Vagnari vicus Preliminary Faunal Report

Vagnari Vicus | 2013 Preliminary Faunal Report Angela Trentacoste University of Sheffield Department of Archaeology [email protected] August 2013 Introduction This preliminary report presents an analysis of the animal bones and shells recovered during the 2012–2013 seasons of the Vagnari vicus excavations. This project, organized by the University of Sheffield, explores the central village of a Roman imperial estate located near Gravina (Puglia, Italy). Excavation has focused on a series of rooms on the northern edge of the vicus and an exterior area immediately outside this space. Animal bones and shell were recovered from the topsoil and archaeological deposits related to the structure. context information is included in the attached appendix (VagnariVicus_FaunalData). No major middens have yet been uncovered, and faunal material most likely represents food and carcass processing debris that accumulated on site. Whether from fills or floors, these remains shape our understanding of the animals raised and consumed by the workers living on this imperial estate. Methods Animal remains were studied in the field during the 2013 excavation season. Nearly all of the faunal material was hand collected, although bulk samples were wet sieved for plant remains (palaeobotanist Matt Stirn has the complete list of sieved contexts). Only one identified specimen was during sieving from context 2010, trench 201 – a vole mandible (see Table 1). The faunal assemblage was composed mostly of small, degraded fragments, which probably suffered significant post-depositional decomposition as result of their harsh environment and proximity to the surface. Difficult specimens were photographed and identified using the University of Sheffield collections. Shells were studied at the University. All identifiable specimens were recorded. The presence of up to six diagnostic zones was noted when specimens were recorded. Measurements were taken to the nearest tenth using a standard set of vernier calipers. Tooth wear stages follow Payne (1973) and Payne (1987) for sheep/goat and Grant (1982) for cattle and pig. Fusion stages follow Silver (1969). Bones are described as ‘fusing’ if any part of the fusion line is still visible. Sex was determined at the time of recording for Sus canines and canine alveoli based on their size and morphology. Sex distinction was also attempted on the pelvis of sheep/goat, cattle and pig using the morphology presented in Boessneck (1969). Data are quantified using the number of identified specimens (NISP) and zone counts. A full explanation of the recording methodology and recording codes is available for download at http://alexandriaarchive.org/bonecommons/items/show/1893. The faunal assemblage Over one hundred animal bones and teeth from the Roman vicus were identified for this report (Table 1). A similar number of shells was also identified (Table 2). Bones and shells collected from the topsoil are excluded from further analyses, as we cannot be certain of their original provenience. Domestic species dominate the assemblage; wild animals, represented by in the Roman material by deer, are rare (4% NISP, 5% zone counts). As in nearly all Roman faunal assemblages, the three main livestock taxa – cattle, sheep/goat, and pig – account for the majority of the remains. Sheep/goat bones are the most frequent, followed by cattle and pig (Table 3). Interestingly, goat remains are more common than those of sheep. Although this data must be treated tentatively, because the sample size is very small, it departs from the norm pattern on Roman sites. Equids (horse, donkey, or mule) and dogs are also present. All parts of the skeleton are represented for cattle and sheep/goats. Pigs are almost entirely represented by teeth and foot bones. The prevalence of less-desirable, low meat yield cuts could suggest a separation of the carcass, with better cuts of meat (like the shoulder and leg) consumed elsewhere. This hypothesis is tentative, however, and more data is needed to confirm this trend. Very little data is available on animal age. Cattle remains are primarily from adult animals. The presence of extremely worn teeth attests to the presence of elderly individuals. Sheep/goat were killed at a variety of ages, and the presence of neonatal/foetal sheep/goat remains also attests to their local breeding. Pigs were slaughtered between immaturity and adulthood. Little information on animal sex was available. One female sheep/goat pelvis was identified, as were four male canines. No female canines were recorded. This small sample may indicate that male pigs were being consumed at the vicus, but a larger sample is needed to be sure. Male pig