The simple idea that every nation should have its own state, accompanied by the corollary that one ethnic or cultural group should not collectively rule over another, has been the most powerful political force of the past two hundred years. While particular nationalisms vary, this basic nationalist conception of an ideal world order has been remarkably unchanged for well over a century. "The world should be split into as many states as humanity is divided into nations," the Swiss international lawyer Johann Caspar Bluntschli wrote in 1870. "Each nation a state, each state a national being." When he wrote, nationalism as a considered doctrine, with its roots in the thought of Rousseau, Herder, Fichte and Mazzini, was already generations old. National sentiments, of course, long predated the doctrine, despite recent attempts to claim that national feelings are purely modern fabrications.

The nationalist ideal has survived one universalist assault after another: the Concert of Europe, which Metternich saw as a way of repressing anti-dynastic nationalism and republicanism; Hitler’s supranational racist imperialism; the doomed Soviet effort to replace national loyalties with commitment to socialist universalism. Even the failure of the European Community to become a genuine federal state was foreseeable long before the troubles afflicting the Maastricht treaty and the crisis of the European Monetary System. It seems unlikely that liberal universalism will succeed where illiberal universalisms failed, in attempting to transfer loyalties from nations to supranational entities.

Despite all the evidence of the enduring power of nationalist sentiment, many statesmen, scholars and opinion leaders continue to treat nationalism as an anachronistic or dangerous relic of a previous age. Translated into policy, this prejudice against national self-determination usually means supporting the efforts of regimes to suppress secessionist movements by national minorities. The widespread conviction that nationalist secession is in itself dangerous and regressive helps explain the vehemence with which many observers blamed Germany for its allegedly premature recognition of Slovenia, Croatia and Bosnia, and the criticism directed at the United States for allegedly engineering the independence of Eritrea.

This prejudice against nationalism, even liberal, democratic, constitutional nationalism, is a mistake. Reflexive support for multinational political entities, especially despotic ones, is as misguided as the automatic rejection of movements that seek the sovereignty of national homelands. For practical strategic reasons, as well as reasons of principle, the United States should identify itself with the most powerful idea in the contemporary world.


Having survived so many setbacks since the wars of the French Revolution, will nationalism now end up in the dustbin of history along with its defeated universalist rivals? Scholars and writers (mostly social democrats and classical liberals, but also a few realists, such as E. H. Carr and James Burnham) have been predicting the imminent obsolescence of the nation state for most of the twentieth century. In most cases, they have rested their argument on the economies of scale made possible by advances in technology, the transoceanic cable of yesterday, the computerized stock exchange and satellite television of today.

But this "interdependence" school, like Marxism, is based on a contradiction. It is simultaneously deterministic and prescriptive. If the world is inevitably growing more interdependent, then there is no reason to oppose particular nationalisms that are doomed in the long run anyway. Why oppose what is bound to wither away? On the other hand, if effort is needed to promote transnational integration, then clearly such integration is not preordained.

The mistake of prophets of a postnationalist world has been to leave out moral and political economies of scale. As a purely technical matter, it has probably been possible since Genghis Khan, certainly since Napoleon, for the earth to be governed from a single capital. That all attempts at world conquest have failed has nothing to do with technology and everything to do with the determination of diverse peoples not to be ruled by the conquering nation of the day. This is true of the latest attempt at world hegemony as well. Superior technology made it possible for the Western alliance to out-innovate and outproduce the Soviet bloc, but it was American, German and Japanese desires to protect national autonomy that kept those countries in a four-decade alliance. Why nations that will fight to the death to prevent surrendering their sovereignty to a conqueror would voluntarily surrender it to a supranational bureaucracy or a global elite of financiers and industrialists is a mystery that interdependence theorists have yet to explain.


