Andre Gunder Frank

The World System and Its Cycles in the Bronze and Iron Ages

INTRODUCTION

I wish here to pursue and extend the earlier explorations by Gills and Frank (1992) into the shape, expansion and cyclical behavior of what I contend was the world system during the bronze and iron ages. In so doing, I join and use the work of a number of archaeologists who are now applying or at least discussing "world systems theory." It is a conceptual aid or framework for the reinterpretation of existing and newly unearthed information on bronze and iron age societies. As part of this task, the main intent here is to try to improve the mapping of the extent and the dating of the growth cycles of the bronze and iron age world system. To do so, I do not pretend to provide new information, but rather to systematize the only infrequently wide ranging systematic reviews and many more often scattered remarks on particular times and places of these same archaeologists and others.

I have previously advanced the thesis that the present world system was born some 5,000 years ago or earlier in West Asia, North Africa, and the Eastern Mediterranean (Frank 1990, Gills and Frank 1990/91, 1992 and Frank and Gills 1992). The argument was similar to that of David Wilkinson (1987) who identified the birth of "Central Civilization" through the establishment of systemic and systematic relations between Egypt and Mesopotamia around 1500 BC. However, the date of the formation of the world system was pushed back farther to at least 3000 BC by analogy to the upstream confluence of two or more major branches to form a single river like the Mississippi-Missouri (Gills and Frank 1990/91). Moreover, we suggested that already in the third millennium BC the world system included not only Egypt and Mesopotamia. It also included the Arabian Peninsula, the Levant, Anatolia, Iran, the Indus Valley, Transcaucasia, and parts of Central Asia. All the regions were in direct "bilateral," or at least chain- linked indirect "multilateral," systematic and therefore systemic contact with each other (Gills and Frank 1992, Frank and Gills 1992).

We also sought to identify "World System Cycles, Crises and Hegemonial Shifts 1700 BC to 1700 AD" (Gills and Frank 1992). In a concluding summary table, we identified the cycles and expansive [A] and contractive/crisis [B] phases as follows:

On this occasion, I wish to confine my review and where necessary revision of the cycle dates only to the pre- Christian era bronze, iron and part classical age period. However, I also extend the identification of the cycles backward through or at least into the 3rd millennium BC. [The latest two millennia are reviewed separately and with little change to 1700 AD in Frank and Gills (1992b) and with special reference to Central Asia in Frank (1992c) and extended to the present day with reference to Latin America in Frank (1992d)].

Already in an "Epilogue" to the earlier Gills and Frank (1992) article written when it was already in press, I noted that David Wilkinson (1992) and Andrew Bosworth (1992) had proceeded independently of each other to test the existence and timing of our cycles [wich were proposed in a draft paper presented at professional meetings in 1991]. Both used data from Tertius Chandler's (1987) census of growth and decline in city sizes. Both confirmed the existence and most but not all of the timing of our cycles. I also pointed to the dating of periods during the bronze age 1st millennium BC by Andrew and Susan Sherratt (1991), which coincided almost exactly with our dating of the up and down phases. Moreover, I referred to Kristian Kristiansen's (1992 and forthcoming) also similar dating of expansions and contractions in Europe during the 1st millennium BC, and to Klav Randsborg (1991) for the 1st millennium AD.

Moreover, now I also have additional evidence available: Some systematic information on dates was already previously published for instance by George Dales (1976) on Iranian-Indus relations and by Shereen Ratnagar (1981) in a tabular summary of her study of the "Encounters" between the Indus Valley Harappa Civilization and points West. On these points West, some more partly systematic information is still unpublished by Christopher Edens and Philip Kohl (n.d.). E.N. Chernykh (n.d.) proposes cycle datings for most of Eurasia in his Russian language Ancient Metallurgy in the USSR: The Early Metal Age, whose English language translation is still unpublished. Other references to more or less precise dates of especially economic expansions and contractions here and there are scattered through previous and new works of various other archaeologists and others. Because of the gap betwen their expertise and mine, I often prefer to let thejm speak for themselves by quoting them directly, instead of my trying to convey the same in my own inexpert way.

I draw on this information below in the attempt to [re]assemble the jig saw puzzle picture of the changing extent and cyclical development of the world system in the bronze and iron ages. However, this jigsaw puzzle assembly differs from the usual kind in several ways that make it much more difficult: 1. the number of pieces in "the box" is indeterminate, indeed infinite [if cut small enough]; and it is possible to place or assemble only a few of them here. 2. There is no original design or intended final picture on top of "the box" to guide the assembly. 3. It is not possible to follow the usual easier procedure of beginning the assembly with pieces that have at least one straight line, which help define the outer margin of the picture. In this case on the contrary, it is the very outer margin that is most difficult to define; and it seems easiest to begin with some pieces that appear to be in the better known "core." 4. The task is not a one time enterprise. The shape of the pieces themselves and their [core-periphery and hegemonic] fit with their neighbors changes constantly over time. Perhaps this change us near- randomly; perhaps it also ocurrs in cycles that should also be identified. 5. One the principal tasks, indeed the main intent here below, is to [re]define such cycles.

Before so proceeding however, I wish briefly to review and take position in two underlying theoretical debates about bronze and iron age economies and world-systems [and their jigsaw puzzle design and assembly].

The "Ancient Economy" Debate

We may distinguish a debate about the "extent of the market" [to recall Adam Smith's phrase that related it to the "division of labor"] in ancient economy. There is also a related discussion about the applicability "world systems" theory or concepts to this economy. Other recent reviews by Kohl (1989), Edens and Kohl (n.d.), Woolf (1990) and Sherratt (1991) of both discussions would make still another extensive treatment of the same here superfluous. Indeed, theirs were really reviews in turn of already lengthy reviews of the first debate by Silver (1985) and others. In the first debate, Edens and Kohl distinguish the following positions: Among historians, the primitivists like Weber [and more recently Finley] vs. the modernists like Meyer. Among anthropologists, the substantivists headed by Polanyi and his defenders who are joined in an intermediary position by Renfrew and his followers among archaeologists who see some pass-me-on chain- linked down-the-line trade, vs. the formalists like Herskovits [and ...], who argue that the market existed and/or modern economic analysis is applicable to ancient economy. The primitivist/ institutionalist/ substantivist denial by Weber, Polanyi, Finley, Renfrew and indeed by Marx before them of the importance of market relations, and a forteriori of capital accumulation, and of the significance of long distance trade in the ancient world had already been challenged de facto long ago. Among historians, Weber's contemporary Werner Sombart (1967,1969) did so, and then so did Gordon Chile (1936, 1942) among archaeologists. Yet even Childe "consistently underestimated" the strength of the opposite case, according to Kohl (1987). Moreover, in a posthumously published essay even Polanyi (1975,1977) wrote that "throughout, the external origin of trade is conspicuous; internal trade is largely derivative of external trade ...either from ... (status motive)- or for the sake of gain...(profit motive)" (Polanyi 1975: 154,136-7).

In the 1970s, archaeologists rejected the earlier Polanyi/Finley views and offered reinterpretations of increasingly available data in the pathbreaking "Anthropological Perspectives on Ancient Trade" by Robert Adams (1974) and "Third Millennium Modes of Exchange and Modes of Production" by C.C. Lamberg-Karlovsky (1975). About the same time, H.W.W. Crawford (1973:273) had also observed "increasing evidence for private ownership of land, property and therefore capital" and suggested that temples may have acted as banks.

Since then, the empirical and analytical refutation of the primitivist-substantivist argument has been almost unceasing. The evidence -- more through the record of archaeological finds than through surviving literary texts -- has been substantial. The related arguments about the importance in and for ancient economy of very long distance trade, market relations, demand and supply related price formation, monetization, entrepreneurship, yes and capital accumulation have been so overwhelming that we can at best only point to some of the tip of the iceberg. Among the more conceptual writings are those of Kajsa Ekholm (1980), Ekholm and Friedman (1982), Rowlands, Larsen and Kristiansen (1987), Philip Kohl (1987, 1989, 1991), and Christopher Edens and Kohl (n.d.). The more empirical reviews include those by George Dales (1976), Shereen Ratnagar (1981), C.C. Lamberg-Karlovsky (19xx), and too many others to recount. However in this regard, Edens and Kohl (n.d.) also refer to Powell (1977), Foster (1977), Gledhill & Larsen (1982), Zagrell (1986), and Charpin (19xx). In these debates, I must take sides with the anti-primitivist "marketeers." latter. This view may be summarized by Lamberg-Karlovsky's (1975) early explicit rejection of the Polanyi position on the role of "profit, price-fixing, wholesaling, supply-demand, or even private ownership of land for surplus production....It is the central thesis ... that all of these existed in a market network at least by the end of the fourth millennium in Mesopotamia." Many records from the 3rd and 2nd millennia also attest to fluctuations in the prices of gold and silver relative to each other as well as over both long and short terms to land, slaves, grains, oil, and wages which also changed relative to each other. If these price changes did not directly respond to supply and demand, they did so through administered prices, which also had to respond at least politically to supply and demand. Moreover, "evidence is abundant of the accumulation of human and material capital, including circulating capital not directly involved in the production process ... and fixed capital" (Silver 1985:163). Documentation from late 3r and early 2nd millennium Mesopotamia analyzed by Mogens Larsen (1987) suggests that public and private accumulation were both simultaneously complementary and that each replaced the other in relative importance back and forth over time. I might suggest that they probably also did so over less and more prosperous phases of the economic cycle.

The refutation of Polanyi and Finley about the importance of market relations as well as the market places he recognized at the local level was one thing. The recognition of the related vital importance of long distance trade and trading networks, beyond Polanyi's local markets and Renfrew's hand- me-down non-trade was another thing. Robert Adams (1974: 247,248) found "little doubt that long-distance trade was a formidable socio-economic force" and also that "we have wrongly deprecated the entrepreneurial element in the historical development of at least the more complex societies." Indeed as observed above, even Polanyi himself came to say that internal trade was largely derivative from external trade. Moreover, as Kohl (1989:228) remarks, "the intercultural trade that developed between resource-poor southern Mesopotamia and resource-rich highland areas of Anatolia and Iran necessarily transformed the productive [and political, social and cultural, in a word civilizational state formation?] activities of all societies participating in the exchange network without the development of an overarching polity or empire."

