Roma Invicta, Africa Felix

Provincial Economic Relations in Late Antiquity

By: Josiah Bloss

Traditionally the Roman Empire of late Antiquity has been characterized as a sick, decaying society, with an inflexible and outdated socio-economic organization. Currency was sidelined by both the private sector and the government due to its increasing instability and unreliability as the imperial government required increasing amounts of resources to support itself. Long-distance trade declined, and wars of all sizes became more common. According to this telling, the Empire’s decaying and dysfunctional social structures were discarded and replaced after its undignified fall to Germanic invaders. This, to some at least, represented the death of the old, imperial, world, and the birth of a new, feudal world.

Of course, this interpretation tells, generously estimated, only half the story. While the West fell, the East lived on, collecting taxes, raising legions, and ruling provinces just as it had done for hundreds of years. This part of the empire had a few more centuries, at least, until it would be shaken as thoroughly as the West was. Even then, there was direct continuity in the geopolitical vessel of romanitas of the fifth century with that of the tenth and eleventh, which cannot be said for the west. Further, the simple narrative of Germanic invasion and occupation is at best half-true for the equally important southern part of the empire. While Africa, Hispania, and Numidia were occupied by Vandals and Ostrogoths, they also were reconquered in part by the still-robust Eastern Empire. Now thornier questions reveal themselves: was the economy of the southern empire radically changed by its Germanic occupation? Or was the “Roman” economic order preserved through the “Vandal parenthesis?” Complicating the picture more, after this part of the empire was restored, it was then permanently lost to Islam in the eighth century. What do these departures mean for the economic order in these provinces? Was the economy more like the East, or the West? Was it preserved or devastated by the barbaric German invaders? More obscurely, did the conquest of the province by Islam radically unbalance the economy, or was the effect the armies had superficial?

Now we reach the point where referencing the northwestern experience is of limited use. While the northwestern empire was being consolidated by the Merovingians, forming the precursor to what would become known as the empire of Charlemagne, itself the precursor to medieval Europe, Africa was reintegrated by the empire, ending its pretensions to independence. Later, while the northern empire was divided by Germanic Anglo-Saxons in Britain and Christian Franks on the mainland, the religious contest southern empire was at first between two sects of Christianity, and then between Christianity and Islam. These differences might be even more pronounced when considering the economies of either region. The Mediterranean-facing south (for this purpose including the Rhone valley and southern Gaul) made for an attractive investment to roman aristocrats and was the backbone of the annona that supported the vaunted urban centers of the Roman world.¹ On the other side of the Alps, the far northern climate was not as suited to the customary crops of the Roman world, and production techniques and patterns change the further north one looks.² Bearing these facts in mind, the southern provinces of the empire should be considered on their own terms.

Unfortunately, this is often more difficult than it appears. Once the historiography is untangled from northern models, it still suffers from the issue of colonialism. Much of African historiography was written in the throes of French decolonization, and sometimes this situation is reflected in the body of work. But the cause of settled agriculture in Africa has been lionized for centuries by different groups at different times, the French are only the latest.³ The ebb and flow of settled agriculture, especially in the context of villa estates, is not necessarily unique to African history, but its extent and volatility is remarkable.

These trends are highly visible in the archeological record and are especially provocative in their periodization. One settlement in Africa Proconsularis, Thugga (alternatively Dougga), is especially interesting. Thugga was excavated by Dr. Mariette De Vos, who presented her diverse findings in several dimensions. On a basic level, her findings document the steady intensification of settlement and imperial investment in the area until the seventh century, when the settlement was suddenly abandoned.⁴ The timing of this abandonment, and its completeness, is provocative because it aligns almost exactly with the conquest of the region by Islamic armies. The natural implication of this is that the settlement must have been directly affected, sacked maybe, during the conflict.

This interpretation poses several problems. First of all, Thugga is inland, and its most accessible route to the coast was the Bagradas (modern Medjerda) river, which still takes a few days of overland travel to reach from the city, before factoring in travel time along the river.⁵ While this isolation does not make military devastation impossible, it certainly would insulate the community from attack. A substantial military force coming from the east would take a few days to reach the settlement, giving locals time to bury their valuables and flee into the hilly, rugged country that surrounds the region.⁵Besides this, the community’s wealth was in low-value high-weight agricultural goods, not in high-value trade goods, as could be found in commercial centers like Carthage. The effort of reaching the settlement, rounding up all the items, and then lugging them back to the coast where they could be sold off was probably not worth the expense. The second hurdle this interpretation faces is, as Dr. Wickham observes, “Cities get burnt by enemies, or hit by earthquakes, throughout history, but if they are not rebuilt afterwards it is because of longer-term causes than one disaster encompasses on its own.”⁶ Meaning, applied to Thugga, that if it had a healthy economy prior to its postulated sacking, it would have been rebuilt instead of abandoned. Therefore, while ostensibly plausible, the theory that Thugga had been sacked must be viewed with a degree of scrutiny.

