The World System: Five Hundred Years or Five Thousand? (review)

Many students of world history have in recent years become critical of the way in which Marxist world history, both generically and in the influential world-systems paradigm of Immanuel Wallerstein, reflects a Eurocentric focus by making European capitalism the motor and story of modern world history since 1500. In Before European Hegemony: The World System A.D. 1250-1350 (New York, 1989), Janet Abu-Lughod described extensive Eurasian trade networks that she said constituted a thirteenth-century world system a hundred years before the European.

In The World System: Five Hundred Years or Five Thousand? Andre Gunder Frank and Barry K. Gills ask us to push back the origins of a world system much further: to the first cities, urban trade networks, and imperial hegemons five thousand years ago. The ancient and medieval periods, they maintain, display the same three characteristics that Wallerstein noted for the post-1500 era: central-core exploitation of peripheral regions, alternating periods (about two hundred years each) of expansion and contraction, and hegemonic rivalries and successions (though pre-1500 hegemons were often multiple rather than single powers).

Frank and Gills would be the first to admit that their thesis is exploratory—a research agenda rather than a definitive conclusion. Neither is a specialist in ancient history. Certain questions—for example, whether they accept Wallerstein's belief in the polarization of core/periphery relations for five hundred or for five thousand years — are only posed, not answered.

A thumbnail sketch of expansive "A" periods and contractive "B" periods from 1700 B.C. to A.D. 1700 is dazzling in its reach, but ultimately disappoints in its grasp. The same sparse evidence is marshaled to support the existence of these periods and to determine if they are expansive or contractive. For evidence, the authors draw on a mixed bag of current specialists, dated secondary sources (V.Gordon Childe from 1942, Frederick Teggart from 1939), estimates of urban census figures, old war stories, and myths of invasions and "dark ages." It is not surprising that in a recent Current Anthropology (34 [1993]) forum on Frank's idea of a Bronze Age world system, seventeen specialists displayed far more awe than agreement with respect to the specifics of Frank and Gills's periodization.

While Frank and Gills should be applauded for their efforts, there are just too many issues left unresolved; many are probably unresolvable. Not the least of these are the methodological problems of defining and characterizing the "A" and "B" phases. The authors seem to imply that these phases are more homogeneous than post-1500 core/periphery imbalances. But when a region seems out of sync, it becomes the exception that proves the general trend. Authorities are pressed into service beyond the call. To show that the period from 100/50 B.C. to A.D. 150/200 was an expansive "A" phase, vague generalities from Marshall Hodgson are cited as evidence of Parthian economic expansion. But Hodgson is writing about the later Sassanian period, which the authors have designated as a contractive "B" phase, from A.D. 150/ 200 to 500.

The question one might ask is why Frank and Gills venture so far into terra incognita. Why not an attempt to push the world system back an additional three hundred years, pace Abu-Lughod? or even five hundred? Why an additional forty-five hundred, all in one sweeping thesis? Their strategy is especially surprising since they quote the sage advice of John K. Fairbank that the historian should "begin at the end...and let the problems lead you back. Never try to begin at the beginning."

Frank and Gills begin at the beginning not merely to extend Wallerstein's work, but to call for a radical revision of its Marxist underpinnings. Unlike a study of a thirteenth...