Animals and Roman Lifeways in Aquincum

210 211

Introduction Archaeozoology is the study of animal bones from archaeological sites. Although an accepted sub-field of archaeology now, there has not always been such enthusiasm for these ‘old bones’. Indeed, while the earliest excavation reports from the Roman Civil Town of Aquincum tantalizingly mention butcher shops in the Civil Town market, no bones were actually studied or even saved from these structures.1 The first research efforts concerned with faunal analysis as a separate field is generally attributed to two Swiss researchers, Ludwig Rütimeyer2 and Jan Ulrich Duerst.3 The first decades of the new century saw a number of

limited attempts to use data from animal bones but archaeological excavation and analysis was not yet focused on the sort of economic and subsistence questions to which this kind of find material is more suited. The years following World War II saw the rise of a few important scholars working in particular on problems of Neolithic domestica-tion and production.4 In Hungary, Sándor Bökönyi began work in the early 1950’s both on questions surrounding the Neolithic development in the Near East and Central Europe as well as on other periods including the first scientific work on the animals of Roman Pannonia (BÖKÖNYI 1974).5 Despite the huge volume of his opus, he actu-ally did very little work on bone material from Aquincum itself, despite the fact that it was situ-ated practically in his own backyard (BÖKÖNYI 1969). In the past twenty years archaeozoology has undergone a minor boom as researchers began to pose crucial socio-economic questions about past societies and their lifeways. Indeed, such data studied together with supporting traditional types of information on material culture can provide strong proof of the presence of certain kinds of agricultural, commercial, dietary and manufactur-ing activities. Variability in dietary and manufac-

6.3. ANIMALS AND ROMAN LIFEWAYS IN AQUINCUM

1 Kuzsinszky, B., Az aquincumi ásatások 1882–1884 és 1889 [Excavations in Aquincum between 1882–1884 and in 1889]. BudRég 2 (1890) 77–160. The erstwhile butcher shop is also touched on in KUZSINSZKY 1934. The only except is K. Torma, Az aquincumi Amphiteatrum északi fele. (Jelentés az ottani ásatásokról) [ The northern half of the Aquincum amphitheater (Report on the excavations there). Budapest, 1881, 100–102.

2 Rütimeyer, L., Die Fauna der Pfahlbauten der Schweiz. Neue Denkschr. D. Allg. Schweiz. Ges. D. ges. Natwiss. 19, 1861. Rütimeyer was the first to recognize that there were differences between the bones of domestic animals and the bones of their wild ancestors.

3 Duerst, J. U., Animal remains from the excavations at Anau and the horse of Anau in relation to the races of domestic horses. In: (R. Pumpelly ed.) Explorations in Turkestan 2. Smithsonian Institution, Washington, 1908, 339–442. Duerst spent three years studying a huge bone assemblage from Anau in Russian Turkestan where he claimed to perceive a reduction in the size and texture of bones during the transition from hunting to domestcation.

4 For example Frederik Zeuner’s basic work on domestica-tion, Charles Reed in the United States who worked on the question of domestication in the ancient Near East, J. Boessneck who established the so-called “Munich School” again to study domestication processes in the Near East. He also developed some of the basic methods used by archaeozoologists today.

5 Bökönyi, S., Animal Husbandry and Hunting in Tác-Gor-sium. Budapest, 1984.

turing behaviors can also provide insight into the kind of ever-changing pluralistic society we know to have been characteristic in provincial Roman territories. Indeed, while it is convenient to speak of ‘Romans’ in the sense of political and social organization, the ethnic composition of ancient provincial capitals such as Aquincum must have been complex indeed.6 Archaeozoology has gradually increased in importance until today there are few archaeologists who would consider excavating without gathering the faunal material as well as ceramics and other conventional artifacts. Screening soil is a regu-lar phenomenon at excavations insuring a more well-rounded picture of what animals were actu-ally consumed. This has especially impacted the recovery of the bones of smaller animals such as hare, rabbit, dove, chicken and particularly fish as well as the bones of immature animals, including those of piglets which are significantly less likely to be found. Areas of interest have expanded to cover research into changing and variable prefer-ences in draught, diet and butchery. Researchers are also conscious of problems of interpretation related to taphonomic loss resulting from natural agents, ancient human behavior as well as exca-vation techniques. The ‘laundry lists’ of species of fifty years ago are now considered somewhat naïve starting points and not the goal of research. In order to produce stronger arguments, greater

attempts are made to coordinate the archaeologi-cal results with those deriving from the analysis of bone materials. The situation at Aquincum is particularly lucky in this regard. With a special bone and plant macro-rest room having been made available students are now abole to come to study bones from individual sites, speeding the analysis of the enormous material. The excavations which have been conducted at Aquincum have all been carried out under the aegis of the Aquincum Museum since its founding well over one hundred years ago as the first Municipal museum in Budapest (PÓCZY K. 1998/2, 219). Bones have been regularly col-lected during excavations since the 1970s even if the techniques used may not have been as fine tuned as could be desired. As the sites are slowly analyzed, the opportunity for real collaboration increases since the faunal analyst and the archae-ologists find themselves under one roof. Areas of immediate concern for the spe-cialists studying find materials from Aquincum include farming practices, distribution networks for meat products, dietary practices within social and ethnic groups, supplying the military and craft activities such as the preparation of hide, horn, glue and manufacture of bone/antler/ivory objects. The proportion of the species present at a given town, village or fort, kill-off patterns related to age, butchery patterns reflected by body part distributions and butchery marks, estimations of size and marks of manufacture and use wear on bone ornaments all contribute to the general understanding of Roman Period human-animal relationships. We know about the Roman farming and dietary ideals in Italy, but it is useful some-times to look away from the center when trying to understand the nitty-gritty necessities of life in the provinces, including the region surrounding the capital of Pannonia. A warning is due, however. The bones found during archaeological excavations usually directly reflect consumption behavior at settlements of most types rather than production related to farming, butchery or even craft activities. Any conclusions concerning farming practices based

6 Fitz, J., The Population. In: (A. Lengyel and G.T.B. Radan eds.) The Archaeology of Roman Pannonia, Budapest, 1980, 141; Mócsy, A., Lakoság és társadalom [Population and society]. In: (A. Mócsy and J. Fitz eds.) Pannonia:régészeti kézikönyve [Archaeological Handbook of Pannonia], Buda-pest 1990, 237–242. Aside from the various aboriginal Celtic tribes, there was a continuous influx of foreign sol-diers and their families/followers from Italia, Gaul, Britannia, Africa and the East into Pannonia. Small groups of so-called Barbarians also found their way into the region. All these groups must have had impact on agricultural tradition, tastes and preferences in food, cults and rituals and designs on bone objects. At the same time, the picture is confused by the need to maintain the symbols and tastes of Imperial Rome itself as the superior, sought after, culture.

213

entirely on bone material from urban settlements or cemeteries must be treated with caution. The most secure picture emerges when faunal data can be combined with other kinds of corroborating archaeological evidence. The converse is true of course true of conclusions based on archaeologi-cal material where burdens of proof also include a look at the archaeozoological picture. The great advantage of working in the Roman period is that there is a wealth of epigraphic and pictorial (Fig. 1) information available about animal hus-bandry practices in the empire.7 Taphonomy of complex sites While there are numerous advantages to studying the bones from well documented complex societies, there are also disadvantages. Material coming from archaeological sites can be modified by contempo-rary cultural behavior such as preferential selection of which animals should be slaughtered, butcher-ing traditions, cooking customs, garbage disposal and clearing work8. Sites in Aquincum range from dwellings of native farmers to upper class Romans and legionary officers to public housing, to villa-farms to ditches near workshops to military forts and their surrounding villages. The different customs of people and the various ways the sites functioned all ultimately affect which bones from which species are present and how different butch-ering traditions affect the form of bones. Since the soils around Aquincum are condu-cive to preservation being largely alkaline, another main source of bias is related to the continual

remodeling which took place at these settlements and the shifting of refuse out of its temporal con-texts. Further, the exigencies of rescue archaeol-ogy and a lack of training have meant that until recently few archaeologists hade the time and energy to learn how to integrate finer excavation techniques, especially screening, into their excava-tion schedules. This is changing now at many of our excavations. Roman animal keeping in Pannonia In general, it can be said that provincial Romans depended largely on domestic animals for a variety of purposes in contrast to the preceding Celtic period during which meat from wild ani-mals such as red deer and wild boar, was much more important in the diet. In Pannonia, cattle is overwhelmingly dominant in faunal assemblages, especially from military establishments. Cattle will always be slightly over-represented because their larger bones are rela-tively weaker than compact bones from smaller animals and tend to fragment into more pieces.9

212

7 Varro T., Rerum rusticarum libri tres. Budapest, 1971. Also, there are numerous stone reliefs depicting animals pulling wagons as well as simple depictions of animals. At Aquincum there are honeycake molds (crustullum) showing Bacchus with a donkey, Hermes with a ram and various triumphal marches with elegant horses.