Since the question of why Thugga was abandoned remains open, it would be folly to not examine Dr. De Vos’ findings further. Understanding the particulars of Thugga requires a review of its wider context. Generally speaking, the economy of the Late Roman Empire was defined by a remarkable degree of economic integration.⁷ Imperial bureaucracy could draw resources and manpower not just to the capitals, but also to the distant frontiers. Roman money, despite its bad reputation, circulated regularly and widely, which was no mean feat for an empire that stretched from Asturias to Antioch.⁷ the most important manifestation of this integration in Africa, where Thugga was situated, was the annona, a massive fleet that sailed from Africa to Rome carrying grain and olive oil.⁶ Africa could support this because it was a naturally rich province, producing olive oil and grain in great quantity for export to the rest of the Mediterranean, but particularly to Italy.⁸ Africa was also relatively secure in the empire, far removed from the major military frontiers of the north and east.⁶ Africa’s accessibility by sea, wealth, and security would have made African estates attractive investments for Italian aristocrats.

So far, Thugga fits almost exactly into this stereotypical characterization of Africa’s economy. Like other African towns, it is highly specialized for agricultural export, maybe unusually so for a community so isolated. As noted above, Thugga is set in the Bagradas river valley, which produced a significant enough grain surplus to contribute to the annona, subsidizing the Roman food supply.⁹ From its position, we can tell that Thugga was an agricultural community among agricultural communities. It can also be said with a fair amount of certainty which of the two African staples were more emphasized in Thugga. That is, from the extensive remains of presses in the area, we can infer that olive cultivation and their conversion into oil was a key pillar of Thugga’s economy. there was probably not at-scale wine production in this case. In the late first century C.E. Domitian issued an edict banning the planting of new vineyards in the provinces, and it does not appear that African wine imports were a major component of Roman civic supply.¹⁰ Instead, it seems like most of the wine supply came to Rome via the wider Italian peninsula, and imported most of its olive oil, first from Spain, and then, later, from Africa.

Excavations in the area (371 km²) around Thugga have uncovered evidence of over three hundred presses.¹¹ This is especially impressive because Dr. De Vos compares this to a similar survey in Algeria, covering an area three and a half times the area, where only about five hundred presses were found (if presses were present in the same density as in Thugga, there would have been over a thousand). This concentration of presses seems unusual, not only because it represents an acute local focus on the cultivation of olive oil, but also because of Thugga’s peculiar, inland situation, as discussed above. Thugga is relatively inaccessible from bodies of water that were the preferred routes for ancient transport networks, commercial and otherwise.

This poses an issue, not only because the number of presses alone indicates intensive cultivation and prosperity in the region, which would imply some kind of access to water, but also this concentration of productive machinery should also be taken as evidence of capital investment and market-orientation in the community.¹² Put less technically, these presses were not installed to meet local demand (for there are far too many to supply a town of this size), they were explicitly intended for making money in markets abroad. These markets would have been quite difficult to access without water-borne shipping, but overland transport was not impossible. In the region around Thugga double and triple presses were much more common (five triple-sized presses around Thugga, compared to just one in Algeria). The installation of such presses was expensive, even in an area where some of the materials could be locally sourced from quarries, like in Thugga.¹³ This indicates not simply market orientation, but wealth stratification and investment in the community from elites. This analysis is reinforced by the presence of a large imperial estate in the survey area, in addition to numerous large private estates owned by non-Africans.¹⁴

The actual organization of these estates remains unclear. While the stereotypical Roman estate is the slave-supported latifundia, recent arguments have been made in favor of a more decentralized sharecropping model, especially as the imperial conquests slow and cease being a reliable source of enslaved labor. A bipartite system has been identified in other places in the empire as far back as the fifth century, particularly in Egypt.¹⁵ However, De Vos identifies a sharecropping system in place in Thugga on the basis of epigraphic and comparative evidence.¹⁶ At any rate, it seems as though many of the inhabitants of Thugga were unfree or at least semi-free, with certain contractual obligations to an absentee landlord in exchange for cultivation and inhabitance on the land.¹⁶It seems likely that Thugga at least had been under both of these models at one point or another, but a precise date is hard to discern. By the sixth century, at the latest, the bipartite model would have been introduced to Thugga by the eastern empire (where this pattern of landholding was well established), provided it did not already exist before the Vandal conquest or wasn’t introduced by the Vandals themselves.¹⁷

