DATA AND METHODS

In the decades since Lauwerier's publication on animal husbandry in the Roman River Area, many new archaeological excavations have been carried out, mostly developer-funded. The size of these excavations varies from small test trenches to large-scale excavations of complete settlements. The quality of the excavations and the archaeological reports also varies. Information for some sites is limited to animal bone data, with little information on the other archaeological finds and structures. Likewise, the size of animal bone assemblages and the detail in which they have been analysed and published is variable. Fortunately, several zooarchaeological colleagues were generous in sharing unpublished reports and primary data.

The selection of sites for this study is based on the presence of animal bones, for which reports or data were available. The description of the archaeological sites in paragraph 2.3 includes some information on the most important archaeological finds and structures, the chronology, any remarkable finds that could be related to production or trade, any relevant botanical data and basic information on the animal bone assemblage. Z.1 lists all the sites included in this study. Site numbers refer to a map of the research area (fig. 2.1), while the codes are used in illustrations in chapters 5 to 7. The table further includes the type of site, the size of the animal bone assemblage (both total size and identified fragments only) and the date of the assemblage. The collection of animal bone data from these sites for this study focused on fragment counts for domestic and wild species, and for the main domestic species skeletal elements, age data, measurements and butchery marks. Not all data could be collected for all assemblages, especially for the smaller ones.

The numbering of the sites is not always logical. The numbers referring to site locations in some cases cover several separate sites. One number has been used where sites are close together, have been excavated by the same archaeologists and are published together. Examples are Huissen-Loostraat Zuid, sites A and D and Zaltbommel-De Wildeman, sites A, B and C. In some cases, animal bone assemblages from what is essentially the same site have been given a separate number, because they were analysed and published by different people.

Since this chapter focuses on rural sites, the emphasis is on agrarian production, although the introduction of new types of food and technology (butchery tools) will also be discussed. This regional study aims not only to provide a general picture of animal husbandry in this region and time period and any developments occurring during the Roman period, but also to study the variation in response that occurred when local communities were faced with a market demand for agrarian products.

Since 45 of the 72 sites included in this study are rural settlements involved in agrarian production, the evidence for production is clearly stronger than the evidence for consumption. There are differences in the size of rural settlements, which vary from one to five or six farmhouses, but apart from that there is no clear hierarchy. Some sites have traditionally been interpreted as villae. Although the definition of a villa is inherent to its function as a large-scale agrarian production unit, interpretations for the Dutch River Area are invariably based on the size and lay-out of structures and the style of construction (stone, tiled roofs, other ‘Roman’ elements). For this study, villa-like settlements are grouped with other rural sites, since they are all agrarian production units. We shall see whether agrarian production differs from ordinary rural sites. The same applies to several sites with a possible military connection. Before we look at the data from Roman sites, it is necessary to discuss farming in the Late Iron Age, to gain some understanding of the agrarian situation before the Roman occupation.

FARMING IN THE LATE IRON AGE

LATE IRON AGE SOCIETY AND SURPLUS PRODUCTION

The Late Iron Age economy in the research area was an agrarian subsistence economy. Mixed farming was practised, with nearly everyone involved in farming the land and looking after livestock. The pastoral side of farming seems to have been especially important, and can be related to the greater suitability of the landscape for animal husbandry (plenty of fertile grassland for grazing) than for cereal farming (suitable arable land limited due to risk of f looding). The so-called pastoral ideology is ref lected by the sharing of living space by man and animal.

The previous chapters have discussed producer and consumer sites separately; this chapter compares the results from both types of site. By comparing the data from rural and consumer sites, it is possible to investigate the hypothesis that most of the food was acquired locally. Furthermore, comparing the data will give insight into production strategies employed at the rural sites and relationships between producers and consumers. The layout of this chapter will be the same as that of the two previous chapters, looking first at species proportions and slaughter ages, then at skeletal elements, butchery methods and biometric data, and finally at archaeobotanical data. The chapter will end with a discussion and the most important conclusions with regard to developments in agriculture and food supply.