Center-Periphery and World-System/s Theory

Some more conceptual writers among the "marketeers" and "long-distance traders" have also sought recourse to at least some aspects of "world-systems" theory. Thus, a new wave in archaeological studies has recently appeared. It applies center-periphery and/or world systems analysis to the study of complex societies of the past. Rowlands, Larsen and Kristiansen, Eds. (1987) entitled a book Centre and Periphery in the Ancient World; Champion (198x) edited one on Centre and Periphery: Comparative Studies in Archeology; Chase-Dunn and Hall 1991 on Pre-Capitalist Core-Periphery Relations; Guillermo Algaze (n.d.) compares several "Prehistoric World Systems, Imperialism, and the Expansion of some Early Pristine States;" Mitchell Allen (1986) discusses "Assyrian Colonies in Anatolia: A World System Perspective;" Greg Woolf (1990) discusses "World-systems analysis and the Roman Empire," and Andrew Sherratt writes of "Core, Periphery, and Margin: Perspectives on the Bronze Age" (n.d.) and asks "What Would a Bronze Age World System Look Like?" (1992).

A half century earlier, Gordon Childe had already written that "if the economy of the Early Bronze Age cities could not expand internally, owing to the over-concentration of purchasing power ... the urban economy must - and did - expand externally" (Childe 1942:139). The center sought "to persuade their possessors to exchange the needed raw materials for manufactures". According to Childe this trade was from the beginning a political trade between elites in the center and elites in the periphery, in which the center sought to induce the periphery to render up a surplus. That is how Childe explained for instance the commercial ventures and associated military campaigns of the Akkadian King Sargon I in 2350 BC.

Recent excavation at Habuba Kabira in northern Syria of a south Mesopotamian colony "represents a deliberate Lower Mesopotamian penetration up the Euphrates ... to secure direct control of vital raw materials and luxuries from the Syrio- Anatolian regions and to regulate exchange of goods from the east and south-east passing this way" (Moorey 1987:44).

More recently, Frederik Hiebert and C.C. Lamberg-Karlovsky (1991),for instance, write of

Egyptian 'colonies' in Palestine, the Uruk 'colonies' in northern Syria, the Proto-Elamite 'colonies' on the Iranian Plateau, and the Indus 'colony' at Shortugai as examples of 'core' regions having a direct impact upon a distant 'periphery.' The almost universal 'explanation' offered for the existence of these 'colonies' has focused upon the need to control and exploit distant resources and trade routes. This is, however, only a partial explanation of the complexity involved. Areas of cultural complexity are constantly confronted by both and internal and external competition that extend beyond the need for resources. Increasing energy expenditures for maintaining an administrative bureaucracy, establishing networks of communication, increasing agricultural and commodity production, as well as sustaining the costs involved in local conflicts, which inevitably emerge in efforts toward centralization, could all lead to expansionist tendencies.... On ancient societies, dependent as they are entirely upon human, animal and plant productivity, the solution to declining marginal returns can be accomplished by territorial expansion and the exploitation of new resources, land, people, etc.... The Central Asian Bronze Age joins the community of Bronze Age civilizations in replicating this process of expanding into a distant periphery (Frederik Hiebert and C.C. Lamberg-Karlovsky 1991:Ms. without pp.).

However, many of these recent appeals to center-periphery categories and to the world-systems or the world system itself are only hinking or halting. Some engage in seemingly arcane discussions with Wallerstein (1974), who never claimed and indeed denies (1991) that his "modern world-system" and its most significant characteristics extend back beyond 1450 AD. Thus, Greg Woolf's (1990) examination of the Roman empire seems to get lost in Wallerstein's distinctions between world- systems and world empires, which I regard as more misleading than clarifying. Notwithstanding their title, Rowlands, Larsen and Kristiansen (1987) is replete with assertions about the limitations and doubts about the utility of core-periphery analysis. [The two last named had however become more enthusiastic about world system categories by 1992, at conferences where they presented papers at panels with the present author. Titles by Kristiansen (1992 and forthcoming) refer to "The European World System in the Bronze Age"].

Philip Kohl (1989) invokes the "Use and Abuse of World Systems Theory," only to argue that "nowhere in the ancient world may one properly speak of 'world' structures of unequal exchange, of 'world' labor markets, or of economic dependence and underdevelopment" (Edens and Kohl n.d.:4). In particular, Kohl emphasizes that manufacturing cores had no special advantages, and especially no technological monopoly, over raw materials exporting peripheries. Therefore, in several publications he also takes special pains to deny any "development of underdevelopment" (Frank 1966) in the ancient world (Kohl 1987, 1989, 1991, 1992, Edens & Kohl, n.d.). However, as Kajsa Ekholm and Jonathan Friedman (1982:90-91) pointed out "center/periphery relations are not necessarily defined in terms of their [raw material-manufactures] import-export pattern.... Center/periphery relations refer, rather, to different structural positions with respect to total accumulation" of capital, from which they derive differential advantages and disadvantages. Nonetheless, Kohl is among those who most demonstrate the de facto existence, albeit with multiple and shifting cores, peripheries and hinterlands, of "the West Asian Early Bronze Age world system described here" (Edens and Kohl n.d. 59 & 60).

This shadow boxing with non-existent opponents who might but do not make claims to the total sameness of the modern and ancient world system seems less than fruitful. It seems better just to use world system categories where and when they can help clarify the "reality" of the ancient world. Gills and Frank (1992) emphasized that through most of history and a forteriori pre-history there have been sets of interlinking hegemonic cores with their respective peripheries and hinterland/s. However, several cores seem to have experienced synchronized nearly simultaneous [cyclical?] ups and downs; and the downs often led to shifts of hegemony to other, sometimes recently emerged, cores. Moreover, following our still earlier essay, we defined hegemony as a "hierarchical structure of the accumulation of surplus among political entities, and their constituent classes, mediated by force. A hierarchy of centers of accumulation and polities is established that apportions a privileged share of surplus, and the political economic power to this end, to the hegemonic centre/state and its ruling/propertied classes" (Gills and Frank 1990/91). To highlight these relations, we selected as one of our epigraphs the following observations by the archaeologist Michael Rowlands:

It has not been sufficiently appreciated that a theory of cyclical change also includes a theory of shifts of centres in space. In other words, expansion and contraction processes have rarely been stable.... This may involve intra-regional shifts in influence between competing centres within a single core area ... [and also] oscillations in intra-core hegemony are interspersed by much larger scale shifts in arrangements of centres and their peripheries.... It is ultimately the temporal that is seen to dominate over the spatial shifts in the waxing and waning of particular centres. This is generally true of all the long cycle theories (Rowlands 1987:xx).

So this paper, following on Gills and Frank (1992), explores - - if perhaps not a "theory" -- some evidence of long cycles and spatial shifts in the waxing and waning of particular centers in the bronze and iron age world system.

Important here is in the distinction between various ancient world-systems, of which for instance Algaze (n.d.) includes two in what is now West Asia/the Middle East, and the one "Central World System," as Chase-Dunn and Hall (1991) recommend we call it by combining the denominations of Frank and Gills, and Wilkinson. All the last named insist that there has been an unbroken historical continuity between this bronze age Central Civilization/World System and our contemporary "Modern Capitalist World-System." To coin a phrase, it is the same system, but it is not the same insofar as there have been some changes and perhaps even development/s.

I argue that this continuous same one world system is characterized by at least the following characteristics since bronze age times:

- 1. The world system itself. Per contra Wallerstein (1974), we believe that the existence and development of the same world system in which we live stretches back at least five thousand years (Frank 1990a, 1991a,b; Gills and Frank 1990/91, 1992; Frank and Gills 1992). Wallerstein emphasizes the difference a hyphen [-] makes. Unlike our nearly World [wide] System, World-Systems are in a "world" of their own, which need not be even nearly world wide.

- 2. The process of capital accumulation as the motor force of [world system] history. Wallerstein and others regard continuous capital accumulation as the differentia specifica of the "modern world-system." We have argued elsewhere that in this regard the "modern" world system is not so different and that this same process of capital accumulation has played a, if not the, central role in the world system for several millennia ( Frank 1991b and Gills and Frank 1990/91). Amin (1991) and Wallerstein (1991) disagree. They argue that previous world-systems were what Amin calls "tributary" or Wallerstein "world empires." In these, Amin claims that politics and ideology were in command, not the economic law of value in the accumulation of capital. Wallerstein seems to agree.

- 3. The center-periphery structure in and of the world [system]. This structure is familiar to analysts of dependence in the "modern" world system and especially in Latin America since 1492. It includes but is not limited to the transfer of surplus between zones of the world system. Frank (1967, 1969) wrote about this among others. However, we now find that this analytical category is also applicable to the world system before that.

- 4. The alternation between hegemony and rivalry.In this process, regional hegemonies and rivalries succeed the previous period of hegemony. World system and international relations literature has recently produced many good analyses of alternation between hegemonic leadership and rivalry for hegemony in the world system since 1492, for instance by Wallerstein (1979), or since 1494 by Modelski (1987) and by Modelski and Thompson (1988). However, hegemony and rivalry for the same also mark world [system] history long before that (Gills and Frank 1992).

- 5. Long [and short] economic cycles of alternating ascending [sometimes denominated "A"] phases and descending [sometimes denominated "B"] phases. In the real world historical process and in its analysis by students of the "modern" world system, these long cycles are also associated with each of the previous categories. That is, an important characteristic of the "modern" world system is that the process of capital accumulation, changes in center-periphery position within it, and world system hegemony and rivalry are all cyclical and occur in tandem with each other. Frank analyzed the same for the "modern" world system under the title World Accumulation 1492-1789 and Dependent Accumulation and Underdevelopment (Frank 1978a,b). However, we now find that this same world system cycle and its features also extends back many centuries before 1492 (Frank 1992d, Frank and Gills 1992).

Indeed, this same long cyclical pattern of ascending and descending phases extends back at least through the 3rd millennium BC in the bronze age.

THE BRONZE AGE WORLD SYSTEM AND ITS CYCLES IN THE THIRD MILLENNIUM BC

The Extent of the World System

Edens and Kohl write

One must define the spatial parameters of one's world system, while at the same time emphasize that the procedure is to some extent artificial in that the system's borders were never hermetically sealed and that they expanded or contracted over time. This difficulty affects the analysis presented above [here below] (Edens and Kohl, n.d.:58).

The farther back we go in [pre]history, not surprisingly, the more difficult it becomes to identify the changing geographical extent and temporal cycle of expansion and contraction of the world system with any kind of confidence. The at least chain-linked interaction has to be great enough so that changes in one part of the whole also affect other parts of the whole. Also, the phases have to be more or less common to enough diverse areas to say that they participated in a single system, which later became world embracing.

One of the important questions -- also here below -- is when and how, not to mention why, this "central world system" grows and changes during the bronze age and then expands during the iron age to include considerably larger areas -- but which? -- well beyond the political confines of the Achaemenid empire. Another and prior question is when, where, and how this bronze age world system came into being. Leon Marfoe (1987) addresses part of this question in his analysis of the emergence of the Egypt-Levant-Syria-Anatolia axis and its extension to Arabia, Mesopotamia and Iran from the 5th and in the 4th and 3rd millennia.

