sociological approaches, but at the same time brought about different "non-governmental" re o rie n ta tio n s in French sociology. Comparable pro­ cesses took place in the West German political science "community", though the lines of debate turned out to be different.

In the early 1960's when the political system was increasingly considered as malfunctioning, in particular with regard to harmonising economic and social development, a debate on the need to reform the governmental apparatus began in West Germany. As earlier in France and Italy, this debate centered on the issue of modernization understood in terms of adjustment to worldwide technological and economic change. Unlike in the other countries, however, the debate concentrated on the issue of restructuring the politico-administrative system and did not really widen to societal modernization in general.4°

Pierre Bourdieu, Jean-Claude Passeron, La reproduction, Paris, 1968 40 Reasons for this may lie in the traditional "Staatsfixierung" (centrality of the state) in German social thought or, much differently, in the smaller need for such a widening given the high com petitiveness of the German economy. At the moment I can only speculate about this. But whatever the reasons are, this political focus may have been decisive for the turn to political science instead of sociology for dealing with this problem. Friedberg/Gremion emphasize th a t in France the "p la n ific a te u rs " were the demanders and addressees o f p o lic y -o rie n te d research, whereas in Germany policy-makers took this role (Erhard Friedberg, Pierre Gremion, La recherche administrative et le reformisme politique, mimeo, P a ris : CSO, 1974) - 20 -

Around 1965 the issue of political reform began to be discussed in the social sciences. In the following years a number of different program- matical positions emerged besides the traditional administrative/jurid­ ical one, which proved to be unable to encompass the empirical analysis of policy-making. Beyond the sociological approach of Niklas Luhmann and an attempt to widen the scope of the traditional administrative sciences (e .g. W. Thieme and F. M orstein Marx) at le a s t three d iffe re n t conceptions emerged within the framework of political science.^

Very much in the tradition of normative political philosophy, Wilhelm Hennis sketched the task of a "Regierungslehre11 (government science) as the study of "the manner in which given the challenges to the modern state the steering, governance and coordination of a policy can be 42 effected" . In a similar way, but more willing to deal with the concrete policy issues of the time, Thomas E llwein argued for studying the processes of political decision-making and their transformation into policy programmes. Thirdly, already a rudimentary alternative research programme emerged as a reaction to the government-centred approach. Its core was the analysis of the entire political process including the way in which changes in the economic system prestructure and define 43 government's capacity to act in an anticipatory way.

The breakthrough from this programmatical debate to a broad range of empirical research projects occurred at the end of the 1960‘s and was made possible by the Volkswagen Foundation, a major social research funding institution, and by a whole set of research contracts issued by the newly-formed project group on governmental and administrative reform and mainly given to the "reform university" in Constance. __ For a description of the development of the debate see e.g. Thomas Ellwein, "Verwaltungswissenschaft: Die Herausbildung der Disziplin", Jens Joachim Hesse (ed.) Politikwissenschaft und Verwaltungswissenschaft, Politische Vierteljahresschrift, special issue 13/1982. 34ff., and Joachim Hirsch, Stephan Leibfried, Materialien zur Wissenschafts- und Bildungspolitik, Frankfurt/M.: Sülirkämp, 1971'. 236ff. ------42 Wilhelm Hennis, "Aufgabe einer modernen Regierungslehre", Politische Vierteljahresschrift, 6. 1965. 424 43 Joachim Hirsch, Parlament und Verwaltung, Vol. 2, Stuttgart 1968, see the discussion in: Ro If-Richar'd Grauhan, "Politikwissenschaftliche Forschung zur Verwaltung", Die öffentliche Verwaltung, 23. 1970. 591 - 21 -

The emerging policy-orientation can best be exemplified in the work of Fritz W. Scharpf. His understanding of administration as political actors and thus having to be researched in the framework of political science is sketched in a paper published in 1971 in which he explicitly refers to the advancing policy research in the USA and advocates this 44 approach as a promising new direction in political science. His publication (together with Renate Mayntz) on the "organisation of 45 planning” collects a number of papers written on political demand for the reform of the Federal politico-administrative apparatus and can be considered as the centrepiece of the early phase of policy-oriented research in Germany. The emphasis of the authors on intra-administrative organisational factors provoked the criticism of a number of resear­ chers, politically conceiving themselves as "left" of the social democratic reformism, who argued that Scharpf and others failed to recognise the decisive importance of external, mainly economic, factors that limited the effectiveness of governmental reform policies and of the scope of reformism itse lf. From this position emerged a "critical" approach which agreed on the need for empirical studies of policy processes. One strand of this approach sought to integrate the viewpoint of the citizens concerned, the other, a much stronger neo-Marxist current in political science, concentrated on determining the extent and the way in which external restrictions lim it the autonomy of the political sphere.