A somewhat more plausible case against nationalism is made by "stabilitarians," or defenders of the present-day territorial status quo. The harmful effects of alteration of existing borders, even peaceful alteration, would, it is thought, outweigh the benefits. Every viewpoint has an address, of course. A national leader will view stability differently, depending on whether he thinks of his state as a status quo or a revisionist power. The belief of the Bush administration that the United States was a status quo power explains its efforts to keep both the Soviet empire and the Yugoslav federation intact.

While the breakup of a multinational state may create a regional power vacuum or a new balance-of-power pattern among its successor states, these results may be strategically desirable for some countries. Britain, for example, sought the independence of the Low Countries, the Hapsburgs the fragmentation of Italy, and successive Chinese empires the disunity of the nomads in the Tarim Basin. A state may easily conclude that a power vacuum in a particular region is preferable to a rival power center. Given the threat the Soviet Union posed to the United States (and the threat its predecessor, the Romanov empire, posed to Great Britain) it is by no means clear that a consolidated entity on the territory of the Soviet Union is preferable to a balance-of-power system of rival successor states.

Assertions that successful secession by one or a few nations will produce runaway disintegration, thanks to the demonstration effect, deserve to be greeted with the same skepticism that should be directed at other straight-line extrapolations. The domino theory of nationalist disintegration is no more persuasive than similar domino theories. Secessionist activity tends to come in limited bursts: decolonization, the disintegration of the Soviet bloc and empire.

The potential for global disorder inherent in a world community with more states than exist at present is easy to exaggerate. To begin with, the number of possible new nation states is in the dozens, not in the hundreds or thousands. While there are thousands of ethnic nations in the world, there are at most only dozens of national groups numerous, unified and compact enough conceivably to serve as the nuclei of sovereign nation states. The impossibility of basing nation states on tiny minorities like Sorbs or Wends in Germany or the Amish in the United States in no way discredits the potential for statehood of the Kurds or the Ibo or the Tibetans.

Even if the number of nation states were to increase by a dozen or two in the next few decades, through the peaceful or violent partition of several multinational countries, the very inequality of power among states would prevent too great a degree of disorder. A world of 200 or 250 effectively equal states would indeed be unmanageable, but not a world of the same number of nominally independent states, in which real power inheres in a handful of great powers, blocs and alliances. The breakup of nineteenth- and twentieth-century empires has produced, to date, almost two hundred states, well short of the 300 independent political units that existed in early-modern Germany, or the 500 that Charles Tilly has identified in the Europe of 1500. If the world survived the rapid expansion of the number of U.N. member states from 52 in 1946 to 183 today, surely it can survive a more incremental expansion by a dozen or two more.

Would the replacement of some of today’s multinational states by new nation states lead to an increase in interstate war? History since the great wave of postwar decolonization in Africa and Asia gives some cause for reassurance in this regard. While many postcolonial states have been riven by ethnic conflict (reflecting the fact that they themselves are often ethnically heterogeneous), major interstate wars have been relatively infrequent.

Although prophets must be careful, it is possible that there would be less interstate conflict in a world of relatively homogeneous nation states than there is intrastate conflict between ethnic groups in multinational states. There are powerful incentives against engaging in cross-border war, whereas the penalties against a dominant ethnic group crushing others in the state it controls are very weak indeed.

Opponents of secessionist nationalism frequently argue that larger minorities, once they gain independence, may in turn oppress smaller minorities in the new national territory. The Quebecers, if independent, might be more inclined to oppress American Indians in Quebec. (The Balkan war is not terribly relevant, inasmuch as Slovenes, Croats and Yugoslav Muslims seceded in the first place to escape oppression in a multinational federation dominated by Serbs.)

Without condoning any injustice, the fact that a secessionist nation engages in oppressive behavior does not mean its complaints about its own oppression at the hands of a central government or dominant imperial ethnic group are not legitimate. Even criminals may be victims of crime. Inevitably, the replacement of a multinational empire or federation by a group of nation states will leave minorities that are too small or too dispersed without states of their own. This, in itself, is no argument for holding the multinational structure together, unless the multinational elite is significantly more virtuous than the successor national elites, which is rarely the case.