8 Sciffer, M., Toward the Identification of Formation Pro-cesses. In: Behavioral Archaeology. University of Utah Press, Salt Lake City, 1995, 177–182. Schiffer discusses the gen-eral processes, both natural and cultural which may affect all artifacts.

Fig. 1. Lithograph from Trajen’s column in Rome’s Forum showing a cattle, sheep and pig being taken away as war bouty. These three domesticates were the basis of animal keeping in Pannonia province.

9 Binford, L. – Bertram, J. B., Bone frequencies and attritional processes. In (L. R. Binford ed.) For Theory Building in

Caprines and pig proportions vary depending on the kind of settlement involved, with more Romanized towns having fewer sheep remains.10 A number of Roman period researchers have concluded that cattle (Bos taurus L.) were used mostly for haulage, milk and manure in the same way as was customary in the Mediterranean area.11 Beef was particularly important in providing meat supplies for the military.12 Large breeds of cattle appear with the Romans although there has been discussion (BÖKÖNYI 1974, 46) as to whether this is a result of better keeping techniques or improved breeding13. It seems clear that these larger breeds are usually found on the large estates of rural Pannonia province where their superior strength would have been important in plowing large fields and hauling over greater distances. Cattle were usually slaughtered at a mature age once they were too worn out for work or their milk production had dropped. Beef, posssibly in a form processed for preservation, was a small but important part of provisioning at military forts and fortresses in the province.14 Manure for

fertilizer must have been a particularly important by-product of this species. Other important prod-ucts derived from cattle included hide,15 glue from hooves, horn and worked bone objects. Sheep (Ovis aries L.) were much more common than goat at settlements in Pannonia. Sheep seem to have been primarily exploited for their wool.16 They were exported from various parts of Pan-nonia to Siscia although sometimes lambs were slaughtered for a better class of meals. The Roman sheep are generally larger than the native Celtic ones in Pannonia (BÖKÖNYI 1974, 178–179) and it is likely they produced a medium fine wool similar to modern day Cotswald sheep. Goats (Capra hircus L.) seem to have been much less important than sheep in the diet although their sizes also increase compared to the native Celtic varieties suggesting that improved forms were introduced into the province as well. The first hornless goats appear during the Roman Imperial period in Pannonia, probably as imports. In Italy, newborn and juvenile pigs (Sus domes-ticus Erxl.) were considered a delicacy with a large number of recipes devoted to pork prepared in

Archaeology. New York, 1977, 77–152. Bartosiewicz, L., Faunal material from two Hallstatt Period settlements in Slo-venia. Arheoloski Vestnik 42 (1991) 199–205; Bartosiewicz, L., Interim report on the Bronze Age animal bones from Arslatepe (Malatya, Anatolia). In: (Buitenhuis, H. – Bartosie-wicz L. – A. M. Choyke eds.) Archaeozoology of the Near East III. ARC Publication 18, Groningen, 1998, 221–232.

10 King, A., A Comparative Survey of Bone Assemblages from Roman Sites in Britain. Bulletin of the Institute of Archaeol-ogy 15 (1978) 207–232. King showed that in Roman Britain there was a progressive decrease of caprine remains relatve to those of pig and that more Romanized settlements tend to have more pig bones. In Pannonia, comparison of material from a wide variety of sites showed similar trends. However, there also tend to be more wild animals at military sites. See: Bartosiewicz, L., Animal bones as indicators of continuity at Roman Provincial sites. Antaeus (1990–91) 103–124.

11 Maltby, M., The animal bones from Exeter 1971–1975. Exeter Archaeological Reports 2, Sheffield University, Department of Prehistory and Archaeology 31, Sheffield (1976); Davis, S., The Archaeology of Animals. B.T. Bats-ford Ltd, London, 1987, 183.

12 KING op. cit. 15.13 DAVIS op. cit. 1988, 183. Bökönyi early on in his career

also identified a large number of wild cattle bones (Bos primigenius Boj.) at the Albertfalva castrum which was probably inflated. Not as much was known about the large Roman breeds of cattle then so that some of the bones may have been misidentified. Never the less, bones of aurochs are more common at military forts and rural settlements than in urban centers. There is only one bone from this wild species in the bone assemblages from Aquincum – a distal femur from an immature animal.

14 For example at the limes forts of Ács-Vaspuszta, Campona and Albertfalva.

15 Fitz, J., Economic Life, In: (A. Lengyel and G.T.B. Radan eds.) The Archaeology of Roman Pannonia, Budapest, 1980, 325. Round wood-lined pits from Aquincum have been interpreted as the remains of tanneries.

16 FITZ J. op. cit. 325. One of the centers of wool process-ing was in Siscia where wool from the Karst region was brought in and processed. Mócsy A., Ólom árucimkék Sis-ciából [Lead labels from Siscia] FolArch 8 (1956) 5–87.

215214

many and varied ways.17 This too seems to have been the case in Pannonia. At one great villa farm at Nemesvámos – Balácapuszta a cellar has been preserved with over 4000 bones from young suck-ling piglets. It is clear that the animals’ carcasses were being prepared in this cellar before it was suddenly abandoned.18 The high regard for pork as a commodity (BÖKÖNYI 1974, 219–220) is perhaps also reflected in the fact that pig bones may be found associated with ritual offerings in cemetery materials. It seems likely that pigs were kept as live animals at military centers in Pan-nonia and slaughtered according to need.19 Horse (Equus caballus L.) and other equids such as ass and donkey (Equus asinus L.) are more commonly found in settlements where haul-ing was important. Since equids were not eaten, except possibly in moments of great food stress, it is hardly surprising that their remains are only rarely encountered in civilian urban centers. Hors-es were kept by the Celts. Their goddess of fertil-ity, Epona, later taken onboard by the Romans, is customarily shown on horseback or surrounded by

mares with foals. Despite their small size, Celtic horses were apparently looked upon with favor by the Romans for their sturdiness. Germanicus received a supply of them in A.D. 15 while he was campaigning against the Germans.20 Romans also imported and bred larger horses21. From depictions on gravestones from some of the tombs lining the main roads in Aquincum it can be seen that these animals had a higher status than other equids, always occurring as the mounts of officers and cavalrymen. A bio-metric study was carried out on three almost complete adult horse skel-etons (A.D. 2nd–3rd century) from the edge of a field outside the limes calvalry fort at Albertfalva.22 Lyublyanovics determined that all three horses were stallions although the large size suggest they may have been geldings. Wither height estimations based on metapodial bones suggest that these horses belong to the large military of horse and that the phenotypic variability results from mixing of large Italian horses with a local horse stock. The horses can be described as heavily built. The average wither height for Roman military horses has been estimated (BÖKÖNYI 1974) at 142.9 cm although the 144 horses studied from Tác-Gorsium had an average withers height of 139.07 cm.23 Mules and asses were important animals for the military in hauling equipment and supplies and in draught in general. Asses definitely can be considered imports from the Mediterranean area since they cannot be found (and by defini-tion neither can mules) in this region before the Roman conquest. They were traditionally used to

17 Apicius, Cookery and dining in Imperial Rome. Dover Pub-lications Inc., New York, 1977. Apicius may have lived in the time of Tiberius [42 BC.–37 A.D.] although there are two other individuals known for their culinary dedication who may have set down the recipes found in this book. Apicius’ chapter on Sumptuous dishes (Lib.VII. Polyteles), including delicacies such as spayed sow’s womb (vulvae ster-iles), cracklings, pork skin, tenderloin, tails and feet (callum, lumbelli, coticulae, ungellae), fig-fed pork with wine sauce (in ficato oenogarum), pork cutlets, hunter style (offellae aprugneo more), suckling pig paunnch (ventrem porcinum) and ham (pernam). Of the meat dishes in that chapter two thirds are made from various parts of the pig carcass.