The capital invested in the building of the presses did not exclusively come from local cultivators who made it big on the olive oil trade or annona grain shipments. A significant portion of investment was a deliberate move from outsiders to develop the profitability of their estates in the region. Given the remote location, this degree of investment and development is surprising. Thugga was seventeen miles from the Medjerda river, which would pose problems for both travel and shipping, but the issues with high-volume shipping would be much more pronounced.¹⁸ Once the goods reached the river, they would have to be loaded onto barges, which would necessarily follow the slower course of the river, reaching the coast near Utica, where the goods might have been sold off to seagoing merchants. The goods might have also followed a longer land route toward Carthage.

Overland travel was significantly more expensive in both terms of labor and capital, not to mention slower than riverine or other forms of water-borne shipping, making for a considerable obstacle for the Thuggan economy. Moreover, shipping facilities at Carthage were much more advanced and may have been preferred by the imperial administration on that account, so the overland route might have been more prominent.¹⁹ While the differing obstacles in modes of transportation might seem marginal, when we consider the sheer quantity of olive oil produced by the community, the complexity of either of these arrangements becomes much more staggering. Dr. de Vos calculates that the community would have to make ninety-seven caravan trips with over fifty animals a year, at minimum, to transport olive oil alone.²⁰ When you recombine the third of olive oil produced that would have to be paid in taxes, grain exported (earmarked for the annona and otherwise), and other goods the region produced, like stone and pottery, it is feasible there would have been significant caravan traffic in the city nearly every day of the year. This traffic may also have intensified depending on the season, with larger caravans of hundreds of animals coming and going at certain times of the year, such as during the olive pressing season or during the wheat harvest, rather than being spread out.

Another important seasonal element to traffic was Mediterranean shipping. A sense of seasonal urgency probably carried this traffic from shipping centers and imperial warehouses. The sea was considered closed to all traffic from early November to early March, and the annona in particular did not sail from the middle of October until the middle of April.²¹

This seasonality also interplays with the harvest season for olive oil, which in most circumstances went from November to January, in the off-season of the annona.²² Working with De Vos calculations of the olive oil crop left over after tax, and for the sake of convenience, supposing the entire crop had to reach a coastal center by the beginning of the sailing season, we might conclude that the caravans would have had to make their first trips as early as the first of January. This would have taken five caravans of fifty dromedaries working constantly but would have had the entire crop in coastal warehouses by the first week of April, and, providing that it took a few days to load the ships, just in time for the first voyage of the annona. Of course, this calculation does not include the portion of olive oil subtracted by taxes, or any other products (notably wheat) which also would have been carried to the coast.

De Vos’s calculation of the number of Caravans assumes that exclusively dromedaries, instead of mules, are used for transport, which can carry more goods. If mules were used in addition to dromedaries, then the number of animals or the number of trips would have to increase. Additionally, once the third of imperial taxation is factored in, the quantity of goods and this number of trips increases dramatically. There is room for a little fuzz around these dates to accommodate these factors. It is likely that not all the oil had to be in port by the start of the shipping season, and caravans might have started coming into the settlement as early as the first oil was pressed in November.

The region produced far too much olive oil to be consumed locally, it is certain to have been sold off elsewhere. The two likeliest shipping routes, one down the Bagradas to Utica, and the other overland to Carthage have already been described above, but it is unclear which of these was more prominent in the case of Thugga. If the province was left to its own devices, Utica seems advantaged due to its convenient location on the Bagradas estuary, but Carthage’s large local population, robust shipping facilities, and large native fleet (due to the annona) probably made it extremely competitive, despite the difficulties with overland transportation. Some archeological evidence strongly suggests that water transport was prominent and very lucrative.²³ But there is no conclusive evidence that olive oil from the surveyed region ended up in Italian markets.²³

Because shipping was so complex, the presence and maintenance of infrastructure and transport networks were essential to the success and prosperity of the community. If there was no transport network, the community would be cut off from transporting and selling their lucrative goods that kept them afloat. During most of the Imperial period, locals could be reasonably secure in their connection to the Mediterranean, there was a legion nearby, and the southern frontiers of the empire were not very active anyways.²⁴ While the frontier was not active, soldiers in this area were not idle. The local military was heavily involved in building and maintaining the roads that ran through the region around Thugga.²⁵

Perhaps the province was too confident in its security because even when the Vandals were moving from the more remote Spanish peninsula into north Africa, the imperial political apparatus did not react until it was too late.²⁶ What does what we know about Thugga tell us so far? We can tell that Thugga was a rich exporter of agricultural goods, which we know to be typical for the region. We can also tell that these agricultural goods were produced by villas, a good number of which were owned by distant investors, including the emperor (until the province was lost). We can also tell that this community was connected to the Mediterranean by a robust logistical network that was maintained by the imperial government. This logistical network was probably complex and bridged several credible geographic barriers. The export of agricultural goods, which seems by far the most important economic pillar of the region, was contingent on this connection. If this connection was severed, surely the community would surely suffer.

