Johan P. Olsen has published an important book. It is basedon empirical findings from the Norwegian "Power project" and structured by organization theory perspectives. The book consists of a short introduction and six substantive chapters focusing on various aspects of the Norwegian political system.

The introduction argues that political institutions have become modified in significant ways since 1945, and that our models so far have not been able to comprehend the changes. Hence, it is possible that the interpretation of institutional behavior as pathological might reflect a theoretical misunderstanding of what is sensible under present circumstances rather than a mistake in behavior: "The institutions have problems, but sometimes the problems are in our models rather than in our institutions" (p. 10). Johan P. Olsen feels a strong need for a new theoretical framework, a concern which is "linked to a 2000 year old search for a theory of governmental forms" (p. 7). "Required is a "new institutionalism" focused on how, and to what degree, behavior and outcomes are molded by political institutions" (ibid). However, Olsen does not purport to present a new theory of organized democracy. His aim is rather to show that organization theory might contribute toward that end, because "contemporary Western politics are organized democracies" (p. 7).

The six substantive chapters deal with the following topics: 1) Citizens' initiatives, 2) The Norwegian parliament (a revised version of a conference paper written with Per Lægreid, 3) The cabinet (revised version of Olsen's chapter in Rose & Suleiman (1980), 4) Top civil servants (revised version of a conference paper written with Per Lægreid), 5) Interest organizations and government (originally published in Nyström & Starbuck (1981), and 6) Policy-making styles (based on the article written with Paul G. Roness & Harald Sætren and published in Richardson 1982). While parts of the book will thus be known to interested scholars, the book as a whole is a unique contribution to the analysis of political systems, because Olsen consciously attempts to draw the pieces together into some sort of overall picture of contemporary Norwegian politics. There is still some way to go in that direction (and beyond that there is the task of revising democratic theory), but the book is more than a collection of articles based on the unusually rich data sources of the Norwegian power project.

What, then, is the message of the six chapters? It is impossible to do justice
to all arguments, perspectives and empirical findings in a short review, but

some main points should be mentioned. First, citizens' initiatives, i.e. singleissue, ad hoc activities organized outside established institutions, are not symptoms of a political legitimacy crisis or a breakdown of representative institutions. They are simply one way of articulating interests and mobilizing support for certain causes. They are "open structures" in contradistinction to the "hierarchical structure" of parties and elections, and to the "specialized structure" of interest organizations and government. The relative importance of these forms of collective political action changes over time. If there is anything wrong today, it is that current democratic theories have not captured this complexity. As Olsen notes: "There is a time-lag between practical politics and political theories — as observed earlier for the assessment of the legitimacy of political parties and interest organizations as participants in the political process" (p. 29). This is a good point. The problem is what to do about it in terms of democratic theory construction. Others have also noted that democratic theory lags behind practice, and it is surely no solution to accommodate citizens' initiatives by resorting to antique theories of small-scale democracy (Dahl & Tufte 1974), as some activists often appear to recommend.

Second, the Norwegian Storting is not an unimportant institution. On the contrary, it has probably become a more rather than a less significant institution. The Storting may not be good at policy-making, but a stagnating economy slows down the number of political reforms and increases the need for what the Storting is good at: finding political compromises. In addition, the Storting performs the (often neglected) symbolic-expressive function of interpreting a situation for which there is also more need in times of slow growth: "There will be a change from developing new policies to attempts at reinterpreting the situation — 'create more realism', 'reduce expectations', 'make people understand the foundations of the modern welfare state'" (p. 73). The chapter on the Storting may be seen as a Norwegian comment on the "decline of legislatures" debate. Olsen does not deny that the Norwegian parliament is less suited for substantive policy-making than the corporate-functional system, instead he argues that other tasks have become more important and that the Storting is comparatively good at performing those functions. On the whole, the Storting is viewed as "an important institution, but it is only one part of a network of institutions, competing and cooperating in processes of policy making and interpretation" (p. 72).

This leads to a third major point: The extent to which politically appointed leaders in the central government are able to give direction to the large administrative apparatus and how they do it. Olsen describes the "anatomy of the executive" as well as the "executive in action". The evidence is then discussed in terms of three possible models of governance: an anarchic or unsegmented structure, a monolithic or hierarchical structure, and a specialized

ized or segmented structure. None of these models fits reality perfectly of course, but: "The main tendency in Norway is toward specialization and segmentation" (p. 116). Even if there are counterforces, such as unsegmented citizens' initiatives and various attempts at hierarchical leadership by parties, the cabinet, and the Ministry of Finance, the government is basically "separated into functional, non-hierarchical coalitions" (p. 118).

