Chapter 19: The Classical World System, Natural Adaptive Cycles, and Societal Collapse at the End of Antiquity


In 97 AD the Chinese general and Protector General of the Western Regions, Pan Chao (Ban Zhou), led a sizeable army (reportedly 70,000) across the Pamir Mountains into Bactria and advanced across the Oxus River all the way to the shores of an unnamed sea, presumably the Caspian. Some fifty kings in this nomadic region submitted to his authority en route. While camped there he dispatched an officer named Kan Ying to investigate further regarding the location of the “Other China.” Kan Ying reached a neighboring sea where he was told by inhabitants that Rome could be reached by a sea voyage that would require three years, round trip, and untold quantities of provisions. Since the Chinese army of Pan Chao had taken the “northern route” across the steppes of Central Asia to the Caspian, it is more than likely that Kan Ying had journeyed across the Caspian to the Caucasus region, perhaps Armenia or Georgia, and had made his way to the eastern shores of the Black Sea. If so he was indeed a long way from Rome. With his commander and an entire army waiting behind him, he thought better of the idea and returned to his encampment. Pan Chao and his forces returned to the Tarim Basin after a four year, roundtrip expedition. A seemingly pointless expedition had brought the Chinese army to the borders of the Roman Mediterranean world. What seems certain is that Pan Chao and his army were trying to gain intelligence about the existence of a distant empire, the one that the Chinese referred to as “Ta Ch’in,” the “Other China,” or the Roman Empire.


Nor was Pan Chao the first Chinese official to explore beyond the frontiers of China. After the Early Han dynasty secured authority in China at the beginning of the second century BC, it needed to address the persistent danger of Hsiung Nu depredations along its northern border. As we noted earlier, the Han first tried a policy of appeasement, hoping to purchase Hsiung Nu compliance with foodstuffs, bolts of silk, lacquered utensils, and other Chinese products and eventually to draw them into the Chinese sphere of influence. Around 130 BC the Han attained a sufficient level of internal control to shift from accommodation with the Hsiung Nu to an active policy of expansion. The emperor Wu Ti dispatched a courtier named Chang Ch’ien to distant nomadic confederacies on the western flank of the Hsiung Nu to persuade them to join the Chinese in war against a common foe. Chang Ch’ien’s journey resulted immediately in his capture by the Hsiung Nu where he was forced to remain so long (a decade) that he took a Hsiung Nu bride. But eventually he escaped and proceeded west, taking the northern route beyond the Tien Shien Mountains into Sogdiana and Bactria. His intended objective was to locate the Yueh Chih, a Turkic-Hunnic people who had been driven westward by the Hsiung Nu into the region of Sogdiana. To western sources these people were known as the Sacae and were linked through confederate relations with the Kushan rulers of India. Although he was unable to convince Sacae tribesmen to join China in a war against the Hsiung Nu, he traveled extensively through Afghanistan and the upper Indus valley gathering crucial intelligence about the urban societies of this region. He observed, for example, that people in the upper Indus were using products (bamboo and silk) that originated from his home country China. According to the natives these goods had arrived from China not via the Tarim Basin but by sea. Returning to the court of Wu Ti many years later, Chang Ch’ien wrote a detailed report of his journey that convinced the Han to explore connections with India via both directions – by establishing a road system through the Pamir Mts in the west and by constructing a second road system through provinces in the south west extending from Canton to within 100 miles of Burma. While the southern route proved impractical, Chinese conquest of the western regions became a viable option. It is important to stress that the motivation for the development of Chinese territories in the west was the unending conflict with the Hsiung Nu and the need to find allies (not to mention horses) to oppose them rather than any overt interest in trade. What the experience of Chang Ch’ien demonstrated, nonetheless, was that trade between India and China was on going at this time and that it flourished sufficiently to warrant the attention of the Han emperor himself. A year did not go by that the Han emperor did not dispatch some five to ten caravans of envoys, each one consisting of hundreds of officials and attendants, to regions to the west and south.


The Emperor Wu Ti also dispatched a general, Li Kuang-li, the brother of his favorite concubine, to undertake a prolonged campaign against the Hsiung Nu, focusing on the conquest and reorganization of the independent city states of the Tarim Basin as a Chinese protectorate. Li Kuang-Li’s initial campaign into western Mongolia nearly ended in disaster, with 90% of his army lost in a disastrous desert campaign. In 104 BC, he was assigned a new army and this time he not only conquered the Tarim Basin, but he successfully invaded Ta Yuan, Sogdiana, defeating a native army outside the walls of the royal capital Er Shih and ultimately reducing the inhabitants to submission. Slightly later when he learned that Chinese envoys had been murdered by a distant king of Chi Pin, he returned across the Pamir Mts., deposed that king and installed a puppet regime more to his liking.  Scholars have identified Chi Pin with a Kushan principality situated in the Kashmir region of northern Pakistan, and the king with a ruler whose name appears on coins as Heraios. The Early Han remained in control of the Tarim Basin ready to strike across the mountains until the end of the first century BC. Chinese policy toward the west remained driven predominantly by considerations about the Hsiung Nu. With the collapse of the Hsiung Nu Empire around 53 BC, Han administrators were free to address other more pressing issues. Some 65 years passed before the Later Han were in a position to dispatch Pan Chao to the region to reconsolidate Chinese control in the Tarim Basin. His seventeen year long tenure as Protector General of the Western Regions culminated in the unprecedented march to the Caspian sea noted above.


