CHAPTER 5 COMPLEX SECONDARY WORLD-SYSTEMS The simplifying notion that world-economies (networks of economic exchange) become transformed into world-empires which fill the region of the economic matrix can only be a rough approximation to what happened with the emergence of the primary empires discussed in the previous section. In none of these cases did the territory of imperial sovereignty expand to include all regions of the economic network. In the subsequent period the above simplifying idea is even less adequate as a characterization of what occured, except for the case of China. While it is true that empires tended to grow along with the increased economic interdependence of regions, there were many periods in which different empires simultaneously were present within a larger economic network, along with independent states of various kinds. It is time we took a closer look at our notions of world-economy and world-empire. The simplest way to characterize the nature of the historical world- systems in the first millenium B.C. would be to say that there were only two substantially self-contained state-based world-systems on Earth: one in China, and a large multi-centric one which linked the Mediterranean, Egypt, the Near East and India. To make this characterization assumes a somewhat looser definition of economic network than that employed by Wallerstein. If the economic network ignores completely prestige goods, including rare metals, minerals, fine cloth, jewels, spices, wood and etc. and includes only such "fundamental" goods as food items or raw materials needed by the masses for everyday life, then world-systems must be smaller. Indeed, the inclusion of Spanish America in the European world-economy of the sixteenth century A.D. would be mistaken by this narrow definition. The empirical reality is that we have smaller circuits of food and commonly used raw materials, larger circuits of rarer foods and raw materials, and then even larger circuits in which valuable (per weight) prestige goods circulate. But this formulation may obscure some important features, especially if we are concerned about fundamental social change. Some prestige goods play only a symbolic role in power-reproduction, but others, as Schneider (1977) points out, function as money and allow mercenaries to be employed and potential enemies to be coopted. Thus we ignore prestige goods exchange at our peril. In addition we must at least mention the diffusion of information, and cultural, technological and religious inventions. Any consideration of social evolution must be cognizant of these things, although the extreme diffusionism of some scholars, e.g. Childe (1946), ought to be avoided. It is not the case that once monotheism or bronze-making was invented it spread uniformly across the globe. Irrigated agriculture clearly "leap-frogged," as McNeill (1963) has pointed out, to those areas which were ecologically suitable. And other inventions took root only in social structures which were prepared to make use of them. The point here is that the units of analysis of evolutionary change must take account of the various circuits which compose an "economic network" and the role that cultural diffusion plays in change. It is laudable that historians have discovered that the everyday life of the masses is more than simply a backdrop for the drama of the great men. But it would be a mistake to reduce our attempt to understand social evolution to a consideration of the history of food. Economic networks and subsystems The consideration of ancient world-systems requires us to pay much more attention to the distinction between fundamental goods and presciousities which Immanuel Wallerstein (1979a) has used to bound world-systems. In Wallerstein's usage presciousities are goods exchanged between relatively unintegrated systems such that each gives something which is fairly worthless in exchange for something of great value. This kind of exchange presupposes not only that each system has items which are unknown to the other, but that there has not been much previous exchange. One thinks of the first glass bead traded to an American Indian tribe in exchange for a beaver pelt. Or the first Chinese silk traded for lapis lazuli. The new item has infinite value because it is unknown, but is also appealing in terms of some already-existing notion of value. The opposite of this is fully-integrated market exchange in which commodities are traded at rates of exchange which reflect their value in terms of substitutable elements, such as "abstract labor" based on equivalent units of socially necessary labor time. Empirically, instances of both types of exchange are rare. The first type requires original contact, and the second type requires a pure commercialization unaffected by contaminating elements of culture or power. The theoretical intent of Wallerstein's distinction is to bound systems which have substantially self-contained mechanisms of reproduction. Any attempt to understand the logic of evolution, the logic of modes of production and their transformation, is vitiated by an approach which asserts that "everything effects everything else." Thus Lenski's (1982) or Wolf's (1982) contention that the whole globe has constituted an inter- actional world system for millenia, although supported by the extreme diffusionists such