The present volume is a translated and modified - not least shortened - version of the author's well-known and much praised, dissertation, Rationalitet magt (2 vols., Odense, Akademisk forlag, 1990, 177+463 pp). The story is, in empirical terms, a case study of a planning process in the Danish city of Aalborg. In considerable detail, we read about 'The Aalborg project' which evolved during the 19705, and which aimed at a significant reconstruction the city center. It is a tale which does not fit very well into the modern view of'rational planning.' In this modern or 'Enlightenment view' on the relationship between knowledge and politics, individuals and collective actors are, of course, expected to behave according to the logic of instrumental rationality. Objective or at least intersubjective knowledge is supposed to be able to 'eliminate politics.' And political decisions made without the explicit support of such knowledge are rarely considered fully legitimate. Flyvbjerg's careful reconstruction of the Aalborg project shows the extent to which such expectations of instrumental rationality were not met in his case. Instead of being pervaded by formal or instrumental rationality, the process was characterized by rationalization - and a considerable of bluffing and deception as well. Instead of 'eliminating politics,' expert knowledge was often unabashedly used by key actors as an instrument of power.

The theoretical interpretation of the case is quite elegant and challenging. The careful reader of both the dissertation version and the present version will no doubt observe some theoretical, conceptual and terminological developments. is still an undisputed hero of the story, but Nietzsche has entered the scene in a much more prominent role. And second-order hero Aristotle from the dissertation has now been outmanoeuvred by Machiavelli. Fittingly, Realrationalität is launched as a new key term denoting power defines knowledge' thesis which is stubbornly pursued throughout the book.

The author thus clearly positions himself in the 'realist' or anti-
Enlightenment tradition. Even if that pre- or postmodern position is still
probably only held by a minority of Scandinavian social scientists, it is of

course far from original today. The interesting thing about an argument in this vein is then, in my view, not the position itself, but rather how it is defended in terms of values of democracy and human emancipation. Some of us would argue that a too heavy dose of 'realism' is not compatible with any acceptable vision of democracy. Flyvbjerg's view is rather that 'real democracy' needs the full insights provided by the likes of Thucydides, Machiavelli, Nietzsche and Foucault:

I maintain that if one takes Enlightenment ideals seriously, one needs to understand the Enlightenment in anti-Enlightenment terms, that is, in terms of Nietzschean will to power and Foucauldian rationality-as-rationalization. We should not conclude that because Nietzsche and Foucault criticized democracy and other Enlightenment ideals, we, as democrats, use their thinking to understand democracy better and to work for more democracy. To be a Nietzschean democrat is not a contradiction in terms but a real and productive possibility, (p. 3)

But we learn little or nothing from Flyvbjerg about the ways in which this 'productive possibility' may be transformed into real world democratic practice. As in many comparable discussions in a postmodernist vein, there is a conspicuous absence of any analysis of concrete democratic institutions and procedures in the present book. Walzer's evaluation of the 'lonely politics' of Foucault applies, I believe, equally well to Flyvbjerg:

Foucault does not believe, as earlier anarchists did, that the free human subject is a subject of a certain sort, naturally good and warmly sociable. Rather there is no such thing as a free human subject, no natural man or woman . . . And so Foucault's radical abolitionism, if it is serious, is not so much anarchist as nihilist. For on his own argument, either there will be nothing left at all, nothing visibly human; or new codes and disciplines will be produced, and Foucault gives us no reason to expect that these will be any better than the ones we currently live with. Nor, for that matter, does he give us any way of knowing what 'better' might mean (Walzer 1988, 202).

We are still waiting, even post-Flyvbjerg, for a convincing portrait of
the Nietzschean democrat. The brand of'realism' espoused in this tradition
comes, I suggest, at a very high price.

Department of Political Science and SCORE, Stockholm University

REFERENCES

Walzer, M. 1988. The Company of Critics. Social Criticism and Political Commitment in the
Twentieth Century. New York: Basic Books.