I welcome the opportunity to clarify a number of points in my article as a result of the four critical comments from representatives of the research projects studied. In my response, I would like to open with some general observations before considering each comment in turn.

Social scientists are very good at objectively analysing others but notoriously and partisan when their own professional organizations are subject to social science analysis. My article is just such a study of social science research in terms of its organizational structure. My concern was to show that it can be fruitful to analyse research organizations in terms of organizational theory, just like any other administrative structure, and that power and agenda-setting play just as important a part in the organization of research as in the organization of any other activity.

The reaction of representatives of the projects subjected to this scrutiny in leaping to their own defence is therefore human and understandable. Much of what my critics say is clearly an attempt to justify the objectivity and independence of the projects. It appears to be a structural characteristic of elite-research relationships that the researchers themselves perceive themselves as being quite independent of, and resistant to, the interests of government and pressure groups even if those they study are not. There is clearly scope here for research on cognitive dissonance and the exercise of power in research agenda-setting.

More substantially, my critics appear to take objection to the lack of detail provided in the article. The presentation of the Norwegian LOS project is described as 'rather insufficient' and requiring 'a more thorough analysis' (Grøholt & Abraham sen 1989), as well as 'superficial' and as 'wild exaggeration' (Baldersheim 1989). Furthermore, the author has exaggeratedthe of the Finnish project' (Tuori 1989). Finally, the analytical frame of reference applied is labeled 'insufficient' by representativesof Danish research project, and 'a more varied, more

deep and theoretically based analysis' is called for (Foss Hansen & Beck
Jørgensen 1989).

It is important to recognize that those deeply involved in their own activities will almost always question any attempt to generalize them. It is not easy to see the wood for the trees when among the trees and it is likely that irrespective of what conclusions an outsider may draw, the woodcutter protest that the wood does not look like that from his perspective.

At the same time, it must be stressed that the article is not the product of an in-depth research project. It is rather a review, carried out at the instigation of the editor of SPS, to provide an overview of a number of social science research projects currently being carried out in Scandinavia. Furthermore, it builds on data provided by leading representatives of the projects, the data only forthcoming often after considerable prompting. And while the corrections they now make do not lend me to modify my broad conclusions, it is to be hoped that the empirical results emerging from the projects will be more reliable than the organizational data provided for my article.

But these are minor matters to preface the substantive arguments made
by the project representatives, to which I now turn.

K. Tuori takes issue with my interpretation of C. Wright Mills, arguing that his analysis is not relevant to the Scandinavian research projects as I have described them. He argues that Mills was principally concerned with relationships between politicians and bureaucracy-based research, whereas my treatment of the research projects concerned primarily their internal organization.

This interpretation of Mills is peculiar to say the least. Contrary to what Tuori argues, Mills dealt centrally with the manner in which academic research was increasingly adapting to the demands of private and public organizations for a specific type of policy-relevant research. He argued that these demands had bit by bit impinged on the traditional organizations of the research community by means, among others, of different kinds of research projects and various mechanisms for channelling academic resources into social science research of a particular nature.

I cannot but conclude that Mills's description provides a most suitable starting point for a discussion of the organization of Scandinavian social science research in relation to different vested social interests. I must therefore assume that either Tuori has misunderstood Mills' thesis or he has missed the central point of my argument.

In the article, I adopted the - scarcely original - viewpoint that universitiesas provide an important degree of protection for the integrity of research. H. Baldersheim, in his elegantly ironic commentary, argues that the thesis needs modifying and complementing. This is because there are additional mechanisms besides universities that contribute to

strengthen or undermine the integrity of research in different respects. While this is certainly true, the author's specific argument, based on a simplistic pluralism thesis, is naive. He argues that in the case of the Norwegian LO centre the political and administrative establishment represents'a of competing interests'. The research agenda is therefore determined by a process of negotiation in which the various vested interests balance each other out and thereby ensure the integrity of researchers.

Such a simplistic pluralism thesis is rarely argued for even among pluralist political theorists. It is widely recognized that not all interests are organized, nor are they all represented in specific negotiation processes, nor do they all have similar weight of influence. It therefore follows that decisionmaking are not as 'pluralistic' as the author suggests, and therefore not provide automatic protection for the integrity of contract research.

Nor in the Norwegian context is that form of semi-pluralism that is sometimes called 'corporate pluralism' (Rokkan 1966) or 'the segmented state' (Egeberg, Olsen & Sætren 1978) effective protection of the integrity of research or the individual researcher. On the contrary, it is just as likely that the Norwegian form of pluralism - which has its equivalent in other Scandinavian countries - results in the co-opting of research by various decision-making segments and is used, just as the author describes, to further the specific interests espoused by these segments (cf. Olsen 1988, 94, 149). Public and private interests do not have to be monolithic juggernauts as the author mistakenly ascribes to Mills' view (cf. Mills 1956) - in order to comprise a threat against the integrity of research.

Mills' analysis of developments within American social science research was based on a broader view of its role. This was to identify the point of interplay between historical and biographical crises in terms of social structure. This view presupposes research imbued with an intimate iteration between theory and empiricism, avoiding 'grand theory' as well as 'abstract empiricism', together with striving after integration between structural factors on the one hand and particularistic or individual factors on the other (see Mills 1959, chs. 7-9).

As I understand H. Foss Hansen and T. Beck Jørgensen, their critique can be seen in the light of the latter part of Mills's analysis. They argue that the basic shortcoming of my article is that it posits a rigid dichotomy between social research as being either autonomous and critical or dependent bureaucratic. A third alternative - combining theory, empiricism and applications across research paradigms and in dialogue between researchers and 'society' - is not discussed in the article.

To a considerable extent I concur with this view, and would add that the
discussion would have been further strengthened had such a third alternativebeen
in the context of Mills's vision for the goals of social

research, perhaps in the context of a research article rather than the review
article I actually wrote. To some extent it would appear that this vision has
inspired the Danish project, to judge by the authors' comment.

However, I would like to make one final point in closing, relating back to one which I made in my opening comments. This is that researchers may have high ambitions, and may even see themselves as being independent and 'free-floating' advisors in a way that outside analysts might be reluctant to conclude. It therefore remains to be seen whether Mills' vision - or as the authors imply, a 'Scandinavian research dream' - can be achieved within the framework of a research project, or whether, like so many dreams, it disintegrates in its collision with institutional realities.

REFERENCES

Baldersheim, H. 1989. 'Comment', Scandinavian Political Studies 12, 187-189.

Egeberg, M., Olsen, J. P. & Sætren, H. 1978. 'Organisasjonssamfunnet og den segmenterte
stat', in Olsen, J. P. red., Politisk organisering. Bergen: Universitetsforlaget.

Foss Hansen, H. & Beck Jørgensen, T. 1989. 'Comment', Scandinavian Political Studies 12,
192-196.

Grøholt, K. & Abrahamsen, T. 1989. 'Comment', Scandinavian Political Studies 12, 189-192.

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

Tuori, K. 1989. 'Comment', Scandinavian Political Studies 12, 196-198.