Late Antique Archaeology 2010

These are some rough notes taken last week at Late Antique Archaeology conference 2010 about Local economies? Production & exchange of inland regions, that took place at King's College, London, Friday 12th to Saturday 13th March 2010. Overall, this conference was interesting, and I had a chance to meet lots of nice people working in Late Antique Archaeology. Inspiration for my PhD research was just great.


Unfortunately I missed Whittow's lecture.


Sarris talked about “Rural production dynamics: autarchy, tax and forms of exchange, seen from papyri”.

Hay and meadowlands weren't actually very important, compared to production of food and textiles (linen, flax) for supplying towns.

Guilds as seen from papyri are much more “medieval” self-organized groups of craftsmen than Roman or Byzantine top-down corporations.

Obviously, Apion's estates were not economicaly isolated.

Early Islamic Egypt is normally described as a “decapitated” society, because of the disappearance of aristocracy. But this situation left room for middle-class landowners, at the same time enabling a stronger central state, efficient collecting taxes.


Macmahon gave a declaratively explorative talk about retail structures, workshops, markets, seeing all these elements as strongly associated with urban landscapes. Their nature is such that if we're not able to find clear evidence for them, they're invisible.

Social shaping of antiquity includes economic factors: for this reason he's trying to use theoretical models to explore ancient society:

  1. <missed>
  2. business, it's not necessarily a modernistic model

Consumption is a social act, in which people do distinguish between goods that are appropriate and not.

A market is defined by a number of factors:

  1. location
  2. income
  3. demographics (age, sex, racial/ethnic mix)
  4. lifestyles (marked for example by clusters of specialised stores in urban areas)

Boundaries between markets, and between societies in general, can be defined in terms of:

  1. institutional boundaries

  2. natural boundaries

  3. spatial density / distribution of households

  4. spatial decay linked to transport costs

  5. competition

    With regard to competition, i.e. Central Place Theory doesn't take into account intra-site distributions and it's too demand-focused (places are primarily consumption sites) rather than retail-focused

The key contrast when studying local exchange is between individual small resellers and large resellers. However, reseller units are part of the urban landscape in the Roman empire, even if they change over time.

Warehousing didn't only mean storage structures, they could have been used also as auction houses. Unfortunately specialisation is poorly visible through archaeology. Literary sources are more important but the two are difficult to integrate.

A single journey to buy or sell a single product can be used as an extreme definition of specialisation. While there's a segmentation of the empire in the later period, autosufficiency in provinces is much more widespread in the early empire.

An important point is that there's apparently no development in transportation networks, that are important in shaping markets. Looking to a contemporary example, the car industry in the US is spatially concentrated in a small area, but distribution is wide. On the contrary, the production of pottery in the Roman world was very scattered. This difference can be explained in terms of different transportation networks.

Competition between retailers, producers, even the State developing productions on its own, is a social element and a major factor of economic development. Luxury goods can survive until there's a market for them.



  • volume of intra-regional trade?
  • evidence for large-scale specialised trade?
  • topography of social control?
  • role of artisans in regional exchange?

Generally speaking, pottery is an easy study object, but what about i.e. the regional distribution of metal objects?


  • regional “markets” in architecture, sculptures and sarcophagi
  • there are some productions that have to be local, like bakers (most bakeries were small-scale enterprises), butchers, goldsmiths

Large coastal cities can have highly specialised artisans (images of Corinth, Philippi, Stobi, Scythopolis, Sagalassos). Being by the coast determines the scale of production and retail. Are things at the same scale or not between coast and inland sites? E.g. textile production.


Spain experiences a collapse in oil prodution during the 3rd c. accompanied by a modest growth in garum. In the 4-5 c. there's a boom of (mostly monumental) villas.


  • is there a regional economy?
  • are villas a straighforward economic barometer?

Existing models for Spain during Late Antiquity:

  1. non-particularity of Spain

    Bowes argues that monumental villas are much more numerous in Spain, Aquitania and Britain in the 4-5 c. Particularity applies, if we consider the empire as a whole, because this phenomenon happens only in some regions <someone commented that this might just reflect a bias in field research>

  2. poverty of Spain

    Bowes sees poverty as nonsense given the amount of monumental villas and the overall richness of material culture.