18 Vörös, I., Állatcsontok [Animal bones]. In: (ed. Bíróné Sey K. et. al.) A balácai villagazdaság fõépületének pincéje. [The cellar of the Baláca villa-farm main building] Balácai Közlemények II, Veszprém, 1992, 270–272. BÖKÖNYI 1974 says that various kinds of hams and sausages were known to have been produced by the Romans.

19 Lauwerier, R., Animals in Roman Times in the Dutch Eastern River Area. Nederlandse Oudheden, 12, Groningen, 1988.

20 Tacitus, Annales II, 5, Budapest, 1970.21 According to BÖKÖNYI 1974 Varro and Columella were

the two Antique writers who discussed the questions of horse breeding most thoroughly, including questions of feeding, testing of stallions, selection of mares and the various diseases afflicting these animals.

22 Lyublyanovics, K, Size variability in three Roman period horses from Aquincum. In: (ed. D. Ruscillo) Aging and Sexing, Oxbow Books, Oxford (in press).

23 Bökönyi, S., Animal Husbandry and Hunting in Tác-Gor-sium. Budapest, 1984, 61.

run reaping machines and mills.24 Asses milk was held to have purgative value and was also used as a beauty treatment for high society women. Dogs (Canis familiaris L.) present a very variable picture during this period in Pannonia depending on whether one considers the animals kept by the native Celtic populations or the vari-ous breeds of dogs used by the Romans them-selves. These latter dogs ranged in size from small lapdogs with thin legs, probably household pets in wealthy families, to strong German shepherd-size animals. It is thought, for example, that up to six breeds of dog may have lived in the town of Tác.25 Most dogs however, were small to medium size and probably looked much like the pariah dogs of Eurasia. Larger dogs were probably used as herding or watch dogs. Many of the larger dogs were certainly used by the upper classes in hunting (in spite of the fact that so few bones from game appear on Pannonian sites). There are indications that the Emperor Hadrian, during his tenure as governor of Pannonian province, hunted boar there26 using hunting dogs. Dogs were also used from time to time in cultic rituals27 and appeared as symbols of death and healing.28 Cats (along with domestic rabbit) are also found for the first time in this region on Roman sites. They were probably used as much for killing ver-min in the close quarters of towns as pets. There were exceptionally large numbers of cat bones recovered during the excavations at the town of Tác-Gorsium.29 Their bones are small and, aside

from the fact that they may not have been par-ticularly common animals in the province, rough excavation techniques mean that they are rarely encountered in faunal materials from this time. Three domestic bird species were exploited by the Romans in Pannonia: domestic hen (Gallus domesticus L.), goose (Anser domesticus L.) and domestic pigeon (Columba domesticus L.). Chick-ens were brought by the Scythians to the Car-pathian Basin and taken up by the Celts. Roman chicken breeds were larger than aboriginal ones which may have weighed between one and one and half kilo. Today, a breed of black chicken found in Romania with a live weight of around 1.5 kilos are thought to be the descendents of Roman chickens.30 The Roman birds could weigh as much as two kilo.31 Varro and Columella wrote down accurate instructions for their keeping, for-age and the construction of hen houses although traces of these are not preserved in the archaeo-logical material. Goose bones are much less common in the faunal assemblages of Pannonian sites than those of chicken. Again, we hear of goose from Varro and Columella and indeed this bird species is mentioned quite often by Antique authors. Romans may have introduced force feed-ing of geese to produce enlarged livers. Goose fat was used in many medicines. Finally, although we know that Romans kept domestic pigeons, their relatively small bones are exceedingly rare on sites in the province. This may be because they were actually rare, their bones were somehow destroyed during food processing or they are not being recovered in the absence of screening. Domestic pigeon remains have been found elsewhere in the empire, however,32 especially where care has been taken to screen.33

24 Toynbee, J. M.C., Animals in Roman Life and Art, Ithaca, New York, 1973, 15, 196, 295. Also see: White, K. D., Roman Farming, Thames and Hudson, London, 1970.

25 BÖKÖNYI op. cit. 18, 23, 66.26 FITZ, J. op. cit. 324.27 Vörös, I., A Ritual Red Deer Burial from the Celtic-Roman

Settlement at Szakály in Transdanubia. ActaArch Hung 38 (1986) 31–40.

28 Toynbee op. cit., 122–124., descibes the votive offerings with depictions of dogs on them left by pilgrins at the cult center for healing of the Roman-Celtic God Nodens in Gloucestershire in Britain.

29 BÖKÖNYI op. cit. 65.

30 Dr. Miklós Mézes (Szent István Agricultural University, Gödöllö), personal communication. This breed of chickens, in addition to its small size, is genetically distinct from all other modern chicken breeds in the region.

31 BÖKÖNYI op. cit. 94.32 Schibler, J. – Furger, A., Die Tierknochenfunde aus Augus-

ta Raurica (Grabungen 1955–1974). Forschungen in Augst, Band 9, Augst,1988, 22.

216 217

The most common wild animal species found at provincial sites is Red deer (Cervus elaphus L.). It is likely that a combination of climatic changes and shrinking environments relative to expanding agricultural territories reduced the deer population. The Celts hunted this animal widely and it appeared to have had some place in Celtic mythology. Sacrifices of red deer have been found buried in pits on Celtic-Roman sites in the province.34 In Roman times, it ceased to be an important meat animal in the province and was hunted by the upper classes for sport. Many of the remains from this species occurs in the form of fragments of the antler rack which stags shed in the Spring. A number of workshops with half-finished objects made from antler,35 as well as microscopic analysis of carved ‘bone’ objects from Aquincum have shown that antler was quite an important raw material for artisans in Pannonia as elsewhere in the Empire36. Organized gather-ing trips, possibly controlled by the producers of

worked bone and antler objects, were probably arranged each Spring to collect antler from partic-ular territories where the red deer stags regularly shed their antler racks. Individual workshops may have had rights to particular gathering territories to insure supplies of this valuable raw material. Another wild species which occurs on Pannoni-an sites is the European brown hare (Lepus euro-paeus Pall.). Hunting this species became quite popular during the Period of the Roman Empire and its meat was considered tasty. There is evi-dence from Augusta Raurica that young hares were kept in enclosures called leporaria and hare bone is found from time to time in the mate-rial of Pannonian excavations.37 So far, however, there is no evidence for organized hare keeping in Pannonia. Again, there would be a bias against finding the small bones of this species in the absence of fine excavation techniques. Bones from domestic rabbits (Oryctolagus cuniculus L.), which the Romans certainly brought into the provinces for the first time, have not yet been recognized in Pannonian Roman faunal assemblages. The Romans themselves did not distinguish between hare and rabbit as separate species. Finally, it is known that larger farm estates in the province had fishponds. Varro describes such freshwater ponds (dulces) on latifundia in Italy. There is a well- preserved dam at Kikeritó in western Hungary which was used to block off a small valley. Another pond bed was found to the north of Tác-Gorsium. Fish were considered delicacies and a staple of Roman cuisine. It is a certainty that wild fish were also caught. Due to rough recovery methods, the only bones usually found come from mature specimens of the larger fish species for example, pike (Esox lucius L.),

33 LAUWERIER op. cit. 22. A sieving experiment was car-ried out by this Dutch researcher on soil from five pits at the Roman perod site of Nijmegen. Various bird species were recovered along with fish remains that belonged to five species.

34 Vörös op. cit. 31, 40.35 Choyke, A. M., Modified animal bone. In: (D. Gabler

ed.) The Roman fort of Ács-Vaspuszta (Hungary) on the Danubian limes. Part II. BAR 531, Oxford, 1989, 624–632. Vörös, I., Campona-Nagytétény római tábor állatcsontmard-ványai [The Animal Bone Remains from the Roman Fort at Campona-Nagytétény]. FolArch 40 (1989) 75–118 here 85. Campona is one of the auxilary forts on the Danube river. It was established by the ala I Tungrorum Frontoniana in the 2nd century, destroyed during the Markomannic Wars in the end of the 2nd century and rebuilt afterwards. The workshop comes from this second period and contains several antler roses and rectangles of antler waiting to be worked into various tools.

36 Deschler-Erb, S., Römische Beinartefakte aus Augusta Raurica. Forschungen in Augst 27/1, Augst, 1998, 59–60. This is the first serious attempt to try and distinguish between the raw materials used to make bone, antler and ivory artifacts.