During a short time around 3000 BC, apparently sophisticated, complex systems ... appeared across an area stretching from the Nile and Aegean in the west to central Asia in the east. It is not impossible that these regional developments may represent a loosely integrated and related series of changes...that may in part be attributable to an interplay between local and external forces. In this regard, one possible effect of their outcome may have been a 'primitive accumulation of capital' and its role as a force for such change. Such conclusion would include a measure of 'market forces' in these periods....Between the late fourth and third millennia, a faint, highly buffered 'market mechanisms' may have operated for different periods of time and in different regions along these networks (Marfoe 1987:25,30,34 in Rowlands ed.)

In the same volume edited by Rowlands, Larsen and Kristiansen, P.R.S. Moorey (1987) pursues the 4th millennium trail from Mesopotamia, via Syria into Egypt, which he notes scholars were already aware of over a century ago. Moorey analyzes at least indirect mid-4th millennium exchange via Syria of Gerzean period Egyptian gold for oil, silver and lapis lazuri from Mesopotamia, which must have received the latter from farther afield in turn. He also asks, but on present evidence is unable to answer, whether and to what extent this long- distance exchange altered the productive activities in Egypt. However, he notes the boon to power and status of those who controlled the distribution of imported luxury items through local exchange networks even before the Gerzean period.

Thus, relations among otherwise distinctive regions have left marks in the archaeological record from the 4th and even the 5th and perhaps earlier millennia BC. Many were ecologically based on differences and complementarities in natural resource endowments, which generated trade, migration, invasion, and in general diffusion. Kohl (1978:475) asks rhetorically if the "world" system already stretched from the Balkans and the Nile to the Indus in the 4th millennium; and in her comment on Kohl, Joan Oates (1978:481) reads the archaeological record to display "international" horizons from at least the middle of the 5th millennium BC onwards. However, it is difficult or impossible yet to maintain that these relations were systematic enough to be called systemic. Nor is there [yet?] sufficient evidence to identify an interregional growth cycle of any significance among them.

Edens and Kohl answer their own above cited admonition to define the spatial parameters of the world system:

Roughly, the area stretches from the eastern Mediterranean in the west to the Indus valley in the east and from southern Central Asia and the great Caucasus range in the north to the Sudan and Arabian peninsula in the south....Intriguingly and perhaps not entirely accidentally, [a millennium later] the political borders of the Achaemenid empire, including those areas, like mainland Greece, where the Persians expanded, coincide fairly closely with the limits of the West Asian Early Bronze Age world system (Edens & Kohl n.d.:58-59).

These limits of the bronze age world system also coincide fairly closely with those set out in Gills and Frank (1990/91, 1992) and in Frank and Gills (1992).

However, in his tour de force on Ancient Metallurgy in the USSR: The Early Metal Age [EMA], E.N. Chernykh (n.d.) argues, as per Kohl's "Introduction," that the development of metallurgy was an increasingly interconnected and shared process throughout most of Eurasia. "The world system itself has turned out to be far more extensive than appeared earlier," (Chernykh n.d.:369) concludes in a fleeting - apparently afterthought - reference to "world system." Nonetheless, already in his foreword he also suggests that "from at least the fifth millennium BC until the third millennium BC, the peoples of the EMA cultural zone seem to have shared the same developmental cycle: the formation and decline of cultures at various levels generally coincided" (p. 2). In his closing chapter, he returns to "the contemporaneity of the decline and formation of various systems over the vast expanse of Eurasia and the Old World as a whole...[when] a whole chain of similar systems arose almost simultaneously, from the Atlantic to the Pacific: the European, Eurasian, Caucasian and Central Asian provinces, along with others outside the USSR" (365).

In the 5th and 4th millennia BC already there was "highly developed commercial exchange" and the export of "huge quantities" of copper and gold from mining and metallurgical centers in Thrace and the Carpatho-Balkan region to oreless consuming regions in Moldavia, the Western Ukraine, the Dnieper and on to the Volga. However, from the middle or second half of the 4th millennium "an extensive chain of Copper Age culture began to break up." With the significant drop of metal production in the early bronze age, the "disappearance [of this complex]...and replacement of a culture ... was as unexpected as its appearance" (87-93). Kohl (1984) also suggests elsewhere that the archaeological record bespeaks some form of pre-historic "silk route" connection to China two thousand years before the classical one. Chinese scholars also refer to the same even earlier at their end (cited in Frank 1992c).

This rather more extensive rendition of the extent of the world system already in the early-middle bronze age is also similar to my own suggestion about the "Centrality of Central Asia" and its Silk Road in the formation, development and operation of this world system (Frank 1992c). Much of the development of the outlying civilizations in China, India, Persia, Mesopotamia, Europe, etc. can only be accounted for and understood through their relations with the peoples of Central Asia, many of whom moreover, migrated to into East, South and West [Eur] Asia. One of its theses was that we should not regard the emergence of Central Asian waves of migration, often also bringing with them technological advances in productive and warfare techniques, as Deus ex Machina intrusions on "civilized" societies around Central Asia. The "Pulse of Asia," to recall the phrase of Huntington (1907), may be in its center (Frank 1992c). Nonetheless, the Centrality of Central Asia is all too neglected -- also in my own examination of the geographical extent and temporal cycles of the world system below!

Of course, our picture of the world system must be derived from survival of textual and the excavation of archaeological evidence. However, beyond the vagaries of what did and did not survive, the pattern of archaeological digs and their analysis is also a function of our own contemporary economic, cultural and political vagaries. Thus, Kohl (1984) remarks, for instance, on the Soviet focus on sites rather than regions [to which the below much cited E.N. Chernykh is a remarkable recent exception] and their preferential access to sites on the territory of the [former] USSR. This lets regions south of their borders fall through today's political economic cracks with little notice, however important their participation may also have been in our world system. Moreover, my own puzzle[d?] assembly below relies more on "economic" trade based than "political" warfare, "social" migration, and/or "cultural" diffusion based pieces of the jigsaw puzzle. The archaeological record, or at least the documentation more available to me, more readily permits the assembly of jig saw puzzle pieces of the early bronze age trading network/s in the West Asian world system. Yet even then, only very few of these pieces can be assembled here.

Kohl (1989:227) notes that "profit-motivated trade extended far beyond the political borders of any state and connected ... [all of these] into a single world system."

Foreign trade in the mid-3rd millennium was an exceedingly complex process, involving the movement of finished luxury commodities, raw materials, and staple products, and was probably conducted both by state agents and by private entrepreneurs....It does show that developments in southwestern Asia were not limited to the alluvial plains and that widely separated communities were linked by complex, well-defined exchange networks (Kohl 1978: 466).

The alluvial plains of Mesopotamia are and were notoriously poor in metals and timber, which they had to import from often very distant sources. Following Larsen (1967), Mitchell Allen (1986) draws a map centered on Assur. It imported gold and silver from Anatolia and tin and other metals from Afghanistan. In turn, Assur exported textiles to both and re- exported Central Asian Afghani and/or Iranian tin to Anatolia. However, Assur also imported textiles and perhaps grain, mainly wheat and barley, from southern Mesopotamia and paid for it with gold, silver and other metals imported from Anatolia and Afghanistan.

Shereen Ratnagar (1981) concentrates on the trade relations between Mesopotamia and the Indus Valley but also discusses some other trade. She discusses partly administered and partly free-lance trade. Copper came from Cyprus and the Levant as well as Oman, Iran and Afghanistan. As already mentioned, the rarer tin came from Anatolia, the Caucasus and Iran. Scattered copper, tin, lead and zinc as well as gold ores also were sources of silver, which was rarely available alone, but whose main source was Anatolia according to Marfoe (1987). Wood came from Meluhha on the Indus, Magan in Oman and Dilmun at Bahrain, as well as the Siwaliks and Punjab regions. As is well known, Egypt imported wood from the Levant. Egypt and/or Nubia in turn were a major source of gold, which was also exported from Arabia, Armenia, and probably the Indus. The latter exported timber, copper, gold, ivory, stones and beads to Mesopotamia and imported food, textiles, silver and earthenware. Earthenware steatite vessels were the most widely traded item, apparently both as a container of other goods and as a trade item in itself. However, Ratnagar observes that pottery did not travel over very long distances, presumably because of its weight, which made cloth and reed containers more suitable.

Ratnagar also itemizes trade in precocities like ivory, steatite, carnelian beads, dice, bird figurines, conch shells, monkey figurines, pearls, and lapis lazuli, the latter from a single source in Afghanistan. Trade in silver seems to have had a special role as a currency medium of exchange, unit of account and store of value. Archaeological finds of weights and seals from distant locations also attest to widespread trading networks, using overland, riverine, coastal and sea transport, individually and in combination. Evidence survives of individual shipments of 20 tons of copper, and exports from Assur to one small town has been estimated at some 100,000 textiles over a 50 year period.

As already noted above, Chernykh (n.d.) examines interconnected but shifting predominantly east-west trade nets of metals and their products across much of Eurasia north of the mountain ranges. As we will note below, they include a 1,000 km long tin trade route in the mid-2nd millennium. However [and despite the political confinement of his research to the fomer USSR], he also gives at least some glimpses of earlier north-south trade relations into Anatolia, Iran, and Afghanistan across the mountains. These, in turn, link the northern Eur-Asian trading and migratory system into the West Asian, Gulf, Arabian, and North African one/s. (ADD MORE, eg CENTRAL ASIA FROM CHERNYKH AND/OR TRADE NET MAP?)

The Cycle[s] in the 3rd Millennium BC

Like Gills and Frank (1990/91, 1992), Edens and Kohl suggest that a major criterion of participation in a single world system is near simultaneity or synchronism of [cyclical?] expansion and contraction, which

suggests the action of an interrelated set of transregional social forces operative over vast regions of western Asia from the mid-3rd through the mid-2nd millennium BC.... The existence of an ancient world system is postulated by the largely synchronous processes of rise and collapse recorded throughout this area; it is difficult to deny that one here is witnessing historically connected processes ... [in] the world system" (Edens & Kohl n.d.:25,61).

Gills and Frank (1992) identified such simultaneous expansive [A] and contractive [B] phases beginning with a B phase from 1700 to 1500/1400 BC. Of course, this implied the existence of a still unidentified A phase before 1700 BC; and we expressed the hope later to be able also to identify such cycle phases through the 3rd millennium BC.