The relatively bloodless dissolution of the Soviet Union into its constituent republics, the separation of the Czech and Slovak republics, the accession of East Germany and, earlier, the Saarland to the German Federal Republic, as well as a number of cases of postcolonial independence, prove that national self-determination need not be accompanied by violence. Those concerned that national self-determination will lead to violence should support the strengthening of peaceful constitutional and diplomatic procedures for increasing the congruence of borders and nations, rather than support the status quo at all costs.


Is there a lower limit to the size of a viable nation state, imposed by the needs of defense or economics or minimum international order? If the viability of a state is defined as its military invulnerability in the absence of allies and economic autarky, then the only viable states would be isolationist, continental superpowers (rather like the Eastasia, Eurasia and Oceania of Orwell’s 1984).

As long as states are willing to cooperate in security alliances and engage in mutually beneficial trade, there is no reason why a small state like Portugal or Croatia should not be as viable as a great power like the United States. In an integrated North American market, an independent Quebec might prosper, even while preserving its distinctive French-American identity (though the transition might be painful). Indeed, smaller states may have advantages over the populous when it comes to economic progress (contrast Singapore and Hong Kong with China). Instead of specializing in one or a few crops, like states in a federation or provinces in an empire, an independent nation can take steps to diversify its economy as a buffer against market shocks. A sovereign state can also have a certain amount of leverage in both economic and military diplomacy, an advantage denied to a region subordinated to a single capital.

It might be thought that the costs of defense for a small nation state would be prohibitive. In fact, experience shows that small nation states do not spend more on defense as a share of GDP than do large countries. Indeed, during the Cold War the United States spent proportionately more on defense than its medium-sized allies like Germany and Japan, or small allies like Denmark and Portugal. A small state can act as a free rider, taking advantage of a powerful neighbor’s interest in defending not only itself but its region. Of course such a neighbor may be a threat as well as an ally, but this is a risk that might be worth taking. After all, Kurds would be safer from Baghdad even in a weak Kurdish nation state than they can ever be as part of Iraq.

At any rate, an argument for the benefits of scale is an argument against small states of any kind, against small multinational states, like Switzerland, as much as small nation states like Slovenia. It is not in itself an argument against making nationality the basis of statehood wherever substantial geographical concentrations of linguistically and culturally similar people exist.


Support in some circumstances for national self-determination need not mean support for nationalism in its tyrannical or imperial manifestations. It is important to draw a distinction between liberal and illiberal nationalism. Liberal nationalists tend to favor a linguistic-cultural definition of nationality and a liberal-constitutional (though not necessarily democratic) organization of the state. Illiberal nationalists (who might more accurately be described as nativists, to employ a term that originated in American politics) favor a religious or genetic definition of nationality, as in Iran or Serbia, and usually (though not always) an authoritarian-populist constitution. It is as great a mistake to confuse liberal nationalism with illiberal nativism as it is to identify social democracy with Leninist communism. Illiberal nationalism is often responsible for terrible atrocities, as the carnage in Bosnia has shown. The problem, however, is with illiberalism and militarism, not with nationalism as such.

Liberal nationalism holds that, far from being a threat to democracy, nationalism, the correspondence of cultural nation and state, is a necessary, though not sufficient, condition for democracy in most places today. Modern democracy presupposes a degree of extra-political community. The linguistic-cultural nation is today generally accepted as the basis for the political community because it is the largest particular community that can still command sentimental loyalty and the smallest comprehensive community that still has features of universality, combining all ages and classes. The nation is a small humanity and a large association. "Few will burn with ardent love for the entire human species," Tocqueville observed. "The interests of the human race are better served by giving every man a particular fatherland than by trying to inflame his passions for the whole of humanity."

