Review of George Modelski’s   (2003) World Cities, -3000 to 2000, Washington, DC: FAROS 2000. 245 pages, $20.00


Christopher Chase-Dunn and Daniel Pasciuti    December 6, 2004



            George Modelski’s main fame stems from his studies of the rise and fall of great powers in the modern world system (Modelski and Thompson 1988, 1996). His recent book World Cities, -3000 to 2000 shifts the focus of attention to human social evolution over the the past 5000 years. It is a formulation and testing of a theoretical model of human social evolution that focuses on the growth of world cities and it is also presents to results of a huge empirical effort to expand our knowledge of the population sizes of the largest settlements on Earth since the Bronze Age. Modelski utilizes the data on city growth to evaluate his new theory of social evolution. The growth of cities is a useful indicator of world system evolution because the ability of a society to produce and maintain a large settlement is a major accomplishment. We can trace the emergence of social complexity by knowing where the largest human settlements are at any point in time. Beginning with Uruk, the first “world city’ five thousand years ago, Modelski traces the emergence and spread of large cities from Mesopotamia and Egypt to East Asia, South Asia, Europe and the Americas.

            Modelski’s evolutionary approach focuses on a single world system that begins with the first cities and states in Mesopotamia 5000 years ago, and spreads out to become global. In focusing on this single system he ignores the differences between regions and civilizations that are of interest to other world historians, but his focus on world cities across large expanses of time allows him to see patterns that other analysts miss.

            Regarding the city population data, Modelski has extended and improved the work of that most eminent coder of city sizes, Tertius Chandler (1987).  For students of urban and world history this work is of immeasurable value. Modelski has labored hard to produce the best comprehensive compilation of estimates of city population sizes now available.        

Modelski’s careful improvement upon earlier efforts to estimate the population sizes of ancient cities is a huge step forward. He uses estimates of the built-up area of a city and a population density factor (see p. 11 and Note 5 on p. 17) to estimate the population sizes. He adds considerable depth, especially to the coverage of the Bronze Age. These data are presented in the new book under review and early versions of the data from the ancient period and from East Asia are available from Modelski’s Evolutionary World Politics web site.

In the “ancient era” (-3000 to –1000) world cities are defined as those that reach a population size of 10,000 or more. In the following “classical era” (-1000 to 1000) cities must be at least 100,000 in population size to count as world cities. And in the modern era (since 1000) the cut-off point is one million.[1] Modelski observes a phenomenon, also noticed by Roland Fletcher (1995), that a few cities are the first to reach a whole new scale, and then a size ceiling is encountered during which cities in other regions catch up to the new scale. The current maximum seems to be around twenty millions and the phenomenon of catching up is now occurring. Some of the world’s largest cities are now in developing countries such as Mexico, Brazil and India.

Modelski’s study of the phases of urbanism is convincing regarding the contention that urbanization has been neither random nor linear. Instead it has followed a recurring pattern of rapid growth followed by slow growth or decline. A phase of fast growth concentrated in one or a few regions is followed by slower growth and the diffusion of large cities to other areas. Rapid and concentrated growth is followed by leveling off and dispersal due to “countervailing forces.” These countervailing forces emerge from what Modelski terms the “Center-Hinterland” divide of a regional world system. The first growth phase emerges in a center that eventually encounters limits to growth from resource exhaustion, environmental stress and “failures of knowledge.” The leveling process occurs as these limits are reached, weakening the old center. Incursions from the hinterland increase, taking advantage of the center’s weakness. This allows the semi-hinterland, a region adjacent to the old center with smaller cities, to catch up to the urban scale of the old center.

            Modelski also compares his phases of urban growth with existing estimates of overall population size and growth. He finds that the overall population growth phases correspond in time with the urban expansions of the three eras. This study leads to what Modelski calls a “manifest case of evolution.” The three phases of urbanization correspond to periods of world system evolution: cultural, social and political. The ancient cultural phase saw the creation of a learning structure based on cities, writing and calendars, resulting in a platform for sustained and intensified human interaction on a large scale. The classical social phase brought about a more extensive, inclusive and integrated system. Expanding during Karl Jaspar’s “ Axial Age,” the cities of the classical period can be grouped according to the world religions that dominated social structures during that era.  The modern political phase poses choices regarding an evolutionarily stable structure of world organization. Modelski predicts that the future fourth phase will be an economic one that will see a “stabilization and consolidation of the economic and material basis” of world society.

While this learning model of human social evolution may seem a bit too functionalist or Whigish to some, it has original theoretical elements, such as the center-hinterland dynamic, that have been absent from most earlier evolutionary models. This, plus its monumental empirical contribution, makes this an extremely valuable step forward toward our comprehension of the human experiment with complexity.


Chandler, Tertius 1987 Four Thousand Years of Urban Growth: The Edwin Mellen Press. Fletcher, Roland 1995 The Limits of Settlement Growth: A Theoretical Outline. Cambridge:  Cambridge University Press.

Modelski, George and William R. Thompson 1988 Seapower in global politics, 1494-1993. Seattle: University of Washington Press.

Modelski, George and William R. Thompson. 1996. Leading Sectors and World Powers: the Coevolution of Global Economics and Politics. Columbia, SC: University of South Carolina Press.


George Modelski’s “Evolutionary World Politics” web site is at


[1] These cut-off points are used by Modelski to determine the list of world cities during each era, rather than the more conventional approach of studying the largest ten or twenty cities on Earth. The only problem with Modelski’s approach is that, at the beginning of each of his eras the number of cities above the threshold are few, and so information is missing from his dataset on large cities that may be just under the threshold but still among the largest cities on Earth.