The passes through the Pamir and Tien Shien Mts., one via Sogdiana (Ferghana) to the north, the other through Bactria slightly south, appear to have fascinated other rulers of ancient world empires as well. In 329 BC Alexander the Great journeyed across the Oxus River to Bactria and Sogdiana. He established his most remote colony, Alexandria Eschata (Alexandria of the Furthest Extent), somewhere in the valley of Ferghana. He conducted this campaign ostensibly to eradicate Achaemenid royal connections to the region and to prevent Persian nobles from provoking uprisings behind the Macedonian supply lines across Afghanistan. Earlier that year, while Alexander was wintering in the region of Lake Seistan in southeastern Afghanistan, the satraps Bessus and Spitamenes had employed a route similar to the one taken later by Pan Chao along the Jaxartes and Oxus Rivers to slip behind his lines into Iran. The mobile detachments of Sogdian and Bactrian horsemen knew the routes through the desert and could seemingly strike at will and disappear without a trace. There was no choice but to root out their bases at the foot of the distant Pamir Mts., despite the hardship this brought on his army. During a winter encampment in Sogdiana in 328 BC, discontent was rife at the Macedonian camp. After a heated exchange during a banquet Alexander in a drunken rage unwittingly murdered one of his best friends, Clitus the Black, after informing him that he would be left behind as the governor of Sogdiana.  The following year when Alexander married the Bactrian princess, Roxane, another conspiracy emerged among his royal pages, leading to the arrest and execution of Artistotle’s nephew Callisthenes. Leaving behind merchants and veterans, Alexander hoped to maintain control over this distant region, and it is interesting to observe that a handful of Macedonian generals managed to sustain “Macedonian” control of Bactria during the confused decades that followed Alexander’s death in 323 BC. The Eucratid dynasty controlled the region until they were overrun by the Yueh Chih / Sacae, mentioned above, in the early second century BC. Heraios of Chi Pin, deposed by Li Kuang-li in the 90s BC, was possibly connected to these rulers as well. Like the Chinese, in other words, the Macedonians seemingly understood the importance of this remote highland region to world affairs.


When invading Afghanistan Alexander followed a road that was first opened up by Darius I of Persia, who journeyed through Afghanistan to reassert Persian authority (first imposed by Cyrus I) in Bactria, Sogdiana, and the Indus around 515 BC. Alexander greatly expanded this road network by establishing colonies along its length. He established irrigated agricultural communities near Lake Seistan to generate food stores for the army, he offered huge incentives for merchants and veterans alike to set up way stations along the route to accommodate the flow of traffic, and he even went so far as to implement the use of camel caravans and wagon trains to move men and material across the central Asian plateau. Phenomenal, if not extravagant accomplishments are recorded during this journey, such as the constant flow of men and weaponry from the Mediterranean and the importation of Aegean Greek wines. While encamped in Bactria his officers allegedly imported wagon loads of fine “wrestling” sand all the way from the Nile river delta. Since Alexander had the benefit of the captured Persian treasury, he spared little expense to move supplies along this route to distant places such as Sogdiana and India. Only rarely did his supply system break down. The transportation nodes he created appear to have survived not only his demise but the relentless political upheavals of the following century. In the early third century BC, the Mauryan Emperor Chandragupta shipped five hundred war elephants to Seleucus I of Syria during his epic conquest of portions of eastern Iran and Afghanistan. Antiochus III (223-287 BC) used the highway once again to invade Bactria and the Upper Indus in 209-205 BC. Then came the Sacae (Scythians), migrating across from the Aral Sea, and the Yueh Chih (176-160 BC), Turkic nomads driven across the central Asian mountains from western China to seize control of Sogdiana, Bactria, Ghandara, and India as the Kushan. After them came Li Kuang-li, and Pan Chao, mentioned above. Regardless of the political chaos and topographical difficulties, the highways leading from Iran to the Tarim Basin, the Indus, and points beyond remained crucial arteries in the network of ancient world politics and trade. Putting aside the questions of the volume of interregional trade (admittedly small) and its content (admittedly restricted to prestige goods), the fact remains that the routes themselves were deemed so important to distant rulers in Persia, India, Macedonia, Syria, and China that they repeatedly dispatched large armies to this seemingly barren region. The importance of these routes is gradually being confirmed by archaeological discovery as well. In Bagram Afghanistan (ancient Kapisa) French excavators in the late 1930s uncovered a remarkable stock of Roman-era Mediterranean finished goods, including glass, sculpture, bronze platters and receptacles, steelyard weights, and alabaster stone vessels. The presence of a statue of Harpocrates and the alabaster vessels indicates likely Egyptian (Alexandrian) origination for many of these commodities. The glassware consist of several major Mediterranean types: millefiori bowls, painted vessels and goblets, cut-glass goblets and diatreta. One of the scenes on the diatreta has been identified as that of the Pharos of Alexandria; others reflect a repertoire fairly typical of Hellenistic glass products in Seleucid Syria, including the battles of Alexander the Great, scenes of gladiatorial combat, one depicting the duel of Hector and Achilles, and others with hunting scenes. Also recovered from the remains were lacquered goods originating from China and ivory plaques, statues, and coffers from India. Assigned dates for the Mediterranean finds range broadly from the first century BC to the fourth century AD. Kapisa would appear to have been a major exchange hub of the Classical world system at that time.