More evidence for simultaneously phased expansions and renewed contractions in the 3rd millennium BC is available. The millennium opens with another of the periodic 200 year long waves of migration out of Central Asia between 3000 and 2800 BC, which was remarked on among others by Gimbutas (Frank 1992c). Periods [or cyclical phases?] of economic expansion and contraction are evident particularly for the Mesopotamian lowland-Iranian highland, where especially Sumer-Elam geo- political rivalry was an [also cyclical?] constant. For instance, findings of shell lamp/cups at Ur and Kish in Mesopotamia peak during ED II and III, that is about 2700 to 2400 BC and decline sharply thereafter "within the general trend of reduced uses of shell products in southern Mesopotamia late in the 3rd millennium and later" (Edens and Kohl n.d.:8). Elsewhere, Kohl (1978) quotes A.L. Oppenheim's earlier observations about Ur to the effect that in the late 3rd millennium "a process of gradual and slow restriction of the geographical horizon marks the entire development of commercial connections. We may well assume that the frequency and intensity of contact had reached a peak early in the third millennium B.C." (Oppenheim 1954:12). Jawad (1965) insists on the ecological, economic, social and political differences of northern from southern Mesopotamia at this time, which however were apparently not sufficient to exclude the North from this same [cyclical?] process.

Indeed, Edens and Kohl note that

numerous commentators have observed a set of fundamental synchronisms across much of western Asia ... in eastern Iran, Central Asia and the Indus ... in the mid 3rd millennium BC; [and] the collapse of these expressions of urban complex society, now extending throughout Iran, into the Gulf and, in a modified way, also southern Mesopotamia, but the opening centuries of the 2nd millennium BC (Edens & Kohl n.d.:23-24).

They note that in the Indus area massive urban growth occurred 2600/2500 BC, and even more spectacular decline around or after 2000 BC. Urbanization in Southern Turkmenistan 2600-2200 was followed by population shifts or dispersion eastward from there and other Transcaucasian areas. Urbanization in Southern Afghanistan also culminated after 2500 and abruptly disappeared after 2000 BC. Settlement in southwestern and southern Iranian peaked around and/or declined after the second half of the 3rd millennium. In and around Oman on the Arabian peninsula peak copper production, best documented at the Maysar I site, was late in the 3rd millennium. In the Gulf region, settlement and economic activity seems to have shifted from the Arabian coast to Bahrain and is evidenced by the growth of Qala' to some 5,000 population also eating Indian rice and the export of grain staples from Mesopotamia to Dilmun in the closing three centuries of the millennium. Harriet Crawford (1991:150) suggests that Gulf states' mercantile rivalry in and for the carrying trade between Sumer and Omani copper as well as with Meluhha may help explain this shift.

'International' relations changed over the greater Middle East during the first half of the third millennium with, the collapse of the proto-Elamite 'hegemony' in southern and Central Iran..." according to archaeological evidence from Central Asia, Baluchistan, southeastern Iran and the Indus Valley...accross the Iranian plateau, in the Gulf area (particularly the Oman peninsula), Mesopotamia, the Anatolian plateau and the Caucasus.... But it is unclear what happened to foreign relations in the later third and early second millennia with the collapse of Akkadian rule and the subsequent rise of and demise of the highly centralized Ur III dynasty. Dales (1977) explained the collapse of proto-urban settlements throughout the Indo-Iranian borderlands (during the so- called urban phase) as due to the cessation of long- distance overland trade and development of direct maritime trade between Mesopotamia and the Indus Valley. His theory only represents an unproven hypothesis but deserves serious consideration (Kohl 1984:242).

Dales (1976) finds a definite rhythm in "The Shifting Trade Patterns between the Iranian Plateau and the Indus Valley in the Third Millennium B.C." He refers to widespread abandonment of sites from Central Asia through Sistan, southern Afghanistan and northern Baluchistan and an almost total break in trade routes and spheres of interaction across their trade routes and then a shift also of maritime trade patterns around 2500 BC. However, the "phase marker" was not absolutely synchronized but had a "sloping horizon" from earlier to later, moving from West Asia to Central Asia and the Indus Valley. Strong trade and cultural interaction between Central Asia, northern Pakistan, eastern Persia, and points west characterized the first half of the 3rd millennium. For instance, the Turkemnian Alty Tepe flourished, but then declined in the late 3rd millennium. At the same time, the Afghan Sistan site at Shar-i Sokhta was destroyed and abandoned in the 3rd millennium, if only because the main stream of the Helmand River changed course [but see and alternative explanation below]. Its Helmand Valley civilization "totally collapsed by 2500 BC." However, the reasons may also have been transregional. For in summary, the three regions of Turkmenia, souther Afghanistan including Sistan, and the Indus Valley had "widespread contacts and interdependencies from the end of the 4th to the middle of the 3rd millennium, after which "they went their own ways." Until then, also "formal maritime trading activities were being conducted between the Indus Valley and southern Mesopotamia via the Persian Gulf," in which the Kulli, the closest westerly neighbors of the Indus Valley Harappa, appear to have played a middle man role.

The Kulli also pay this role in Shereen Ratnagar's (1981) account of Encounters: The Westerly Trade of the Harappa Civilisation. In her Table 4.4 on p. 213, she summarizes the "relative chronology of the third millennium" for each of the sites/regions of Barbar, Umm An Nar, Shahdad, Yahya, Shahr-i- Sokhta, Bampir V-VII, Kulli, and Harappa. The respective periods of maximum recorded activity in her table begins between the Mesopotamian periods ED I and II in 2750 for each of these sites/regions except Shahr-i-Sokhta and Harappa, where they begin about one to one and a half centuries later. The endings of the periods of activity marked by solid lines on her table, followed by dotted lines ending in question marks, fall mostly within the Mesopotamian ED III period. In five of these sites/ regions, the fall off date is between 2450 and 2350, the latter for Kulli already in the Akkadian period. For the others, the period of decline, followed by a question mark culminates between the Akkadian and post- Akkadian periods around 2250 BC. Only Harappa continues until 2000 or perhaps even 1800 and Barbar until 1800 BC. However, "the archaeologically attested trade contacts of the Harappa and Mesopotamian civilizations are the most numerous in the ED III and Akkadian periods" from 2500 to 2250 BC (Ratnagar 1981: 204). Elsewhere in an unpublished paper on agriculture, Ratnagar (n.d.) notes "dramatic" declines of both the sown area and the yields of wheat in the Lagash area of Mesopotamia first between 2400 and 2100 BC and then still further declines to 1700 BC. She also notes the time of troubles in Egypt from 2250 to 2035 BC, when starvation and foreign incursions made all Pharaohs' hold on power short lived. Since the invaders included especially Lybians, the implication is that also they were in or were entering this world system.

Moreover,

The pre-Indus connections were all overland with Afghanistan, Iran and Turkmenia...[and] gave rise to intermediary settlements in Afghanistan and Iran - such as Tepe Yaha, Shar-i-Sokhta and Mundigak. All these three sites lost their importance and came to their last stages of their life when Mohenjo-darians or the Mature Indus people began to forge ahead in their civilizational advance by capturing the world trade market of that time. They now directly approached the mineral sources areas.... [Then] it was the direct contact with the Gulf countries which led to a new phase of expanding trade and consequent enricheness by both the Mohenjo-darians as well as by the people of the Gulf.... There was shift in the trade routes from north to south in this period, and greater reliance was seen on sea route.... The coasts from Mesopotamia down the Gulf all along the Arabian sea to Gujarat is littered with sites bearing evidence of the Mature Indus period....The Oman sites not only point to connections with the Indus valley, but they have a marked relationship with southeastern Iran sites.... But the rise of the Gulf sites seem[s] to coincide with the rise of Indus cities in the east.... The important thing is that sea connections were evident only in the Mature period, neither before not after it.... As a matter of fact it was because of the Indus-Mesopotamia contacts that there was a rise of ... intermediate sites along the Persian Gulf making them international.... (Shaikh n.d.: 2-23 passim).

My only doubt would be whether it was the Harappans' direct contact with the Gulf that brought expansion and richness as Shaikh says; or the other way around whether it was not general economic expansion that brought on the contact with and therefore richness of the inhabitants both of the Indus and the Gulf.

In summary, the datings by Edens and Kohl, Dales, and Ratnagar, as well as others scattered through the archaeological literature, all attest to something of a generalized or generalizing A phase expansion during the first half and a major B phase crisis spreading from west to east through all of the West Asian world system in the late 3rd millennium BC. E.N. Chernykh (n.d.:371) notes destabilization also farther north throughout the Early Metal Age communities in the second quarter of the 3rd millennium. The 27th to 25th centuries were "one of major culture-historical change ... reflected in various spheres (political, ethno-cultural, productive and technological), themselves clearly interrelated" in Western Asia, Asia Minor, and the more northerly regions of the Circumpontic area stretching from the Adriatic to the Urals and Volga in the east and the Aegean and Asia Minor in the South (158). "The period of greatest disruption was probably the twenty-sixth and twenty-fifth centuries BC (on the basis of a series of calibrated radiocarbon dates)" (371).

Kajsa Ekholm already observed that

in the period around 2.300-2.200 B.C. there occurred serious economic crises that affected much of the Middle East and the Eastern Mediterranean. Everywhere there is indication of decline in quality and quantity of production that was usually state monopoly and oriented to export. Correlatively, there was an increase of local violence often culminating in obliterating warfare and destruction. These large scale crises are often explained by barbarian invasions, but it is just as likely that the violence is internal, the only migrants being "capital" and labor forced out of their homes by acute survival problems. The collapse of "supralocal" space leads to accelerating competition between and within political units, that is to warfare and intensified class struggle (Ekholm 1980:165).

Indeed, the "external" and "internal" problems need not have been unconnected let alone mutually exclusive. Invaders were more likely to succeed when their target was already economically and politically / militarily weakened by its own and regional or system wide crisis. Moreover, the invasions themselves were often generated by survival problems in their own areas of precedence and/or by falling domino like invasions and other pressures from still further beyond, particularly in Central Asia. These considerations, of course, raise largely still unanswered questions both about common ecological/ demographic changes or cycles and the extent of the extension of the "world system" into and the "centrality" of Central Asia (Frank 1992c).

In a similar vein, E.N. Chernykh also argues that regarding

the conexion between large-scale movements of people and the collapse of the system, I would like to point out that displacement by another group is by no means always the root cause of the formal alteration in and destruction of a culture. Much more often, perhaps, movements of population are brought about by reason of the internal conditions of a society, and an increase in deep and hidden processes that require changes in a number of social structures. Another cause is related to ecological changes (Chernykh n.d.:366).

Unfortunately, Chandler's (1987) data on the number of cities and their sizes as recorded by Bosworth (1992) and Wilkinson (1991) [his 1992 test follows Gills and Frank and only begins with the 2nd millennium] do not [yet?] reflect these changes: They only begin in 2250 BC with 8 cities in the region, record 9 in 2000 BC and 8 again in 1800 BC, plus one in India, for which however none were recorded during its Harappan civilization prime in the 3rd millennium.