All of these connections, the annona, the transport of goods from Thugga to the coast along imperial transport networks, the maintenance of the transport networks by the imperial administration, the investment in the area by the elites in the form of estates, and the continued administration of these estates should be described as “Relations of Production” speaking in roughly Marxist terms. Relations of production can be described abstractly as the rules of an economy, the current and the water in which economic agents (productive forces, namely labor and capital) swim. When relations of production are taken with forces of production, they become modes of production, which would be analogous in abstract to a wider oceanic biosphere.

Many of the most important relations of production affecting Thugga have already been described. Thugga’s reliance on its connections to the Mediterranean economy has been demonstrated; Both the microeconomic imperial investments in infrastructure, as well as imperial macroeconomic policy of annona subsidies to Italy, contributed to Thugga’s prosperity, and their effects were even more pronounced in compound. Further, nothing suggests that Thugga’s experience was atypical, except perhaps the degree to which these relations of production have defined the economy, and how prosperous they made the community. On a provincial level, these policies made Africa easily one of the most important provinces of the empire, and probably the second most agriculturally productive, behind only Egypt. Surpluses, taxes, and consumer goods flowed into Italy from Africa, propping up the perennially dysfunctional western empire.²⁷ The Bagradas valley was certainly an important component of this prosperity, a secure and fertile hinterland providing a source of export goods and sustenance to the larger coastal cities. These coastal cities (especially Carthage) being economically integral to the Mediterranean, not to mention the entire empire, the risk of disconnection, and hence economic collapse, seems very slim. The existence and importance of this connection was even recognized by contemporaries, who summed up the economic relationship between Rome and Carthage (and by extension Rome and Africa) ominously: “INVICTA ROMA FELIX CARTHAGO” Translated by Dr. Richard Miles as “As long as Rome remains unconquered, Carthage remains prosperous”.²⁸

However, we already know that Africa was severed from the empire, at least in some ways. In the early decades of the fifth century, the province was occupied by the Vandals, beginning, in Dr. Wickham’s view, the end of the Western Empire.²⁹ Wickham pinpoints this moment as the watershed moment in Imperial fortunes precisely because of how codependent Italy and Africa were. Further, Wickham predicts that a break from the empire would affect Africa severely.³⁰ However, curiously, this period actually marks a peak in the inhabitation of Thugga itself.³¹ If Thugga was truly dependent on relations of production fostered by the empire, how could this be? So far it seems that imperial administration, investment, and shipping provided the bedrock backbone of the Thuggan economy! It seems perfectly clear that Thugga’s economy, while productive, was not self-supporting. The fate of the annona after the vandal occupation is unclear; African goods continued to reach mainland Europe, but the exact method is obscure.³² Some historians have seen the annona shipments continuing under the vandals via treaty with the Western Roman rump, while others have suspected that shipping to Italy continued on a private basis.

The latter situation seems to be more likely to cause the trends of rising inhabitation that have been observed. In the short term, the uptake of shipping by private individuals under the supervision of the Vandals would have allowed greater flexibility in where the African products, namely olive oil, could be sold. Additionally, the annona was a subsidized state service, which kept prices down.³³ Further, it has been argued that the state was a major buyer of agricultural goods, especially in rural areas removed from trade networks.³⁴ While Thugga is not so far removed from trade networks to make state patronage necessary, the horrea warehouse along the Bagradas twenty-five miles north of Thugga seems a likely candidate if such transactions did take place.³⁵

Without the subsidized price controls of the annona and the spike in the demands of a wider market, the price olive oil could be sold at, logically, would have jumped significantly. This means more money could be funneled back into the community, which might make settlement in the region more attractive. In addition to this fresh influx of capital, the Vandal occupation of Africa would have also severed any obligations to absentee landlords, who certainly owned a significant segment of the land in Thugga. The most prominent of this number would be the emperor himself, but other Italian families certainly lost their local estates.36 Freed of obligations to distant but exacting landlords, some of the inhabitants could have escaped rent for a few years at least until new obligations were imposed by the new Vandal administration. Another prominent form of relief that is identified in the realm of taxation. Ostensibly, the Vandals continued to collect taxes, but the main items on the expense column of Roman administration had always been annona and the Army. Once these expenses were no longer necessary, there was a massive budgetary imbalance, and as Wickham argues, a massive accumulation of wealth on the part of the Vandal kings. Following this accumulation of wealth, the urgency of maintaining tax records disappeared, and it appears that tax collection slipped.³⁷