We raise all this by way of introduction to address questions of the interconnection of ancient world systems and the reasons for their simultaneous collapse. As noted in the last chapter, the Roman imperial administration in the Mediterranean, the Kushan dominance in India, and the Later Han Dynasty in China all collapsed toward the end of the second century AD, when all three societies were seemingly at their height with respect to population size, productivity, and general prosperity. Through the implementation of more disciplined, possibly more efficient, administrative practices, the Romans and the Gupta Dynasty in India were able to stabilize things by the late third century AD. They sustained their urban societies for another two centuries before collapse overtook them in the form of external invasions by nomadic peoples from the north. In China, the most distant of the three world systems, a period of confusion, the Era of Disunity, set in with the collapse of the Later Han Dynasty (220 AD). The political situation did not right itself until the emergence of the Tang Dynasty (618-907) in the seventh century AD. During the interim repeated invasions by nomadic peoples offset each new threshold of stability [the Three Kingdoms (220-280), the Jin Dynasty (265-420), the Sixteen Kingdoms (304-439), the Sui Dynasty (588-618)]. Collapse came swiftly in each region, in other words, but it was not necessarily similar in form or duration. For example, the eastern half of the Roman Mediterranean where most of the productive capacity of the empire lay was able to sustain itself relatively intact as the Byzantine Empire for several hundred years. But the overall effect was a decline in production and living standards throughout the three regions of these ancient world systems. This fact can be demonstrated in a number of ways, archaeologically as well as textually. Perhaps the most compelling evidence comes from the ice core samples obtained in Greenland, already used in Chapter Three to demonstrate the impact of the Ice Age. According to these records the levels of airborne pollutants such as heavy metals that are released during the smelting of metals, having risen gradually during the course of antiquity, dropped to nearly Neolithic levels after the fifth century AD. The production of metals world-wide clearly declined at this time. A heavy reliance on firing technologies across Eurasia to produce metal tools and weapons, not to mention the kiln firing necessary to generate massive quantities of roof tiles and ceramic wares, suffered significant decline at this time. In some regions such as northwestern Europe where evidence for the reintroduction of thatched roofs and wooden utensils is certain, firing technology practically went out of use altogether. Empirical evidence such as this raises two important questions, accordingly. One, did the trade routes across central Asia have anything to do with the high level of prosperity enjoyed by ancient world systems in three distant basins? Two, in view of the evidence of urban prosperity in all three regions, why did they collapse in such a seemingly coordinated manner?


To answer the first question concerning the relative importance of the trade routes between the Mediterranean, India, and China, one has to consider the nature of the trade itself. Skeptics who believe that transcontinental trade in antiquity was marginal to the economies of the regions insist that the trade itself represented “high value, low volume” commerce in prestige goods that inevitably did not affect general well being in any of the regions in question. They argue that the volume of this trade at any given time was minimal and that the sorts of commodities being transported, precious metals, gems, incense, exotic plants and animals, coral, and spices, were non productive in character. At best they represented prestige goods obtained by ruling dynasties in distant basins of the world to be distributed as rewards among their subordinate elites. Even in the Roman Mediterranean, where luxuries were distributed independent of the imperial court at Rome, the difficulties entailed in transshipping these commodities across vast distances ultimately rendered them too rare and expensive to be enjoyed by the general public. Accordingly, their impact economically would have been marginal. As we have noted, recent archaeological investigations in East Africa, south India, and Afghanistan have revealed the presence of relatively inexpensive Roman commodities, such as ceramic finewares, metal wares, glass. Amphoras laden with eastern Mediterranean, wines were equally components to this trade. The weight and size of Mediterranean amphoras recovered in southern India all but necessitating that they traveled in large ocean going merchant vessels. In other words, Roman merchants appear to have sailed to India to exchange what were essentially Mediterranean staple goods. In exchange they obtained commodities that were unavailable in the Mediterranean, such as silk, spices, aromatics, and precious stones. Much like Roman wine and oil, many of these products were neigher rare nor expensive in the regions of their origination. What affected their prices at the opposite end of the ancient world were the inherent costs of transport and the manifold layers of tariffs and commissions that were incurred along the way. To some degree the widely varying climates of ancient world production regions rendered one region's staple goods another region's luxury commodities. Accordingly, the international luxury trade of the Roman era needs always to be considered within the context of various socio-economic and political systems in the geographic regions through which it passed. Finds of Chinese jade statuettes in Pompeii and Roman coin hoards in Vietnam indicate that regardless of the actual volume of this trade, adventurous merchants found the necessary incentives to make it occur.