At the 1971 congress of the German Political Science Association, held in Mannheim under the significant title "Social Change and Political Innovation", an intense debate and dispute took place between represen­ tatives of the different approaches, who could present their first findings. While this debate was s till considered as being led in a "constructive" atmosphere, in the following years a "polarisation" of __ Fritz W. Scharpf, "Verwaltungswissenschaft als Teil der Politikwissenschaft", reprinted in: Scharpf, Planung als politischer Prozess, Neuwied: Luchterhand, 1973 45 Renate Mayntz, F ritz W. Scharpf, Planungsorganisation, München: Piper, 1973 46 As a useful reader in English see John Holloway, Sol Picciotto (e d s .), State and C a p ita l. A M arxist Debate, London: Arnold, 1978. For a summary of the critique of the approach taken by Scharpf and others see Wolfgang Fach, "Verwaltungswissenschaft - ein Paradigma und seine K a rrie re ", in : Hesse, op. c i t . , 1982. Note 41. 55-73 - 22 - political science in Germany emerged dividing the "governance-oriented" approach from the neo-marxist one. This "polarisation" was most visible at the national congress in Hamburg in 1973 and fo r some tim e i t pushed all traditional approaches into the background.

After having briefly sketched the scientific positions which had come up during the 1960's, the task of the following section w ill be to trace the politico-scientific interactions which lay behind the successful appearance o f new research programmes.

Political Innovativeness as a Topic: Interactions between Social Scientists and Policy-Makers

The changes which occurred between, broadly speaking, the period of the 1950's and the one o f the 1960's and which we termed the emergence o f a policy orientation were conceptually least significant in Italian sociology. With no real change in theoretical and political assumptions, a shift from industrial relations to policy issues is mainly to be explained by a growing debate in this period on the need for public intervention in the various policy fields. To understand why political influences could so strongly reorient the entire scientific discipline, one has to recall the low degree of academic institutionalisation and recognition. "Sociologists constitute a reference group of yet low relevance and cohesion, and thus Italian sociologists address themselves only in part to a public of scientific colleagues, but value much more high the reference group which is formed by the larger cultural envi­ ro n m e n t."^

In the years around 1960 a striking convergence of interests emerged between social researchers and policy-m akers in the "c u ltu ra l environ­ ment" of the North, especially of Milan and Turin. Left-catholic, socialist and modernization-oriented lay groups transcended party boun­ daries on the basis of the common idea of politically regulated and planned social change. Social scientists offered their support as

Martinotti, op. c it., 1972. Note 33. 143f. In this section of a paper written in 1971 Martinotti on these grounds rejects the proposition that Alvin Gouldner's critique of sociology can be applied in similar terms to the Italian case. - 23 -

scientific experts able to analyse social problem constellations and to indicate necessary political interventions: "In these years, sociology identified completely with the social project of changing Italian 48 society." The political and intellectual elites from the North joined forces to achieve this project. A number of initiatives flourished during the second h a lf o f the 1950's providing contacts between people from different professional groupings and p o litica l.orientations. The cultural and political association 11 Mulino in Bologna was one of the most important of these initiatives, starting mainly as the editorial committee of a journal and later on expanding to include one of the most important social science publishing houses and a research centre. Another one was the Centro nazionale della prevenzione e difesa sociale (CNPDS) in Milan which gave birth to the most important sociological re­ search institute of the 1960's, the Istituto Lombardo per gli Studi Economici e Sociali (ILSES). CNPDS was instrumental in founding the Italian Association for the Social Sciences in 1958 and in organising and financing its firs t conference. In 1960 debates at CNPDS led to the idea of setting up a research centre, after the Milan congress on the impact of technological progress on society. Informal talks in Milan with the new Centre-left coalition administration sufficed, ILSES devel­ oped into the place where almost all sociologists of the first and second post-war generation worked and which gave an enormous impetus to sociological research in Italy. To fu lly grasp the closeness of science and politics, it is useful to quote Alessandro Pizzorno, one of the founding fathers, describing the project in retrospect: "It was a study centre of the anticipated Centre-left government, where political forces met with a group of independent scholars. ILSES was to be the symbol of . 49 new politics, one which should be based on analyses and research." This concept was transferred to the national level after 1962, when the Centre-left alliance was formed in Rome and again brought about both a number of ambitious political reform initiatives and the setting up of policy-oriented research institutions.

48 Guido M artinotti, L'istituto superiore di sociologia di Milano, mimeo, M ilan, 19847“ 3 49 Quoted from: Silvia Giacomoni, Miseria e nobiltä della ricerca in I t a li a, Mi 1 an: Fel t r i n e l1i , 1979. 1Ö5 - 24 -

In her analysis of the relation of sociology and politics in Italy, Diana Pinto asserts that the interest in modernization translated itse lf in that period into a recourse to American political, economic and social models and, as such, into an interest in American social science. This science model, on the other hand, was a precondition for sociology becoming

"the new lingua franca of the Centre-left, capable of bridging the historical and cultural gaps which separate the Christian democratic tradition from that of lay socialism. Ahistorical and international in scope, the social sciences smooth out any major c o n flic t between the two p o lit ic a l tra d itio n s , p rim a rily by treating Italy's problems as the product of qualitatively new social and economic transformations, for which past solutions (and therefore conflicts) would be of little use."