as Childe (1946), makes it exceedingly difficult to ask, let alone answer, questions about modes of production and transformation. Wallerstein's theoretical intent is, thus, important, but his distinction is both empirically fraught with difficulties, and theoretically inadequate. Wallerstein's original purpose was to separate the European economic network from its exchanges with other areas. The trade with China was portrayed as one of presciousities, and this was alleged to include prestige goods which are used or consumed by elites. It is implied that these exchanges do not have important consequences for reproduction or transformation of modes of production, although Wallerstein does not himself use these terms. The first thing which a consideration of stateless and ancient world-systems implies is that we must separate the institutional form of exchange from the social uses to which the exchanged items are put. Thus salt or birdfeathers may be sold or given as gifts, and, after the transaction, may perform fundamental and necessary functions or merely epiphenominal ones. We cannot bound world-systems by distinguishing market trade from other forms of exchange. Market trade did not exist, or was very rare, in stateless world-systems, whereas in the modern world-system it is the most frequent and important form of exchange. Using the institutional nature of exchange to bound world-system economic networks only adds confusion. Rather we must look at the circuits of material exchange regardless of the institutional forms by which they are accomplished. So what is a theoretically useful method for bounding world-systems? Let us look first at the empirical nature of characteristic spatial exchange patterns. Any spatial exchange network consists of a series of circuits which are nested and overlapping. At the spatially smallest level an individual or local collectivity consumes what it produces. This we call subsistence. Exchange may not be involved at all, unless we want to consider the intercourse between the hand and the mouth. Actually, exchange relations within such local circuits can be of extremely varied form, but whatever they are, it is notable in all economic systems that bulky and difficult-to-store food items with a low value per unit of weight tend to be confined to the local circuit. This generalization is affected by transport costs as well as institutional structures. Where transport of bulk items is "economical" in the sense that the cost of transport is substantially less than the value of the good at a distant location from its point of origin then the bulk good may be carried long distances. But generally, bulky, spoilable goods have small circuits. A larger circuit is composed of less-perishable, more valuable foods, and consumer durables such as pots, tools, etc. These may be produced at certain artisan centers for distribution across a larger area than that of spoilable food. Live animals for food, wool or as work-animals also may be transported over wide areas, as they can be herded rather than carried. The limit on their distance of trade is the cost of forage or feed. Once this exceeds the value of the animal the circuit halts. Strategic military or religious-symbolic value can lead to the long-distance movement of animals. One thinks of the usage of Indian elephants by Greek armies. In addition, the movement of slaves is economically similar to the movement of animals, as they can go under their own power, but must be fed. The next larger circuit is composed of precious metals which are used as money. The fact that these metals may be used both for decoration and as means of payment may confuse our interpretation of the importance of bullion circuits. But as Schneider has pointed out, silver and gold have long been used to pay mercenary armies, and thus to support or undermine states and empires. An also potentially very large circuit is formed by those prestige goods which operate within a cultural context to symbolize high rank. Certain types of Incan cloth, lapis lazuli, turquoise, hummingbird feathers, are easily transportable over great distances and have great "value" within the political and cultural contexts in which they symbolize high rank. Marriage exchanges, the sending of women, is also a form of exchange which may easily be accomplished at very long distances. The use of wife-giving and wife-taking to cement political alliances or empire- formation is a very important element in the building of states and empires. Of course we have only considered material items of exchange above. These nested circuits are also linked with important cultural and information flows, as well as social and political structures. But here we want to utilize the Wallersteinian focus on "economic network" to define world-system boundaries in a more empirically valid and theoretically useful way. The real intent of such boundary definition is to separate epiphenominal from important exchange. But none of the types of circuits defined above are epiphenominal. Even purely prestige good economies have important consequences, causing uneven development, the development of money and often leading to other sorts of exchange. And of course it is not just a question of the existence of exchange, but also of its quantitative extent. One deal does not a system make, no matter what