Questions regarding the volume of this traffic remain more speculative. Although archaeological data for the commodities from the eastern ends of this trading network, silk and spices, remains inadequate, the amphora evidence at Arikamedu raises the possibility that Roman maritime trade with India during the first centuries BC and AD was vibrant, and that Mediterranean staple commodities circulated in the Indian Ocean at a pace relatively similar to that which occurred in the Mediterranean itself. Roman merchants who established the trading contacts in India utilized those "finished goods" that the Roman world was most adept at producing. The Auximite kings of east Africa used Roman weapons to assert local supremacy; the Tamil princes of southern India used their geographical position at the center of ancient oceanic sea lanes to raise standards of living of an otherwise unproductive region. Traders from the Far East made the difficult journey to India bearing commodities that, according to available documentation, were readily available in their home regions. No one would argue that commonplace wines from the eastern Mediterranean bore intrinsic value commensurate with those of spices or silk from the East. Asymmetrical values of the commodities in circulation may have required much larger shipments of Mediterranean staple goods to obtain sufficient amounts of the goods in question. This would explain not only the presence of so many Roman amphoras at Arikamedu but the perception as well that the trade with India imposed an unfavorable balance of trade on the West. The repeated references to silk garments in Roman literature and the widespread finds of artifacts such as pepper shakers throughout Mediterranean archaeological contexts indicate, nonetheless. that however expensive these luxuries may have been their prices were affordable to wealthy elites outside the city of Rome. Assuming that eastern luxury goods were relatively plentiful in the Roman Mediterranean, the volume of traffic across the Indian Ocean may very well have been significant, at least at times.


Based as it is on archaeological discoveries at a handful of sites (Kapisa, Arikamedu) which may or may not be representative, any argument for a significant volume of interregional trade during the Classical era remains speculative. If we accept the prevailing “reductionist” argument and assume that international trade at this time was minimal and restricted largely to limited quantities of exotic prestige goods, it is still possible for long-distance luxury trade to have stimulated the simultaneous development of urban societies in all three basins. The explanation lies with the effect that globalism had on world systems development in general. As we have demonstrated, each of these regions developed highly integrated world systems capable of moving locally produced commodities throughout the topographical limits of society. The Mediterranean Sea enabled inhabitants in that region to move heavy commodities and staple goods from Egypt to Rome, Spain to Syria, and Libya to the Aegean; iron mines near the mouth of the Ganges furnished materials for metals production throughout the Indian subcontinent; rice, grain, and metals produced in the lowlands of the Huang Ge and Yangtze Rivers in China furnished the necessary foodstuffs and materials for imperial courts, such as Ch’ang An, situated in the highlands and ultimately enabled the the Han dynasties to project force as far as the Tarim Basin to the west and Vietnam and Burma to the south. Most production was probably consumed locally in a subsistence manner. Another significant portion appears to have circulated in delimited regional networks. However, some portion of the materials produced in each basin was transported across vast distances. It is interesting to note, for example, how often the breakdown of these transport systems, particularly the overland supply of foodstuffs in China, forced the ruling houses to relocate their capitals from the western highlands to locations more proximate to the centers of production themselves. In short, the progressive development of world systems enabled stratified urban societies to harness resources throughout their respective regions in a manner necessary to sustain more prosperous ways of life.


Globalism, or in this instance the access to exotic prestige goods from regions beyond the immediate world system, appears to have acted as an accelerant to the pace of production in each respective basin. As anyone who has witnessed the worldwide appeal of contemporary music, Hollywood movies, American blue jeans, or Asian produced electronic equipment will acknowledge, modern global inhabitants display an enormous appetite for exotic goods and the latest technologies. The authors themselves have repeatedly marveled at the incongruous display of satellite dishes perched atop rustic cinder block houses otherwise devoid of running water or toilets in the mountains of southern Turkey. Some people will forego basic necessities, in other words, to acquire the latest consumer goods. The first tenet of globalism, therefore, is that the availability of exotic prestige goods in and of itself tends to drive demand. It does not matter how rare or how expensive a commodity was when it reached the opposite end of the globe. Provided that an exotic prestige good was available some limited element of the population, the wealthiest elites, could afford it. Hence, the relative volume of long distance trade was less important that the existence of the trade in and itself. The driving force to this principle rests with the means used to purchase expensive imported goods. To make these purchases the inhabitants of interconnected ancient world systems made use the most widely accepted media of exchange, money in the form of coins, bullion, or goods such as silk or incense whose ease of transport and intrinsic net worth enabled them to serve as “liquid capital.” To generate the liquid capital necessary to obtain expensive imports, local property-holding elites in all three regions raised productivity by investing in their landholdings. They opened up virgin terrain to agriculture, they constructed kiln sites and forges, they built ships and wagons, and ultimately they generated surplus staples commodities intended for wider consumption within the world system. This pattern is visible in all three regions, along with evidence of population growth both in terms of its expansion spatially across the landscape and in terms of density at urban centers. In addition, all three regions exhibited a gradual transition from subsistence production by dispersed rural populations to highly efficient agricultural systems (oftentimes slave driven) intended for the production of surplus commodities. These commodities were circulated at local, regional, and world systems scales of the trading network. The availability of expensive prestige commodities, that is, trade globalism, had the capacity to quicken the economic pace in distant corners of the world by furnishing wealthy elites with the necessary incentive to expand the base of production in their respective localities. The rate and volume of exchange in distantly traded commodities did not “flow” so much as it “pulsated” with variable ebb and flow depending on political conditions along the trade routes and the relative success of regional staples production at any given time. The essential point to recognize is that globalism, the availability of exotic prestige goods from opposite ends of the world, had the capacity to accelerate productivity and to raise living standards in distant regions of the globe. The Chinese general, Pan Chao, knew relatively little about Rome, its location, its organization, or its size, but he returned to China from the shores of the Caspian Sea with the report that the inhabitants of Rome lived prosperously. It was the rumored prosperity of the Romans and the fact that a large civilization so far removed from China could afford to import its locally produced prestige goods that had attracted his and his ruler’s attention in the first place.


