MULTITUDE: War and Democracy in the Age of Empire. Well before 9/11 and the Iraq war put the idea in everybody’s mind, Michael Hardt and Antonio Negri had popularized the notion of a modern empire. Four years ago, they argued in a widely discussed book — titled, as it happens, ”Empire” — that the globe was ruled by a new imperial order, different from earlier ones, which were based on overt military domination. This one had no center; it was managed by the world’s wealthy nation-states (particularly the United States), by multinational corporations and by international institutions like the World Trade Organization and the International Monetary Fund. This empire — a k a globalization — was exploitative, undemocratic and repressive, not only for developing countries but also for the excluded in the rich West.

Hardt and Negri’s new book, ”Multitude,” argues that the antidote to empire is the realization of true democracy, ”the rule of everyone by everyone, a democracy without qualifiers.” They say that the left needs to leave behind outdated concepts like the proletariat and the working class, which vastly oversimplify the gender/racial/ethnic/ class diversities of today’s world. In their place they propose the term ”multitude,” to capture the ”commonality and singularity” of those who stand in opposition to the wealthy and powerful.

This book — which lurches from analyses of intellectual property rules for genetically engineered animals to discourses on Dostoyevsky and the myth of the golem — deals with an imaginary problem and a real problem. Unfortunately, it provides us with an imaginary solution to the real problem.

The imaginary problem stems from the authors’ basic understanding of economics and politics, which remains at its core unreconstructedly Marxist. For them, there is no such thing as voluntary economic exchange, only coercive political hierarchy: any unequal division of rewards is prima facie evidence of exploitation. Private property is a form of theft. Globalization has no redeeming benefits whatsoever. (East Asia’s rise from third- to first-world status in the last 50 years seems not to have registered on their mental map.) Similarly, democracy is not embodied in constitutions, political parties or elections, which are simply manipulated to benefit elites. The half of the country that votes Republican is evidently not part of the book’s multitude.

To all this Hardt and Negri add an extremely confused theory, their take on what Daniel Bell labeled postindustrial society, and what has more recently been called the ”knowledge economy.” The ”immaterial labor” of knowledge workers differs from labor in the industrial era, Hardt and Negri say, because it produces not objects but social relations. It is inherently communal, which implies that no one can legitimately appropriate it for private gain. Programmers at Microsoft may be surprised to discover that because they collaborate with one another, their programs belong to everybody.

It’s hard to know even how to engage this set of assertions. Globalization is a complex phenomenon; it produces winners and losers among rich and poor alike. But you would never learn about the complexities from reading ”Multitude.” So let’s move on to Hardt and Negri’s real problem, which has to do with global governance.

We have at this point in human history evolved fairly good democratic political institutions, but only at the level of the nation-state. With globalization — and increased flows of information, goods, money and people across borders — countries are now better able to help, but also to harm, one another. In the 1990’s, the harm was felt primarily through financial shocks and job losses, and since 9/11 it has acquired a military dimension as well. As the authors state, ”one result of the current form of globalization is that certain national leaders, both elected and unelected, gain greater powers over populations outside their own nation-states.”

The United States is uniquely implicated in this charge because of its enormous military, economic and cultural power. What drove people around the world crazy about the Bush administration’s unilateral approach to the Iraq war was its assertion that it was accountable to no one but American voters for what it did in distant parts of the globe. And since institutions like the United Nations are woefully ill equipped to deal with democratic legitimacy, this democracy deficit is a real and abiding challenge at the international level.

The authors are conscious of the charge that they, like the Seattle anti-globalization protesters they celebrate, don’t have any real solutions to these matters, so they spend some time discussing how to fix the present international institutions. Their problem is that any fixes are politically difficult if not impossible to bring about, and promise only marginal benefits. Democratic institutions that work at the nation-state level don’t work at global levels. A true global democracy, in which all of the earth’s billions of people actually vote, is an impossible dream, while existing proposals to modify the United Nations Security Council or change the balance of power between it and the General Assembly are political nonstarters. Making the World Bank and I.M.F. more transparent are worthy projects, but hardly solutions to the underlying issue of democratic accountability. The United States, meanwhile, has stood in the way of new institutions like the International Criminal Court.

It is at this point that Hardt and Negri take leave of reality — arriving at an imaginary solution to their real problem. They argue that instead of ”repeating old rituals and tired solutions” we need to begin ”a new investigation in order to formulate a new science of society and politics.” The woolliness of the subsequent analysis is hard to overstate. According to them, the fundamental obstacle to true democracy is not just the monopoly of legitimate force held by nation-states, but the dominance implied in virtually all hierarchies, which give certain individuals authority over others. The authors dress up Marx’s old utopia of the withering away of the state in the contemporary language of chaos theory and biological systems, suggesting that hierarchies should be replaced with networks that reflect the diversity and commonality of the ”multitude.”

The difficulty with this line of reasoning is that there is a whole class of issues networks can’t resolve. This is why hierarchies, from nation-states to corporations to university departments, persist, and why so many left-wing movements claiming to speak on behalf of the people have ended up monopolizing power. Indeed, the powerlessness and poverty in today’s world are due not to the excessive power of nation-states, but to their weakness. The solution is not to undermine sovereignty but to build stronger states in the developing world.

To illustrate, take the very different growth trajectories of East Asia and sub-Saharan Africa over the past generation. Two of the fastest growing economies in the world today happen to be in the two most populous countries, China and India; sub-Saharan Africa, by contrast, has tragically seen declining per capita incomes over the same period. At least part of this difference is the result of globalization: China and India have integrated themselves into the global economy, while sub-Saharan Africa is the one part of the world barely touched by globalization or multinational corporations.

But this raises the question of why India and China have been able to take advantage of globalization, while Africa has not. The answer has largely to do with the fact that the former have strong, well-developed state institutions providing basic stability and public goods. They had only to get out of the way of private markets to trigger growth. By contrast, modern states were virtually unknown in most of sub-Saharan Africa before European colonialism, and the weakness of states in the region has been the source of its woes ever since.

Any project, then, to fix the ills of ”empire” has to begin with the strengthening, not the dismantling, of institutions at the nation-state level. This will not solve the problems of global governance, but surely any real advance here will come only through slow, patient innovation and the reform of international institutions. Hardt and Negri should remember the old insight of the Italian Marxist Antonio Gramsci, taken up later by the German Greens: progress is to be achieved not with utopian dreaming, but with a ”long march through institutions.”

Francis Fukuyama, a professor of international political economy at Johns Hopkins University, is the author of ”State-Building: Governance and World Order in the 21st Century.”