Pots, presses, people and land (2011-2015)

Pots, presses, people and land. The role of overseas export and local consumption demand in the development of viticulture and oleoculture in central Adriatic Italy (250 BC - AD 200). (Doctoral research: Dimitri Van Limbergen)

The link between demography and agriculture in economies preceding the industrialisation era is a widely acknowledged feature of historical studies focusing on 13th-19th century Europe. Bus despite Bruce W. Frier’s almost desperate call for the deeper integration of population issues in Roman scholarship, a recent contribution by Neville Morley could do little more but to address the general lack thereof in socio-economic studies on classical antiquity. However, the work of Walter Scheidel is a good reminder of how such constraints necessarily must have cast their shadow on the agro-economic developments in many parts of the Roman Empire.

Despite this notion, interpretative models of Italy’s economy in the Roman period – and in particular those concerning the Italian wine and oil industry – have to a great extent stressed the importance of the volumes of food transported over longer distances; i.e. on extra-regional or even extra-Italian levels. As a result of this rather one-sided point of view, archaeological proxy indicators for the economic performance of the Italian wine and oil sector – in the first place amphorae and press installations – have been studied almost exclusively from said perspective. In contrast, there has been nearly no attention for how the developments in these two categories of archaeological evidence might relate to local events, such as population dynamics and their associated food resources.

This research project set out to explore these issues for the central Adriatic area of Italy in the Late Republic and the Early/High Empire (c. 250 BC – AD 200). The phraseology ‘central Adriatic’ is here used to delineate an area of ca. 11,600 km2 that comprises the Marche region and the northern part of Abruzzo (Abruzzo Teramano). It is situated between the Apennine hills and the central Adriatic coast and is characterised by a comb-shaped geomorphologic structure in the form of a series of parallel river valleys descending towards the Sea. In Roman times, this area mainly coincided with the fifth administrative district of emperor Augustus’ Provincia Italia (Picenum), while the most northern section was part of the sixth Regio (Ager Gallicus).

In order to tackle this topic, I first examined two of the main proxy indicators for commercial wine and oil production – amphorae and presses – in order to see which patterns emerged. Next, an attempt was made to reconstruct urban (and rural) demographic patterns through time on the basis of the available structural remains. With said population figures in mind, both a consumption and a land requirement model were then outlined through the use of ancient textual accounts and comparative evidence from pre-industrial central Adriatic Italy. Finally, these models were then compared with the size of the study area in order to assess what portion of the local agricultural produce in cereals, wine and olive oil was consumed by the country- and urban dwellers in the region, and to which extent possible surpluses could be absorbed by consumption groups further afield over time.

The model that was built as such over the course of five chapters was thus set-up from one point of view; i.e. with the intention of creating an interpretative framework for evaluating the possible impact of local demographic conditions on the evolution of the wine and oil business in central Adriatic Italy between Late Republican and Early/High Imperial times. The main question was hereby whether regional demographic developments had anything to do with the diminishing exports of wine in amphorae from the Augustan period onwards, as well with the minimal involvement of the area in the Mediterranean oil trade in the entire studied time period. This model was essentially constructed through an in-depth reading of the archaeological evidence regarding wine/oil production and transport; and by positioning the changes in this evidence against a theoretical population/land model inspired by the archaeological record.

On the conditions that all these calculations with regard to population dynamics, food intakes and agricultural productivity have provided us with a useful scale with which to explore the possible effects of lower and higher population levels on the wine and oil trade in this area, it would seem that the conditions for the prosperity of the latter were more favourable in the Late Republican period than in the Early/High Imperial one. Indeed, there is a high degree of probability that between 40% and 85% of all land in the study area – i.e. all aggregate land and not actual agricultural land – was not needed for local food requirements in the former period. On the other hand, if local land requirements – and in particular local cereal requirements – rose as significantly as has been implied, and if this increase was mainly met by local production, the effects would have been as such that the amount of land available for export-oriented wine and oil production was reduced until 67 to 5% of the total territory when population- and consumption levels both reached their theoretical peak in the High Imperial period.

In other words, in the first case, Roman farmers found themselves in a very favourable position with ample space for expanding and intensifying vine plantations in the wider Adriatic coastal area by selecting good viticultural lands from which maximum profits could be obtained. As such, these circumstances would have allowed for the area to participate to the full in the wider phenomenon of extra-regional wine commerce in the Late Republican period. It is this on-going intensification and expansion process that we see so well reflected in the amphora evidence discussed in the first chapter. In the second case, however, matters were rather different, with population pressure in town and country rendering it unlikely that in the 1st and 2nd century AD large amounts of local wine and oil could be exported on an annual basis towards external markets. This is not to say that such developments would have actually stopped the existing commercial wine and oil trade, but under such circumstances the scale of this trade was bound to be much more modest, with a major portion of the surpluses being trade towards the towns in the region.

Taking into account some of the patterns that have been examined with regard to the ceramic and archaeological evidence, such a process would actually bode well with an increasing co-existence of amphorae and barrels – with the latter eventually becoming the only carriers for wine (and oil?) in the central Adriatic area – and with the main chronological trends individuated in the press evidence, i.e. continuing wine and oil production on a substantial scale in the 1st-2nd century AD. Indeed, the 1st century AD numeral peak in the latter evidence – and the only minimal decrease in numbers in the following century – seemingly portends the ongoing vitality of a business, whose manufacturing process became more visible in the archaeological record than its distribution modii because of a major shift in market focus. This seems especially the case for olive oil production and containerization. In fact, while the available amphora evidence indicates minimal involvement of the study area in the overseas oil trade in the first two centuries of the Common Era – and even less so in the preceding centuries – about half of the product-identified presses in the area turned out to have been producing olive oil in the first place. If we take the theoretical oil needs by the urban population – in particular those proposed for the Early/High Imperial period, which could have been as high as two million litres – as a major incentive for their existence, much of this oil probably circulated in perishable skin containers, and not in amphorae.

Finally, the constructed model also stresses how we should be open to interpreting the evolution of commercial viticulture and oleoculture in the area not as an isolated phenomenon, but rather as part of a much wider agrarian system, whose evolution was determined by local events in general, and by its capacity to sustain the local population in particular, especially with regard to cereal requirements.

The main aim of this research has been to stimulate the debate on the fundamental modifications that shaped central Italian agriculture in the period under consideration. Admittedly, the speculative component of the story line that has been set-out is considerable, but it was never the intention to claim historical exactitude. Instead, I have argued that the cross-fertilisation between the amphora evidence, the press remains and the reconstructed population/land model implies that at least ‘something’ happened in the course of the 1st-2nd century AD that significantly reshaped the central Adriatic wine and oil sector; and that said ‘something’ may somehow be linked to developments in intra-regional demography and food supply, rather than merely to the disappearance of the provincial markets. In terms of agricultural practices, those changes may have comprised an increase in the distribution of arbustum (a typical extensive way of cultivating the vine by the Romans in central-northern Italy) and low-density olive orchards in the area in the Early/High Imperial period.


Dr Dimitri Van Limbergen