37 TOYNBEE op. cit. 200. Varro describes these enclosures on the country estates of late republican times as being carefully fenced in and carefully planted in grass and shrub in which the hares could hide. Organized hunts with bow and arrow sometimes took place. Hare meat was considered a delicacy and they were sometimes made into pets. Hares are occasionally depicted on Pannonian Samian ware and on stucco molds.

carp (Cyprius carpio L.) catfish (Silurus glanis L.) and even sturgeon (Acipenseridae)38 There are many Roman fish recipes, and given the proxim-ity of marshland and rivers to settlements it does seem likely that with the introduction of screening and flotation techniques a clearer picture of fish exploitation ponds as well as freshwater fishing customs in Pannonia will emerge. Luxury foods such as oysters, and song birds would also have bee consumed by the Roman elite in Pannonia as elsewhere. However, the latter have not yet been found anywhere in Pannonia and will not be until dry and wet screening and even some flotation is seriously and regularly employed on excavations. However, there is an example from Augusta Raurica in Switzerland where the archaeozoologist carefully sampled soil from a elegant house in the town center. Among the bones of the usual meat animals appeared remains from tiny song birds whose tongues were supposedly eaten by the discerning39

Aquincum – the town and its territorium Aquincum can be thought of as having existed in three parts: The Legionary Fortress around with the Military Town, the Civil Town and the surrounding territorium.40 Examination of the ani-mal bones from a variety of locations indicate that animal consumption customs differed according to social status and ethnic affiliation. For example

inhabitiants of ‘suburban’ villages slaughtered their own animals and butchered them in a way that would have been familiar to their Celtic ancestors. The bones from these village sites represent all body parts of the slaughtered animals indicating that they were butchered on the spot. The bones themselves display the spiral fracturing character-isic of prehistoric fractured animal bone. (Fig 2a) In the Military Town, meat was procured from a centralized butcher. Thus, body parts are selected and the bones display regular chopping marks (Fig. 2b) The presence of half-finished worked bone products shows that there were bone manu-facturing workshops scattered in the more indus-trial sections of the town. (Fig. 9) The faunal remains found in the Civil Town come mainly from older animals who must have been brought to the edge of town to be slaugh-tered once they were too old to be useful. Cattle predominates. After being slaughtered, skinned and roughly divided they were brought to centralized butchers in the town to be chopped and filleted into the parts and sizes which fit the traditional pots employed in Roman cuisine, as exemplified by Epicius. Only the house of Marcus Antonius

Fig. 2. a. Bone broken in traditional manner resulting in spiral fracture compared to b. a bone chopped by a Roman butcher

38 Bartosiewicz, L., Animal remains from the fort. In: (D. Gabler ed.) The Roman fort of Ács-Vaspuszta (Hungary) on the Danubian limes. Part II. BAR 531, Oxford, 1989, 600–623, here 614–615. Fish species found at this castrum located near the Danube river include the common carp (Cyprinus carpio L.), pike (Esox lucius L.), pikeperch (Sti-zostedion lucioperca L.) and great sturgeon (Huso huso L.). Although, this latter fish species is now found only in the lowermost reaches of the Danube it is known to have been fished extensively in medieval times in Hungary, prior to the construction of mid 20th century hydroelectric dams.

39 SCHIBLER – FURGER op. cit. 225.40 Detailed information on the layout of Aquincum can be

found in this volume in the “Topography” chapter (5.)

218 219

Victorinus41 contained the remains of piglets and lambs, reflecting his wealth and a more refined taste. Bones have also been studied from the area of the southern townwall where the so-called Fireman’s collegium headquarters42 was located. A small amphitheater north from the Civil Town was built during the AD 2nd century to hold theater performances, gladiator fights and political gather-ings. Serious excavations there in the 19th century43 by Károly Torma44 recovered a large number of remains from animals, mostly outside the walls of the arena. While bones from upper layers probably post dated the Roman occupation, three groups of bones found clustered together are worth mention-ing: dog bones by the 3rd supporting pillar, horse and oxen bones by the 7th supporting pillar and autopoia bones from cow, mule, roe deer, wolf and wild boar together with scattered bones from horse, domestic pig and dog. The presence of wild animals in exactly this spot is surely of some sig-nificance since such finds are virtually non-existent elsewhere in Aquincum. The author also suggests that some of the bones may be from European bison which could not be indentified because of the lack of comparative specimens in the University comparative collections. The bones themselves have long since disappeared. The area stretching from the eastern edge of the Civil Town to the port on the Danube was an industrial area where activities such as pottery manufacturing and blacksmithing, crafts employing fire and therefore dangerous in the close confines of the own itself, were carried out. There is a little evidence of local bone or antler working in the form of half-finished pieces. It may be presumed that glue and hide processing were also important crafts although again, little direct evidence has been preserved. A series of middle-sized (between 1 and 10 km2) villa-farms and villa-manufacturing locals were situ-

ated on the floodplain of the Danube stretching towards the Buda hills. With the exception of areas south of Gellért hill and north of Arany hill from the Buda hills to the Danube there was only a limited amount of land truly suitable for agriculture (PÓCZY 1970/2, 85,91).45 At some of these villa’s it appears that whatever meat was being eaten came from the centralized butchers of the town rather than being slaughtered on location. These villas are not to be confused with the luxury buildings located in the Buda hills which were places of relaxation for the Roman elite. One notable example is the villa of the aforementioned Marcus Antonius Victorinus which was found by chance during earth-moving work in the hills in a suburb of Budapest some 12 km from the Civil Town46 where he also had a house. While one would expect a variety of luxury foods to come from the kitchen middens of such a villa given what was found at his house in the Civil Town mentioned above, however, the site was excavated at a time when animal remains were not being carefully saved so that we can only imagine the kinds of foods which would have been consumed at this wealthy, powerful man’s table. Forts near Aquincum along the limes Aquincum with its legionary fortress occupied an important position along the Danubian limes or border of the Roman Empire for 400 years. In addition to the protection offered by the forces of the legion there were a series of forts spaced out along the banks of the Danube which played an instrumental role in the Romanization of the

41 Unpublished excavation O. Madarassy 1989–1991, Plan ?42 ZSIDI 1997/4.43 The analysis of the material was carried out by Mr. József

Szukáts and the identification of the bones by dr. Tivadar Margó

44 TORMA op. cit. 100–102.

45 On the basis of finds of inscriptions and villa remains it would seem that the estates were from 5 to 8 km2. After the Markomann wars many of the inscriptions from villa-farms are oriental.

46 Póczy, K., Aquincum: The Roman Town in Budapest, 2nd edition of the guide to the permanent exhibition, Budapest History Museum, 36. He and his wife Ulpia Spectata raised an altar to Terra Mater, a fertility goddess. A hoard of some 2236 silver coins was also found buried here, almost certainly the family fortune.

local population. The soldiers stationed at them would have brought some of the culinary tastes and traditions from their own native lands. Each of these forts was surrounded by its own settlement or vicus. Two of these are worth par-ticular mention because they have been extensively excavated and the faunal materials analyzed. Both provide an interesting counterpoint to dietary hab-its at civilian settlements in and around Aquincum itself. The fort at Albertfalva (the original Latin name is not known) held one thousand eques-trian troops Remains of meat provisions and an increased number of wild animals were found here (BÖKÖNYI 1974, 351). Domestic equids were also relatively common in the faunal assem-blege from the vicus. The remains of a number of horses were found buried around what would have been an open field near the fort. A metric analysis has been carried out on these skeletons.47 There was a civilian settlement which spread out around the fort from which we also have faunal material. Excavation has been carried out in areas northwest, southwest and west of the fort.48 The Roman fort at Campona was also located on the west bank of the Danube several km upstream from the Albertfalva encampment. The fort itself measured 178 by 200 meters and was also surrounded by a vicus. A similar situation in terms of meat consumption practices was found here as at the neighboring Albertfalva fort. The remains of a small antler tool workshop represents an outstanding find from this site. Recent excavation work and the faunal material Faunal material from the spate of rescue exca-vation either awaits analysis or is currently under analysis but the simple mass of finds means that inevitably backlogs will buildup. However, the poten-tial for comparing the faunas exploited at contempo-

rary sites of different kinds has perhaps never before existed. Furthermore, sudents have begun to take on individual site materials for their diploma work in the newly established ‘bone’ laboratory which should eventually help alleviate some of these problems. At the same time, as a result of the pressures of rapid excavation, a separate trend may be observed. On the one hand, it is sometimes neces-sary to work in the mud and cold – no chance here for the niceties of excavation techniques. On the other hand, archaeologists are adopting new recording techniques49 and screening as a response to the need to producing reports quickly. There is reason here for a great deal of optimism for the more efficient processing of the animal bone material as well. Animal keeping in the territorium around Aquincum It cannot be sufficiently emphasized that saying anything about animal keeping around Aquin-cum and the role of animals in the economy of the surrounding middle sized villa-farms is very problematic if the only source of information is faunal remains. Faunal remains directly reflect modes of consumption and only very indirectly tell us something about production traditions. Before they arrive on the archaeozoologist’s table, bones pass through numerous filtering processes: primary and secondary butchering, re-distribution to indiviual households and workshops, natural destruction and excavation methods. All of these may bias any attempt to extapolate back to the original live animal in a herd. The faunal assemblages which have been ana-lyzed to date, originate from sites in the Civil Town, a Mithraeum in the Legionary Fortress50, a cemetery51, a south of the Military Town52, the

47 LYUBLYANOVICS op. cit. 48 See in this volume the chepter “Castellum and Vicus at

Albertfalva” (5.2.2.1.)