Much of this analysis would also hold for Michel Crozier's political and sociological ambitions. His article on "the cultural revolution", which has already been quoted above, can be read as an attempt to identify his conceptions of required political advances in French society and of the role of the intellectuals in these processes with the dominant tendency 51 in reality. But in France, it proved to be much more d ifficu lt to get this position accepted in and against the much stronger and more self-referential intellectual life . Thus, sociologists intending to adopt the prescribed new role of the intellectual had to turn outside for support and acceptance.

Personal contacts were vital, and the main reference group was high level administrators and managers, both in the public and the private sector. As in the Italian case, a major catalytic influence can be assigned to private associations around journals or simply as meeting

Pinto, op. c it.. 1981. Note 28. 680. Franco Ferrarotti saw as one of the insidious dangers for Italian sociology in the mid-1960's its being considered as a "deus ex machina" for evidently complex social problems" ("Changement social et sciences sociales en Ita lie ", Revue frangaise de sociologie, 7. 1966. 29) 51 From a different standpoint Lucien Goldmann wrote: "Future historians w ill probably identify the years 1955 to 1960 as the so cio lo g ica l tu rn in g p o in t in France between c r is is ca p ita lism and organised c a p ita lis m , accompanied by a tra n s itio n from philosophical, historical and humanistic sociology to the a -h is to ric a l s o cio lo g ica l th in k in g o f tod ay." (Sciences humaines et Philosophie, preface to the new edition, Paris: Gonthier, 1966. 6) - 25 -

places, such as the Club Jean M oulin, the A ssociation d 1 etudes pour 1'expansion de la recherche scientifique, or the journals Esprit and Prospective. Unlike in Italy, these groups could not yet refer to a specific and emerging political project but in a less clearly defined way assembled members of the political, economic and intellectual 52 e lite s .

Looking back, Jacques Lautman has recently recalled the importance of "personal conjunctions which have played a capital role in the formation 53 of the spirit of state patronage" for social science around the Fourth Plan. He notes that Alain Touraine had close relations to Claude Gruson, head o f the s ta t is t ic a l in s titu te INSEE between 1962 and 1967. In fa c t, Gruson, who in 1964 argued for the integration of sociological expertise into the planning process—though on a position subordinate to economics--seems to have offered Touraine the post of a director of a newly to be created sociological section at INSEE, plans which did not 54 materialise due to lack of financing possibilities. Michel Crozier's research profited from the patronage of Jean Ripert, at that time high functionary in the Planning Commission, and was conceptually strongly 55 oriented towards the sociological analysis of the planning process. Touraine himself now sees the constellations around the colloquium at Caen in 1956, one of the major science policy initiatives of the Mendes-France government, as a confused coalition of all modernization- oriented elements. "This coalition of the Mendes-type for a number of years constituted the universe in which the social sciences have 56 developed." Its strength even increased with the return to power of de Gaulle and the reformism which then set in; after a great number of more or less informal meetings a major event in this phase of intense

52 See e.g. recently Drouard, op. c it., 1983. Note 23. 68, see also Bourdieu, Passeron, op. c i t . , 1967. Note 21. 187, and P oliak, op. c i t . , 1978. Note 15. 54 53 in: Drouard, ib id ., 78

54 Claude Gruson, in: ib id ., 147. See also his programmatical paper: "Planification economique et recherches sociologiques", Revue frangaise de sociologie, 5. 1964. 435-446 55 See his: "Pour une analyse sociologique de la planification francaise", Revue frangaise de sociologie, 6. 1965. 147-163 56 A. Touraine, answer to a questionnaire, published in: Drouard, op. c i t . , 1983. Note 23. 180 - 26 -

interaction between social scientists and modernization-oriented administrators was the 1965 congress of the French Sociological Society at which a group of high-level bureaucrats participated, among them Gruson, P. Masse, d ire c to r o f the Planning Commission, and R. Gregoire, active in economic policy-making.