canines are much larger and more robust than those of females, and their recovery may be linked to their greater size and chance for survival. Very few traces of butchery or burning, and no instances of carnivore gnawing were recorded. Evidence of these modifications was most likely obscured by the poor surface preservation of many remains. One work bone fragment, a pin or awl, attests to some craft production at the vicus. The presence of marine shell fragments illustrates links between Vagnari and the sea. However, the presence of many shells in the topsoil complicates an interpretation of their presence. Shells probably provided a raw material rather than a food source, since the site is some distance from the coast. Land snails are more abundant than marine molluscs. Many of the recovered snails are common throughout the Mediterranean and attest to the local environment and indicate the area’s open country, grasses, and loose/sandy soil. Snails may have been consumed in antiquity, as they still are in Gravina today. For both types of shell, but especially those of land snails, it is difficult to separate post-Roman remains from archaeological material. Considering the abundance of snails in the area today and their proclivity to seek shelter in loose and cracked soil, many snail shells probably become incorporated with the archaeological material through burrowing or disturbance. Further excavation should resolve this issue, as intrusive material is less likely to become mixed with deeper strata or under packed floor layers. Lastly, joins between bones in contexts 2005 and 2006 in Trench 200 were found. The distribution of joins between these two contexts could suggest they are part of a single deposit or that they share material derived from the same origin. Discussion and future directions Although the assemblage is small, it still allows for some contextualization of animal use at the vicus. Species frequencies are similar to those previously recorded at Vagnari (Table 4). Sheep/goat are the predominant taxa in all studies at Vagnari, a trend likely linked to production on the imperial estate and its location near transhumance routes. Sheep and goat pastoralism increases in southern Italy during the Imperial period with the rise of large-scale farming and long-distance transhumance (MacKinnon 2004). However, this emphasis on sheep/goat husbandry was not present on all sites, as the marked presence of pigs at San Giovanni (Ruoti, Basilicata) illustrates (MacKinnon 2002). Cattle at Vagnari would have been an important source of labour, especially in agricultural production. Pliny expresses the importance of cattle when he calls them ‘a partner in labour and in husbandry’ (NH 8.70.180). The presence of very old cattle in the assemblages attests to their role as beasts of burden; elderly cattle would have been killed and consumed at the end of their working lives. Adult would also be able to supply milk. Sheep and goats would have provided milk, meat and wool, but more information is needed to understand their role on the estate. Pigs would have provided another source of meat and fat. The small number of dog and equid remains is to be expected, as these animals were not normally consumed and their remains would have been treated differently than food debris. Dogs may have been kept as pets or guard animals. Horses, donkeys and mules would have aided in travel and transport. The presence of deer bones indicates that hunted animals also were brought to the vicus, although they made a very small contribution to the diet of the inhabitants. While the material included in this brief report fits well with broad trends in southern Italian Roman animal husbandry, the assemblage also raises new questions and suggests new avenues for future research. The assemblage studied for this project provides some suggestion that the inhabitants of the vicus had access primarily to less desirable types of meat – elderly cattle beyond their best years, a higher proportion of goats than sheep, and the predominance pig heads and feet. However, these patterns may be a product of low sample size. Indeed, MacKinnon (2011) suggests the opposite for the vicus, having found a more diversified and perhaps elite pattern of meat consumption, including the presence very young sheep and pigs, in his earlier study. The best way to resolve these issues is through systematic integration and comparison of the faunal data from all the Vagnari excavations. A larger study that draws together all the faunal data from these excavations could also address differences in how the labourer community interacted with animals, versus those in higher social positions. Together with botanical evidence, zooarchaeology