It would of course be desirable if others more qualified than I would refine the dates, places, and geographical movement of the economic cycle and to explore its relation to shifts in "hegemony" within the World System in West Asia and elsewhere during the 3rd millennium BC. Perhaps the same could also throw further light on the center-periphery structure and regional changes in who was able and not to sit on which of its musical chairs at what times -- and why. Indeed, such research and analysis might also assuage rather than further deepen Philip Kohl's continuing doubts about these center- periphery relations. [Andrew and Susan Sherratt (1991) begin this task for the late 3rd millennium but concentrate on the 2nd millennium, as we will note below].

In conclusion of this review of the third millennium BC, we may ask with Shereen Ratnagar

whether the eclipse of the sea trade can explain the collapse of the Harappan urban system....It may well have been a significant factor....If the efficient and wide- reaching urban system of the Harappans was generated by trade mechanisms and dominated by a merchant class become powerful by its successful participation in an extensive trade network, and if the 'markets' for this mercantile urban system dwindled, as has been argued before, the wealth and power of rulers would have been seriously affected. Repercussions of a fall in the quantum of trade could also have been felt by the rural population," [if only through the move back to the countryside of newly unemployed urban dwellers. That may be how Harappan civilization] "was phased into oblivion" (Ratnagar 1981:237,253).

In that case, perhaps Harappa was phased into oblivion also as a delayed consequence of an already earlier more generalized late 3rd millennium long B phase crisis throughout most of the world system in West Asia. For "in conclusion, the evolution of the Indus Valley civilization must be explained historically: that is, by reference to those larger processes which all interacting societies of West Asia were experiencing in the latter half of the third millennium B.C. (Kohl 1984:356).

SECOND MILLENNIUM BC LATE BRONZE AGE CYCLES

The extent and shape of the world system at the beginning of the 2nd millennium and its expansion/contraction as well as hegemonial shifts through the late bronze age still remain less clear [at least to me] than would be desirable. Kohl (1987:23) contends that there was no direct contact from one end to the other, indeed more, that "there was not a single Bronze Age world system." A late 3rd millennium gravitational shift to the Gulf region, which continued into the 2nd millennium, was noted above. In Mesopotamia, activity shifted northward and became more decentralized with many smaller political units until the rise of Babylon. Then, "the central area of the Near East, from the Zagros to the Mediterranean, and from the Gulf to the Taurus and sometimes beyond to the Black Sea, appears to have formed a natural unit ... and there was a developed network of routes and exchanges within the region. Egypt is conspicuously absent" but there may have been connections to Cyprus and/or the Aegean (Larsen 1987:53). Kristiansen (1992:34,31) goes further: "Regional interaction between empires of productive irrigation agriculture in the Near East, commercial city states in the Mediterranean, nomads to the north, and ploughland agriculture and mineral exploitation in temperate Europe, created a rather unique world system from appr. 2.000 B.C. onwards ... characterized by the intensification of connections ... forming a regional hierarchy of indirect C/P [center/periphery] relations."

On the other side, with the decline of Harappan civilization, the Indus Valley seems to drop out for about a millennium, at least in regular contacts with the west. However, there is some evidence of an Indian turn southward, as Dales suggests, and eastward instead. Jonathan Friedman (n.d.) links the latter in with the emergence of the trading and migratory system between the Indian East Coast and Southeast Asia and the Lapita expansion into Melanesia and Polynesia. On Harappa's Central Asian frontier, late 3rd millennium settlements appear in Bactria, Margiana, and the Kopet Dagh piedmont in perhaps previously less settled areas. New evidence shows that, excepting perhaps in the last named, urban settlement continues and only shifts location through the late bronze age and increases into the iron age. According to Kohl (1984, 1987), this evidence contravenes the previous impression of an urban collapse, to which I return in the discussion of the next B phase below. He also argues for the probable expulsion of Harappans due to competition for minerals from Afghanistan.

The A and B long cycle phases and their dating suggested by Gills and Frank (1992) seem to be largely confirmed by evidence from Chandler's city census and many recent accounts by archaeologists. Before the rather firm 1200 BC date for the final crisis of the 2nd millennium bronze age however, they neither inspire additional confidence in our dating, nor do they offer sufficient guidance to a definite alternative dating.

A Phase 2000 - 1800/1700 BC ?

The identification of cycle phases in the world system by Gills and Frank (1992) began with a "B" crisis phase from 1700 to 1500/1400 BC. This implies a previous "A" phase of expansion, especially if the 3rd millennium ended with an earlier long B phase. The evidence, however, is not unambiguous. Chandler's city census remains at 9 in 2000 BC and in 1800 BC. However, there is an addition of a city in India, that Wilkinson regards as spurious -- and that comes at the time of the extinction of the Harappan civilization -- which may also have continued its decline during these first two centuries of the 2nd millennium. The decline of (southern) Mesopotamia is marked by the loss 3 of its 6 cities in the Chandler census, but Egypt increased from 3 to 5 cities. Related to Egypt and the Levant, economic activity increased in Cilicia and Cyprus and then also Crete and the Aegean, which began developing Minoan civilization. More of the Mediterranean and its coasts are incorporated into the world system. Larsen (1987) describes a trading system centered on the middle-man role of the relatively small Mesopotamian city of Assur, which flourished apparently independently during the 19th century BC and then was absorbed into a larger political unit until Hammurabi unified the whole area around Babylon.

On the other side, beginning around 2000 BC a region centered around Bactria and Margiana in Central Asia flourishes, but for no more than 250 years ending in 1800-1700 BC according to Hiebert & Lambert-Karlovsky (1991). I noted above, however, that according to Kohl, evidence now disputes the thesis of total collapse of urban settlement there in the next period, to which we may now turn as the first B phase identified in Gills and Frank (1992).

B Phase 1700-1400?

Gills and Frank noted simultaneous crises of interlinking hegemonies and the conquests of Anatolia and Mesopotamia by Hittites and Kassites, while the Hurrians and Hyksos overran the Levant and Egypt. That is, this was another of the recurrent [cyclical?] 200 year long periods of massive migrations out of Central Asia, which I noted in my study of the latter (Frank 1992c). This period of simultaneous disintegration of hegemonies was accompanied by inevitable economic disruptions. Silver notes the onset of the 'Dark Age' (1600-1347 BC) and says that "during this era urban life and legal documents relating to private commercial activities decline steeply" (Silver 1985:161). The Dark Age is also marked by the "disappearance .... of all vestiges of social reform - or experiments - of the Hammurabi era" after his death about 1750 BC (Oppenheim and Reiner 1977:159).

Similarly, Chernykh (n.d.:165) notes that in the 18th to 17th centuries "there is another noticeable increase in destructive phenomena manifested in the majority of Eurasian cultures." He remarks on "the destabilization of the ethno-cultural and political systems ... between the eighteenth century BC and the sixteenth century, when obvious signs of universal cultural crises and mass migrations can be observed ... throughout the Eastern European steppe and forest-steppe ... [as well as] in the eastern Mediterranean" (372) to which Gills and Frank already referred. Like we, Chernykh also remarks on the simultaneous collapse in distant China, followed later in the 16th century by the emergence of the Shang state. He says that a whole chain of cultures disintegrated and new ones were formed in their place (230) and that in the 16th century "in all of Eastern Europe as in other regions, completely new ethnocultural communities established themselves" (165).

Bosworth and Wilkinson both find that Chandler's city census confirms a B phase during this period, especially in Egypt, which drops from 5 major cities in 1800 BC to 3 in 1600 BC, and in India. Increases are registered, however, for Asia Minor and the Aegean. Wilkinson notes 4 political economic "peaks" in the period 1600-1400 BC and six peaks in the succeeding period from 1400 to 1200 BC, which we identified as an A phase. The Sherrats (1991:369-70), who concentrate more on westerly regions, witness "an increase of scale and tempo, with the corresponding friction of growth, between 1700 and 1400 BC....The political consequences of this enlarged scale of activity were to create new, expansive power-centres on the edges of the system, which sought to achieve independence and extend their control over the centre." This increase in scale, which is not uncommon in periods of crisis also in the modern world system (Frank 1978a), also helped set the stage for the next A phase expansion.

A Phase 1600/1500? - 1400 - 1200BC

Gills and Frank (1992) dated the next expansion from 1400 to 1200 BC. However, there is some evidence that it may have begun earlier. From 1360 BC to 1200 BC, the total number of major cities declines by one, and also by one each in each of the Aegean/Mediterranean, Asia Minor, Egyptian and Levant regions. Nonetheless, Bosworth interprets Chandler's city data to "lend strong support" to this period to its dating as an A phase by Gills and Frank, and Wilkinson treats it as a "rally," despite some misfitting data and his observation of an increase in "peaks," as noted above.

Chernykh (n.d.:290,313-4) also finds increased interaction between Eurasian and European provinces between the 14th and 12th centuries. However, he also regards a longer period, beginning with the 16th century BC and including this period until the 13th and 12th centuries BC, as one of "stability." Then also "a whole chain of new metallurgical provinces, stretching from the Pacific to the Atlantic, was formed. The technology of casting thin-walled tools and weapons, and the production of tin-bronzes spread explosively through this entire area."(n.d.: 372-3) In the 16th and 15th centuries, from the Dnieper eastward there was a sharp increase in the amount of mining in new copper and tin ore areas in the Urals, Kazakhstan, the Altai and both sides of Lake Baikal. "The huge scale of mining in a number of the mines is astonishing" (231). The "volume of mining reached truly fantastic proportions": For instance 2 million metric tons of ores were mined and some 100,000 tons of copper were smelted at just two Kazakh copper-ore deposits. There were "specialized settlements of professional miners, metallurgists and metalworkers" (231-35). There was a 1,000 kilometer trade route for tin (235), and copper ore was transported 300 km through the mountains (243). "Steppe and forest-steppe peoples had a uniform economy and were very closely connected to one another. cultures, it appears, were not isolated but consisted of open systems: economic, ideological, and kin-based interconexion and exchange were not only possible but, very probably, actively encouraged" (236). Strangely however, Chernykh also remarks on a "significant territorial rupture between ... Eurasian and European systems" at this time (288).

Kristian Kristiansen (1992) also diagrams an upward phase of Tumulus culture in North-Central Europe between 1600 but especially 1500 and 1250 BC. In Europe, Kristiansen (19xx:30) notes "an expansion phase suddenly, within a generation at about 1500 B.C. the fully fledged chiefdom structure emerged in northern Europe ... [in] a period of conspicuous wealth [that] lasted, with some ups and downs, from 1500 to 1100 B.C., but, already in the later part ... declined." Kristiansen (1992:22) also remarks on expansion from 1500 onward, along with a shift in trade of northern Europe with the Mediterranean area from an eastern axis via the Danube and Black Sea towards the Western Mediterranean and Italy instead via the Rhone Valley.