If taxation in Thugga indeed slipped, as it appears to have done in the rest of the province, then that would again mean that a larger share of each year’s crop could be marketed at competitive rates, instead of taken as tax. Again, more money remains in the hands of the local cultivators, meaning more could be reinvested into production facilities and the community at large. It’s also important to keep in mind that while it appears that the Vandals ruled with a relatively light hand, compared to the Imperial administration that came before, that hand could not have been too light. The basic infrastructure outlined above that enabled Thugga’s connection to the Mediterranean must have been maintained throughout the period, and the disorder that came with the administrative transition also must have been relatively minimal, because any interruption in trade links that lasted too long would have done more damage to the market-oriented agricultural economy than any tax relief could repair.

Despite the ostensible political estrangement the province underwent, economically African goods, the same goods that were produced in the province before the Vandals, continued to reach the remainder of the formal imperial territories. The physical infrastructure established by the imperial government was maintained by their “Germanic” successors, and while the more prominent characteristics of the Imperial administration faded away, such as absenteeism and the annona, it has been stressed that the Vandal state followed the Roman example closely in many other ways.38 Arguably, the annona continued on in the form of private enterprise, and inarguably in many cases, Vandal aristocrats replaced Roman aristocrats if there was any replacement at all.38 In short, the economic relations of production that were established by the Roman empire long ago continued, and the kind of economy fostered by the presence of the empire in Africa, focused on export to the Mediterranean core, remained in place. This evidence of overwhelming continuity in relations of production I would argue constitute evidence for continuity in the mode of production. That is, there was not a significant economic transition taking place in Africa on account of the Vandals. When discussing economic changes during and immediately after the Vandal hegemony, it seems more helpful to view the changes as consequences of the niceties of statecraft rather than sea changes.

While this situation goes against the grain speaking historiographically, from the Vandal point of view it is extremely straightforward.³⁹ Many historians, even Dr. Wickham, have concluded that in most places the mode (Wickham now maintains it was a sub-mode) of production that had existed under the Roman empire died during the tumult of the fifth century when one province after the other was occupied by Invaders, who landed their armies and abandoned the state-tax structure.⁴⁰ However, by any objective measure, the regional economy of Africa was working quite admirably under the system of the Imperial sub-mode, especially under the considerable burdens put upon it by supporting the Western empire. Why would the Vandals change the fundamental logic of this economy intentionally? By cutting Africa loose from the wider burdens of empire, but still maintaining the advantages that Mediterranean integration brought, it was possible to get very, very rich.

Furthermore, there is the fact that the province was reconquered by the Eastern empire one-hundred years after it was lost, which could threaten the prosperity built up in the community under the Vandal’s program of privatization. Since the Eastern Roman empire had not yet undergone its’ traumatic eighth century and was ostensibly still operating under the economic conditions that it had been for the last few centuries, does this constitute a regional flip-flopping between modes or sub-modes? If so, shouldn’t this cause at least a minor degree of economic disorder that might be observed in the archeological record? In Thugga, at least, there does not seem to have been much change. In Thugga, inhabitation seems to have slipped a little, but the community seems to have remained quite prosperous.⁴¹ While the slight erosion stands out compared to the remarkable growth of previous centuries, there are some other factors that should be taken into account. Most importantly, the new administration was dealing with the very same pressures of the empire that the Vandals had helped Africa escape. The Eastern empire could not afford to let taxation slip, or idly let those taxes accumulate in a state treasury as the Vandals did.⁴² The Eastern empire, after the Justinianic reconquest, reinstated the funneling of state wealth away from the province and into the mouths of Imperial cities and soldiers.⁴³

The reimposition of this policy meant that Thugga could no longer enjoy the benefits of the relatively inattentive Vandal state. This inattentiveness benefitted Thugga in some ways, in the lighter taxation and laissez-faire export policy explored above, these policies had serious implications for the remainder of the formerly Imperial economies. Without Africa providing economic support for the other provinces the economies of those provinces quickly fell into disrepair.⁴⁴ Italy, in particular, was reliant on Africa for economic support, after centuries of close, subsidized, trade links.44 The hike in prices for grain and olive oil that brought in fresh wealth to Thugga would have ruined poor urban families in Italy, and any urban governments beholden to care for them. Rome, in particular, began to shrink dramatically.45 While at first this did not have major implications for the economy of Thugga, once the eastern empire reestablished control of the territory, and attempted to restore the old order of things, economic reality began to set in.