SPECIES PROPORTIONS

TOTALS PER PERIOD

For each period, total numbers of fragments for the four main domesticates for all rural and consumer sites have been compared. In all periods, cattle have a higher percentage in the consumer sites; this is most noticeable in the Middle Roman period (fig. 7.1). The percentage for sheep or goat is higher in the rural sites, especially in the Early Roman period. In this period, the percentage for pig is much higher in consumer sites. In the Middle Roman period, percentages for pig are almost equal, while in the Late Roman period, the percentage is much higher in the rural sites. The percentage of horse fragments is higher in rural sites in the Early and especially the Middle Roman period, but lower in the Late Roman period. The near absence of sheep in Nijmegen-Valkhof fits in well with the low proportions of sheep in Late Roman rural sites.

SHEEP / GOAT IN CONSUMER SITES

The two 1st-century watchtowers in Utrecht-Leidsche Rijn both show a high proportion of sheep or goat. This was not seen in other military sites. Perhaps this is related to the small number of troops stationed at the watchtowers. Supplying them with whole cattle or parts of cattle would be more than they could eat. It is more logical to supply joints of smaller livestock and beef taken off the bone. Indeed, evidence has been found for the supply of joints of meat.

This chapter describes the main methods used in collecting and analysing zooarchaeological data for this study. Nearly all these data were taken from existing publications or unpublished reports; only one assemblage was analysed specifically for this study. The variety of the publications and authors means that there are differences in the methods that were used and in the detail in which assemblages have been analysed and described. In many cases, the lack of primary data limited the use of the data set. The size of data sets further affects their usefulness, with larger data sets naturally having more to offer than smaller ones. Nevertheless, smaller assemblages were included in order to fully utilise the potential of the zooarchaeological archive for the Roman Dutch River Area.

TAPHONOMY

Taphonomy has a huge impact on zooarchaeological data. To investigate the extent of this impact, and whether it has affected the sites in the study area in similar ways, a taphonomical study was carried out. This study was limited to cattle, for three reasons. First, in theory, each species could have a different taphonomical history, depending on whether it was consumed, how the carcass was butchered and how refuse or diseased animals were dealt with. Second, cattle is the most common animal in most assemblages, and therefore provides the best data. For less common species, or those that fluctuate strongly over time, the number of fragments may be too small. Finally, since this study was carried out in the first place to help understand skeletal element distribution, which focuses on cattle, it seemed logical to limit the taphonomical analysis to cattle as well.

For 40 assemblages from 24 rural sites and five assemblages from military or urban/military sites, data on two major taphonomic agents were collected. By looking at these two taphonomic agents (butchery and dog gnawing), we can get an indication of the impact of taphonomy. Butchery by humans will cause fragmentation of bones into smaller pieces. It is well-known that gnawing by dogs has a major effect on skeletal element distribution, since dogs favour porous ends of long bones, such as proximal humerus and distal femur. The percentage of loose teeth is the third variable which was recorded, and is taken as an indication of the degree of fragmentation and preservation of the animal bones.

FRAMEWORK, SCOPE AND RESEARCH QUESTIONS

INTRODUCTION

Little is known about the way in which previously self-sufficient rural communities responded to market demand for agrarian products in the past, or about the resulting changes in agrarian strategies.

This topic will be studied through a case study of animal husbandry in the Lower Rhine area in the Roman period. With the arrival of the Romans to the southern and central part of what is now the Netherlands, a substantial group of consumers was introduced into what was basically a subsistent agrarian society. In earlier periods, with the exception of perhaps a small minority of religious or craft specialists, everyone was involved in agrarian production. The arrival of the Roman army constituted a large group of people that depended on others for their food. Moreover, in the town of Nijmegen, we find administrators, traders and craftsmen, most of whom did not or only to a limited extent produce their own food. The Roman occupation thus introduced a separation between producers and consumers.

The Roman army and administrators, as well as an inf lux of traders, stayed in the southern half of the Netherlands for nearly four centuries. While some products were imported, other foodstuffs were of local origin. The presence of imported material culture in rural sites is an indication that local people participated in trade. With farming as their economic basis, an agrarian surplus is the most likely form of goods that could be traded for imported products such as pottery. This means that the local farmers managed to produce more food than they needed for themselves. How they managed this is the main question of this study. The aim is to trace developments in animal husbandry from the Late Iron Age to the Late Roman period, investigate whether these developments can be related to market demands, and gain an understanding of the impact of the Roman occupation on the agrarian economy.

Although this is a case study, it is expected that the results will be of wider relevance to other researchers with an interest in agrarian societies faced with major economic changes or in the organisation of food supply to towns and army.