If one accepts the idea that globalism functioned as an accelerator of economic production, then the last question remains why these world systems failed and failed contemporaneously as they did at the end of the second century AD? If the trade between them was minimal to the point of being intangible, is it really possible for its demise to have precipitated economic and political collapse in all three basins sumultaneously? To address this question we need to return to points raised in our initial chapter concerning the building process of world systems and their natural adaptive cycles. In a useful book, The Collapse of Complex Societies (Cambridge, 1988), J. A. Tainter identifies eleven different factors that contribute to societal collapse. These include


Ø the depletion or cessation of a vital resource(s) on which the society depends

Ø the establishment of a new resource base that renders current procedures obsolete

Ø the occurrence of some insurmountable catastrophe

Ø a failure to respond adequately to changing circumstances

Ø unsuccessful competition with other complex societies

Ø the arrival of foreign intruders 

Ø the emergence of class conflict

Ø societal contradictions, or elite mismanagement or misbehavior

Ø a general social dysfunction

Ø the existence of debilitating mystical factors

Ø a chance concatenation of events

Ø unforeseen economic influences


Although the range of variables inevitably dilutes his argument, Tainter concludes that collapse is fundamentally a sudden pronounced loss of an established level of sociopolitical complexity. According to Tainter the ancient state emerged as a problem solving organization due to changing circumstances. Ancient states required legitimacy and the ability to mobilize resources. These sociopolitical organizations encountered repeated problems that required increased investment merely to preserve the status quo. This investment typically came in the form of:


Ø an expanded bureaucracy

Ø an expansion to the specialized departments of a bureaucracy (increased stratification)

Ø cumulative organizational solutions that grow bewilderingly unwieldy (consider the example of modern US tax codes)

Ø increased costs of legitimizing activities

Ø the increased costs of internal control and external defense


All of these must be borne over time by levying greater costs on the supporting population, often to no increased advantage. As the number and costliness of organizational investments increase, the proportion of a society’s budget available for investment in future economic growth invariably declines. Tainter defines the overall effect as one of “declining marginal returns.” Eventually declining marginal returns from state institutions precipitate collapse, sometimes sudden, sometimes in the form of the slow disintegration of a regime. 


Although Tainter’s interpretation tends to focus inordinately on the role of central authority or governing institutions in the process of collapse, the model of declining marginal returns does seem to offer crucial insight into the patterns seen occurring in the Mediterranean, India, and China at this time. By placing appropriate emphasis on the tendency toward growth it enables us to identify essential patterns to the growth cycles of past societies. Then as now, growth was viewed as a requirement in the construction of world systems. Growth yielded an array of benefits including better administration of resource storage and distribution, a better investment in agricultural energy and minerals production, improved internal order and external defense, and greater information processing and public works. In so many words, these improvements represent the seven criteria to civilizations noted at the outset of this book. One additional insight furnished by Tainter’s analysis is that growth in human societies has ceilings or thresholds that confront continued growth in ancient societies. To move beyond a threshold or ceiling requires innovations, new technologies, or new means of conservation. The ancient historian Polybius observed that ancient empires were always in a state of expansion or contraction; they never attained “steady state equilibrium” because conditions were always changing. At the same time, it seems clear that no society had the resources to grow indefinitely. “Sustainability,” thus, becomes an important part of the problem. One must recognize, however, that the further one pursues “collapse theory” in the ancient record, the more speculative our arguments inevitably become.


Questions about “sustainability” have received enormous attention in the current era, particularly the ecological argument of resource depletion. It is entirely possible that modern historians of the ancient world have underestimated the impact of firing technologies on the landscapes that we have visited in the Classical world. As the ice cores in Greenland demonstrate, the reliance on firing technologies to generate metalwork in the Classical period was so great world wide that its heavy metal by-products are visible in the sampled remains. Moreover, these cores factor in only the heavy metals released through smelting; they do not address the quantities of fuel exhausted by ceramic kilns, heated baths, and household uses that were employed extensively throughout the Classical world. Unlike modern energy supplied by various means such as oil, gas, coal, and hydro-electricity, in antiquity the principle source of fuel was wood harvested from forests. To generate the necessary firing temperatures for smelting and glazed ceramics, charcoal was needed. Given the limited means of charcoal producing technology, it typically required seven to eight units of wood to generate a single unit of charcoal. In other words, the ancient reliance on charcoal for high firing temperatures placed even greater demand on timber extraction from ancient forests. When combined with the expanding land clearance and conversion to agricultural production that is visible in every landscape of the ancient world, questions arise regarding the sustainability of urban societies of ancient world systems that were overly reliant on wood-based fuels for sustained growth. Many recent scholars have argued that resource depletion played a role in the collapse of ancient societies. J. Hughes, for example, categorically indicts Roman failure to adapt their society and economy to the limits of the natural environment. Hughes points to erosion resulting from deforestation, exhaustion of the most readily accessible minerals, and the overgrazing and excessive utilization of agricultural terrain as evidence of environmental failure. Food shortages and population decline sapped the Roman Empire’s strength and ultimately led to collapse. Geoarchaeologists investigating evidence of eroded landscapes (pollen counts in alluvial soil) have in many instances observed a disastrous change toward the end of Roman Empire. This transition is characterized by a major decline in pollen counts of cereals, arable plants, and pasture weeds, combined with a rise in tree pollen. This would seem to indicate that the agricultural landscape went out of use and reverted back to a more natural state with woodlands encroaching on formerly cultivated fields.