The temporary boost that the privatization of the annona gave to the regional economy of Africa paled in comparison to the economic damage it wreaked on the rest of the empire. The economy of the entire empire up to this point, especially the economy of Africa, was oriented towards providing Italy with cheap goods. Without Italy able to serve as the focus of this economy, dysfunction resulted. This dysfunction predictably manifested in Italy first, but without a healthy Italian market for African goods, the African economy would suffer too. Still, though, the kind of economy that existed in Africa, or in Italy for that matter, has not changed.⁴⁶ The focus of Africa’s economy remained the export of agricultural goods to the Mediterranean, and this relationship continued to be maintained by the Vandals, and then even more aggressively by the Eastern Roman state.

The invasion of the Vandals removed the economic imperative that was moving African wealth out of the province, and instead, that wealth began to accumulate in the Vandal treasury. This shift is not insignificant; however, it could be argued that the accumulation of economic resources in the Vandal holdings was simply previously unrealized economic power that had been siphoned off by the Roman state. Africa had been central to the Imperial economy for centuries, only more so since African olive oil displaced Spanish olive oil in the third century.⁴⁷ The economic significance of Africa was inherent to the relations of production that defined the Economy of this period; we can tell this precisely because when these relations of production are tampered with slightly, jury-rigged by the Vandals to support a little graft, the entire system is threatened. Despite this, the system continued to operate in roughly the same character that it had as before. The annona had been revived officially, Constantinople serving as a substitute for Rome, taxes continued to be collected (albeit more acutely), and Thugga continued to produce olive oil for the market.⁴⁸

This pattern seems to hold until the province was conquered by Islam in the seventh and eighth centuries. There is considerably less to say about this period for Thugga because Thugga practically ceased to exist.⁴⁹ Why? Just a century earlier, it seemed like the Vandals had brought the African economy to a fever pitch, with Thugga itself reaching the height of its inhabitation, and probably its economic significance.⁴⁴What else could this population have been up to besides producing agricultural goods for export?⁴⁴ Even once its growth stalled out under the Byzantines, its population never shrunk smaller than it had been before the Vandals.

This sudden collapse paints a bleak picture, not just for Thugga, but for the entire provincial economy, even if sites closer to the river or coast suffered less. How could this have happened? Historians have traditionally characterized this debate as a “failure” of settled agriculture in the face of pressure from nomadic pastoralist Berbers coming in from the mountainous interior of the province.⁵⁰ According to this interpretation, while in some parts of the province, Agriculture was too established to be dislodged, in the “marginal” regions in the interior of the province, the land was turned over to waste or pastoral uses.⁵⁰ The most contested part of this characterization is the chronology, with some historians placing the first signs of failure beginning to show as early as the Roman period, and others as late as the eleventh century.

It should be insisted right away that this framing is mistaken. While the broad contours of the narrative conform with evidence, many parts do not. We can observe from our own archeological evidence in Thugga that the regional economy as it had been thriving right up to the seventh century when it collapsed nearly completely. This collapse cannot even be attributed to the Berbers. While many of the dominant families of the region were not from Africa, some were native, and archeological evidence implies that at least some nearby land was inhabited by a Berber community.⁵¹ The fact that native Africans, including Berbers, were attached to the regional agricultural export economy engenders scrutiny of the idea that the economic devastation was the result of a “decolonial” revolution against Roman hegemony.

At any rate, Thugga could hardly be described as economically “marginal”, with the degree of productivity and integration it had with the rest of the Mediterranean. We have already identified extensive links to the Mediterranean on two axes, one along the Bagradas river, and another overland to Carthage.⁵² Besides this, we have already identified estates belonging to high-profile aristocratic families, up to and including an estate belonging to the imperial house.⁵³ Thugga’s extensive participation in the African export economy and its reliance on relations of production that define the economy of the Roman empire extensively outlined above should be enough to indemnify it against any claims that it, and any places like it, were not economically critical to the very health of the imperial economy!