Some claim that national loyalty is irrational and atavistic, compared to "rational" patriotism or allegiance to a state that is not identified with any predominant linguistic or cultural group. There is nothing at all "irrational," however, about making the suprafamilial community with which one identifies the cultural nation, rather than the territorial state in the abstract. Quite apart from the psychological reasons, there are practical reasons. In most of the world, nations are considerably older than states. One is usually born into a cultural nation for life, but the state to which one owes allegiance may alter its borders, change its constitution, change its name, even cease to exist through conquest or merger. National communities, while by no means immortal themselves, tend to be more stable and long-lived. This being the case, to identify primarily, not with a historic linguistic-cultural nation, but with a possibly transient government or a paper constitution, would be the height of irrationality.


The evidence that democracy almost never works in societies that are highly divided along linguistic and cultural lines is overwhelming. Examples of multinational countries that have failed are numerous: Cyprus, Lebanon, Sri Lanka, Sudan, the Soviet Union, Yugoslavia, Czechoslovakia (India increasingly looks like another failure). Nevertheless, many persist in arguing that multinational democracy not only is possible but represents the wave of the future. Multinational despotisms, they argue, should not be partitioned into nation states that (in some cases) may become democracies. Rather, they should be transformed from multinational despotisms into multinational democracies.

As examples of successful multinational federations, proponents of multinational democracy usually point to three countries with elaborate ethnic power-sharing arrangements: Switzerland, Belgium and Canada. The very fact that only three successes can be found, out of dozens of multinational states, in itself suggests the difficulty of getting linguistically and culturally distinct nations to cohabit peacefully under a common democratic constitution. In reality, these three examples hurt the multinationalist case more than they help it. Switzerland, for example, is better described as a confederation of relatively homogeneous territorial nation states (the cantons) than as a truly multiethnic society. Belgium is a society deeply troubled by its linguistic and political divisions, and Canada recently almost came apart over the Quebec question. The two "founding nations" of Canada may yet go their separate ways, like the Czechs and Slovaks did.

The fact that Switzerland and Belgium are small countries (and Canada a huge country with a small population) tends to contradict another argument made in favor of holding multinational entities together: the argument that more populous states benefit from economies of scale. While there are economic and military returns to scale, these may be neutralized if they are accompanied by the costs of increased ethnic diversity accompanied by ethnic conflict. All other things being equal, a large homogeneous nation state may well be preferable to a small homogeneous nation state. But a small nation state may be better off, in terms of prosperity and governability if not necessarily defense, than a gigantic state riven by ethnic and linguistic conflicts.

Those who call on nondemocratic multinational states to adopt Swiss- or Canadian-type power-sharing arrangements as an alternative to partition seldom describe the policy to be pursued if their constitutional panaceas fail (as they have in most cases). If elaborate power-sharing schemes are rejected, or tried and found not to work, is partition or secession then in order, as a second-best option? Or should multinational states like Iraq that cannot be held together by democratic and federal means be held together by force and terror?

Those who seek to promote democracy and at the same time to preserve multinational entities intact will discover that in many cases these goals cannot be reconciled. A world of liberal nationalist states, including many that are nondemocratic, is much more likely to develop into a world of democracies, as Franco’s Spain would attest. For this reason, proponents of democratization are justified in encouraging liberal nationalism even where democracy is not yet possible. This might be the case, for instance, in Algeria or Egypt. Conversely, it may be a waste of time to try to hold together and democratize a multinational state, even a relatively liberal one, where a common national identity is lacking. This might be the case in the future in South Africa. Instead, it often makes more sense to promote liberal and constitutional nationalism, with or without electoral democracy. First comes the nation state, then a liberal constitution reinforced by a liberal political culture, and only then, if at all, democracy. For many, living as the citizen of a liberal but nondemocratic nation state is preferable to being the subject of an illiberal multinational despotism that can only be held together by force.