The orientation and behaviour of individual social scientists have to be seen against the background of a changing socio-economic and political situation in which a strong demand for certain types of social knowledge was a major influence on the social sciences. But they have to be understood also as individual strategies to achieve or secure a reputable position in a scientific field which was undergoing rapid transformations. The importance of the traditional Parisian intellectual community diminished and political recognition gained in significance. This development has been described as the emergence o f a second pole of reference fo r socia l s c ie n tis ts , the pole o f power as opposed to the intellectual pole.^^

The social processes which lay behind the emergence of a policy-orienta­ tion in German political science seem to suggest that here the pole of power predominantly structured the field on which different scientific approaches competed

Without doubt, the launching of the in itia l research projects and the opening up of the administrative apparatus towards empirical research was closely linked to the political intentions of the social democrats who became minority partner in government coalitions in 1966 and majority partner in 1969. A key figure in this process seems to have been Horst Ehmke, himself professor of law, state secretary in the Ministry of Legal Affairs and minister in the Chancellor's Office, after W illy Brandt, a close friend of his, took this office. Ehmke was one of the programmatic thinkers of the SPD, who wanted to develop long-term perspectives for social democratic government, and saw a primary task in restructuring the politico-administrative apparatus to increase

57 Michael Poliak, "L'efficacite par 1'ambiguite", Sociologie et societes, 7. 1975. 36 58 In fact, such an interpretation has been given in an early assessment by H irsch, L e ib frie d , op. c i t . , 1971. Note 41. - 27 -

rationality and efficiency in political processes. His preparatory work led—after some political compromises—to the formation of a reform cabinet which soon created the Project Group on .Governmental and Administrative Reforms, s till under the Great Coalition government in 1969.59

This inter-m inisterial group and the Department of Planning established in the Chancellor's office by the subsequent social-democrat-led government drew a great number of social scientists into the policy process as advisors. In a second stage, and not least due to this advice, a demand was voiced for social science analyses to be pursued in the m inistries. Many of these research tasks were contracted out to the newly-formed "group for comparative administration research" at the University of Constance. Given the political interest, it is unsurpris­ ing that the research group closely followed the "government-centred" approach, dealing with issues such as internal control in planning administration, medium-term budgetary planning or reform of the ministe­ rial organisation.

However, this was only one step towards changing the political science agenda. Already in 1969 and sim ilarly influenced by the political debate on reform needs, the Volkswagen Foundation raised the issue of funding concentrated research efforts on public administration. After long discussions about the appropriate conceptual framework, a research committee was created for reviewing funding applications. This contained representatives of the traditional administrative sciences, of the reformist political scientists and of the emerging neo-Marxist current. Thus, a funding source came into existence whose decisions were not fin dominated by purely political criteria.

Both the reformist proponents of an "active policy-making" (R. Mayntz/F.W. Scharpf) as well as the neo-Marxists presented their research at meetings of the political science association. Given the state of the discipline, which was not homogenous, no integrative effect 59 See Heribert Schatz, "Auf der Suche nach neuen Problemlösungs­ strategien: die Entwicklung der politischen Planung auf Bundesebene", Mayntz, Scharpf, op. c it., 1973. Note 45. 29ff. 60 For a similar view see Friedberg, Gremion, op. c it., 1974. Note 40. 8 f f . - 28 - could be expected. Rather than the search for a common research programme or paradigm, a polarisation of the discipline occurred, as discussed above.

The emergence o f s c ie n tific currents c r it ic a l o f the p o lic y o rie n ta tio n , which fo r some years was or seemed to become dominant, is a common fea­ ture in the social sciences of the three countries analysed here. Guido Martinotti suggests for the Italian case (and with some qualifications this holds true also for France and West Germany) that the social scientists were for some time guided by a mythical figure, the "in- novative policy-maker" . Subsequent disillusionment and the ocurrence of social revolt against the policy model they had advocated led to a reorientation not only of political conceptions, but also of scientific approaches.

Political Crisis and Social Revolt: Social Scientists in Search for a New Coalition

With the changed political environment of the late 1960's and early 1970's a rethinking of conceptual approaches started or accelerated among social s c ie n tis ts . The re v o lts o f 1968/69; the abandonment of economic planning in Italy; disillusionment about the capacity to steer economic and social developments in France; and emergencing financial and political restrictions on the reformist capacity of government in West Germany: these destroyed the political and "epistemological optimism" and brought about or strengthened less directly policy-centred conceptuali zati ons.

In Italy, the sociologists' withdrawal from the modernization project had set in already in the early 1960's, but became generalised at the end o f the decade w ith increased working class m ilita n c y , student rebellion and the observation of political impotence in the face of an aggravation of social problems. Of great significance was the concept of action research, conricerca, as developed by the Quaderni Rossi group. Quaderni Rossi was a le ftis t political group organised around a journal which criticized the traditional workers' organisations for their

61 Martinotti, op. c it., 1972. Note 33. 146 - 29 -

integration in the capitalist system and their class collaboration. In the firs t half of the 1960's—six issues of the journal appeared between 1961 and 1967—the concepts developed by the group had a considerable influence on the strategic debates in the reviving workers' struggles. Central to these discussions was a notion o f autonomy which indicated not only a disassociation from the established organisations for the representation of workers' interests, but also a rejection of the entire capitalist state, society and way of life and the attempt to organise self-determined spheres of work and life .