Another much discussed possibility is climate change. Even some slight variation in climate could result in significant deterioration in ancient human well being. Preliminary models do suggest that the Mediterranean basin during the Roman era was slightly wetter and warmer than it is today. Assuming that the same was true throughout the temperate regions of Eurasia, even mild climatic variation, such as a slight drop in temperature, could potentially have placed inordinate stress on all ancient societies including nomadic tribes that for some reason became increasingly mobile at this time, moving southward and westward throughout Eurasia. Such an environmental fluctuation could potentially have overwhelmed existing production systems, political hierarchies, and social organization. As Tainter has observed, however, the available environmental data for antiquity is extremely limited, too much so to verify an ecological argument for ancient world systems collapse. A compelling argument requires precise data regarding climate, population, available crop production, and other resource yields alongside data for the annual requirements of the population and of the sociopolitical system and the adaptive capabilities of the society in question. This sort of comprehensive data simply does not exist for the Classical world, where many economic assumptions, such as the 5% income tax imposed by the Athenian tyrant Peisistratus, are based entirely on anecdotal testimony. Given the limits of ancient technology (particularly the limitations of land transport for the shipment of bulk commodities such as timber, metals, or stone), the development of so many urban populations, built environments, agricultural landscapes, and fire-based production centers simultaneously across the globe may conceivably have resulted in resource depletion. If this is true, however, it left little trace in the historical literature.  Apart from the generally apocalyptic tone of “gloom and doom” recorded in most societies at the end of antiquity, there is little in the eye-witness accounts that points directly to evidence of an ecological disaster, certainly nothing that compares with the current debate about the impact of global warming, the depletion of fossil fuels, and the rise in population in the modern global world. The ecological argument may ultimately bear weight, in other words, but the fact remains that the research is in its infancy and the verdict is out.


Despite the inadequacy of empirical data, the ecological argument remains valuable for another reason altogether, namely, its ability to place human societies within the proper order of the natural ecosystem. This has become increasingly evident to specialists who focus on sustainability and human growth. This argument proceeds as follows. Humans as organisms are actors engaged in a wider ecosystem that supports numerous additional species of plants and animals. Like humans all the species that we see around us, including many that have disappeared, successfully survived the irreversible winnowing process of the Pleistocene Era. Unlike most other species, humans emerged from the Ice Age with no natural predators. From a natural perspective, that is, humans as organisms have the capacity to expand to the limits of their ecological niche. During the Classical era that niche was restricted by and large to the river valleys and temperate zones of the northern hemisphere, extending along an east west trending line from the Mediterranean to the shores of Asia. Within the limits of ancient technology, this was the region most capable of sustaining large urban populations, and centuries of significant technological innovation were required before any similarly sized populations could emerge in other parts of the globe. If we look at patterns of growth and collapse in ancient world systems, what scientists refer to as sociopolitical organizations, what we find is a rhythm, an ebb and flow in human population growth, that closely resembles the adaptive cycles that are commonplace in the wider ecosystem.


Natural Adaptive Cycles and the Ancient World Systems

The theory of natural adaptive cycles is best explained in the book Resilience Thinking, Sustaining Ecosystems and People in a Changing World (Washington Island Press 2006), by B. Walker and D. Salt. Their concepts are actually modeled on the teachings of the noted economist, Joseph Schumpeter, who studied recurring patterns in economic growth (the “Schumpeter curve”) and coined the expression “creative destruction.” Its chief premise is that change is constant and that for any system within the natural order to survive it must be resilient to change. Human systems, accordingly, must be sufficiently flexible and adaptive to withstand shocks and disturbances that occur naturally the wider ecosystem. As we noted in the introduction, natural adaptive cycles progress through four visible states: rapid growth, conservation, release, and reorganization. An examination of each phase may be useful. As Walker and Salt explain, early in the cycle a human or social-ecological system is engaged in a period of rapid growth as people exploit new opportunities and available resources. These actors make use of available resources to exploit every possible ecological or social niche. The system’s components are weakly interconnected and the internal state is weakly regulated. In the human historical experience the rapid growth cycle can be identified with the process of state formation and the growth of early empires.


Eventually, as resources become harnessed, the opportunities for rapid growth diminish and a transition occurs to the phase of conservation. During the conservation phase energy gets stored and materials slowly accumulate. Connections between the actors increase and elite elements seek to secure their position at the top of society. By the end of the growth phase few if any new actors are able to establish themselves. The competitive edge shifts from opportunists to specialists who reduce the impact of variability through their own mutually reinforcing relationships. Conservation strategists tend live longer and are more conservative and efficient in their use of resources. As strong competitors they operate across large spatial scales and over longer time periods. In growing economies this often translates into a move toward greater specialization and greater efficiency through large economies of scale. As the system’s components become more strongly interconnected its internal state becomes more strongly regulated. Conservation results in the establishment of a status quo, in other words, that excludes new entrants or new ways of doing things. Capital (human and material) continues to grow but it is increasingly harder to mobilize. Efficiency and connectedness increase but the growth rate tends to slow. The cost of efficiency is a loss in flexibility; the system becomes increasingly rigid and resilience declines. Different ways of performing the same function (redundancy) are eliminated in favor of doing something in the one most efficient way. Increasing dependence on existing structures and processes renders the system increasingly vulnerable to disturbance. In the human historical experience the conservation phase is typically the phase of large urban civilizations.