49 For example, the Harris matrix recording technique. Regu-lar examination of macro remains of plants adds a new dimension to our understanding of the dietary customs of different peoples.

50 House of the tribunus laticlavius KOCSIS 1991.51 Budapest, Ladik street: ZSIDI 1997[1998].

220 221

area of a 1st century fort and stone building locat-ed north of the Military Town53 and two nearby military forts on the limes.54 Some of these mate-rials have been published others not. Differences between different kinds of sites may be seen in the graph. (Fig. 3) Quite a number of the bones have been identified but not analyzed so many of the remarks must rely on impressions of special features gathered during identification work. What little information can be gleaned con-cerning animal keeping practices around and in Aquincum. A hint about cattle keeping at any rate comes from the bones found at the Firemen’s Headquarters in the Civil Town. In the first place,

the overwhelming major-ity of the cattle bones recovered here come from mature animals, as is the case in gen-eral on Pannonian sites. Furthermore, where sex could be determined (on horncores and complete metapodials) most of the animals brought to Aquincum were cows. Eleven complete metatarsals, preserved in full length,55 could be used in estimating the stature of Roman Period cattle from this site. According to the small values of rela-tive proximal breadth (expressed as the % of greatest length, “Nobis

index”), all metatarsals originate from cows56 which were of small and medium size respec-tively. Only two of the horncores found at the Firemen’s Headquarters site came from bulls. None were from oxen. These proportions as well as the dearth of juve-nile individuals strongly suggest that cattle were not raised exclusively for meat. That cows were also used in draught is shown by two horncores from cows which came from a trench running along the southern wall of the Civil Town. These cores had indentations on their bases resulting from intensive yoking. A relatively few horncores

52 Budaújlak 6 Kolossy Square Unpublished excavation from E. Márity, 1993

53 Budapest, Filatorigát ZSIDI 2002/2.54 Albertfalva SZIRMAI 2002/2, 2003/1., Campona, excava-

tion of Ferenc Fülep, 1949–1957, 1960, Fülep F., Panno-nia Research in Hungary, Aquincum. In: (A. Lengyel and G.T.B. Radan eds.) The Archaeology of Roman Pannonia, Budapest, 1980, 33–56.

55 Measurements after Driesch, A., A guide to the measure-ment of animal bones from archaeological sites. Peabody Museum, Bulletin 1, Cambridge, Massachusetts, 1976.

56 Nobis, G., Ur- und frühgeschichtlicher Rinder Nord- und Mitteldeutschlands. Zeitschr. f. Tierzüchtg. u.nd Züchtungs-biol. 63 (1954) 155–194. The value of the Nobis index is greater than 24% for bulls and should be at least 21% for oxen, the presence of these latter, however, could not be identified on a morphological basis.

Fig. 3. The percentile contribution of domesticates at three different types of sites

from this same trench came from oxen which would have been valuable in hauling heavy loads as well. The withers heights of the animals from the Firemen’s Headquarters show that, with the exception of three animals, these animals were small, falling below the minimum withers height calculated for Roman breeds in Pannonia (between 120–140 cm)57 but well within the range of Celtic cattle. These measurements suggest a scenario in which local peasants brought their old dairy cows (these animals would also have been used for hauling) to the edge of the town where they were slaughtered, a messy job not suitable for the densely inhabited town. The roughly sectioned carcass could then be brought into a butcher-shop where it could be further processed. The picture differs somewhat from the medium and large Roman cattle breeds more typically found

in rural settlements related to large estates.58 The floodplain between the city and the Buda hills was large enough to support small farms but perhaps the smaller Celtic breeds would have been, by and large, strong enough to do the work on them as well as to ultimately provide a supply of meat and milk to the town. This would explain the generally small size of cattle from refuse deposits in the Civil Town. (Tab. 1) Small numbers of sheep and goat were also kept in the territorium around Aquincum. The former were certainly kept for their wools and

Tab. 1. Distribution of the size, age and withers height calculations based on the basis of the grea-test length of complete bones from the Fireman’s headquarters site in the Civil Town. Withers heights were estimated using coefficients developed by Calkin,59 Matolcsi (withers height 1)60 and Boessneck (withers height 2)61 respectively. The Nobis index reflects the sex.

Period and Size category

Age Greatestlength, mm

Proximal Breadth, mm

Nobis index, %

WithersHeight 1, mm

Withersheight 2, mm

Period 1SmallMedium sizePeriod 3SmallSmallSmallSmallSmallPeriod 4Medium sizeSmallPeriod 5Medium sizeSmall

AdultMature

MatureAdultAdultAdultAdult

MatureMature

MatureAdult

210.0240.3

215.4203.6201.5212.2210.0

244.2211.1

227.0204.5

46.047.5

43.642.041.845.044.5

51.739.0

48.641.3

21.9019.77

20.2420.6320.7421.2121.19

21.1718.47

21.4120.20

1121.41283.2

1150.21087.21076.01133.11121.4

1304.01127.3

1212.21092.0

1182.31352.9

1212.71146.31134.41194.71182.3

1374.81188.5

1278.01151.3

Mean value, mmStandard deviation, mmCoeffi cient of variation

1155.377.60.067

1218.081.80.067

57 BÖKÖNYI op. cit. 128,

58 Vörös, I., Animal bone finds from the Imperial settle-ment of Balatonaliga (in Hungarian). A Veszprém Megyei Múzeumok Közleményei. 19–20 (1993–1994) 195–214. He found that juvenile cattle had a withers height of between 120.0–130–0 cm, adult cows had a withers height between 125.4–127.0 cm, the single bull had a withers height of 137.7 and their was an oxen with a withers height of 136.7 cm. These were stong boned animals of medium to

222 223

both species for their meat. These animals had a certain economic importance as meat animals for the elite since bones from more elegant dwellings in the Civil Town show that lamb meat was also purchased. The practical Romans undoubtedly skinned these after they were slaughtered. These hides (and bones for tools) would then have been processed for leather. There is no evidence that this was a regular activity in the same way as for cattle. The relative lack of skull elements from sheep or goat also shows that the animals may even have been slaughtered at the farm and the butchered carcasses brought into the town for sale. The meat from sheep seems to have been prefered among certain ethnic enclaves in Aquincum as seen from food offerings left at a cemetery for people from Northern Italy which will be discussed in detail later.

Pig is exclusively exploited for its meat. Typi-cally, pigs are kept and fattened as household animals although the pigs may be gathered and taken out to for age rather than being penned. Pork was considered an important element of the diet, especially the meat from young animals. The large litter sizes of pig made it easier to procure this latter delicacy. An interesting aspect related to animal keeping on villa-farms concerns a small but very sugges-tive bone material from the Kaszásdűlő – Csikós street site.62 (Fig. 3) This site was discovered during ground leveling work by bulldozers lying about 1/2 km from the Aquincum Civil Town. It belongs to the category of small industrial villa-farms from the AD 3rd and 4th centuries. These more industrial type villas produced goods with slave labor such as metal objects, bricks and ceramics for the markets of Aquincum. In this case, a metal workshop or smithy was discovered adjacent to the main building. The sparse animal bones come from outside the walls of this build-ing. Considering that this villa complex was occu-pied for at least 150 years there are remarkably few bones from this part of the site. The speed and intensity of deposition was very low63. Most interesting as far as the question of animal keep-ing is concerned is the fact that the majority of the bones from food refuse were from parts of carcass low in meat quality. The bones of cattle show the unmistakable chopping marks character-istic of carcasses prepared in the butcher shops of the town. As far as it is possible to tell the cattle bones were mostly from small local animals. While sheep, goat and pig may have been kept at the farm in small numbers it is clear that someone else other than the people working the smithy were getting the higher quality meat cuts. Common people in any case were probably more dependent on plant foods such as grains, legumes, fruits and vegetables in their daily diet.

large size chiefly used in plowing and transport. Interest-ingly enough an actual iron plow blade was found in the excavations. It was badly worn with a shovel shape typical for heavy plows of this period.