The concept of conricerca was considered by Quaderni Rossi, which consisted mainly of intellectuals, as an essential instrument to get into contact with workers on the shop-floor and to intervene in industrial relations. The group was aware of the American origins of the 62 idea of action research, but intended to redefine it politically. Drawing on Marx's workers' inquiry, worked out in 1880, the aim was to use such sociological techniques for a number of purposes. First, the responses would allow an assessment of the development of capitalist work organisation and thus help to refine the theoretical analysis of the c a p ita lis t system. Second, i t would make the workers aware of the exploitative conditions under which they were living and working and would initiate or advance their reflections about counter-strategies. Third, it would establish or improve contacts between revolutionary 63 groups and the workers. The actual research pursued by members of Quaderni Rossi hardly reached the high theoretical and methodological levels which were sought. But these reflections were important in preparing the ground for a reorientation in sociologists' work concerning both the users of research findings and theoretical developments. ‘ In a cultural environment, where in particular the (communist) left was suspicious of modern sociology as being an instrument of class domination, this approach (re-)conciled Marxism and sociology by defining Marxism as sociology.^ ------See Dino de Palma, V ittorio Rieser, Edda Salvatori, "L'inchiesta alia Fiat nel 1961", Quaderni Rossi, No. 5, March 1965 63 Dario Lanzardo, "Intervento socialista nella lotta operaia: l'inchiesta operaia di Marx", Quaderni Rossi, No. 5, March 1965 64 "Marxism . . . originates as sociology, what is 'Capital', as a critique of political economy, else than sociology? . . . In my view . . . sociology is not a bourgeois science, . . . we can use - 30 -

Among the firs t generation of Italian sociologists, who were already professionally established in the early 1960's, the impact of the rising political and social tensions can best be exemplified by the work of Franco Ferrarotti. Acting as a sort of spokesman for Italian sociology he proclaimed as e a rly as 1966: " I t is necessary to advance from 65 sociology as a palliative to sociology as a means of participation." In the early 1970's, a programmatical essay on "the alternative sociology" summarized his critique of sociology, and he published the findings of a study of the victims of modernization in the shantytowns fifi around Rome. Organised expression of the irrita tio n and desire for new conceptual orientation among sociologists was the 1971 conference on "Sociological Research and the Role of the Sociologist" in Turin, which focused exclusively on attempts to clarify past errors, assess the development of the discipline in Italy and abroad and define a new role 67 for the profession in society.

The problem of an adequate interpretation of the notion of "society" now became a central conceptual issue. Society in the early 1960's had been seen in a very restricted sense as a "force to be moulded and regulated 6R by planners and technicians from above" , whereas not least through the revolt it had constituted itself as an actor in its own right. S im ilarily, at the Second Congress of the French Sociological Society in 1969, Robert Castel argued that the methodologies adopted in sociology had strengthened the importance of empirical statements, of formal schemes of organisation and of superficial equilibria, and had tended to misconceive the social significance of change and of social conflicts.

and criticize sociology in just the same way as Marx did it with classical political economy." (Raniero Panzieri, Contribution to the seminar "Uso socialista dell'inchiesta operaia", Turin 1964, published in Quaderni Rossi, No. 5, March 1965, quoted from the reprint in: Claudio Pozzoli (ed.) Spätkapitalismus und Klassenkampf, F ra n k fu rt: Europäische V e rla g sa n sta lt, 1972. 105 and 108) 65 F e rra ro tti, op. c i t . , 1966. Note 50. 31 66 Franco Ferrarotti, Una sociologia alternativa, Bari: De Donato, 1972, subtitled "From sociology as a technique of conformism to critical sociology", and Roma da capitale a periferia, Bari, 1971 & See Rossi, op. c it., 1972, Note 33

6R Pinto, op. c it., 1981. Note 28. 680, see also Rossetti, op. c it., 1982. Nöte"'2'9.' - 31

This was thought to have led to similar concepts of political action now reduced to the management of social order and manipulation of the subjects to adapt to the order. A la in Touraine summarised the s ig n if ic ­ ance of the social crisis for sociology: "The intellectual problem which is posed to the sociologist is the political problem which is posed to the society; and the political awakening of the society ... or at least, the calling into question of its orientations and of its forms of organisation allow the sociologist to rediscover the unity of the object he studies and, thus, of his own approach."^ In his own works since

1968 Touraine has continuously tried to draw conclusions from this diagnosis. The approach which he had set out in "la sociologie d'action" in 1965, seeing social actors as creating instead of purely responding to situations and transforming society, is further developed by his turn to social movements as the principle agents of history. Starting with his analysis of the French May revolt^ he not only chooses to study different social phenomena but links this to a concept of sociological intervention. The interventionist sociologist as an analyst in interaction with social actors seeks to stimulate the autoanalysis of groups actin g in social movements. This autoanalysis, i t is assumed, enables the actors to give meaning to the confrontation and their position in it and thus contributes to advancing the movement as a whole. In later years, Touraine applied this approach in sociological interactions with a student movement of the 1970's, the anti-nuclear movement and the Polish workers' movement around the Solidarity union.