The transition from the conservation phase to the release phase can happen in a heartbeat. Moreover, the longer the conservation phase persists the smaller the shock required to precipitate collapse. A disturbance that exceeds the system’s resilience breaks apart its web of reinforcing interactions and the system comes undone. Resources that were tightly bound are now released as connections break and regulatory controls weaken. The loss of structure continues as linkages are broken and natural, social, and economic capital leaks out of the system. Although the release phase tends to be brief, the dynamics are typically chaotic. However, the destruction that ensues exhibits a creative energy, something Karl Schumpeter described as “creative destruction.” Tightly bound capital is released and becomes a source for reorganization and renewal. In the chaotic release phase uncertainty rules and all options are open. Typically, the release phase leads quickly into a phase of reorganization and renewal. In economic or social systems new groups may appear and seize control of an organization. The reorganization phase begins to sort out the players and to constrain the dynamic. The end of the reorganization phase and the beginning of the new rapid growth phase is marked by the appearance of a new attractor, a new identity. This phase of the cycle may lead to a simple repetition of the previous cycle, the initiation of a novel pattern of accumulation, or it may precipitate a collapse into a degraded state. Usually but not always, a system passes through an adaptive cycle by moving through the four phases in the order described here but not necessarily in every instance.


The similarities between the phases of the natural adaptive cycle and the repeated pattern of rise and fall of ancient civilizations seem remarkably evident. In addition, the progression through natural adaptive cycles demonstrates a pattern of “diminishing marginal returns,” as described by Tainter. As we have observed repeatedly with ancient world systems, ancient empires typically progressed through phases of rapid growth, during which they conquered neighboring peoples, expropriated their wealth, and incorporated the energies and skills of their productive population to expand the empire. Imperial consolidation was typically followed by a phase of conservation where an established hierarchy attempted to maintain its supremacy by implementing greater efforts at efficiency throughout the system. Landowners adapted from subsistence forms of agriculture that were highly redundant (since each household produces most of its own necessities) to more efficient use of the landscape to produce agricultural surpluses. This enabled them to generate the kinds of products that the landscape in that region was most capable of producing as surpluses intended for distribution elsewhere. Egypt generated grain exports, the Aegean wine and oil, Italy, livestock in addition to wine and oil, and so forth. Increasing interdependency made the system more efficient and better able to sustain a larger population, but it placed the overall population at the mercy of a technologically constrained “distribution system.” This process of integration and heightened efficiency can be applied to nearly every aspect of the “seven criteria for civilization,” which, in the end, function in this textbook as signposts to the key components of developed civilizations that existed in the “conservation” phase of the cycle. Collapse of the system due to some unforeseen disturbance such as a barbarian invasion would result in the breakdown of the distribution of necessities throughout the system. At the very least, inhabitants of surplus producing landscapes would have to revert back to subsistence practices in order to survive, some more successfully than others (redundancy). Marginal landscapes that had gradually been colonized and made productive would be abandoned (release). The ruling hierarchy would be defeated or collapse on its own. Stored wealth, such as the Persian treasure conquered by Alexander the Great at Persepolis and Ecbatana, would be released. Having eliminated the hierarchy and plundered its wealth, newcomers such as migrating nomads became the “new actors” who reorganized the system and initiated a renewed cycle of growth.


This model not only captures the essence of world systems growth (conquest), consolidation (conservation), retrenchment (“diminishing marginal returns” -- consider, for example, the Roman adaptation to the military emergency of the third century AD with the doubling and quadrupling of the size of the army), and collapse (barbarian invasions), but it even goes so far as to identify the chronological patterns visible in the “life cycle” of the ancient world system. A characteristic pattern in the natural adaptive cycle is that of “fore loop” and “back loop.” A fore loop is characterized by the accumulation of capital stability and conservation; a mode that is essential for the system’s well being to increase. It represents the part of the cycle in which levels of human well being can be raised. The back loop is characterized by uncertainty, novelty, and experimentation. The back loop has the greatest potential for the initiation of destructive or creative change in the system. It is also the phase in which human actions can have the biggest impact. However, capital cannot accumulate during periods of release and renewal. What is most significant about the pattern of fore loop and back loop in natural adaptive cycles is that the fore loop tends to be represented by slow predictable patterns of growth. Resources slowly become locked up in the most efficient way of doing things. Back loops on the other hand are represented by sudden collapse and release of resources. Similarly, civilizations seem to grow to sustainable levels over long periods of time, but they seem to collapse relatively quickly. As a result, urban societies reaching the end of the conservation phase invariably try to sustain themselves, their built environments, and their human material capital, at growing levels. Conversely, since the back loop is a time of uncertainty and great change, that is, a time in which wealth and elite status undergo significant and unpredictable rearrangement, it is feared and resisted by those in power. Ruling hierarchies may attempt to stem the tide by embarking on even greater measures of efficiency or greater levels of interconnectivity to the same effect. This phenomenon closely resembles the pattern of the Late Roman Empire, for example. Inevitably, the system itself becomes increasingly less resistant to shocks and disturbances that are a part of the natural order. Eventually it progresses into the back loop phase of release and reorganization. What is most intriguing about this process is that the longer ruling hierarchies in a given system delay the back loop, the greater the likely losses that will occur in the release phase. In addition, in these circumstances the subsequent forward loop of the cycle is likely to develop in a far different manne than it might have had the reorganization phase occurred earlier. In other words, if the collapse phase is bigger, it is likely that the start of the next cycle will witness a period of significantly reduced human well being. In addition, the greater and the more widespread the level of “creative destruction” that occurs, the less likelihood there will be of preserved “memory” necessary to reconstitute a system similar to the last. Widespread clear-cutting of forests can result in an absence of pollen necessary to regenerate the same kind of forest. In a similar manner, the destruction of the recursive institutions of a world system (the educational systems necessary to generate leadership and skilled artisans) can result in the development of new culture without the necessary “memory” to reconstitute the skilled infrastructure of the one that preceded. These observations help to explain the dramatic, sudden collapse of the Roman Mediterranean world system, as opposed to the repeated but seemingly less chaotic transitions in Chinese hierarchies at the end of the Classical era, not to mention the rapid pace at which the Tang Dynasty was able to reassemble the constituent parts of the Chinese world system.