is dealt. But how many deals do? What proportion of what constitutes a single system? The cutting points are arbitrary, but, keeping in mind that we want to separate epiphenominena from systemically important exchange, I shall contend that any regularized and repetitive exchange of material goods which substantially affects the welfare of a population or the maintenance of a power structure constitutes part of a single world-system network. Indirect links, such as when two, non-exchanging core areas compete with one another in a peripheral zone, or two peripheral zones are indirectly connected through links to a single core area, will constitute parts of the same world-system. It will be useful to designate more densely interconnected regions as "subsystems" within larger multicentric world-systems. The Eurasian and Chinese world-systems This section deals with the two complex world-systems which emerged in China and in the Near East (including the Mediterranean and India) during the first millenium B.C. The inclusion of India in the Near Eastern- centered world-system is somewhat problematic, as it participated in an exchange of some prestige goods with the West but not fundamental goods, and the diffusion of cultural traits was weak, facilitating the emergence of very different social and religious institutions. The Chinese world-system actually fits most neatly into the category of a world-economy composed of states which was later almost completely encompassed by a single-over- arching political empire. On the other hand, the Near East does not fit this idea at all. In the Near Eastern centered Eurasian world-system empires were always partial, often existing simultaneously within the same larger economic network with other empires. And the phenomenon of specialized semi-peripheral trading nations emerged in the context of this complex world-system. To further specify the focus of this section, we will include, in China, the emergence of the Ch'in and Han empires. In the world-system centered on the Near East we will include the Assyrian, Median-Persian, post-Hyksos Egyptian, Alexandrian and Mauryan empires as well as peripheral and semi-peripheral states such as the Indian Mahajanapada states, and the Phoenecian and Greek city-states. For purposes of comparison we include another figure from Taagepera (1979) even though it only partly overlaps the period we are here considering. (Figure 2 About Here) As did Figure 1 in the previous section, this shows the relative size in land area of empires in different regions, but the time period covered in Figure 2 is from 650 B.C. to 700 A.D. Comparing this to Figure 1 (3000 B.C. to 600 B.C.) we can see, as Taagepera points out, that the size of the largest empires increased dramatically around 600 B.C. with the coming of the Median-Persian (Achaemenid) empire. A similarly large size increase was attained by the Ch'in and Han empires of China about 300 years later. Taagepera supposes that these great size increases had to do with innovations in the delegation of authority, but when we compare the political forms used by the Persians in their rule of various satrapies and relatively autonomous cities (Cook, 1982: Chapter 8) we don't find anything radically different from the smaller Assyrian warrior empire or even the primary empires. A broad range of organizational forms, from feudal fiefs to multitiered bureaucracies, were used in various parts of the Persian empire, depending on local circumstances. The one policy which seems to stand out as different from earlier unifying empires was the Persian support of local religious leaders. Darius funded the reconstruction of the temple at Jerusalem and restructured the state of Israel along more hierarchical lines. This type of policy gained the Persians valuable allies in many regions and undermined local mililtary leaders, but such divide and conquer tactics had been used before. It may be rather that it was the degree of economic and cultural integration which had been attained over a wider area which accounts for the growth in the size of the largest empires. The evidence on which we must base our understanding of the complex world-systems is again archeological, supplemented with historical documents. Ironically, the transition to writing on paper instead of clay means that in some cases less documentary evidence survived. For the Persians we learn most from stone inscriptions on monuments (Cook, 1983). We are quite dependent on those documentary sources which did survive, especially the Greek historians, but also Egyptian material. Because of the dryer Egyptian climate, papyrus documents were better preserved. The foremost characteristic of the large empires which emerged in the Near East is their completely military nature. They are termed "warrior empires" by Diakonoff (1969). Utilizing the horse-drawn chariot as a tool of war, the Assyrians, based in Assur, were able to conquer a very large portion of the existing economic network of the Near Eastern world-system. According to Schumpeter (1951) this war-machine state expanded according to an internal logic which drove it to conquer areas for which there was no economic motive at all. He maintains that a psychology of conquest drove the Assyrians, that they required warfare in order to sustain their self- conception as a conquering