59 Calkin, V., Izmenchivost metapodii i eo znachenie dla izuchenia krupnogo rogatogo skota. (Metapodial varia-tion and its significance for the study of ancient cattle). Biull. Mock. Obshch. Ispit. Prirodi, Otdel. biol. 65 (1960) 109–126.

60 Matolcsi, J., Historische Erforschung der Körpergrösse des Rindes auf Grund von ungarischem Knochenmate-rial. Zeitschrift für Tierzüchtung und Züchtungsbiologie 63 (1970) 155–194.

61 Boesseck, J., Ein Beitrag zur Errechnung der Widerristhöhe nach Metapodienmasse bei Rindern. Zeitschrift für Tier-züchtung und Züchtungsbiologie 68/I (1956) 75–90. These works have been summarized Bartosiewicz, L., Sexual dimorphism of long bone growth in cattle. Acta Veternaria Hungaricae 32/3–4 (1984) 135–146. The coefficients devel-oped by the first two authors for Kalmuk and Hungar-ian Grey respectively result in somewhat smaller estimates than the values recommended by Boessneck. This is due to the fact that while Calkin and Matolcsi used series of unimproved cows, Boessneck worked with the mixed group of reference animals that also included modern animals whose distal extremity segments are relatively shorter even in cows.

62 The material from this site has been analyzed but not pub-lished by the author. All the information in the following section come from these notes.

Suprisingly, only a single dog bone was found in the refuse bone of the villa. At the same time, there is other bone material from the stone build-ing located near a roman cross-road, on the mod-ern Kolossy square, 1/2 km south of the Military town amphitheater, where a good number of dog bones was recovered. These were Airdale terrier-size animals who must have been used as guard dogs. Dog bones are encountered from time to town at sites in the Civil Town. Most of these are small-medium size dogs with thin legs. Since there is no evidence for dog’s gnawing bones in the faunal assemblages from the towns, it is clear that dogs did not run free in Aquincum’s streets. There is one bone from a large dog, German shepherd size, and a few from lap dog size animals. Only one cat bone, an extremity bone, has been recov-ered from all the Aquincum faunal assemblages. However, as can be seen on the compara-tive graph (Fig. 3) there were large numbers of ‘horse’ bones present in the Kaszásdűlő-Csikós street villa. While none of the bone was from ass it is not possible to say with absolute certainty whether the elements present come from horse or mule.64 These animals were certainly used to carry or haul goods around and to and from the villa. They were not eaten and the fact that they died at the villa is shown by the presence of bones from all parts of the skeleton. During the course of the analysis of animal bones from a small Avar settlement65 faunal material from a small 2nd–3rd c. bath building from settlement north of the Military Town were also identified.

Material from the nearby 1st c. timber fort has not been selected and awaits identification and analysis. Interestingly, the proportion of domestic animals is remarkably similar to what was found at the Kaszásdűlő-Csikós utca villa with the meat coming from the centralized butcher shops of the town and relatively more equids, perhaps mules. Finally, a few general words should be said about bird keeping and fish exploitation around and in Aquincum. Bird bone seems to have to come from whole carcasses, that is, bones from all parts of the carcass may be found on sites where finer excavations techniques were employed. However, there is no way of telling whether birds were kept around the town and smaller settle-ments since they represent easily ‘movable’ goods. There have been no finds of eggs, even from burials as grave offerings, although we know that eggs were a part of the diet. There is no evidence for carp farming or other form of fish keeping in ponds from Aquincum. Altogether, a grand total of two pike bones have come to light from a Celtic-Roman settlement west of the Civil Town and a settlement (as well as one sturgeon bone from this same area) just south of the Military Town respectively. However, numerous Cyprinidae (carp family) remains were recovered from a site along Bécsi Street where screening was regularly carried out by the excavat-ing archaeologist. The lesson is clear. These were all most likely fished directly out of the nearby Danube river by this Romanized native population to supplement the meat diet from domesticates. Animals in the diet One of the aspects of the faunal material from the more recent excavations at Aquincum is the wide variety of sites it derives from. Because assemblages exist from such variable sources, in principal it should be possible to say quite a bit about differences in the meat diet between differ-ent social groups. Cattle, sheep/goat, pig and chicken provide the basis of the meat diet. (Fig. 1) The informa-tion available to date makes it clear that in the

63 This is expressed by the restricted area where bones were found and their weahered surfaces showing they were not bur ied immediately as was typical within the town of Aquincm.

64 WHITE op. cit. 295. Larger measurments must surely have come from horses but the bones in the smaller range may have come from mules produced by crossings between mares and jack-donkeys. Crossings between jennies and stallions result in smaller hinnies.

65 Daróczy-Szabó L., Pásztorhagyományok a Filátorigáti településen. Régészeti szakdolgozat, Eötvös Loránd Egy-etem, Budapest (2002), personal communication.

224 225

Aquincum towns, remarkably little game was consumed. Fish may have been impor-tant for certain dishes or as a supplement to the diet for all but that is still unclear. Horse and dog were not eaten at Aquincum. In addition to the fact that bones from these two species bones are very rare they are usually unbroken and show no sign of butchering although horse bone may have skinning marks on the extrem-ity bones. Differences seem to exist between the various parts of Aquincum, the villa-farms and the close by military forts of Albertfalva and Campona, although a great more work needs to be done to clarify why these difference exist. The Civil Town Three sites from the Civil Town serve to illus-trate the eating habits of better off people who lived there in the end of the 1st, 2nd and 3rd centuries.66 Table 2, which shows the skeletal part distribution from the Civil Town shrine site provides a typical example of refuse bone in the town.67 (Tab. 2)

At all three sites cattle was most important, comprising up to 70% of the everyday diet, as for example, at the Firemen’s Headquarters. (Fig. 8) The same picture is also shown by a small material from the canabae of the Military Town (BÖKÖNYI 1974, 351) where cattle domi-nates.68 Although there are bones from every part of the skeleton, they tend to be chopped up with heavy metal tools. (Fig. 4) These large bones can be chopped lengthwise and even the dense short bones are chopped across. (Fig. 5) The butchers working in the meat shops of the Civil Town must have been immensely strong. The idea is to

66 The Civil Town shrine was excavated in the early 1990’s but has not been published due to the death of the exca-vator, Erzsébet Marity, Plan The dwelling of Victorinus was excavated slightly earlier (see note 41.). While the archaeological and faunal material has been identified actual analysis has been put off until the stratigraphic clarifica-tion of the faunal material is complete. Both excavations were carried out prior to planned reconstruction of the 2nd–3rd century structures. The analysis of bones from the Firemen’s Headquarters (ZSIDI 1997/4) is complete.