For the Sherratts (1991:370) also, "these two centuries [1400- 1200 BC] are somewhat arbitrarily separated from the preceding phase, and mark the climax of the palatial trading system and the political frameworks within which it was carried out." Like Gills and Frank (1992), the Sherratts underline the expansion of the Hittites and Assyrians, but also a major phase of urbanization in Cyprus, the importance of Rhodes, and a shift from Crete the Greek mainland. They also remark on the related "intensive diplomatic activity" among blocs and others. Similarly, Mario Liverani (1987: 67) also remarks on the exceptionally high frequency of paritetical treaties in the 15th to 13th centuries.

Gills and Frank (1992) had observed that dominant but inter-linking hegemonies were the Hittite empire, based in Anatolia and the dominant in northern Mesopotamia, and the empire of New Kingdom Egypt. The period was clearly marked by the prominence of inter- linking hegemonies, including Babylon, Assyria, and Mitanni, all of which took a full part in the well developed diplomatic discourse of the period. There was for a time something like a concert of powers among these inter-linking hegemonies. The Mycenaean trade supplanted the Minoan in the East Mediterranean.

The Sherratts summarize:

This [1400-1200 BC] period represents the climax of bulk maritime trade in the Bronze Age. It differs from the 15th century patterns in the disappearance of Cilicia and Crete as major centres in their own right , and the emergence of Cyprus as a major international trader....The system thus seems to have differentiated into two components: a major long-distance international route marked by port towns and emporia such as Tell abu Hawam, Ugarit, Enkomi, Ialysos, Kommos, and stations like Mersa Matruh, and operated by ships with large cargoes...heavily capitalises and partly state dependent; and a series of cycles operated by long-distance ships of smaller capacity in the west, some controlled from mainland centres but many under less centralised control.... Not all this activity, however, should be imagined as the peaceful growth of commerce. Factors of international competition are evident in the insecurity of the Levant revealed by Amarna letters ... [and] Egypt and the Hittites clashed at Kadesh in 1284 (Sherratt 1991:372, 373).

The beginnings of the introduction of iron, especially in weapons but also in tools, initiated the beginnings of a major transition.

B Phase 1200 - 1000 BC

This B phase is most remarkable and remarked as a major period of crisis, indeed of "dark ages," which also spelled the end of bronze age civilization and its definitive replacement by the new iron age. Gills and Frank (1992) recalled how Gordon Childe (1942: 185) already remarked that "the Bronze Age in the Near East ended round about 1200 B.C. in a dark age.... Not in a single State alone but over a large part of the civilized world history itself seems to be interrupted; the written sources dry up, the archaeological documents are poor and hard to date." Liverani (1987: 69, 71) also comments on "the collapse of Near Eastern Civilization...[whose] crisis is rather extended and takes place at roughly the same time over a large area." He also observes, this scarcity of surviving documentation "is not fortuitous ... [but] is itself an effect of the crisis (eclipse of scribal schools and palace administrations)." For instance, 576 years of Kassite domination in Mesopotamia came to and end in 1171 BC. The Cretan based Mycean civilization came to an end about the same time. It was again a time of [for?] another 200 year long wave of migration, this time of Indo-Europeans eastward toward the Tarim Basin and of Arameans, Dorians and others southwestward into the Levant and Greece (Frank 1992c).

Chernykh echoes this same theme when he writes that

these processes of widespread migration and the related collapse of cultural systems reached a peak in the 11th- 10th centuries BC. Archaeologically speaking, this is the boundary between the Bronze Age and the Iron Age - and one of the most significant and critical periods in the history of the peoples of the Old World (Chernykh n.d.:373).

"There was a sharp decline in the production of bronze artefacts throughout the Eurasian steppe at the end of the LBA" [Late Bronze Age] (322). From 1200 to 1000 BC, he also notes a "collapse of the system" in the Irano-Afghan province. "Settlements disappear... the settled way of life apparently changed to mobile pastoralism" (326). There was a similar falloff in metal production throughout the northeast Balkans and Carpathans (323) and a sharp increase in mobile subsistence strategies (299).

Kristiansen (1992:23) relates the "collapse" of the Mediterranean and Near Eastern regional systems shortly after 1200 also to Europe and refers to evidence of mercenaries and later southward migrations from Central Europe and the Balkans. In Europe, agriculture became more dominant and political organization less chiefly and more "populist and with a more "democratic ideology."

These "dark ages," are however often dated to have lasted up to 350 years. We may return to Peter James' et al treatment of this problem, which is his principal concern in Centuries of Darkness, in connection with discussion below of the therefore uncertain dating of the beginning of the next A phase. In the meantime suffice it to note his agreement and moreso that

the term 'Dark Age' seems like an understatement when the archaeological remains from Babylon ... examined.... There can be no doubt that in many parts of the Old World there was a dramatic collapse at the end of the Late Bronze Age. The centralized economies controlled from the palaces disintegrated, the old trading markets broke up, diplomatic contacts were lost and major settlements were abandoned....Ten separate interpretations of the events at the end of the Late Bronze Age can be discerned ... [including] cultural decadence [a la Toynbee] ...climactic catastrophes...and invasions by outside barbarians - notably the Sea Peoples [and others].... Such 'external' causes are rarely convincing because they cannot themselves show why the civilised society was unable to cope (James 1991:279,311).

Kristiansen (1987:84) notes that "with the decline of international exchange networks of prestige goods at the transition to the Iron Age, the whole system of center/ periphery relations collapsed. the various regions developed autonomous cultural and economic traditions." Sherratt (n.d.:13) also remarks how in the final centuries of the second millennium ... the long-distance north-south links, however, temporarily slackened: The Nordic regions developed on its own, without plentiful supplies of metal from further south." However,

the collapse of large-scale inter-regional trading systems began in the most heavily capitalised areas, and its effects reached outward to involve all the palace economies which were dependent on them.... Ugarit ... was permanently destroyed some time shortly after 1200 BC, along with neighboring Atchana and Carchemish. The recession in Syria which followed these destructions had further effects on the Hittite hinterland, and as economic difficulties exacerbated local unrest, leading to the destruction of BogazkÀ?
Ày itself at this time. The Assyrians under Tglath Pielser I (c. 1100) took advantage of this unstable situation to invade Syria and the Levant, before the shift of power to semi-nomadic Aramean tribes caused a fundamental decentralisatiion of local economies (and, incidentally, a new set of inland routes made possible by use of the dromedary) (Sherratt 1991: 373,374).

Gills and Frank noted that at the same time, the Mycenaeans in Greece and the Levant were overrun by new waves of invasions, which included the Dorians, Aramaeans and Phoenicians. The Hittite empire disintegrated. The Kassite dynasty in Babylonia collapsed to Chaldeans and Aramaeans. Political power almost everywhere was unstable and short lived.

Chandler's data on cities also support the B phase designation according to Bosworth and seem to mark "a genuine" B phase for Wilkinson. The growth of cities stagnates; the number of major ones declines a bit; and Hittite and Aegean cities disappear from the list altogether. Wilkinson notes a marked decline of political economic "peaks" from 6 between 1400 and 1200 BC to only one between 1200 and 1000 BC and observes "a more noticeable character of disintegration ... than in preceding centuries."

FIRST MILLENNIUM BC IRON AGE WORLD SYSTEM EXTENSION AND CYCLES

The extent and development, as well as the precise dating of the cycle phases, of the world system during the first millennium are also still subject to dispute. This uncertainly is increased by two problems in particular: One problem is the length of the last bronze age crisis and the transition to the iron age, which may have differed among regions. Another problem, paradoxically, is the vast expansion of the world system, probably midway through the millennium, thence definitively to include most of Eurasia and more of Africa. India also returns into active participation in the world system, after a near millennial absence - at least in my review.

For these and other reasons, also, Chandler's city data as analyzed by Wilkinson and Bosworth display some ambiguity and possible misfits with the direction and/or dating and regionality of the cycle phases proposed in Gills and Frank (1992). I will note both the fits and these misfits as we review these phases below. However, at least in the western part of the world system it might be suggested that the structural divergences created during the first millennium B.C. between northern Europe, central Europe and the Mediterranean, determined the later course of european history by the establishing structural foundations upon which it came to rest, e.g. the limits of the Roman empire in Europe" (Kristiansen 1992:35).

Ghirshman summarizes events in West Asia, where

The first half of the first millennium B.C. was a turning point in human history. The centre of 'world politics' or of the age shifted ...[from alluvial valleys in the south] more to the north...that the struggle for world power was centered...[among] three principal actors on the drama: the Semitic Assyrians with their vast empire; Urartu, a powerful kingdom of Asiatic origin, tenacious opponents of the Assyrians...and finally the Aryans, the Iranians who, after a long and arduous struggle, triumphed over their two adversaries and, with the spoils, founded the first World Empire [under the Achaemenid kings from the 5th century onwards]. There was a shift in the centre of gravity of exporting countries. Assyria, which was a great consumer, had no iron mines; for a time, especially during the earlier half of the eight century B.C., it was denied access to the mining centres of the southern coast of the Black Sea and Transcaucasus by the neighboring kingdom of Urartu. Inevitably it turned its attention to Iran....[which obtained this metal from regions inaccessible to Assyria] (Ghirshman 1954: 75, 88).

To begin with, the bronze age crisis and the previous B phase seems excessively long by some accounts. Peter James (1991) and his collaborators devote a recent book to demonstrating inconsistencies and hiatuses in regional datings of the crisis, which sometimes leave 350 year gaps during which on the evidence as good as nothing seems to have happened. In his review Sherratt (1991b) accords them more success in demonstrating the dating problems than in resolving them by their readjustment of the relation between dating sequences in Egypt and elsewhere. However, to the extent that the shortening of the crisis by James et al is well taken, Gills and Frank's (1992) otherwise more uncertain dating of an A phase beginning about 1000 BC also becomes more credible.

A Phase 1000 - 800 BC

The Sherratts (1991:375) remark that "the system was revitalised ... in the 10th century," particularly along the spice route from Arabia and by Levantine centered trade with "pan-Mediterranean scope." Kristiansen (1992) refers to Phoenician expansion through the Atlantic to France and Britain in the 9th and 8th centuries BC. Gills and Frank had similarly underlined the Phoenician expansion through the Mediterranean during this period and noted the rise of and then challenge to Assyrian power on the mainland. However after 1000 BC, metal supplies also increased again in distant England and Scandinavia (Kristiansen 1992:24). Indeed, the increase was of such enormous proportions in the West in its final phase as to suggest "overproduction" and the use of Amorican axes as currency to Kristiansen (1992:5). However, "it can hardly be doubted that large scale metal consumption and inflation in the west was somehow related to the decrease of metal production in the East," where its consumption declined drastically during the next Hallstatt B2-3 period and never recovered (pp. 6,5).