people. Diakonoff (1969), on the other hand, maintains that the Assyrians were able to consolidate their empire because of the central location of Assur near the main trade route between the Tigris-Euphrates and the Mediterranean. The Assyrians politically united the Nile and the Tigris-Euphrates, temporarily consolidating the two older core areas into a single world-empire. Diakonoff describes the motive of the Assyrian conquests as "forcible exchange," at first simply the accumulation of booty, but later evolving into more regularized tribute-taking. For the first time a world-system became primarily organized around a coercive core/periphery relationship.14 The organizational problem for the conquerors, once they had extracted all the booty they could within their range of conquest, was to convert pillage into tribute-taking. And if they succeeded at this, they became "civilized" men trying to keep together a large political entity through a combination of force and cooptation. Military cadre settled down on the lands they acquired, had children, learned the pleasures of a peace-time life, and became reticent to personally heed the call to arms. Mercenary soldiers from outlying provinces were hired to fight, and these soon learned that the old rulers had become weak. This set the stage for conquest by a new semi-peripheral marcher state, the Hittites, Medes or Persians. Other factors, of course, entered in. At first a warrior-state tried to convert all trade to state-administered exchange. But eventually, in part because of the enforced peace and the differentiation of power which allowed private accumulation, a long distance market reasserted itself. This created sources of wealth outside the state apparatus, which may have been the basis of political strife over succession to the monarchy as well as challenges from regional centers of power. Within the warrior states many areas such as old Babylon maintained a certain autonomy which became the basis of challenges to the empire. Another factor which apparently played a role in the success of marcher states was changes in military technology. The bronze age reliance on noble charioteers was undercut by the development of cheaper and stronger iron weapons, first perfected in Anatolia. The chariots could not match the new heavily armed infantry, and thus warfare became less an honorable profession of bronze-age aristocrats and more a mass-based confrontation between hordes of infantrymen with iron weapons (McNeill, 1963). The consequences of the policy of looting were not easy to mend. Vast areas were ruined and depopulated. In order to convert to a tribute-taking basis these areas had to be reconstructed as productive. The Assyrians accomplished this by the mass deportation of peoples from peripheral areas to the old centers of high production (Oded, 1979). Schumpeter's claim that a kind of imperialism for its own sake was based primarily on psychological and cultural elements may be in error, but his contention that modern imperialism is only a weak and atavistic version of "real" imperialism may be closer to the truth. Weak or not, the question remains as to how necessary imperialism is to any mode of production. There is little doubt that it was necessary for the warrior empires. Diakonoff (1969: 31) develops his argument about the economic basis of the warrior empires as follows: The main economic role of the warrior empires was, as we have tried to show, the establishment of "forcible exchange" in the interests of agricultural and handicrafts in the more developed countries, such as Babylonia, partly Assyria, and later on probably also Phoenecia. This means that behind the activities of the warriors we must seek for a social stratum which profited by these activities. It was not merely the top military bureaucracy, hoarding wealth from their part in the loot and receiving from the king estates immune from tax- ation and corvee, and peopled with deportees. It was also such privileged cities, centres of handicrafts and agriculture, as Harran, Assur, Babylon, Sippar, Nippur, Uruk and later also the Phoenecian autonomous cities. These cities continued to be governed by a Council of Elders and an Assembly; their citizens thus enjoyed full political rights and, moreover, were exempt from empire taxes and levies. These citizens were priests, merchants, owners of handicraft shops with slave labor, or of land, being in all cases either free land-holding producers or slave-owners depending on slavery economy. Diakonoff contends that the problem of the relationship between the autonomous cities and the royal power was one of the main political problems of the warrior empires. In a footnote he clarifies the interaction and shows that Schumpeter's characterization is not totally without foundation. The political relations between such cities and the royal power were subject to change. There seems to have existed a party (represented e.g. by Sennacherib) which did not favor city and temple privileges and preferred uninterrupted looting of the peripheral regions (as a means of keeping the warriors content) to the relative stability of the political situation needed by such trading centres as Babylon, Harran, etc. For this party the wars of the empire were an economic goal in itself, irrespective of the causes which had started them (Diakonoff, 1969: 