Politically motivated action research also was a topic in German sociological debates in the wake of the student revolt. Compared to Italy and France, it was, however, seen less as a new and alternative political and scientific approach than as a critical reorientation, complementing and adjusting existing research programmes and reform gg-— Quoted from: Riccardo S c a rte z z in i, " I l d ib a ttito metodologico in Francia", in: Rossi, op. c it., 1972. Note 33. 257

A lain Touraine, Le mouvement de mai ou le communisme utopique, Paris: Seuil, 19"5S

For recent c ritica l assessments of his approach see: Remi Hess, La sociologie d'intervention, Paris: Presses Universitaires de France, 1981, 140-149, and Peter "kivisto, "Contemporary Social Movements in Advanced In d u s tria l S ocieties and S ociological In te rv e n tio n : An Appraisal of Alain Touraine's Pratique, Acta Sociologica 27. 1984. 355-366 - 32 - strategies (which were s till pursued by the social-liberal government at that time). In the field of political science, which is traditionally focussed on institutional analysis, the latently counter-institutional approach of action research had no influence. The politicization of the research environment strengthened the "le ftist" position in political science, but very few studies tried to advocate a participatory approach. Most research e ffo rts were used to counter the "p o lic y - oriented" researchers on the system level by developing the concept of systemic limits to state action in capitalist society based on a 73 re co n stru ctio n o f the Marxian c ritiq u e o f p o litic a l economy. In it s extreme versions, the neo-Marxist current almost denied any relevance of intra-organisational factors in the politico-administrative system for the outcome of (reformist) policy-making. By contrast, the proponents of the "active policy-making" laid almost exclusive emphasis on the study of these intra-organisational factors thus, at least by implication, placing high value on their importance. The dispute can partly be described in terms of a too restrictive conceptualisation of the political system and especially of the notion of "policy" which had been sheared from its historical and societal context. As with the influence of functionalism on Italian and parts of French sociology, this approach in German political science was shaped by the American understanding of policies as discrete institutionally and temporally exactly defined actions pursued by an independent political system. This concept easily lent itse lf in support of voluntarist reformism. During the 1970's, the experience of rising difficulties in and a partial withdrawal from reformist policy-making, on the one hand, and the exhaustion of ever more barren attempts to determine purely theoretically the restricted characters o f the s ta te 's autonomy under c a p ita lis t co n d itio n s, on the other hand, led to convergence of both approaches on the field of an empirical, theoretically more open study of policy processes.

72 Early contributions were: Werner Fuchs, "Empirische Sozialforschung als politische Aktion", Soziale Welt, 21/22. 1970/1971. 2-17, and F r itz Haag, Helga Krüger, W iltru d Schwärze!, Johannes W ild t (eds.) Aktionsforschung. Forschungsstrategien, Forschungsfelder und Forschungspläne, Munich: Juventa, 1972. See also Klaus Horn (ed.) Aktionsforschung: Balanceakt ohne Netz?, Frankfurt: Syndikat, 1979, and re c e n tly Horst Kern, op. c i t . , 1982. Note 17. 261-272 73 The, at that time unexpected, revival of Marxism both in the social sciences and in society which occurred also in France and Italy would require an own analysis. - 33 -

After these periods of high engagement and dispute, inner-scientific discussions were less conflictual in recent years in all three 74 countries. This may be not least due to the fact that such intense alliances involving the merging of political and scientific goals no longer attracted large groups of social scientists. But that the question how to develop the social science-politics interaction is s till high on the agenda can be seen for example from the recent debates at a round table of the German Political Science Association on "policy studies and traditional political science". This had at its centre the fear of a fragmentation of the discipline due to rising political demand 75 for specialized analyses in different policy fields. Similarly, at the National Colloquium in January 1982, which was to inaugurate a new and better era in science policy in socialist France, a strong resistance on the part of the social scientists could be observed, rooted in the expe­ rience of a tendency towards subordination to political interests in the 1960's and 1970's.

