The 1980s have been characterized by marked changes within research policy,
the research system, and research organizations, in Denmark as well as in the

other Nordic countries. Presently, new forms of control and organization are being tried in research policy. To analyze these changes and elucidate their consequences for research and society, positive as well as less positive, is important in order, among other things, to develop consciousness and create learning in relation to the formulation of research policy in the 19905.

In this context, the article 'Nordic Social Science Research and the American is a good initiative. The article raises important problems related to the consequences of the new forms of organization as to the autonomy social science research and its societal anchoring and potential.

The conception of research and science reflected in the article is, however, rigid and the analytical frame of reference applied insufficient. Therefore, the conclusions of the article are weak. In what follows, these weaknesses are commented upon and some conceptions of the sociology of science are presented, which may further develop the comparative analysis. The organization COS and its network is used to illustrate these ideas. Thereby, it is hoped to contribute to the important discussion about the development of social science research, which Scandinavian Political Studies has made an issue by publishing the 'Nightmare' article.

New Forms of Organization

The article distinguishes between four forms of organization: the Department model, The Network model, the Contract model, and the Center Model. It concludes that the models involve an increasing degree of separation of social science research from the traditional researcher control of the departments of higher education concurrently with an increasing degree of involvement with and dependence of the elite groups of society. COS is characterized as being based on the Network model. Together with the Norwegian LOS, which is based on the Center model, COS is characterized as a clear example of empirical, practice-oriented social science research integrated into rational policy-making.

First of all, seen from within the focal organization, COS is a center as well as a network. From the time it was established, COS has consciously aimed at developing both forms of organization. The size of the funds allocated to COS and COS's ambitions are the background for the choice of organizational

The total annual budget of COS amounts to approximately USSO.3 millions and not $0.5 million as stated in the article. The involved university departmentsand Social Science Research Council (SSRC) co-finance the majority of the amount equally, each contributing approximately $0.14 million.The sum, approximately 0.02 million, is financed externally by, among others, the elite groups mentioned in the article. The funds allocatedby

catedbythe elite groups are applied for, but only within fields where the more practice-oriented projects supplement theoretical projects already in process. This is in accordance with COS's ambition, which is to develop independent research, independent in the sense that it is disengaged from both isolated paradigms and segments within the researcher community and unilateral dependence of practice, the public, and society.

Where COS has chosen a combined model, LOS has, on the basis of funds ten times the size of COS's funds, chosen the Center model. The strength of the combined model is a visible organization as well as a large research potential. It may be said that the SSRC gets much more for the money than it actually pays for. Through the principle of financing a few researchers, they obtain participation by many, who are employed and financed by the initiating universities.

Forms of Control

Development of research policy is not only a question about local forms of organization and interest groups around research, but also a question of changes in forms of control. The establishment of COS, as well as the establishment the macroeconomic sister center 'The Research Group for The Public Sector and the Economy' (Institute of Economics, University of Copenhagen), is an attempt by the SSRC to obtain experience with new forms of control. Compared to the normal practice of the SSRC, COS reflects a development towards: (1) allocation of funds in the shape of grants adjustable within certain limits instead of isolated project grants; and (2) a decentralization the decision-making competence by transferring it from the Council to a group of competent researchers after a tender phase based upon a combination of market control and peer review.

It is our opinion that the article would have gained by applying a more varied, more deep, and theoretically based analysis of the changes in forms of control, manifested by the new local organizations. Inspiration might have been taken from recent research within the sociology of science (see, for example, Clark 1986; Premfors 1986; and Foss Hansen 1988).

In relationship to COS, different forms of control have been of varying importance over time. During the phase of establishment, the control form of market combined with the political climate of the SSRC was very important. decision to establish two concrete centers led to the creation of a buffer against the political climate of the SSRC, and COS chose to develop a meritocracy and enter into a cross-paradigmatic research dialogue as well as into a dialogue with society, inclusive of its elite groups.

Freedom and Dependence

A statement of the most characteristic changes, perceived as changes in the
ties experienced through research work, is as follows:

• Research and teaching have been separated to a certain extent, as
additional funds allocated to research to some degree are used to buy off
senior researchers from their obligation to teach.

• A certain detachment from the general political-administrative pinpricks
within universities has taken place because additional funds naturally open
up for the Center's independent priorities.

The above changes imply that a more concentrated research effort is possible.
This has, undoubtedly, been of great importance to some of those participating
COS.

It is not only a question of detachment and disengagement, however. Certain
have also been established:

• A certain amount of dependency of fiscal sources outside of the university sphere is created. In the case of COS it is almost exclusively the SSRC. This dependency is, however, modest, because the SSRC has allocated larger grants which are adjustable within certain limits, and, not least, established a small coordination group that functions as liaison committee between the SSRC and COS. This committee also functions as a buffer with the result that the SSRC does not feel called upon to check up constantly.

• It cannot be denied that a modest obligation to practice exists, but compared the funds allocated by political-administrative agents it is, as mentioned, little importance. However, what does obligation to practice imply? This is, in itself, worth contemplating. It should be stressed that COS has, independently, chosen an orientation towards practice primarily to be inspired and secondarily to make itself useful.

• The strongest obligation is perhaps that to other researchers and that connected an enlarged level of ambitions. As to the first one, it is probably more a function of the size of the project than of external funding. In any circumstances, it is a real and perceptible obligation which is not at all mysterious. The level of ambitions may be a problem in the sense that larger projects, centers, and so on are more visible and thus also forced to legitimize themselves even more strongly. This causes an increase in the level of ambitions, partly forced by external research colleagues, partly generated from the inside.

• This leads to the fourth type of obligation, namely a more or less problematic with the mother institutes. One the one hand it is a question of separation, on the other hand this has not been completed. Therefore, the center is dependent on interest and sympathy, and the

relationship between the Center and the mother institutes has actually
been very time-consuming. It is difficult to be inside and yet a little outside.

We do not deny that the more substantial conclusions of the 'Nightmare'
article may come true on the long view, but on the short view and as members
of COS, we experience different obligations.

The View on Research and Science

Basically, we find it a weakness of the article that it concludes exclusively on the basis of two alternative views on social science research. Social science research is either autonomic and critical or dependent and bureaucratic. The social science researcher either develops theories, but is often isolated in his or her ivory tower and dependent on the frames of one single paradigm, or he or she is one-sidedly oriented towards empiricism and practice and dependent on the elite groups of society. At COS, we believe in a third possibility: a social science research which combines the development of theory and practice, couples theoretical and empirical analyses, works across existing paradigms, and is developed in dialogue with research milieux and society, aiming at, among other things, contributing to the development of society in the 19905. This model for social science research is not even considered the 'Nightmare' article.

REFERENCES

Clark, Burton R. 1986. The Higher Education System. Academic Organization in Cross-National
Perspective. Berkeley: University of California Press.

Hansen, Hanne Foss 1988. Organisering og styring af forskning-en introduktion tilforskning
om forskning (Organization and Control of Research - An Introduction to Research on
Research). Copenhagen: Nyt fra Samfundsvidenskaberne.

Premfors, Rune 1986. Svensk forskningspolitik (Research Policy in Sweden). Lund: Studenterlitteratur.