In the fourth and fifth chapters a closer look is taken at important members of these coalitions, viz. bureaucrats and interest organizations, respectively. The message is clear enough: Top civil servants are important actors, but they do not act as a united force against politicians, they have developed a common style and been quite effective in protecting their interests as civil servants (the merit principle and protected careers) but in policy-making they "play on different teams, and when conflicts arise they take sides in predictable ways" (p. 146). Interest organizations have become integrated in government, but we should not fear for a "corporate era where integrated organizational participation replaces parties, electoral systems and legislatures" (p. 187). Integrated participation is attractive only for some organizations, for some policies, and in some situations. It is more widespread in Norway (and Scandinavia in general) than in, for example, Britain and the United States, and mainly involves economic organizations with respect to well defined material interests and technical matters. Some (promotional) organizations refuse to establish permanent, cooperative links to government because of the costs involved, such as, for example, reduced ideological purity and autonomy. The conclusion therefore is that integrated participation supplements rather than replaces other forms of participation.

The book finally presents an analysis of the Norwegian "policy styles". This rather fashionable term really refers to certain characteristics of the process by which policies are formulated and implemented (cf. p. 188). In Norway, the predominant styles since 1945 have been problem-solving, bargaining, and self-governance. These patterns mainly reflect the dominance of economic cleavages and the importance of economic interest groups allied with different political parties. They represent regular or routine policymaking. In recent years, however, citizens' initiatives and styles of mobilization and confrontation have challenged and supplemented stable routine policy-making styles, the divisive EEC-issue and the Alta River issue being the most spectacular examples. Olsen expects that the traditional policy styles will continue to dominate, although mobilization and confrontation will also be used. If political leaders decide to become organizers they can influence future developments to some extent. However, major reorganizations are not likely to occur; "ejection" of issues is a possibility, but cooptation of interested groups or their causes is perhaps the most promising strategy.

Thus, the final discussion of the concluding chapter reflects a concern for

political leadership in an advanced welfare state, a topic pervading most other chapters as well. There are numerous suggestions as to what political leaders could or should do. They face a strongly segmented, specialized or sectorized political system structure (although some counter forces are also operating), but could increase their influence and control by conscious regulation of ' 'the ways in which streams of participants, problems, solutions and choice opportunities flow together or are kept apart" (to quote p. 207; there are similar expressions passim). Given the opening pages, the conclusion is somewhat disappointing, although perhaps both predictable and reasonable in view of the genesis of the book. The Norwegian power project and organization theory emphasize the complexity of modern government not yet grasped by democratic theory. This is a major contribution. But where do we go from there? Perhaps the book, after all, also reveals some limitations of organization theory?

If the ultimate concern is with democratic government, and I suspect it is, we must return to democratic theory — with the empirical knowledge gained through the efforts of numerous scholars — establish fundamental principles of democracy in the modern world, become aware of theoretical problems and dilemmas, and evaluate existing institutions and procedures on that basis. Norway is a "pluralist democracy" par excellence in Robert A. Dahl's sense (as are the other Scandinavian states), and thus presumably not unaffected by the four defects of organizational pluralism which Dahl (1982) discusses at great length (stabilizing political inequalities, deforming civic consciousness, distorting the public agenda, and alienating final control). Johan P. Olsen's book (and several other publications of the power project including those dealing with business firms, multinational corporations, and the mass media, NOU: 1982:3) provides very rich material for such evaluations. At the same time the book provides important information on the nature of a pluralist democracy without which theorizing about democratic government would be less fruitful.

REFERENCES

Dahl, R.A 1982. Dilemmas of Pluralist Democracy. Autonomy vs. Control. New Haven:
Yale University Press.

Dahl, R.A. & Tufte, E.R. 1974. Size and Democracy. Stanford: Stanford University Press.

NOU 1982:3. Norges offentlige utredningar. Maktutredningen, Sluttrapport. Oslo: Universitetsforlaget.

Nystrom, P.G. & Starbuck, W.H., eds., 1981. Handbook of Organizational Design, vol. 2.,
pp. 492-516. New York: Oxford University Press.

Richardson, J.J., ed., 1982. Policy Styles in Western Europe. London: Allen & Unwin

Rose, R. & Suleiman, E.N., eds., 1980. Presidents and Prime Ministers. Washington, D.C.:
American Enterprise Institute.