31, Note 3). From where did the conquering peoples who created the warrior empires come? The domestication of horses, the invention in Sumer of the chariot and the spread of bronze technology to the peoples of the desert, steppe, and mountains surrounding the valley civilizations created a grave threat to the cities. At the same time the increased density of inter-regional trade and the spread of a more productive agriculture to the upland, rain-watered areas, as well as the previous political integration carried out by the primary empires within regions, created the organizational possibility for larger empires. The conquerors who created these empires were of different backgrounds. Assur, the city from whence the Assyrians came, was on the main trade route between Sumer and the Mediterranean. Merchants carrying goods along this route are thought to have brought civilized elements to the people of Assur (McNeill, 1963: 48, Note 28). The Medes and the Persians were upland people, nomads, who had only recently adopted agriculture when they began their campaigns of conquest. The Persian King Darius "trans- formed an Aryan feudal kingship into an oriental despotism" according to Cook (1983: 76). The later nomadic Scythian and Mongol horsemen who showered the infantries of civilizations with arrows were less involved in the exchange networks of civilization before the time of their conquests and thus may be best categorized as peripheral rather than semi-peripheral. Of course they rapidly adopted many of the ways of the conquered after establishing their power over old core areas. Egypt maintained a studied isolationism from the other centers of the Near East before the invasion of the charioteering Hyksos. There was tribute-taking from Nubia, cedars were received from Byblos, and there was some trade with the Minoans and, through Syria, with Mesopotamia. But the inward-looking culture of the Egyptian system considered the outside world to be inferior and not worthy of much consideration. This attitude was rudely challenged by the barbarian Hyksos, and after a long struggle to expel the aliens, the Egyptians developed a more outward-looking policy. They engaged in an imperial expansion which subjected a part of the Asian continent to their rule, and the Pharoah Ikhnaton tried to suppress the older Egyptian gods in favor of the single great god of the Sun, which shined on all lands, not just Egypt. The expansionary phase established more formal domination over Nubia, as described by Adams (1977) in the previous section. But Egypt's experiment with imperial expansion and cosmopolitan culture was partial and temporary. The more-efficient alphabetic form of writing was successfully introduced, but the party of monotheism was overthrown by a resurgence of the local priests. Ikhnaton's policy of military expansion was stemmed, and a traditionalist revival of the old ways triumphed. Even the king of tiny Byblos refused to supply cedar logs on demand (Herm, 1975). Egypt's move to become a significant player in the arena of the warrior empires failed, and it became only a regional center, sometimes independent, often tributary to the larger empires of the East. Not until it was Hellenized was Egypt to play again the role of great power in the Eurasian world-system. Lattimore (1940: 531) graphically describes a somewhat similar characterization of the cycles of Chinese dynasties: Although the social outlook of the Chinese is notable for the small honor it pays to war, and although their social system does not give the soldier a high position, every Chinese dynasty has risen out of a period war, and usually a long period. Peasant rebellions have been as recurrent as barbarian invasions. Frequently the two kinds of war have been simultaneous; both have usually been accompanied by famine and devastation, and peace has never been restored without savage repression. The brief chronicle of a Chinese dynasty is very simple: a Chinese general or a barbarian conqueror establishes a peace which is usually a peace of exhaustion. There follows a period of gradually increasing prosperity as land is brought back under cultivation, and this passes into a period of apparently unchanging stability. Gradually, however, weak administration and corrupt govern- ment choke the flow of trade and taxes. Discontent and poverty spread. The last emperor of the dynasty is often vicious and always weak--as weak as the founder of the dynasty was ruthless. The great fight each other for power, and the poor turn against all government. The dynasty ends, and after an interval another begins, exactly as the last began, and runs the same course. Lattimore goes on to qualify this characterization for different periods. Here we are concerned with the Ch'in and Han empires and their relationship with peripheral regions. It was not until the first century B.C. that the Silk Road became a safe route for caravans with the establishment of a chain of empires linking Han China, Kushan, Parthian and Roman Empires. Before that diffusion was limited to down-the-line trade among oases settlements and/or the carrying of cultural elements by migrating steppe nomads. Some have argued that original Chinese state formation by the Shang was spurred by an invasion of chariot-making barbarians from the West (McNeill, 1963). It is true that bronze-making