Conclusions for a Political Sociology of Science

This paper has described and analysed the emergence of an intense collaboration between social scientists and political actors, a reform 76 coalition , its demise, and the search for a new coalition, a new definition of social relevance on the part of the social scientists. Coalitions are joined for certain goals. In very broad terms, the aim of

74 The debates on the role of the social sciences in society during the 1970's and early 1980's cannot be dealt with in the framework of the argument proposed here. To contribute to the s till lacking analysis of this period of "social seientification" of politics is the intention of the above mentioned research on the development of p o lic y research, see Thurn et a l. , op. c i t . , 1984. Note 2. 75 See the proceedings in: Hans-Hermann Hartwich (ed.), Policy-Forschung in der Bundesrepublik Deutschland, 0piaden: Westdeutscher Verlag, 1985 76 The term "reform coalition" in this context has been used by Hellmut Wollmann, "Policy Analysis—Some Observations on the West German Scene", Policy Sciences, 17. 1984. 44. I heard the much nicer term "honeymoon■' from Guido M a rtin o tti and rediscovered i t in K atrin Fridjonsdottir's contribution to Blume, Bunders, Leydesdorff, W hitley (e d s .), op. c i t . 1987. Note 11. Its advantage (or disadvantage) is that it may be considered to have an im plicit assumption about an e a rly end. - 34 - the reform coalitions discussed here, their "common social project", can be called modernization and social reform??, whereas the goal of the action research projects which followed was support for the victims of capitalist modernization and the construction of a societal alternative. The following remarks explore some of the implications of these processes in terms of the interaction between scientists and non-scien- tific groups.

Social demand for knowledge, or, in the case of the social sciences, also for interpretations of social reality which are adequate and necessary for some actors' interests can be considered as a general condition for such types of collaboration. However, the movement of social scientists towards doing p o litica lly usable work cannot simply be seen as a "selling out". To me it seems important to note that since the end of the Second World War the "mythical promise of social renewal 78 through the social sciences" was one of the most important motives for becoming a so cia l s c ie n tis t. This promise was not only u n fu lfille d at the turn of the 1960's, rather more urgent, as the political restoration had le ft open many social questions. Thus, the pressure on the social sciences to become societally relevant stemmed from a larger socio­ p o lit ic a l environment and must have been in te rn a liz e d by many social scientists. Their identification with the political project thus went fa r beyond the mere acceptance o f a social task, and can only be explained in this context.

But the social scientists also had an academic project. In all three countries, in France and Italy more than in Germany, the degree of institutionalisation of the social sciences was insufficient for expanding research activities and for consolidating the disciplines. The promise of immediate political u tility could be used to claim resources and to fa cilita te academic establishment. In France and in Germany the phase o f intense in te ra c tio n between socia l s c ie n tis ts and policy-m akers paralleled expansion and full institutionalisation of the social sciences at the universities. In Italy, delays in a planned university

For an evaluation of the German experience in these terms see: Ulrich Beck, Wolfgang Bonß, "Soziologie und Modernisierung", Soziale Welt, 35. 1984. 381-406 78 Lentini, op. c it., 1974. Note 16. 21 - 35 -

reform due to resistance by representatives o f established academic in te re s ts , meant th a t the f u l l re co g n itio n of sociology as an academic subject did not occur before 1968, but with the "Committee for the Political and Social Sciences (COSPOS) other institutional solutions could be found to bridge this gap. From the viewpoint of disciplinary interests, then, the interaction with influential non-scientific groups served this aim of achieving fu ll academic institutionalisation.

A third consideration concerns the role of the disciplinary community as a reference group for the individual scientist and the implications for innovatory strategies. In West German political science the normative commitment to Western parliam entary systems was i n i t i a l l y shared by almost all members of the community, since it was almost a constitutive element in the emergence o f the d is c ip lin e . The la te 1960's p o lic y - orientation introduced an unfamiliar perspective into the disciplinary debates. The focus was now on the empirical study of substantive policy issues and to a number of the "traditional" political scientists this "modern" viewpoint overemphasised technical issues and disregarded the central political problems. It can thus plausibly be argued that without external support providing research funds and influencing science policy decisions besides giving necessary access the new approach would have had much more d ifficu ltie s in gaining acceptance.

In France, the intellectuals in Paris doubtless formed a sort of commu­ nity, in which every member had to refer to the works of other members and to write and act according to certain standards. These intellectual circles, as strongly as they shaped the reemergence of sociology, did not themselves, however, form or intend to form a scientific discipline. Thus, much of the sociological work of the 1960's bears signs of the tensions between the strong grip of the traditional reference group and the attempt to build one's own positon by distinguishing oneself from the philosophies of the past and by establishing a real "science".

The issue of "scientificity" as a precondition for becoming a discipline had least relevance in Italian debates about sociology in the 1950’s and early 1960's. The sociological approach identified almost perfectly with political reformism and not much care was taken to emphasise differences between science and p o lit ic s . For s o c io lo g is ts i t could be eq ually or - 36 - even more important to establish close links with relevant Centre-left party politicians than to engage in intense discussions about scientific progress in the field .