67 Because this end of the first beginning of the 2nd century material has not been published before, skeletal part distribu-tions are also provided here. Only cattle remains were found

Tab. 2. Skeletal Part distribution at the Civil Town Shrine and Adjacent Refuse Pit

Pit Shrine

cattle cattle sheep goat sheep/goat pig dog hen human

horn coreskull fragmentMandibulaVertebraeRibScapulaHumerusRadiusUlnaMetacarpusPelvisFemurTibiashort boneMetatarsusPhalanges

348110

1

1

3544121242

3253104

1111

11

2

1

131

21

12124111

1

1 1

NISP total 28 64 4 2 3 21 1 1 1

in the pit next to the shrine and over 1/3 of the bones was made up by rib fragments. While cattle also dominated in the assemblage from the shrine, bones from this animal were only twice as many as from pigs. Anatomically, the cattle remains not only included remains of high meat value: numerous short bones, metatarsals and phalanges of the metapodium belong to the commercially less important of the animals. Once the shrine ceased to function the area was leveled and filled in with debris from the nearby shrine complex. The bones here may very well represent the remains of food offerings.

pot-size cuts of meat for stews and to allow the rich tasty nutrient marrow to get into the food.69 Sometimes the butchers chopped the meat from the leg bones to produce fillets. The evidence for this are overlapping shallow chop marks running

down the diaphyses of long bones of cattle legs, particularly humerus and femur. (Fig. 6) One of the most common bones from cattle are ribs which have cut off from the vertebrae. (Tab. 2) Sometimes these ribs even have long striations where diners scraped the last bits of tasty meat from them. Finally, a total of three complete scapulae from cattle have been recovered from the Civil Town. All have a hole through the blade where the shoulders were hung, probably for smoking. (Fig. 7) Curiously enough there is no evidence of such smoked meat being eaten in the Military forts studied here. This latter is strange because such bones are relatively common at Roman military sites in England and Holland.70 Smoked meat exports from Pannonia and Gaul were well known in the Roman Empire and these specimens show it was consumed locally as well. Most of the bone remains from cattle come from adult or mature animals which must mean that Romans, even the better off ones, tolerated rather chewy meat. The bones from sheep are not particularly common at these three sites in the Civil Town, amounting to between 10 and 15% of the total bones. (Fig. 3) Goat bones are even rarer. At the Firemen’s Headquarters there even seems to be a slight decline in their numbers which mirrors the situation in the province. With the exception of the faunal material from Victorinus’ dwelling, not quantified here, the bones are from adult indi-viduals. Over half the sheep or goat bone from this house however, did not have fused epiphyses. In other words they came from lambs, a much greater delicacy and therefore more expensive. The bone parts found in the Civil Town shrine come from meaty regions, the pelvis and ribs. (Tab. 2 and Fig. 3) Pig in the Civil Town shows a more variable picture. It is dangerous to draw conclusions on 68 While the numbers here are truly very small and the mate-

rial was not screened the trends in species proportions are repeated.

69 BARTOSIEWICZ op. cit. 630–632. also see: Grant, A., Some Observations on Butchery in England from the Iron Age to the Medieval Period. Anthropozoologica Premier Numéro Special, Paris, 1987, 53–58.

70 LAUWERIER op, cit, 195. Also see: Izard, K., 1997, 369. At the military fort at Birdoswald on Hadrian’s Wall 17 of the 65 cattle scapulae had such holes in them. Izard also interprets this as a sign of smoking and further that the shoulders were smoked at the fort.

Fig. 4. Roman iron axe of the kind that might have been used to chop through bones by but-chers in the town

Fig. 5. Chopped bones and boneswith marks of filleting

226 227

the basis of such small numbers but it is safe to say that pig was eaten slightly more often than mutton in the town. (Fig. 3) The percentages of pig in the Firemen’s Headquarters appear small because cattle is so dominant. This dominance is not entirely related to gastronomic activities in this building as we shall see later. The bias toward cattle distorts the percentages. The animals butchered in the shops were generally young ani-mals although bones from new born or suckling pigs have not yet been recovered, perhaps because of sampling problems. All parts of pig were eaten as can be seen by the distribution of body parts from the Civil Town shrine site. As noted previ-ously Apicius’ recipe book is dominated by pork dishes. Pork was generally a favored Roman meat. However, it can be said that one encounters pig bone in Aquincum assemblages less often than would be expected based on other provincial Roman sites.71 It is very tempting to suggest that

this was a reflection of the dietary preferences or prohibitions of the Syrian soldiers and merchants who came to Aquincum in the AD 3rd and 4th centuries. The people living in the Civil Town ate chicken regularly. Probably the number of bones from this species would be found more often with screen-ing. (Tab. 2 and Fig. 3) Their bones are slightly more common in the refuse from the Victorinus house. Bones from goose were also found there. The diet of the Romanized native population Unfortunately, not very much of the large bone material from settlements72 of the Romanized native population has been analyzed although some of it has been identified.73 Beef certainly dominated in the diet but species proportions compared to sites in the towns are not available. One thing is clear. The people living in villages outside Aquincum in its territorium brought animals into the habita-tion area to slaughter them. All parts of skeleton appear but even more significant, the bones rarely display any trace of the chopping marks typical of the town material. Furthermore, the long bones at these village sites have been broken in the old traditional manner which produces a spiral frac-ture on the diaphysis. A series of huts dating from the first period of the military occupation74 was discovered in an area just south of what was to be the canabae and the area of the future Military amphitheater. The faunal material from these huts was striking in its similarity to that found in the villages of the native population which was later moved into this area. The soldiers at this point in the pro-

Fig. 6. A cattle bone bone with scalloped marks of filleting

Fig. 7. Cattle shoulder-bone with hole where the shoulder was suspended during smoking.

71 For example, Augusta Raurica in Switzerland where pig percentages can reach 35% as opposed to 20% at Aquin-cum. Schibler– Furger 1988, 175.

72 44–46 Bécsi Road BERTIN 1998, BERTIN 1999 73 This is largely because the stratigraphic situation has not

been clarified because of the tragic death of the excavator, Erzsébet Márity.

74 Excavation of E. Márity, 1990 Budaújlak, Lajos Street and Budaújlak Csemete Street, Plan

cess of Romanization seem not to be relying on military supplies but depended on the local Celtic population for their meat. Not strictly related to food is the find here of a wolf crania from one of these huts. It is tempting to think that the skin and cranium of this wolf ultimately decorated the uniform of the Roman living in this hut. The Military Diet There are two Military forts south of Aquin-cum, Albertfalva and Campona, for which faunal analyses exist (BÖKÖNYI 1974, 351; CHOYKE 1993, 133–135; CHOYKE 1994–1995, 51–60).75 The forts themselves mirror the situation found elsewhere in the province. Cattle bones at both sites seem to have been brought into the castrum area in a processed form. Sheep or goat and pig seem to occur in roughly equal numbers within the fort however pig bone was much more frequently encountered in the vicus surrounding Albertfal-va. (Tab. 3) Soldiers at both forts seem to have supplemented their meat diet with game including wild cattle (Bos primigenius Boj.), red deer (Cervus elephus L.), roe deer (Capreolus capreolus L), hare, wild birds and fish.

Animals and Draught work Cattle were used in plowing and hauling work. At Aquincum there are cow horncores from the Civil town mentioned above which are direct proof of yoking with inden-tations at their bases caused by long term contact with the yoke. The horncores from oxen also show that this domestic species would have been used in plow-ing and hauling heavy loads where speed was not needed.

Fig. 8. Lithograph of a panel from Trajen’s column in the Forum in Rome showing mules being used for military transport.

75 VÖRÖS op. cit.

Tab. 3. The Species Distribution from the Albertfalva Vicus. all wild animal NISP is smaller than 10%

Periods1 and 2 mixed

Period2

Period3

Period1 to 5 mixed

Modern or

mixed

CattleCaprineSheepGoatPigHorseDogDomestic hen

785224562653

18112

181

1

351

45203125412

1782

1111

NISP domestic 226 51 9 101 49

Red deerRoe deerWild pigSturgeonSturgeon

11

11

1 1

11

NISP wild 4 1 2 1

NISP total 230 53 9 103 50

HumanRodentFishUnioNon-ident.

2119

1

15 2 6

2

3

228 229

Those settlements discussed above where draught and move-ment of men and goods would have been important include the Kaszásdűlő-Csikós utca villa where an extraordinarily high proportion of the bones came from smallish equids, probably mules. This picture is mirrored at another unpublished indus-trial villa at Mocsárosdűlő 2 km from the Kaszásdűlő Csikós utca villa site.76 This villa seems to have produced bricks. The yard of this second villa was strewn with the complete bones of horse and mule. Complete equid bones were also relatively common in the faunal materi-als of both of the Military forts discussed above, Albertfalva and Campona. These animals were not eaten but the skins were probably removed and so the carcasses remained close to the settlements. Mules and mule breed are known to be important in the Roman military system (Fig. 8) in the forum in Rome showing mules drawing a wagon in a military setting). Horses were ridden by officers and calvalrymen. They may be seen on numerous gravestones of soldiers with proudly arched necks compared to the more plebeian look of mules shown pulling wagons on family gravestones from around Aquincum. Goods from animals There are three by-products made from slaugh-tered animals whose traces can actually found in Aquincum. These include hide and hide process-ing as well as bone, antler and ivory carving. A large ditch excavated by the southern wall of the Civil Town contained the bones of many

animals, mostly cattle. Typically for some kind of hide processing area are the accumulations of horncores and metapodials which is what was found here, mixed with some other food debris.77 There has been some speculation that this was a hide processing area which was in use during one of two periods after the Civil Town had been attacked and before the town had really recovered. At the same time, there is suggestive and similar material coming from the Firemen’s Headquarters which is located by the southern wall. The building functioned as a collegium for almost 400 years and the faunal material in it is quite separate in terms of its composition from that in the trench. Aside from the food refuse bone already discussed in this paper, there are numbers of metapodia and connected foot

Fig. 9. The percentile contribution of animals at the Firemen’s’ Headquarters.