did probably diffuse from its point of invention in Sumer across the Asian continent to China. But little evidence supports the conquest theory, and most scholars believe that primary state formation in China was autochthonous. The relative isolation of Chinese civilization prevented the diffusion of alphabetic writing. By the time contacts with the West became more common, the Chinese already had a substantial investment in a great literature written in their complicated ideographic script. On the Indian subcontinent the early Harrapa/Mohenjo-Daro civilization in the Indus valley had strong trade ties with Mesopotamia and Egypt. About 2000 B.C. the wars between Hammurabi and Elam interrupted this trade (Chandra, 1977: 36). By about 1500 B.C. the urban civilization of the Indus was gone, most believe as a result of the Aryan invasions. A process of state formation which appears very similar to that which we have described in Sumer began in the Ganges valley during the first milennium (Thapar, 1980). In the seventh century B.C. northern India was divided into sixteen Majanapadas or "great states" (Chandra, 1977: 50). Actually some of these were probably still chiefdoms. In a slow process of alliance and conquest this interstate system became dominated by the Mauryans of Magadha. George Modelski's (1964) analysis of the third century B.C. Indian interstate system is based primarily on the Arthasastra, allegedly written by the Brahmin "Machiavelli," Kautilya. Kautilya's prescriptions for the successful prince reveal some of the principles of the Hindu interstate system as it operated before the emergence of the Mauryan empire. Implicitly comparing the Hindu system with the European interstate system of the modern era, Modelski notes the relative lack of development of an institutional apparatus for relations between equal states. The Indian ideology assumes that relations are always hierarchical and Kautilya prescribes the methods by which an aggressive prince (uijigishu) can improve his power vis a vis other states. A list of the types of sandhis, or agreements, accords between a dominant state and a weaker one reveal some interesting insights into interstate relations in the context of a tributary model of production (Modelski, 1964: 552). Sandhis relating to the army: Atmamishasandhi, agreement whereby the king and a fixed number of his troops present themselves when called by the superior king. Purushanturasandhi, whereby the commander in chief or the crown prince alone (but not the king) will present the army when called. Adrshtapurushsandhi, whereby no person is specified to serve with the army called by the superior king. Sandhis relating to wealth: Parikrayasandhi, offer of wealth (tribute) whereby other elements of the state remain free. Upagrahasandhi, offer as of much money as can be carried on a man's shoulder. Kapalasandhi, paying an immense amount. Suvarnasandhi, (golden peace), where the amount paid is tolerable. Sandhis relating to territory: Adistasandhi, cession of part of territory. Uchchhinnasandhi, cession of the whole territory except for the capital. Avakrayasandhi, agreement to hand over the produce of the land. Paribushanassandhi, promise to hand over more than the land yields. The extent to which India was linked with the Near East in this period should not be overemphasized. The conquests of Alexander extended trade and cultural contact through to the Ganges Valley, but in many respects the Hindu world remained an autonomous subsystem. Wallerstein has pointed out that the boundaries of political-military arenas and the boundaries of economic networks sometimes do not coincide. This seems to apply to the Indian subcontinent, which remained a semi-autonomous subsystem of the multicentric Eurasian world-system with its own core/periphery hierarchy and interstate system. This is confirmed by Modelski's (1964: 555) description of the boundaries of the Hindu interstate system: That the international system within which Kautilya's princes exercised their status-seeking existed in a "world" of its own is indicated in a number of passages. The Arthasastra refers several times to those who will conquer "the earth," or "the earth bounded by the four quarters." And in another place it indicates that this earth is to be understood roughly as the Indian sub-continent south of the Himalayas. This is the phyical setting of Kautilya's international relations, and the arena in which the vijigishu could exercise his ambition. Thapar (1980) makes a persuasive case against the use of the concept "Asiatic mode of production" to characterize the Indian civilization. In addition to communally held village lands there were privately owned lands, and both intraregional and long distance trade were important. The regionally varying forms of caste institutions, the struggle between Brahminism, Bhuddism, and Jainism, and the later expansion of economic and political influence over peripheral areas in Southern India, Sri Lanka, and Southeast Asia suggest a dynamic social system rather than a stagnant evolutionary cul de sac. Repeated invasions by Near Eastern emperors affected Northwestern India and brought it within the tributary orb of the Western empires. But the Persians and Macedonians never conquered the Ganges valley, reaching only as far as the Punjab. During the first millenium