Written for a cultured public as well as for political scientists and theorists, this is a book about what politics were, are, and could be. Dunn makes politics an object of knowledge (the first question he asks is: “What exactly is politics?”) (p. ix) because he believes that knowledge can make human agency “potentially effective” and “a locus of value” (p. 111). In order to understand modern politics, we must analyze how the classical political categories have changed. In order to solve the problem of contemporary political science, we must keep the general norm and the specificity of the actual context together. Thus, although he acknowledges, realistically, that interests are powerful factors in human agency and a limit to politics, Dunn wants to preserve the leading role of politics. The goal of modern representative democracy is to keep “a polity in good order” by “the selection as rulers of those with the intelligence, moral commitment and strength of character to rule as they should” (p. 268). To attain this goal, politics needs the “great imaginative force” of the ideal of impartiality to rectify the partial in political judgment stemming from economic interests (p. 357).

The Cunning of Unreason is divided into three parts and combines historical, empirical, and analytical approaches. The first part discusses the transformation of the classical political categories. Paraphrasing Benedetto Croce, Dunn asks: “What is dead and what is alive of Aristotle's vision?” His conclusion leaves no room for doubt: “Whatever we may in the end choose to agree with Aristotle, we can hardly hope to see eye to eye with him” on many ideas (pp. 16–7). Aristotle's conception is no longer viable because it demands that the city be “virtually self-sufficient” and makes women, slaves, and manual workers into pariahs. We live in a globalized economy and have a more comprehensive view of political inclusion. To be sure, Aristotle's world was also global, albeit in a limited way. Greek city-states were engaged in a network of commercial relations with other Mediterranean populations and were familiar with imperialism. Aristotle conceived of “self-sufficiency” as political autonomy, the basic condition for public decisions. Perhaps this general criterion can still be useful, at least to evaluate the character of our democracies, whose scope of action, as Dunn notes, is on many occasions defined by a nonpolitical (thus, free from political control) domain such as the market.

Nonetheless, Dunn is right to trace modern politics back to the less noble and more instrumental view of politics that sprang from Christianity. Starting with the original sin meant to give politics both a coercive and a salvationist role, Christianity promoted both realistic and eschatological perspectives. Predictably, Dunn endorses the former: Hobbes not Marx. Hobbes, Dunn says, was right to treat cruelty and vainglory as “ordinary vices” reason can control (but not eradicate) by transforming them into the cornerstones of a secure social environment. Hobbess' legacy is to have conceptualized the content of the kind of politics we deal with today: interests, instrumental rationality, and the individual potential for good and evil. These are the foundations of the main characteristics of modern politics: the juridification of state authority and the depersonalization of economic relations.

The second part of the book is a critical analysis of an empirical case of modern politics: the “structure of domination” of the Thatcher regime, which was a unique mix of “free economy” “protected by and in the last instance enforced by a ‘strong state’” and “belligerent chauvinism” (p. 144). What can possibly explain the longevity and popular support of a political program that was clearly not in the interests of that majority? As an Italian compelled to obey Silvio Berlusconi's government, I sympathize instinctively with Dunn's skepticism about the voters' ability to make rational decisions. Yet, citizen “incompetence” is an artifact that begs other questions about how private interests control and manipulate information as well as about the role of ideology, rhetoric, and passion in politics.

In the third part of the book, Dunn outlines the theoretical bases of his reasonable hope. He singles out two possible explanations for the “unimpressive” performance of modern democracy: the Platonist (which “attributes the misdirection essentially to incomprehension” of both ordinary citizens and career politicians) and the sociological (which “attributes the performance of politicians, above all, to the cognitive facilities and distortions which result from, and the more or less perverse incentives which are offered by, the institutional settings and social milieu”) (pp. 300–1). Dunn declares his preference for the latter and proposes a vision of democratic politics that is anti-Habermasian, realistic, and contextual in its value. This is a Shumpeterian view of politics, consistent with Dunn's previous choice of Hobbes over Aristotle. Electoral competition is vulnerable to corruption, creates a “gap” between “enfranchised citizens and career politicians,” and suffers “the sheer difficulty of forming and implementing rational strategies” (p. 285). Yet, it is all we have if we are to make government democratic and “do the best we now can to modify the consequences of our action and inaction for the better” (p. 341).

Dunn endorses Benjamin Constant's paradigm: The moderns have to cope with the constraints that private liberty imposes on participation. Representative democracy, like capitalism, is our only choice, but dropping the Enlightenment's grand vision does not imply abandoning “the strategy of rectification” modern rationalism has perfected (p. 360). Modern politics is a complex concoction of “fatalism” and “voluntarism,” “routine” and “crisis,” realism and hope. It demands that reality be accepted, not glorified or demonized.