In this latter case it is obvious that the degree to which a self- referential disciplinary community existed was connected to the degree o f external o rie n ta tio n . W hilst in I t a ly , th e re fo re , the absence o f a scientific community was connected to the extreme importance assigned to external political processes, the two factors are not necessarily identical. In France, on the contrary, it was a key feature of the intellectuals that in being self-referential in their interactions they were h ig h ly engaged in commenting on social developments and takin g political positions, the ideal being the "intellectuel engage". The direct and utilita rian contact to political, economic and administrative elites was considered a break with the traditional ideal, endangering the critical function of intellectuals by reducing their distance to power. It seems to be not least the definition of their relation to society that emerged in the 1960's which today distinguishes the different "schools" in French sociology, mainly grouped around dominant individuals.

The interference of these disciplinary and individual considerations with the macro-politically influenced desire to conduct socially relevant research, in my view considerably modifies the notion of "shared social purpose" which Stuart Blume assumes to be the basis for such in te ra c tio n s between s c ie n tis ts and n o n -s c ie n tis ts . In p a rt, the different actors use a common situation for their different purposes.

Such a feature in these coalitions, one might want to call it the "honeymoon syndrome", is at the root of their demise. Coalitions are entered with high expectations and programmatic statements at the beginning o f a c o lla b o ra tio n u su a lly show high enthusiasm. Regarded in retrospect, both the political and the scientific outcomes are highly exaggerated, and sometimes it seems as if a more sober contemporary might well have recognized this overstatement. But exaggerating the opportunities of the coalitions might have been a useful way of strengthening the p o s itio n o f the c o lla b o ra to rs in both fie ld s . However, once an innovative scientific approach has been established with the - 37 -

help of external support and once a group of political actors has come to power proclaiming rational policy-making on the basis of social scientific advice, different logics in the two’ fields reassert 79 themselves and interests diverge.

In the analysis above i t has been suggested th a t the demise o f the modernization-oriented reform coalition led to the emergence of the critical action research approach. This shift, of course, should not be regarded as a clear sequence. As can be seen most clearly from the discussions about conricerca in Italy, this conception emerged while modernization-oriented sociology was s till in full strength. But a second differentiation has to be made regarding the success and the stability of such coalitions. Policy-oriented social sciences addressed themselves to identifiable actors who would ultimately dispose of large resources. The in te ra c tio n between social s c ie n tis ts and policy-makers might become institutionalised in social science policy and thus reach a 80 certain continuity and stability. The action researchers, by contrast, had much more difficulties in identifying their addressees and, even if they could, in obtaining a stable relationship of communication and in te ra c tio n . T h eir engagement very much depended on the existance of active social movements or other social groups having a definable political perspective and interest. Moreover, even if these conditions were met, continuity can only be expected if institutional settings are available which allow researchers to concentrate on these issues. In the

79 As a profound discussion of the idea of logics—unitary, diverse or analogous—shaping social science-politics interaction, see: Björn W ittro ck, "S ocial Knowledge and P ublic P o lic y : Eight Models o f Interaction", Helga Nowotny, Jane Lambrini-Dimaki (eds.), The D ific u lt Dialogue between Producers and Users o f Social Science Research, Vienna, European Centre fo r Social Welfare Training and Research, 1985. 89-109 80 For the German (and American) development in political science, this relative continuity contributed to researchers staying in the same logic with the policy-makers by shifting emphasis from the processes of policy formulation and planning to implementation analysis and policy evaluation during the 1970's. But this relatively stable relationship to the political system, by which larger and larger shares of the total research budget were allocated, can also be considered as responsible for the loss of a critical perspective and of the potential for fru itfu l intra-disciplinary dispute on conceptual and theoretical issues. For a more detailed analysis see: Thurn et a l. , op. c i t . , 1984. Note 2. - 38 -

absence of these preconditions, the action research approach as a broad movement of social scientists proved to be very short-lived, though it survives in other forms.

Finally, the problem of the epistemological consequences of coalitions between social scientists and non-scientists is to be raised briefly. On the basis of the previous analysis this problem cannot be fu lly analysed here, but the discussion of the changing concepts of "society" and "policy/political system" suggested clearly that significant epistemol­ ogical shifts in the social sciences can be observed depending on the 81 type of actor-orientation which is dominant at a specific time. Though these shifts cannot be assessed against any baseline (of purely internal scientific progress), the concept of "epistemic d rift", introduced by 82 Aant Elzinga seems to be a valuable one. The task for a political sociology of science would thus be to specify in detailed analyses the character of specific "drifts" by studying both the socio-political environment and potential external audience of a discipline and the internal structure of the scientific field as a precondition to understand the behaviour of specific actors in the field.

See for a suggestive historical account, which, however, obviously lacks in detailed analysis, Paul Diesing, Science and Ideology in the Policy Sciences, New York: Aldine, 1982 82 Elzinga, op. c it., 1985. Note 8.