the states of the Ganges traded prestige goods including cotton and wool textiles, and elephants across the Great Road and by sea to the Near East, and this trade was in the nature of equal exchange. As mentioned above, the tension between independent cities and the royal institutions of empire was a central one in the ancient Near Eastern world-system. Diakonoff (1969: 31-32) says: The Achaemenian kings vacillated between regarding an autonomous city as a menace to the stability of the empire, and regarding the former as one of the chief economic and political mainstays of the latter.... The Hellenistic monarchs were staunch adherents of the second point of view, which resulted in the foundation of hundreds of new privileged cities after the pattern of the Green poleis... moreover, alongside the Hellenistic poleis with a Greek constitution there existed--and were being founded--a number of cities with a local constitution but enjoying the same rights and privileges as a polis (the constitution being usually that of a temple-city). The existence of a polis or a temple-city safeguarded the households and the political rights of the slave-owners and the free citizens in general from arbitrary actions of the royal power; their constitution guaranteed free development of handicrafts and commerce, of private agricultural estates, and of the new money economy. In the Eastern Mediteranean early sea-going trade had been carried on by the Minoans on Crete, and perhaps also the Mycenaeans in the Aegean. During the second millenium B.C. the Mediterranean was upset by migrations from the north, barbarians and pirates who the Egyptians called the Sea Peoples. These sea-going raiders disrupted the sea-borne trade relations between Egypt and the Levant. The early Lebanese town of Byblos had long been a tributary supplier of cedar to Egypt, but the arrival of the pirates interferred with this trade. Semi-peripheral capitalism Eventually the new peoples settled on the Lebanese and Palestinian coast and mixed with the older Semitic tribes there. Out of this admixture the Phoenician cities of Tyre and Sidon emerged, transforming piracy into merchant capitalism. The Phoenicians specialized in trading across the whole Mediterranean, especially in exchanging with primitive peoples to whom the products of the old East were, at first, preciousities. They also maintained a certain autonomy as carriers of goods between empires, especially the Egyptians and the Assyrians. The Phoenicians established a colony for the production of copper on Cyprus, and began a manufacturing industry which provided inexpensive glass vases to mass markets throughout the Mediterranean. They had stolen the Egyptian secret of glass production and improved upon it for use in the carrying trade (Herm, 1975). Thus the Phoenicians began a process of exchanging unequals, that is trading between systems not previously integrated by exchange. Buying (or producing) cheap and selling dear was their specialized purpose, and these independent temple-cities of Lebanon were perhaps the first capitalist states in which profit-taking through commerce was the dominant raison d'etat. Their domain spread as far as present-day Cadiz on the Atlantic coast of Spain, and they traded down the Red Sea to Africa and India. They had alliances with the land-based Hebrew states of Judah and Israel, building the temple at Jerusalem under contract. There are many definitions of capitalism. The one we are using here is a modification of Marx's which is suggeted by the world-system perspective. Chapter above discusses this in detail, but here suffice it to say that there may be intermediate forms between a completely uncommodified exchange network and one in which commodification of every element is perfect. Land, wealth and labor are never perfect commodities in any historical system. Rather commodification is a matter of degree. The Phoenicians engaged in merchant capitalism by buying and selling products of different, relatively isolated areas around the Mediterranean littoral (the exchange of unequals) and by doing this they brought the separate economies of these areas into interaction, subjecting the division of labor within each to the market forces generated in the network as a whole. The Phoenicians also produced manufactured goods for export, and these were not only prestige goods. They specialized in producing cheap copies which could expand the market demand by lowering prices. The Phoenician glass implements industry and textile industry were not merchant capitalism, but rather production capitalism which imported technical knowledge and applied it to the mass production of commodities (Herm, 1975: 80). We don't know how the logic of costs was applied to raw materials, technique and labor in the cities of Tyre and Sidon, where these industries were located. The commodification of labor power may have been constrained in many ways, by kin networks or state regulation. But it seems reasonable to assume that these forms of labor regulation were subjected to the competitive logic of the larger Mediterranean market economy. Dominant powers at sea, the Phoenicians built their cities on promontories where they could be easily protected from land attack. Tyre was built on a small rock island, which was only conquered after Alexander's army built a causeway to