Zimbabwe's crisis has incited an unsettling feeling of déja vu . The reason is clear: it is, thankfully, no longer fashionable to decry colonialism's evils in assigning blame for every national misfortune. The imperial statues are toppled, cities and streets renamed, the vestiges of foreign rule either abandoned or adapted. With the sole exception of Zimbabwe, no leading politician in any post-imperial country has made a notable speech in recent years attacking colonialism. That great staple of political rhetoric appears to have been buried across the developing world.

Internationally, colonialism is even more passé. Once, the votaries of one kind of new international order or another decried the evils of imperialism (sometimes, but not always, prefixed with a ``neo-'') in justifying demands for a more just dispensation. That theme has died out in diplomatic discourse. Yet followers of world affairs would be unwise to consign colonialism to the proverbial dustbin of history, for it remains a factor in understanding our world's problems and dangers.

To begin, residual problems from the end of the earlier era of colonization, usually the result of untidy exits by the colonial power, still remain dangerously stalemated. The events in East Timor in 1999 remain fresh in memory, and difficulties linger. But at least closure seems in sight, unlike those messy legacies of European colonialism: Western Sahara, Cyprus, and Palestine.

Moreover, fuses lit in the colonial era could re-ignite, as they have done, to everyone's surprise, between Ethiopia and Eritrea, where war broke out over a colonial border that Italy's occupiers had failed to define with precision. In Zimbabwe, colonial land ownership patterns that gave most of the viable farmland to white settlers are at least one root of the country's current crisis.

But not only the direct results of colonialism remain relevant: the indirect ones also matter. The intellectual history of colonialism is littered with many a willful cause of recent conflict. One is careless anthropology: Belgium's classification of Hutus and Tutsis in Rwanda and Burundi, which reified a distinction that had not existed before, still haunts Africa's Great Lakes region.

A related problem arises from sociology: how much bloodshed do we owe, for instance, to the British invention of ``martial races'' in India? One can never overlook the old colonial administrative habit of ``divide and rule,'' exemplified, again, by British policy in the subcontinent after 1857, which led almost inexorably to the tragedy of Partition. Such distinctions were not merely pernicious; they were often characterized by an unequal distribution of the resources of the state within a colonial society. Belgian colonialists favored Tutsis, leading to Hutu rejection of them as alien interlopers; Sinhalese resentment of Tamil privileges in colonial-era Sri Lanka prompted the discriminatory policies after Independence that fueled the Tamil revolt.

A ``mixed'' colonial history within one modern state is also a potential source of danger. When a state has more than one colonial past, its future is vulnerable. Ethnicity or language hardly seem to be a factor in the secessions (one recognized, the other not) of Eritrea from Ethiopia and the ``Republic of Somaliland'' from Somalia. Rather, it was different colonial experiences (Italian rule in Eritrea, British rule in Somaliland) that set them off, at least in their own self-perceptions, from the rest of their ethnic compatriots.

A similar case can be made about the former Yugoslavia, where lands that existed under Austro-Hungarian rule for 800 years were joined to lands that spent almost as long under Ottoman suzerainty. The war that erupted in 1991 in no small measure pitted those parts that had been ruled by the German-speaking empires against those that avoided such colonization.

Boundaries drawn in colonial times, even if unchanged after independence, still create problems, especially in Africa. Where colonial constructions force disparate peoples together by the arbitrariness of a colonial map-maker's pen, nationhood becomes elusive. Older tribal and clan loyalties in Africa were mangled by the boundaries drawn, in distant cities like Berlin, for colonially-created states whose post-independence leaders needed to invent new traditions and national identities. The result was the manufacture of unconvincing political myths, as artificial as the countries they mythologize, which all-too-often cannot command genuine patriotic allegiance from their citizenry.

State failure in the wake of colonialism is another source of conflict. The collapse of effective central governments - manifested in recent years in Sierra Leone, Liberia and Somalia - could unleash a torrent of alarming possibilities because ``weak states,'' particularly in Africa, seem vulnerable to collapsing in a welter of conflict.

Underdevelopment in post-colonial societies itself causes conflict. The uneven development of infrastructure, as a result of priorities skewed to benefit colonialists, can lead to resources being distributed unevenly, which can lead to increasing social fissures. Underdevelopment in many countries of the South, which are faring poorly in their struggle to remain viable in a globalizing world, creates conditions of desperate poverty, ecological collapse and rootless, unemployed populations beyond the control of atrophying state systems.

We will not create a better world in the 21 ^st century by forgetting what happened in the 19 ^th and much of the 20 ^th . This is not to say that our responses to the dangers emerging from the legacies of the past must be rooted in the past. There is a greater need than ever before for innovative, forward-looking approaches to global governance. As we embark upon the still-new millennium, though, it seems ironically clear that tomorrow's possible disorder might be due, in no small part, to yesterday's colonial order.

I have no wish to give politicians in post-colonial countries whose leadership has been found wanting any reason to find excuses in the past for their own failures. But in looking to understand possible future sources of conflict, the best crystal ball may be a rear-view mirror.