In short, the application of the model of adaptive cycles to social-ecological systems such as human societies suggests that large urban civilizations are not sustainable indefinitely. Attempts to make human systems more efficient cannot be done in isolation from the rest of the natural order. When societies attempt to hold one part of the ecological system constant (e.g., agriculture), the rest of the system adapts around these changes and typically loses resilience in the process. There is always “feedback,” in other words. Human societies have demonstrated repeatedly that they can hold parts of a system in a static condition temporarily; however, the broader system is largely beyond human control. Conceptually, resilience theory and the model of adaptive cycles of social-ecological systems offer a useful construct for the problem of world systems emergence and collapse. It posits that the rise and fall of civilizations are a natural process and that attempts to resist the cycles in some ways only makes matters worse, particularly insofar as the well being of the inhabitants of the social-ecological system is concerned. Once we accept this explanation for the collapse of the Classical world systems, the only remaining question concerns the simultaneous timing of the “release” or collapse phase of these world systems at the end of the second century AD. Why did world systems at distant ends of the Eurasian world experience societal collapse simultaneously, particularly when the trade occurring between them is generally dismissed as “low volume, high value” trade in nonessential prestige goods with marginal impact on the well being of the inhabitants? At this point it might be useful to recall the point made at the end of the Bronze Age collapse phase and to remind ourselves that a similar pattern of simultaneous collapse was visible not only in the eastern Mediterranean, but as far away as Denmark, Central Asia (1300-1200 BC), and China (the Shang collapse occurring ca. 1028 BC). Given the limited scale of the urban economies at that time, the lack of interdependency between these societies, and the vast distances trade in material goods had to cover, so similar a pattern at the close of the Bronze Age seems even more difficult to explain. The anwer to this would appear to lie with the principal of globalism, in this instance, the likely economic integration of distant world systems. Conceivably, long distance trade in prestige goods not only functioned as an accelerator of local production, as we noted earlier, but also as a synchronizer to the rhythm of the adaptive cycles of interconnected world systems. Since the availability of prestige goods stimulated increased productivity at distant ends of the trading network, the respective economies of distant civilizations, once linked into the global system, would have grown at approximately the same time and pace. Typically, interregional trading networks took hold at the “consolidation phase” of the rapid growth cycle in state formation, as hierarchies began to accumulate sufficient wealth to engage in the risk-prone and costly process of engaging long distance trade. In other words, they were already well advanced toward the phase of conservation. As an accelerating factor, interregional trade enabled distant trading partners to stimulate enhanced productivity through greater efficiencies. The interregional trading networks thus helped to extend the conservation phase of the cycle, to sustain the status quo, and to improve the general well being of the inhabitants of the respective world systems. In one respect, globalism extended the process of greater efficiency across a wider portion of the globe, by drawing into the cycle inhabitants of distant world systems even though they had no genuine knowledge of one another. But it did so by synchronizing the economic cycles of these populations and ultimately by putting them on the same trajectory for “release” (collapse). This would explain why all three world systems appeared to collapse simultaneously, around 200 AD for the first wave, and then again around 500 AD in Rome and India.


The theory of natural adaptive cycles thus appears to explain the recurring pattern of “rise and fall” in ancient world civilizations. This theory has the advantage of placing human societies in their proper place in the ecological order. Humans are undeniably the “exceptional” species of the current phase of ecological existence, unsurpassed in their ability to utilize the resources furnished by the natural landscape. However, they remain biological organisms nonetheless and depend like all others on the sustainability of the ecological systems that surround them. Humans have demonstrated remarkable resilience at adapting “strategies for survival” to fit any number of environmental conditions and at exploiting and inventing ways to utilize natural resources to enhance well being for large populations. The most important skill of human societies remains the “recursive process,” that enables humans to preserve the “memory” of past and present skills and insights so that they can recognize where they stand in the adaptive cycle and proceed accordingly. However much humans and their societies work to construct systems that enable human populations to grow beyond the earth’s carrying capacity, the “fore loop / back loop” pattern of natural adaptive cycles will inevitably restore a sense of balance.





J. D. Hughes, Ecology in ancient civilizations, Albuquerque 1975


J. A. Tainter, The Collapse of Complex Societies, Cambridge, 1988


B. Walker, D. Salt. Resilience Thinking, Sustaining Ecosystems and People in a Changing World, Washington Island Press 2006