Wilkinson finds this phase "in sync" with and "not challenged" by the data. Bosworth also agrees and finds corroboration for Assyria, but suggests that "perhaps this A phase might be extended as Niniveh, the seat of Assyrian power, peaks somewhere between 800 - 650 BC, when it reaches 120,000 people -- the first city on Chandler's list to break the 100,000 mark." The number of major cities from the Mediterranean to India remains the same at ten, however, and increases from 3 to 5 in China.

B Phase 800 - 550 BC ?

It was particularly problematic to identify and date this phase by Gills and Frank (1992), who adjusted the opening and closing dates several times and sometimes left them ambiguous as well. Therefore, I am not surprise to find that Wilkinson and Bosworth single this phase out among bronze and iron age ones as the least confirmed or most challenged by the Chandler data on cities. Wilkinson says the phase is not reflected in the Chandler data. Bosworth does find an apparent "period of 'contraction' and 'fragmentation,' but only in the western part of the Old World. Between 800 and 600 BC there is little growth for Babylon, Jerusalem or Van, and other cities drop from the list entirely. By contrast, Chinese cities roughly double in size, and another Indian city appears." For this phase, Gills and Frank (1992) noted increased competitive pressures in the Mediterranean and of rivalries in West Asia, as well as the presumably related collapse of the Assyrian empire in the 7th century BC. However, we also noted technological / economic development in India and new rivalries in China during this period. The growth in the East apparently foreshadows its approaching inclusion in the "Central World System," which is also reflected in city data problems with the subsequent phases and their dating by Gills and Frank.

The Axial Age

This period was called the "Axial Age" by Karl Jaspers (1949/1953), who regarded it as the turning point in human history. He also noted, like Frederik Teggart (1939) and William McNeill (1963) more recently, that the great religious movements and their prophets were born at almost the same time in the 6th century BC: Pythagoras in Italy, Thales in Greece, Ezekiel and the second Isaiha in the Levant, Zoroaster/ism in Persia, Buddha and Mahavira, the founder of Jainism, in India, and Lao-tse and Confuscius/anism in China. The three above named scholars and others suggested or at least implied that this simultaneity was probably no accident. McNeill (1963:338) suggested "if social and psychological circumstances of the submerged people and urban lower classes were in fact approximately similar in all parts of Western Asia, we should expect to find close parallels among the religious movements which arose and flourished in such a milieux. This is in fact the case."

Indeed, Gills and Frank considered whether these similar "social and psychological circumstances" may not also have reflected similar economic circumstances and a at least just previous common economic crisis. Moreover, we added, the emergence of universalist religions may also be an indication of the high level of real economic inter-linkage and perhaps the attainment of a new level or stage of economic integration, which characterized this period. For it is also in this "axial" period that, it may realistically be argued, China first became permanently incorporated into the Central World System. (ADD EVIDENCE HERE?)

Less researched is the apparent incorporation also of Southeast Asia. Coedes (1968:7), following van Stein Callenfels, dates the arrival of bronze in Indochina around 600 BC and in the islands around 300 BC, but xx dates iron finds already from 750 BC. Archaeological finds also establish significant contacts and trade of tin and gold between the islands and the Malayan Peninsula and mainland from the middle of the 1st millennium BC (Rhaman 1991). That is also the recorded beginning of the "Indianization" of Southeast Asia (Coedes 1968, Clover 1991). Indian texts attest to "speculative mercantile voyages for commercial profit, financed by merchant guilds in many parts of India" in the 4th century BC (Clover 1991). At the same time, according to Chinese texts, their merchants travelled and carried silk over the "southwestern route" from Szechuan, through Yunan, across Burma into India. This route was also prominent again in the 1st centuries BC and AD. Moreover there was "considerable trade" between Chinese and YÀG
Àehs to the South in China and Indochina before the end of the 3rd century BC. Then a Ch'in emperor sent five armies of 500,000 men against the YÀG
Àeh to secure economicÔ

A Phase 600/550 - 450/400 BC ?

During this phase, Gills and Frank (1992) noted the economic development in Greece, replacing that of the Phoenicians, and especially in Persia. This period witnessed the rise of the Achaemenid Persian empire, which stabilized much of West Asia by re-imposing a more unified political order in that part of the world economy and system. The Achaemenids from Darius to Xerxes achieved at least a regional position of hegemonic accumulation in the world system, on the basis of the imperial tribute system. The Persian empire exceeded even the Assyrian in the degree to which it succeeded in incorporating the most important economic zones of the world system in West Asia. There was at this time a shift in the center of gravity of the world economy of very great historical importance. The key area of logistical inter-linkage in the world economy/system shifted from Syria and the Levant to Central Eurasia. Achaemenid control of Central Asian cities, such as the great city of Bactra, and the northwest India trading center of Taxila were very important elements in consolidating Persian hegemony and accumulation. The Persian investment in infrastructure included the 1677 mile Royal Road, which Darius built from Ephesus to Susa and the road from Babylon to Ortospana (near Kabul). Persian cities, like their Assyrian predecessors, were cosmopolitan; and its armies were multi- national. "It was in this period that the great caravan cities of Syria - Aleppo, Hama, Homs (Emesa), and Damascus, in particular - truly came into their own, receiving goods form the Silk Road as well as spices and perfumes from Arabia's Incense Road and other luxuries brought by sea from India. Arameans ...were such active traders on these caravan cities that their speech became the common commercial language" (Franck & Brownstone 1986:65).

Since this phase falls within a longer period between two of Chandler's city censuses in 650 and 450 BC, Wilkinson observes that the data are ambiguous but "at least not out of sync." Bosworth sees the data as broadly confirmatory, but suggests that "Frank and Gills' focus on Central Asia as the locus of this A phase seems to be misplaced, as events there are eclipsed once again by those farther east" in China and Korea. The Bactra and Taxila, signalled as trading cities by Gills and Frank, do not appear on Chandler's list for this period; though I might retort that commercial importance is not necessarily always reflected in population size, eg. Hong Kong today. However, Bosworth notes that "China begins as practically a footnote in Chandler's list and by 430 BC it has seven of the world's largest 25 cities and the second largest.... This dramatic rise cannot be over-emphasized." Whether Gills and Frank were guilty of a "western," that is Central Asian, bias at the expense of China is worthy of consideration. However that may be, the "dramatic rise" of China in the city leagues also speaks for its growing commercialization and probably relations with and incorporation into the Central World System at this time, whose world historical significance "cannot be overemphasized." p In the West, Kristiansen (1992) focuses on important events between about 600 and 450 BC in Europe. It was re- or more fully integrated into the Mediterranean and it in turn into the West Asian world [system?]. "A new axis of exchange emerged (during Ha[llstatt] B2-3), stretching from northern Italy over Switzerland to the Lower Elbe and further on to Scandinavia, Northern Germany and Pommerania" (pp.6-7). Dietler (1989) argues for the important intermediary role of Rhone valley inhabitants in articulating and perhaps even initiating such long-distance north-south trade and to "foster dependent relations of a center-periphery nature." The Rhone corridor trade to Hallstatt Europe broke down again in the early 5th century. As a result also in Central Europe, the Hallstatt cultures that had first "climaxed," then "declined" as trade routes again shifted and/or they overexploited their peripheries, according to Kristiansen.

B Phase 450 - 350 BC ?

Although Gills and Frank inveighed against excessively Greco- centered readings of this period, we did identify this relatively short B phase largely on the basis of symptoms of economic crisis in Greece and its relations with Persia. Intensified class struggle and wars symptomatic of an underlying economic contraction or slow down in expansion. Rostovtzeff had characterized the fourth century as one marked by increased proletarianization, landlessness, unemployment, and food shortage. It was marked by a contraction in the market for manufacturers and the ruin of "free" petty producers. Wealth was over-concentrated in the hands of the commercial and landed ruling classes. Livy notes a series of famines in Italy in 490,477, 456, 453,440, and 392 BC. The Celts invaded Italy and sacked Rome, while setting up the kingdom of Galatia in Asia Minor. The hegemonic disintegration of this period is in evidence from the Peloponnesian wars and the successful revolt of Egypt against Persia c. 400 BC and the breakaway of the Indus from the Persian empire c. 380 BC.

The phase is too short to be well reflected by Chandler with a longer time span between city censuses. Nonetheless, Bosworth notes that the size of Athens decline; but that of Rome increase several fold. Kristiansen (1992:11,9) remarks that "the apparent correlation betweencompetitive changes in Greek, Phoenician and Etruscan trade routes with the geographical movement and collapse of princely centers [in North-Central Europe] has been seen as a confirmation of the dependence on long distance trade and the supply of prestige goods.... This question is ultimately linked to the problem of explaining the collapse of the Hallstatt centers ...[and their being] able to monopolize trade with the Mediterranean to such a degree" at this time, when they gave way to the Celtic La Tene culture after 450 BC. Chernykh (nd.:373) also terms this "the next 'destructive' period, " when Celts moved from Western Europe towards the Balkans and Asia Minor in the 5th century and Sarmatians in the opposite direction in the 4th century. Chernykh also notes the political destabilization of the Warring States period in China and new confrontations with Central Asian pastoralists. "the high-water mark of these destructive processes and the disintegration and reformation of cultures in Eurasia was the fourth to third centuries BC" (Chernykh n.d.:374).

A Phase 350 - 250/200 BC ?

Gills and Frank's identification and dating of this A phase rested primarily first on the Alexandrian expansion through West Asia into Central Asia and India. It was followed by the economic expansion in India under the Mauryas in the 3rd century after Alexander's death and failure; and the simultaneous consolidation of China under the Qin dynasty, as well as the increase in trade between the last two of these regions. Here again, Chandler's census dates are not very helpful. Nonetheless, Bosworth notes the prominence of Alexandrian cities, including Alexandria itself as the third largest in the world, during this phase. However, he also suggests that, again, the East may deserve "top-billing": In 200 BC, for the first time the world's largest city is in China (Changan with 400,000 inhabitants) and the second largest is in India (Patna with 350,000). China and India now also have similar large shares of urban population the world's top 25 cities.