could provide evidence for local foodways, and relationship between those working the land the food they produce. Were workers primarily fed primarily from a surplus of the estate or did they manage they manage their own food sources? For example, pigs may have been kept by the workers and fed of scraps and food rubbish (cf. Plin. NH 8.77.206). Goats could also be kept in small groups, and hunting may have supplemented their normal diets. Old draft cattle would have been slaughtered after their working lives; this tough and less desirable meat may have been destined for workers tables rather than those in higher classes. Overall, further investigation of the site’s animal remains would help illuminate the relationship between the products of the imperial estate and those living and working at Vagnari. Lastly, I would like to close with a recommendation that sieving be integrated with excavation strategy. Un-sieved faunal assemblages are inherently biased against small bones and small animals (Payne 1972), a problem which complicates the analysis of the material. Dry sieving dirt buckets with a c. 1cm garden mesh, or something similar, would minimally interrupt excavation while ensuring higher recovery of valuable animal remains (as well as other important, but hard to spot finds like, slag, etc.). Wet sieving of bulk samples should also continue, in order to recover the remains mirco mammal and birds. Works Cited Boessneck, J. (1969). Osteological differences between sheep (Ovis aries Linne) and goat (Capra hircus Linne). Science in Archaeology. D. Brothwell and E. Higgs. London, Thames and Husdon: 331–358. Grant, A. (1982). The Use of Tooth Wear as a Guide to the Age of Domestic Ungulates. Ageing and Sexing Animal Bones from Archaeological Sites. W. R., C. Grigson and S. Payne. Oxford, BAR British Series 109: 91-108. MacKinnon, M. (2002). Excavations of San Giovanni Di Ruoti. Vol. III. The Faunal and Plant Remains. Toronto, University of Toronto Press. MacKinnon, M. (2004). Production and consumption of animals in Roman Italy: Integrating the zooarchaeological and textual evidence. Portsmouth, Journal of Roman Archaeology. MacKinnon, M. (2011). The faunal remains. Vagnari. Il villaggio, l'artigianato, la proprieta' imperiale. A. M. Small. Bari, EdiPuglia: 305-328. Payne, S. (1972). Partial recovery and sample bias: the results of some sieving samples. Papers in Economic Prehistory. E. Higgs. Cambridge, Cambridge University Press: 49–64. Payne, S. (1973). "Kill-off patterns in sheep and goats." Anatolian Studies 23: 281–303. Payne, S. (1987). "References codes for the wear state in mandibular cheek teeth of sheep and goats." Journal of Archaeological Science 14: 609–614. Silver, I. A. (1969). The aging of domestic animals. Science in Archaeology. D. Brothwell and E. Higgs. London, Thames and Hudson: 283–302. Small, A. M., B. Roe, et al. (1994). "A pit group of c. 80-70 BC from Gravina di Puglia." Papers of the British School at Rome 62: 187-246. Watson, J. (1992). The mammals. Gravina, An Iron Age and Roman Republican Settlement on Botromagno, Gravina in Puglia. Excavations of 1965-1974. Vol. 1. The Site. A. M. Small. London, British School at Rome: 92-120. Table 1 Number of identified specimens and zone counts for animal bones One vole remain was recovered from wet sieving. Roman Topsoil NISP Zones NISP Zones Cattle 28 17 6 1 Sheep/goat 48 42 6 4 Sheep 2 2 0 0 Goat 3 3 1 1 Pig 19 15 1 1 Dog 1 1 2 2 Small canid 1 1 0 0 Equid 1 1 0 0 Deer sp. 1 1 0 0 Red Deer 3 3 1 1 Hare 0 0 1 1 Vole (Microtus sp.)* 2 2 0 0 Rat 1 1 0 0 Cattle/Red deer 1 1 0 0 Large mammal 3 1 0 0 Medium Mammal 2 0 0 0 Small mammal 1 1 0 0 117 92 18 11 Total Table 2 Number of identified specimens and zone counts for shell Roman Topsoil NISP Zone NISP Zone Eobania vermiculata 26 17 8 5 Rumina decollata 32 28 2 2 Theba pisana/Cernuella virgata 22 24 2 2 Pomatias elegans 1 1 Helix sp. 13 1 1 1 Land snail fragment 1 Land snails Marine Bivalves Ostrea edulis 1 Glycymerididae 1 1 Arcidae 3 3 1 1 Veneridae 1 1 5 5 Pecten sp. ? 1 Bivalve fragment 4 9 2 Other Dentaliidae 1 Total 105 76 30 18 Table 3 Roman relative frequencies of cattle, sheep/goat, and pig (excludes topsoil) NISP Zone N % N % Cattle 28 28% 17 22% Sheep/goat 53 53% 47 59% Pig 19 19% 15 19% Total 100 79 Table 4 NISP values for other southern Italian sites Site Date (centuries) Total n Cattle % Sheep/goat % Pig% Reference Vagnari c. 1–3 AD 167 20 41 39 (MacKinnon 2011) Gravina villa Gravina villa – cistern Gravina villa – pit F202 Gravina settlement Hellenistic/ Republican 562 16 52 32 (MacKinnon 2011) Republican 374 1 89 10 (MacKinnon 2011) Republican 290 2 73 24 (Small, Roe et al. 1994) c. 2 BC–1 AD 555 22 58 20 (Watson 1992)