76 Mocsárosdűlő: E. Márity excavation, Plan ?, ; Kaszásdű lő- Csi kós utca: ZSIDI 1991.

77 Serjeantson, D., Animal Remains and the Tanning Trade. In: Diet and Crafts in Towns: D. Serjeantson and T. Waldron eds., BAR British Series 199; Oxford (1989) 129–146, here 136.

bones as well as skulls with attached horn cores which is associated with hides. In fact, one of the means of fire fighting (SZILÁGYI 1984, 51, 68) included raw hides soaked in water or vin-egar and raw animal hides filled with chaff and soaked. These skin bags were called cento. It is very likely that the extremity bones would have been left in the legs of these skins for easier handling. As the skins would have gotten burned or worn out they would have been discarded – perhaps into the very trench previously thought to be part of a hide processing operation. The bones from the Firemen’s Headquarters itself would be from the last phase of each building stage. (Fig. 9) Another type product regularly manufactured in Roman towns are objects made from bones and antlers. By the Roman period, bone was carved by hand or on a lathe into ornaments, fittings, bits of inlay and parts of complex objects in centralized workshops which were undoubtedly tied in to the butchershops which provided them with the appropriate raw material. 78 While the concrete location location of these workshops is unknown, remains of debitage or half-finished pieces indicates their general locations. (Fig. 10) Horse and cattle autopodial long bones were most often worked in addition to red deer ant-ler. A brief review of the Roman objects from Aquincum79 and the Roman period objects in the National Museum of Hungary80 show that they range from the very elaborate to crudely carved imitations made at home from available bones. This stands in interesting contrast to the bones from the Roman fort of Ács-Vaspuszta which more closely resembled prehistoric or Sarmatian

tools81. There were few ornaments and the tools retained many of the features of the bone such as the epiphyses. Although these objects from Aquincum are often so heavily worked that all identifying features of the raw material are removed from the bone, new research has shown that some of the microscopic features related to the structure of the particular material are retained.82 On this basis it has been determine that at least 30% of the objects are made from red deer antler. Red deer was appar-ently not hunted at Aquincum but, as mentioned 78 Maltby, M., Urban Rural Variations in the Butchering of

Cattle in Romano-British Hampshire. In: (D. Serjeantson and T. Waldron eds.) Diet and Crafts in Towns., BAR British Series 199, Oxford, 1989, 75–106, here 89.

79 This research into the Aquincum bone tools is ongoing by the author and Mária Bíró.

80 Bíró, M., The Bone Objects of the Roman Collection. Catalogi Musei Nationalis Hungarici Series Archaeologicica II, Budapest, 1994.

Fig. 10. Find spots of debitage from antler and bone working may mark the general area of for-mer workshops

81 Choyke, A., Worked animal bone at the Sarmatian site of Gyoma 133. In: (ed. by A. Vaday) Culture and Landscape Changes in South-East Hungary, Budapest, 1996, 307–322. As well as Choyke op. cit. 1989, 624–632.

82 DESCHLER-ERB op. cit. 33–47. A microscope with mag-nification of up to 80 X should be sufficent to make these identifications.

230 231

previously, there must have been organized gath-erings of its antler each Spring. Interestingly, half finished pieces made from antler do appear scat-tered throughout Aquincum but no concentrations of workshop debris have yet come to light here except for the antler workshop discovered in the military fort at Campona (VÖRÖS, 1989, 85).83 It contained antler burrs with the beam sawn off as well as plaques carved from the beam and ready for further working, possibly into comb sections. Animals in rituals Animals played their own part in religion and ritual at Aquincum. From time to time a site is excavated where the bone material is clearly related to this type of activity. In the 1980s the house of the Tribunus Laticlavius was excavated in the Military Town. Attached to this luxury house was a mithreum. The cult of Mithras was brought from Persia originally by the influx of eastern people mentioned previously. One of the scenarios of this cult was the image of Mithras, representing light, killing and therefore over-coming the bull, representing dark. Bulls were regularly sacrificed in the underground shrines dedicated to this God however, it seems the prac-tical Romans ate the animals afterwards because accumulations of cattle bones are never found near these Mithraeum. However, a number of extremely interesting animal bone remains were recovered from a sealed-off context – the altars by the northern wall and in particular the snake altar within this particular cult place. The animals found here included domestic hen, pig, domes-tic goose, food remains of cattle, hare and fish. Most significant were the 16 chicken bones found together.84 As previously mentioned, a roe deer

(Capreolus capreolus L.) was possibly sacrificed at the altar of Nemesis by the amphitheater of the Civil Town. Other ritual contexts with related bone finds include three skulls from large dogs found under a threshold at the stone building excavated at Kolossy Square south of the Military Town. The building included human burials within the walls, a distinctly non-Roman custom. This and some of the finds led the excavator to suggest that this building had been occupied by Romanized Celts who still followed some of the old ways.85 The cemetery/shrine complex located near the northern edge of the Military Town was first excavated in 1984 (ZSIDI 1997[1998]). The small sample of bones from food offerings placed in the graves and offered at the associ-ated shrine indicated that this area was used by people from northern Italy. The offerings seem to be the remains of stew-like meals. As opposed to elsewhere in Aquincum the remains of sheep or goat dominate over pig. This patterning is probably related to some kind of food preference. Since other grave offerings (CHOYKE 1998/2, 150) clearly point to northern Italy it is safe to say that people taking part in the graveside rituals brought foods which would have been eaten by the deceased in life The analysis of the bones and archaeological material offers a rare but happy example of the use of faunal and material culture data to strengthen the interpretations concerning the ethnicity of of the people buried in this cem-etery parcel. Finally, as increasing numbers of graves are excavated, finds of special bones are being recov-ered such as the goose foot bone tucked in the swaddling of a baby buried in the late 4th century, located on Vályog Street near the former Military Town.86 The symbolic meaning of this bone has been lost to us.

83 Vörös I., Campona-Nagytétény római tábor állatcsontma-radványai [The Animal Bone Remains from the Roman Fort at Campona-Nagytétény] FolArch 40 (1989) 75–118, here 85.

84 Vörös I., A Tribunus Laticlaviusok háza az Aquincum 2–3. sz.-i legio-táborban – A mithraeum állatcsont-leletei [The

House of the Tribunus laticlavius in the legionary fortress of 2nd–3rd century Aquincum – the animal bone remains from the Mithraeum]. BudRég 28 (1991) 118–132, here 132.

85 Erzsébet Márity, personal communication.86 Vályog Street, MADARASSY 2002

Conclusions The work on the many sided relationship between the human population and their animals is ongoing. As more and more material is studied, it is expected that the picture of how animals were used in the town and on the villa-farms in the surrounding territory. The various social classes and the different ethnic groups that made up the pluralistic society of most Roman towns also left their mark on dietary traditions and, with luck, patterns in species proportions and body part distribution will emerge which can be related

to discrete parts of the town and the settlements surrounding it. Altogether, animals were an integral part of Roman life. A better understanding of the way they were exploited and viewed at this particular site will help us to escape the ‘ideal’ and nor-mative picture offered by antique authors and develop a three-dimensional view of how Romans and the people they conquered adapted, lived together and thrived for almost 400 years in the Pannonian environment.87

Alice M. Choyke

87 The author would like to gratefully express her thanks to Dr. László Bartosiewicz for his truely untiring efforts to weed out inconsistencies and other errors in the text. Thanks as well to my colleagues in the Aquincum museum who have provided me with material and explanations. To all of you, many thanks.

7. KUNSTDENKMÄLER AUS DEN JÜNGEREN FORSCHUNGEN