  3. villa-driven economy

    Bowes thinks that the concentration of landholdings in the hands of a few is not a convincing model.

The Serpa survey showed a high density of villas <even though I couldn't get if all of them are contemporary>, there are very rich buildings like at La Olmeda, but in general the spectrum is very wide and there's a sort of regional architectural “discourse” (between either artisans or elites). Mosaics have lots of inscriptions, and personal names (representing display of the self).

Villas are found in clusters, they are not everywhere and particularly:

  • not on the coast
  • not in the Mediterranean region

Are these symptoms of a regional economy? Monumental villas are spatially concentrated, while productive sites are also on the coast. But why these local differences?

Fortifications and state investments in the N and W regions are the context for these local disparities, together with the presence of troops. There is some overlap between villas' clusters and roads <this didn't look very convincing to me>, and there are some theories about this overlap:

  1. an annona route in the context of supplying the Rhineland limes by seaborne trade in the Atlantic, and villas exploiting the annona route (e.g. there are granaries near roads in villas). Two new provinces get created (Gallaecia, Novumpopuloniae). Is Spain producing horses for the Rhineland limes ? Theodosian AE coins have a similar distribution with villas <why in the world they should not have a similar distribution? why is this relevant?> Lots of milestones account for importance of roads. There are problems with this theory, for example no Spanish amphorae in Rhineland forts.
  2. routes as arteries of competitive opportunity: State investment is selective, so there is social competition around it. Fortifications are not necessarily signs of danger, but certainly <a display> of power. There are diverse elites at work: dux, governor, curiales, landowners all competing and not just in cities (Spain is much less urban than other provinces). Villas may be agents of social competition. TSHT is concentrated in this same area.

Bowes thinks that the engine of the villa boom is not regional, but rather it's the State.

Previous models for villas as an economic indicator include:

  1. tenurial relationships and rise of colonate (Carandini)
  2. self-display

Instead, the proposed model defines villas as places for competition, not a straightforward economic indicator. Other competitive landscapes of the same kind are then concentrated bureaucracies (either military or civil) and supply regions.

Major previous models for Roman economies and State are:

  1. Finley, who refused to accept state-driven economy as “real economy”
  2. those who see taxation and coinage as an engine for economy

In Bowes's words, both seem to fail, because they assume a fully working and discrete state.


Mattingly observed that Spain has always been a huge producer of metals.

Sarris commented that generally we see an “expanding state” during Late Antiquity.


Major events in history of Sicily are the conquest of Egypt and the creation of Constantinople.

Sub-coastal sites:

  • Segesta
  • Himera
  • Platani survey


  • Milena: continuity from 2 c., increase in 4-5 c.
  • Margi valley: peak of settlements density in Late Roman period
  • Ramacca area
  • Simeto valley
  • Morgantina: peak in Late Antique period
  • Campanaio

In general, Late Antiquity in Sicily sees a widespread increase in number and size of rural sites.

Sofiana is a small agro-town founded in the Augustan age, during Late Antiquity it is still 21 hectares. Vaccaro thinks that the old theory about a satellite settlement depending on the Villa del Casale should be re-evaluated.


Mulvin presented archaeological evidence for rural settlements in Pannonia, including villas and other productive sites.


The economy of Numidia was heavily based on the presence of the army.

If one has to pay taxes in cash, it means there's a “free market” of some sort where one can actually get cash.

The pastoral element of economy was pushed away from areas now exploited with intensive agriculture.

Speaking of the chronology of sites as seen by survey, and of the apparent peak of sites in the 5 c., it's important to state again that we cannot be sure that we see all the early pottery from site surveys.

This 5 c. peak has to do with textile production, it's something that could have been done in domestic contexts.

Diana Veteranorum's position would fit in a network with cash-producing provincial towns where bureaucracy is dwelling. Horses and slaves too are an obvious production for Numidia.

In the 6 c. there's a general abandonment of sites, a phenomenon that Fentress describes with Wickham's “drop of complexity”.


Bruce Hitchner observed that during the Kasserine Survey a large number of animal enclosures were found.