A more plausible narrative for the decline of Thugga involves what Wickham describes as “an involution of an agricultural export economy.”⁵⁴ The dating around the collapse of Thugga indicates the Islamic Conquest as the watershed event, which is likely, but misleading. As proven above the prosperity of Thugga was a direct result of its connections with Mediterranean markets, and Africa’s position in the Imperial economy. By the time of the province’s permanent divorce from Roman power, these markets, the Imperial economy itself, simply no longer existed. Italy, particularly Rome, had been devastated for generations by a succession of wars and plagues.⁵⁵ Trade connections with Italy were replaced by Constantinople and its armies, but by the ninth century, Constantinople had retreated on account of its own crisis.⁵⁶ Carthage was sacked, reducing its capacity as an entrepot, but even if it wasn’t, where was the export-based economy supposed to export to now? While the armies of Islam had a hand in the retreat and decline of the Byzantine Empire, the last viable market connection for the region, the resulting devastation of the African economy could hardly be devised. If anything was planned for the African economy at all, then it was probably something like how the Vandals ran the province; the same Export-focused agricultural economy that had existed before, but with an extra element of graft added on. While Arabic armies may have caused the economic collapse, their actual culpability in Africa’s devastation certainly seems to be minimal. The essence of this narrative is that the regional prosperity of Thugga was a result of an economy that no longer existed. The relations of production that defined the community, which were the very lifeblood of its economy, could no longer be supported. While it could be argued that the economic trouble in other parts of the empire caused by Vandals seizing Africa signaled the beginning of the end of this style of Mediterranean economic organization, in Africa, at least, things still went on as before.

The relations of production that characterized the Imperial sub-mode were maintained, and were relatively successful, right up until they were rendered completely obsolete. This conforms with what Wickham identifies as the “catastrophe theory of mathematicians”.⁵⁷ In this model, a system, in our case, an economy, a mode of production, can sustain itself until conditions slowly change until the system can no longer support itself. In our case, this point when Thugga could no longer support itself probably lies when making the long trip to the coastline was no longer worth it. As Mediterranean markets shrunk, and they were shrinking, this trip became less and less profitable, and eventually, they would have shrunk to the point where African wares were no longer necessary, or at least no longer profitable enough to make a special trip for.

Interestingly, the collapse of an economic system due to the breakdown of trade links with the empire is not unique to Africa. In Spain, at least a similar problem developed in the fifth century.⁵⁸ However, in that case, instead of dissolving, the former was re-occupied by groups of peasant families. It is possible that the difference in outcome in these examples comes down to simple geographic accessibility. While Thugga, with its fertile soil but the arduous overland connection to local commercial centers, the Spanish sites are more accessible from regionally significant cities and are able to serve as part of their hinterland.59 This possibility seems remote for Thugga and other cities in Africa, whose agricultural production was destined not just for local urban centers in Africa, but also for the capital city itself, neither of which was in any shape to serve as a viable market for an agricultural community that was as productive, but remote, as Thugga.