To substitute indiscriminate support for national self-determination for reflexive defense of the territorial status quo around the world would be a mistake. Rather than strict principles, a few rules of thumb are in order. To begin with, the United States should refrain from making gratuitous statements in favor of state unity, like President Bush’s lamentable "Chicken Kiev" speech. Even if American policy is to favor state unity in a particular case, the United States might lose some of its leverage if this preference is too obvious. Even worse, the United States may appear to license vicious repression, as the Bush administration’s statements in favor of Yugoslav unity may have convinced Serb nationalists that they would not be penalized seriously for attacking Slovenia and Croatia.

In civil wars where ethnic or cultural differences are at issue, partition should no longer be considered the last resort. It might sometimes be wise to stress national self-determination above free elections, during the terminal crisis of a state that is both multinational and undemocratic. Oppressed nations seeking to escape from a multinational empire should not be told that they will be free to vote on everything except their independence. Since democracy and liberal constitutionalism work best in relatively homogeneous nation states, in most of the world democratic constitution-writing should follow national independence, not be promoted as an alternative to it. Indeed, it is not only futile but insulting for policymakers and academics in Western capitals and campuses to design democratic federal constitutions like the Vance-Owen plan for Bosnia and try to impose them on Kurds, Kosovars or Kashmiris as alternatives to national independence. Where a multiethnic federation has utterly collapsed, it may be better to create two or more new, relatively homogeneous nation states than to try to piece the wreckage together with ingenious but unworkable power-sharing schemes.

The corollary of support for national self-determination in the form of secession is support for the enlargement of nation states through peaceful and democratic accession or annexation, like the unification of Germany. The United States enlarged itself in this manner as recently as 1958 (with the statehood of Alaska and Hawaii); President Bush called for statehood for Puerto Rico. If east Germany can join west Germany, by what reasoning can the 90 percent majority of ethnic Albanians in Kosovo be denied accession to Albania, if they choose and can make their choice effective? How can compact populations of Bosnian Croats be forbidden by the international community from voluntarily merging with Croatia (the very international borders of which are recent and fluid)? States should be allowed not only to shrink but to expand, so long as the expansion is undertaken peacefully and with the consent of majorities (or perhaps supermajorities) of those affected. The difficulties attending European unification suggest that we need not fear the creation of possibly overpowerful bureaucratic superstates like the Third Reich and the Soviet Union through purely voluntary mergers.

The United States may legitimately refuse to support nationalist movements that define the nation in narrow racial or religious terms, rather than in inclusive linguistic and cultural terms, as well as movements that threaten minorities with persecution or genocide. Also, as a condition of admitting new nation states to the international community, outsiders may legitimately insist that states protect the rights of association of individual members of cultural minorities, such as private religious or language instruction. It would be a mistake, however, for the international community to attempt to promulgate a general duty of states not only to tolerate but to subsidize and promote minority cultures. Such policies, whether undertaken as a result of international or purely domestic pressure, tend only to inflame majority resentment without accomplishing any important goals that cannot be achieved by less intrusive, more voluntary means.

The wave of disintegrative nationalism that ripped apart the former Soviet Union and the Yugoslav federation will not be the last. In all likelihood, the next few decades will see increasingly determined secessionist movements in the multiethnic successor states of the European empires: India, Pakistan, South Africa, Iraq, perhaps even the Russian federation. In such countries, as dominant elites, seeking new formulas for legitimacy to replace fading secular and socialist philosophies, make more concessions to the national and religious sentiments of ethnic majorities, minority nationalisms may grow more bitter and intense in response. The fact that in many, perhaps most, cases central authorities will prevail will not prevent secessionist nationalism from being a major source of terrorism and civil war in the 21st century. The United States does not need to become an exporter of secession. Washington should recognize, however, that in particular cases American values, as well as American interests, may be served by those who seek to break up multinational states rather than by those who seek to preserve them.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Michael Lind is Executive Editor of The National Interest. This article is adapted from his forthcoming book, The Next American Nation.
  • More By Michael Lind