Garamantes are a good example of trade and production outside the Roman empire - something that already Wheeler had recognized when facing with terra sigillata in India or Scandinavia.

The basic question he poses is whether or not there is trade in Sahara prior to the Islamic period.

Oasis' network demands regular contacts between them, allowing movements of goods, people, animals, ideas along trans-saharian routes.

The Garamantes are an urban, agricultural society that represent the emergency of a powerful polity in Central Sahara. Their location is central for movement from the Mediterranean to sub-saharian regions.

Three families of traveling goods:

  • Mediterranean imports (including Punic)

    1 c. tombs at Germa have some TSI. There's a sherd with a graffito reading “Nimir”, maybe a trader. Later tombs have a few imports, like Keay 62 amphorae and a Hayes 67 dish. Glass is a very valuable item. Some amphorae are enormous for travelling by camel!

  • Fazzan exports

    Beautiful textiles.

  • sub-Sahara imports (including e.g. gold)

    Migrations of people from these regions are well seen in graves, e.g. a woman buried following Garamantes' ritual but with characteristic objects of sub-saharian peoples.


Haeverbaeke talked about oil production in Anatolia, focussing in particular on the hinterland of Sagalassos.

In a few points, she thinks that:

  • oil production was for local consumption
  • presses weren't used just for olive oil
  • continuity of village culture is the most important factor for explaining the diffuse wealth of this region in Late Antiquity

According to Mitchell, olive is an important cultivation in the region during the Hellenistic and Roman period. Olive cultivation requires a stable political and economical condition, because of the time needed before the trees are adult. Mitchell believes the during the Late Roman period oil is instead produced for export.

Instead, palynology has shown the prevalence of mixed orchards, instead of olive monoculture. This would better fit with a production intended just for local consumption.


Bonifay's lecture started with a question: “To what extent do ceramics reveal markets?”

In Late Roman Africa, there's a completely different situation between:

  • Tunisia vs other modern states
  • coastal vs inland sites
  • quantified and stratified (mostly in the region of Carthage and Nabeul) vs poorly indagated sites

Needless to say, the former cases are respectively much better known than the latter. In inland regions, cities are mostly studied by means of excavation, but there are some large surveys.


There are lots of them on coastal sites, from Spain, Crete, Sicily and of course the majority from Africa itself.

Inland regions have some Mediterranean amphorae, but they are very rare, only a few sherds. Most amphorae are African, but from where ?

  1. continental
  2. small globular
  3. storage amphorae

Fine wares

Imports are very rare everywhere. ARS was produced in many workshops, the main source for Africa was Northern Tunisia. Tripolitania has its own RSW, but was also supplied by the continental workshop of Sidi Aïch.

Inland sites are mostly served by central Tunisian products, and almost no Northern RSW. There are lots of local and regional products, generally classified as “ARS D” with very bad results for the general understanding of production.

Lamps and other

Same as above.

Tiles and bricks

Surprisingly, there are bricks from Italy in Africa, also in some inland sites (Haïdra).

In general, inland looks very different from coastal regions, both in quality and quantity of imports.


Poblome gave a very interesting and inspiring lecture about Asia Minor and Greece, focusing on problems of complexity and multi-scalarity in understanding the nature of local and regional economies.

Most examples were referred to either Sagalassos or Beotia.


Vokaer's lecture showed archaeological evidence from Syria. The overall number of published archaeological sites is very low, but there are interesting patterns in the spatial distribution of both imported wares (from Mediterranean and from Parthic region) and local wares.

Brittle ware, thanks to archaeometric analysis, shows distinctive patterns of distribution, with significant turnover among the most important production centres over time. In particular, the most active regions are the Euphrates area and North-Western Syria, but the city of Antiochia was always served by a local production.

In the later period, the same forms of some cooking vessels are produced by different workshops.


At the end of this last session, I made two related questions, namely:

  • how much room is there in our interpretative models for choice and taste by the “final user” ?
  • to what extent did final users know the source of pottery they were buying? and did artisans know about the destination of their work ?

Poblome answered briefly to both questions, mostly focussing on taste (rather than other paradigms for choice) and local trade.

After that, it was 6 o'clock and discussions went on at the nearest pub.