Notes

Michael McCormick, Origins of the European Economy: Communications and Commerce, AD 300-900 (Cambridge: Cambridge University Press, 2001), 97.
Annalisa Marzano, “Capital Investment and Agriculture” in The Roman Agricultural Economy, eds. Alan Bowman and Andrew Wilson, 107-141 (Oxford: Oxford University Press, 2013), 109-11, 117.
Christopher Wickham, Framing the Early Middle Ages: Europe and the Mediterranean, 400-800 (Oxford: Oxford University Press, 2006), 18-19.
Mariette de Vos, “The Rural Landscape around Thugga: Farms, Mills, and Transport” in The Roman Agricultural Economy, eds. Alan Bowman and Andrew Wilson, 143-218, (Oxford: Oxford University Press, 2013), 198.
ibid, 174, 201.
Wickham, Framing the Early Middle Ages, 13, 20.
ibid, 79-80, 74-5
M McCormick, Origins of the European Economy, 100; Wickham, Framing the Early Middle Ages, 87.
de Vos, “Rural Landscape” 146.
Annalisa Marzano, “Agricultural Production in the Hinterland of Rome: Wine and Olive Oil” in The Roman Agricultural Economy, eds. Alan Bowman and Andrew Wilson (Oxford: Oxford University Press, 2013), 85-106; Marzano, “Capital Investment and Agriculture” 94-95.
de Vos, “Rural Landscape” 154.
Marzano, “Capital Investment and Agriculture” 108-9.
de Vos, “Rural Landscape” 150.
ibid, 183-4.
Peter Sarris, Economy and Society in the Age of Justinian, (Cambridge: Cambridge University Press, 2006), 91-92.
de Vos, “Rural Landscape” 146, 184-5.
Peter Sarris, “The Origins of the Manorial Economy: New Insights from Late Antiquity.” The English historical review 119, no. 481 (2004): 282, 311.
de Vos, “Rural Landscape” 174.
The port in Carthage was circular, which allowed for a greater number of ships to be docked and loaded at the same time. This would have been highly advantageous for seasonal high-volume traffic, like the annona. M McCormick, Origins of the European Economy, 86.
de Vos, “Rural Landscape” 174.
McCormick, Origins of the European Economy, 450, 454.
Although some kinds of olives were harvested in later months, the more conventional kinds seem likely to have been preferred in Thugga so as to fit the shipping season discussed above. Marzano, “Capital Investment,” 99.
de Vos, “Rural Landscape” 207, 210.
Wickham, Framing the Early Middle Ages, 20.
de Vos, “Rural Landscape” 202.
Wickham, Framing the Early Middle Ages, 87.
ibid, 17-8, 20.
Richard Miles, “Vandal North Africa and the Fourth Punic War,” Classical Philology 112, no. 3 (2017): 398.
Wickham, Framing the Early Middle Ages, 87-88.
ibid, 80.
de Vos, “Rural Landscape” 198.
McCormick, Origins of the European Economy, 103-4.
ibid, 111.
Wickham, Framing the Early Middle Ages, 74-76.
de Vos, “Rural Landscape” 207.
ibid, 148, 183-6; Wickham, Framing the Early Middle Ages, 88.
Wickham, Framing the Early Middle Ages, 91-2.
Wickham, Framing the Early Middle Ages, 88-9, 91.
Christopher Wickham. “The Other Transition: From the Ancient World To Feudalism.” Past & Present 103, no. 1 (1984):24.
Historians have described the mode of production that existed under the Roman empire variously as the ancient mode, or the tributary mode, or as a submode of the tributary mode. Since we are only concerned with the economies that existed during and immediately after the Roman empire, it is not necessary to be too particular with terms. Henceforth, I will call the mode of production that existed under the Roman empire The Imperial submode and what comes after the medieval submode; Wickham “The other transition” 24; Wickham, Framing the Early Middle Ages, 90.
de Vos, “Rural Landscape” 163.
Wickham, Framing the Early Middle Ages, 91-2.
McCormick, Origins of the European Economy, 103-104.
Wickham, Framing the Early Middle Ages, 87, 92.
McCormick, Origins of the European Economy, 66.
For the state of tax collection in Italy, Wickham, Framing the Early Middle Ages, 86.
McCormick, Origins of the European Economy, 100.
Wickham, Framing the Early Middle Ages, 124-5.
de Vos, “Rural Landscape” 163.
Wickham, Framing the Early Middle Ages, 18-19.
de Vos, “Rural Landscape” 198.
ibid, 174.
ibid, 183-4.
Wickham, Framing the Early Middle Ages, 19. Wickham’s framing of Africa seems generally accurate, remarks about the retreat of settled agriculture notwithstanding.
ibid, 34-5.
ibid, 31.
ibid, 13.
Saul Martin Gonzalez, “From Villae to Villulae: Settlement and Social Organization in Late Antique Hispanic Countryside” in New Perspectives on Late Antiquity, ed. David Hernandez De La Fuente, 173-187 (Newcastle upon Tyne: Cambridge Scholar’s Publishing, 2011): 182.
ibid, 186.

Bibliography

De Vos, Mariette, “The Rural Landscape around Thugga: Farms, Mills, and Transport” in The Roman Agricultural Economy, eds. Alan Bowman and Andrew Wilson, Oxford: Oxford University Press, 2013, 143-218

Martin Gonzalez, Saul. “From Villae to Villulae: Settlement and Social Organization in Late Antique Hispanic Countryside” in New Perspectives In Late Antiquity, ed. David Hernandez De La Fuente, 173-87. Newcastle upon Tyne: Cambridge Scholar’s Publishing, 2011

Marzano, Annalisa, “Agricultural Production in the Hinterland of Rome: Wine and Olive Oil” in The Roman Agricultural Economy, eds. Alan Bowman and Andrew Wilson, Oxford: Oxford University Press, 2013, 85-106

Marzano, Annalisa, “Capital Investment and Agriculture” in The Roman Agricultural Economy, eds. Alan Bowman and Andrew Wilson, Oxford: Oxford University Press, 2013, 107-141

McCormick, Michael. Origins of the European Economy: Communications and Commerce, AD 300-900. Cambridge: Cambridge University Press, 2001.

Miles, Richard. “Vandal North Africa and the Fourth Punic War.” Classical Philology 112, no. 3 (2017): 384-410.

Sarris, Peter. “The Origins of the Manorial Economy: New Insights from Late Antiquity.” The English historical review 119, no. 481 (2004): 279–311.

Sarris, Peter. Economy and Society in the Age of Justinian. Cambridge: Cambridge University Press, 2006

Wickham, Chris. “The Other Transition: From the Ancient World to Feudalism.” Past & present 103, no. 1 (1984): 3–36.

Wickham, Chris. Framing the Early Middle Ages: Europe and the Mediterranean, 400-800, Oxford: Oxford University Press, 2006