B Phase 250/200 - 100/50 BC

Gills and Frank noted another brief B phase in the Mediterranean region, including Egypt and Greece during the 2nd century BC. It was again marked by signs of crisis and contraction and slower expansion of the market. In Egypt, the second century was characterized by all the signs of economic decline, such as over-taxation, official corruption, increased debt, and unrest and brigandage. The Rosetta stone characterizes the period by: "pressure of taxes, rapid accumulation of arrears and concomitant confiscations, prisons full of criminals and debtors, public and private, many fugitives scattered all over the country and living by robbery, compulsion applied in every sphere of life." There were signs of crisis and slave revolts in Rome, but also imperialist expansion westward, which presaged Imperial Rome during the next long period of expansion after 100 BC. While this phase is again too short to be reflected in Chandler's census dates, which jump from 100 BC to 200 AD, Bosworth writes that "there is consensus that this was a period of 'contraction' and 'decline'." However, China already began its period of expansion under the Western Han Dynasty after 200 BC. Since Imperial Rome did as well, should the beginning of the next major A phase be moved forward a century or so? That would convert this B phase into a still shorter and more localized phenomenon, hardly worthy of the name.

A Phase 100 BC - 200 AD

I do not wish to prolong this examination of the bronze and iron ages to treat this phase -- and the perhaps beginning again of a new period in world system history -- here in further detail. Suffice it to recall that this period was the high-water-mark of expansion phases in Gills and Frank (1992), with the simultaneous and well known rise to imperial grandeur of Han China, Kushan India, Parthian Iran, Axium East Africa, and Imperial Rome. [Their rise was followed by their again simultaneous decline in the major B phase from 200 to 500 AD, which was also again accompanied by another major wave of invasions, including that of the legendary Attila the Hun]. Moreover, we cited both Roman writers like Pliny and more recent ones like Teggart to the effect that what happened at one end of this chain also substantially affected what happened at the other end. The writings of Ptolemy and the famous Periplus of the Erythrean Sea attest to regular maritime trade to the west coast of India, but the same was equally or more intensive also onward from the Coromandel east coast and Ceylon to Southeast Asia and China. Chinese Eastern Han Dynasty documentation attests to significant trade with Southeast Asia in the 2nd century AD, and there is also evidence of the same from the 2nd century BC.

Jacques Gernet notes

Just as the power of the great nomad empire of the Hsiung-nu in the steppe zone was probably created and strengthend by import of iron and silks from China, Han expansion in Asia was certainly due fundamentally to the economic upsurge of the Chinese world. Not only were Han China's strength and prestiege abroad based on this economic prosperity, but it was also the trade with Mongolia, Korea, central Asia, South China, and northern India... (Gernet 1982:129).

Agreed, except for the [also Western] Sino-centric perspective that sees the "economic upsurge" in all these regions and even the power of the Hsiung-nu in Central Asia as propulsed fundamentally by "the Chinese world." Apparently the A phase economic upsurge was not confined or due fundamentally only to China, or else it would not have so easily included all the other areas Grenet mentions, not to mention many more across Eurasia.

Moreover,

the great expansion of Southeast Asian, and particularly Island-Mainland exchange which is evident in later prehistory is, I believe, closely connected with this Indo-Roman commerce and can be explained in part, at least, by a rising demand....[Recent] finds have been made or recognised and these are enough, I believe, to permit us to argue that regular exchange links between India and Southeast Asia commenced earlier than Wheeler or Rashke allowed.... By the early Christian era these trade routes reached out to bring together the previously rather separate Southeast Asian exchange systems, linking them into a vast network stretching from Western Europe, via the Mediterranean basin, the Persian Gulf and the Red Sea, to India, Southeast Asia and China...[in] what has been called the World System (Clover 1991:n.p.) However, writing of course quite independently, Chernykh (n.d.:374) also remarks that "the period of renewed stability, which lasted from the second century BC to the second/third century AD, was related to the existence of three major empires: the Han Empire on the eastern flank of the Eurasian landmass; Rome on the western flank; and the Parthian and Kushan kingdoms in the centre" before the all gain succumbed to the "truly colossal" "destructive processes and migrations," which were "linked to the destruction of this system."

CONCLUSIONS

"Clearly there is substance to the Gills and Frank proposal. One would not want to ... reverse the labels of the phases: the fit is too good. Still, it needs more measuring, hewing, trimming .... Considerable refinement of phase time- boundaries, and data collection for crucial but unmeasured years, is called for" (Wilkinson 1992: 11,30). So writes the author of one of two empirical "tests" of the identification and dating of long cycles in the world system by Gills and Frank (1992).

This new essay has sought, if not yet to collect more data, at least to begin systematizing some of the other relevant data, other information, and their analysis, which is now available to me as a non professional on short order. Practically, in two senses of the word however, my selection of the pieces to place in this jig saw puzzle are largely derived from the documentation by professionals, which my friends among them have kindly supplied to me of their own and other writings. [Eg. Philip Kohl supplied me his own and others' still unpublished writings by Hiebert and Lamberg-Karlovsky, and the translation of Chernykh's book from the Russian]. It would of course be desirable that professionals more qualified than I assemble this puzzle much more competently and fully. They would also have to give much more due weight to the recurring and related migrations, invasions and wars in this picture, which have received excessively short schrift in mine above. In the meantime and on the basis of the evidence summarily presented above, I may conclude as follows.

Substantial archaeological evidence and important analyses thereof seem to confirm the existence of long cycles in what may also for this reason be called an at least 5,000 year old world system. Alternating expansive A and contractive B phases reach back through the 3rd millennium BC and probably into still earlier times. They are so synchronic over so large and ever growing part of the world that it truly appears as a true world system -- if the socio political economic mechanisms that generates this "cycle" is at least also partly endogenous to this system and the part of the world that it includes.

Interestingly, increases and declines in city sizes, as recorded by Chandler and reproduced by Wilkinson, also characterize the "Western" hemispheric regions of the "New" World in the "Americas" before Columbus "discovered" them 500 years ago. However, their phases and cycles are totally unsynchronized with the fluctuations in city size and/or the phases of expansion and contraction suggested and dated in the "Old" World "Eastern" Hemisphere by Gills and Frank (1992) and refined above.

This implies that the substantial confirmation and occasional adjustment of cycle phases by city sizes in Eurasia is not fortuitous and does reflect some underlying reality. But what is that? If that underlying reality is climactic change, which is only common to all parts of the "system" and to which they independently react in tandem without significant interaction with each other, should we then still call them part of a single system? Perhaps an ecological system, but not a social one?

Yet, I hope that I have marshalled sufficient evidence above to find that "different" regions and people/s have also been so socially interlinked through economic, political, migratory, cultural and other both cooperative and conflictive relations as to meet the criterion of systemic participation in a single world system. The participation of the parts is so inter-active, even if sometimes only indirectly chain-linked, that no part of this system would be as it is or was if other parts were not as they are or were -- albeit they may all have also reacted to, and on, the global ecological constraints. [That was the system criterion proposed in Frank 1990 and Gills and Frank 1990/91).

Notably, major periods of near world system-wide major migration accompanied several of the phases I have identified as "B" contractions. This was especially so for the major B phase "dark ages" in 1700-1400 BC, 1200-1100 BC, and 200-500 AD which initiated a "Dark Age" in Europe, as well as the mid first millennium "Axial Age," discussed in Frank (1992c). The significance of this "coincidence" itself remains a dark mystery. However, Chernykh (n.d.:367) also notes that "these are critical periods in human history. The migration at the end of the second and start of the first millennium BC defined the boundary between the Bronze Age and the Iron Age; the later migration [in the 200-500 AD B phase] defined the boundary between Antiquity and the Medieval period or, in Western historiographic terminology, feudalism."

So the question must finally be raised - were all these regional trends and dramatic historical events somehow interlinked? Were Europe, Asia and the Mediterranean so interdependent that major changes in one region would lead to predictable changes in the other regions, forming a kind of interrelated world system?... But how did these changes [in Northern Europe] relate to changes in Central Europe and the Mediterranean. Were they somehow connected? This is, in the last instance, dependent on our chronologies, where I do not feel able to decide (at least not at the present moment) if there is a significant time gap or not between these regional changes." So ask and answers Kristian Kristiansen (1992: 20, 28).

"We suspect that such explosions follow some regular rhythm; that, in accordance with this rhythm, various provinces at the same time either collapse or emerge; however the nature of this rhythm is equally unclear.... We are going into the unknown when we try to understand the reasons for such explosions and successions, and to try to identify the hidden driving force behind such phenomena" (Chernykh n.d.:354).

Both of these archaeologists and others of course also refer to possible climactic constraints and dynamics. "There have been several attempts to link oscillations in historical development in particular regions with various natural phenomena, which in the final analysis, are related to periodic changes in global climate. Hypotheses of this kind, are to a certain extent contradicted..." Chernykh (n.d.:376) writes, because the same kind of changes are not observable in Stone Age cultures, which experienced the same climactic changes at the same time. On the other hand, Neolithic Stone Age peoples were more affected by environmental change. Although elsewhere (Frank 1992c) I have also emphasized the possible role of climactic cycles particularly in Central Asia, I might here add that the problem Chernykh mentions also works the other way around: The same climactic change can affect different regions differently. For instance the same period of warming can affect agricultural possibilities in arid and humid, and/or high and low lying, areas in exactly opposite ways. Yet we have found substantial evidence for very generalized simultaneous economic up and down swings over geographically very different areas. Moreover, these economic cycles seem to have become progressively shorter over the millennia [or has our identification of them just become finer?]. All the more so therefore, "the reasons for the changes in rhythm during the 'metal era' should probably be sought primarily among the socio-economic conditions of the period," including in particular their interactive participation in a single but cyclically developing "world system" (Chernykh n.d.:376).

An intermediate position, or "model" as he calls it, is proposed by Jack Goldstone (1991), writing, however, about early modern history. He notes recurrent waves of state breakdown and prior population increases across all of Eurasia, for which it is not possible to account in terms only of local conditions or particular cultural patterns. So he offers a "structural/demographic model" to explain complicated recurrent [cyclical?] interactions: Demographic pressures derived from temporarily falling death rates impinge on overstretched resources and lead to bureaucratic paralysis. That in turn is derived from elite in-fighting and large sale social rebellion, which lead to state breakdown. In Goldstone's "model," all of these events are thus ultimately - - but also socially and not simply physically -- also generated by climactic change. Goldstone only analyzed early modern large bureaucratic states. Perhaps however, mutadis mutandis his approach may also prove fruitful in the analysis of crises involving other political economic institutions in earlier periods as well.

In [literally] any case,

No single factor may account for the observed structural changes, although some factors, such as climactic change, may reveal striking patterns of parallelism with changes in settlement structure.... Climate thus represented both potential and constraints to subsistence, but social and economic forces remain the prime movers when the environment is exploited not only close to, but often beyond its carrying capacity.... In such situations climactic fluctuations may trigger the collapse of an unstable economy. [That is] one of the lessons we may learn from Bronze Age sequences..." (Kristiansen 1992:26).

It is a lesson that sounds particularly pertinent also to our own times as I write during the "Earth Summit" in Rio de Janeiro in June 1992.