it. Although fairly autonomous, as mentioned by Diakonoff above, the Phoenicians were not adverse to paying tribute to whatever land based empire was currently responsible for the peace. They allied with the Persians in their war against Greece, and thereby gained the permanent enmity of the Greeks. The Phoenicians never achieved political unity amongst themselves, and one of their split-off groups founded Carthage, to become the center of an independent trade network in the Western Mediterranean. The Carthaginians long battled with the Greeks over Sardinia, and in one of their temporary conquests of a Greek colony they discovered the superior art and craft products of Greek culture. After this, statues of Greek gods appeared in the gardens of Carthaginian aristocrats, although when they were direly pressed by a Greek conqueror from Sardinia, they flooded back to their old religion, carrying out a hugh human sacrifice to try to sway the old gods. The Greek city states invented a new type of polity and combined sea-borne trade with commodity production in agriculture. The Phoenician temple-cities were so specialized in trade that they did not concern themselves with revenues or profits from agricultural land. Their colonies, like the original cities on the Lebanese coast, were set up on promontories, and oriented toward the sea. The Greek polis, on the other hand, was an association of land-holders. Its uniquenes was not in its autonomy from imperial rule. This was shared by other autonomous cities of the Near East. What made the Greek cities different was the small and inconsequential nature of the state sector compared with the private sector (Diakonoff, 1982). Sparta, of course was a partial exception. The polity of most of the Greek city-states was primarily an association of private land-owners rather than a major land-owner and economic sector in itself. This, and the high level of commodity production, were the features which made Greek society different. Diakonoff puts these developments in the perspective of the larger Eurasian world-system as follows: This was the reason why individual slave-owning market- oriented economies could develop only here. And that was the basis of the splendid growth of ancient democracy, arts and science. But the existence of a world commodity market is a question of diachrony and of previous development of other and older ancient societies, so that in principle, the existence of a "one-sector" structure of economy (with only the "communal and private" sector!) does by no means prove that Pericles' Athens belonged to another and "higher" formation in comparison with the societies which alongside of a private-communal one also had a state sector (Diakonoff, 1982: 93-94) (emphasis in the original). Unlike the Phoenicians, who only set up trading cities, the Greeks had traditional attitudes toward expansion and conquest. The purpose of conquest was booty, and the purpose of colonies was to settle citizens who could farm, and provide food for the mother city. The cultural disapproval of merchants did not prevent Greeks from engaging in trade, but it may have partly accounted for the large number of foreigners (metics) who lived in Athens. While most Greek colonies were established for the purposes designated above, there were a few that specialized in trade (Curtin, 1984). And as the cultural Hellenization of the Near Eastern world-system proceeded, the Greeks, or at least many of them, became long distance traders. Even Polanyi (1975: 87) admits the existence of price-setting markets at Athens. He contends that coinage in small denominations indicates a real peoples' market, and that for the first time it became possible to buy a prepared meal. Greek generals undertook to provide victuals for their troops on the road by encouraging local provisioners to set up markets ahead of the army's arrival. A generalized commodity economy was emerging, at least within the Greek sphere of influence. How can we describe the political economy of the complex Near Eastern- centered world-system? Within the great empires trade and production for exchange grew, but two tendencies limited it. As of old, the empire, and most of the competing polities within empires, tried to monopolize exchange. This was accomplished by converting long-distance trade into tribute-taking and local trade into royal monopolies producing revenues utilizing a dependent population receiving rations or subsistence plots. Nevertheless private trade and markets tended to grow, but they were stifled by the lack of a generalized monetary system. Shortage of money was reinforced by and, in turn stimulated, usury, and by the efforts of emperors and kings to horde. In the interstices between empires, as the Lebanon, and in other peripheral and semi-peripheral areas, market forces were less constrained. And here we see for the first time the emergence of capitalist states, the purest case being the Phoenecian cities. The Greeks, especially at Athens, demonstrated that a state can adapt to market forces and benefit from them (rather than trying to consume them) while still remaining a predominantly land-based society. This trick, the coexistence of a tributary mode of production with a large measure of commercialization, was invented in semi-peripheral Athens, but was later learned by whole core empires.