7 Bird remains from Late Medieval and Postmedieval sites in Brussels, Belgium

Transcription

1 Part I Birds and the provision of food 7 Bird remains from Late Medieval and Postmedieval sites in Brussels, Belgium Sofie Thys and Wim Van Neer Royal Belgian Institute of Natural Sciences, Vautierstraat 29, B-1000 Brussels, Belgium and Katholieke Universiteit Leuven, Laboratory of Animal Diversity and Systematics, Ch. Debériotstraat 32, B-3000 Leuven, Belgium; wvanneer@naturalsciences.be Abstract The avifauna is presented of nine sites within the town of Brussels that are dated between the 13th and 20th centuries. The bone material consists of consumed birds and of individuals that can be interpreted as urban scavengers. For the various sites, the proportion of birds versus mammals is considered as well as the frequency of wild versus domestic birds. The proportion of the domestic species is also documented through time. The comparison of various types of sites shows status related trends: noble sites dated to the Late Medieval and Early Modern periods seem to be characterised by high percentages of wild birds and wild mammals, combined with low proportions of, probably domestic, goose. Finally, some remarks are made on the provenance of the birds and a few special finds are described. Key words Archaeozoology, birds, consumption, status, cockfighting Introduction Rescue excavations carried out in Brussels since the 1990s yielded abundant faunal remains, including bird bones, that permit a discussion of the role that birds played in this town during Late Medieval and Postmedieval times. The avifauna of the nine sites considered in the present overview, comprises a total of 2832 identified bird remains and consists of hand-collected material. The fauna from four of these sites has not yet been published, whereas the data from the five others was compiled from the literature (see below). The sites date between the 13th and 20th century AD and are all located in the central part of Brussels, the so-called Pentagon (fig. 1). The faunal assemblages were dated on the basis of the archaeological material with which they were associated. Certain sites yielded a series of successive, smaller, chronological units, while from others only a single, sometimes broadly dated, assemblage is available. Most of the sites are of a secular nature, but there are also religious and noble contexts. This paper will focus on the role of birds in the consumption pattern of the inhabitants of Brussels. Additionally, the provenance of the birds and their possible place in the urban environment will be investigated. Finally, some special finds will be discussed. Material Below, a brief presentation is given of each site and its chronology, followed by a short description of the assemblages considered in the present overview. The identified bird remains are listed in tables 1 to 9. In several cases the numerous chronological units distinguished by the archaeologists were retained during the analyses. In some cases, however, for clarity material is combined into larger phases in the tables provided here. The bird remains from the site of Hoogstraeten mainly derive from refuse layers and can be grouped into three broad chronological phases (S. Modrie, pers. comm.), i.e. 13th-15th, 16th-18th and 19th-20th centuries AD (table 1). From the end of the 14th century onwards this quarter of town was inhabited by members of the nobility (Van Eenhooge & Celis, 1988; Van Eenhooge, 1999). The court of Hoogstraeten reached its maximal extension at the beginning of the 16th century when several Late Medieval buildings were merged and the complex was inhabited by the count Antoine de Lalaing (Modrie, 2000). Although subsequent phases of plundering and destruction by fire occurred, the court remained

2 72 Part I Birds and the provision of food Figure 1. Location of the nine sites within the Pentagon of Brussels: 1. Hoogstraeten; 2. Zavel; 3. Zuidstraat; 4. Rijke Klarenwijk; 5. Arme Klarenwijk; 6. Eenmansstraat; 7. Sint-Katelijnewijk; 8. Treurenberg; 9. Dinantstraat. occupied by the nobility until the first quarter of the 19th century. The only other site that potentially contains refuse from noble consumers is Zavel, with bird bone assemblages dated between the late 13th and 18th century AD (table 2). The faunal remains from the site Zavel can be grouped into three broad chronological units (Degraeve, 2001). The oldest material comes from layers dated mainly to the second half of the 13th to the first half of the 15th century. During this period the area was the property of noble inhabitants, of which one family founded at the end of the 14th century a hospice for poor elderly people (Henne & Wauters, 1968). The site also has occupation levels dating to the 16th-18th centuries, including a refuse pit dated to the 16th century. During this period, the quarter was inhabited by a high class population that built large private manors. Finally, there are a large number of destruction layers and a surface level that mainly yielded material from private buildings occupied during the 19th and 20th century. These contexts have not been considered here because there were almost no bird bones present. Faunal remains from a religious context were found in the Zuidstraat, an area where the remains of a tower belonging to the first town wall were discovered. The upper part of a ditch fill (Nachtergael, 2001: p. 78) yielded ceramics, faunal and botanical remains dated between the 14th and 16th century AD that were derived from the kitchen of the nearby Carmelite convent. The bird bone sample, identified by Gautier (2001a), consists of 92 identifiable bones (table 3). Excavations have also been carried out in the Rijke Klarenwijk and Arme Klarenwijk, the quarters of the Rich and Poor Clares, that were inhabited from the early 14th century until the 20th century AD. Eleven chronological units were retained during the initial faunal analyses at the Rijke Klarenwijk (Gautier, 1995a), of which seven, that are relatively well dated, have been retained here. Those assemblages were in turn combined into an early phase (second half of the 13th first half of the 14th century) and a later phase (16th first quarter of the 17th century) (table 4). The faunal remains from the latter phase mainly came from recycled material, brought in from elsewhere and used during building activities for the monastery. It is therefore doubtful whether this material can be considered as consumption refuse of an ecclesiastic nature, an interpretation that seems to be supported by the archaeological material, which yielded only limited evidence of monastic life (De Poorter, 1995: p. 173).

3 7 Bird remains from Late Medieval and Postmedieval sites in Brussels, Belgium 73 Table 1. Bird remains from the site of Hoogstraeten, grouped into three chronological phases. centuries AD Wild birds Grey heron (Ardea cinerea) Common teal (Anas crecca) Tufted duck (Aythya fuligula) Ducks (Anatinae sp.) Swan (Cygnus sp.) Black kite (Milvus migrans) Black or red kite (Milvus migrans / M. milvus) Grey partridge (Perdix perdix) Wader, size of Eurasian golden plover (Pluvialis apricaria) Northern lapwing (Vanellus vanellus) Eurasian woodcock (Scolopax rusticola) Common wood pigeon (Columba palumbus) Eurasian jackdaw (Corvus monedula) Songbird, size of Eurasian blackbird (Turdus merula) Fowl (Gallus gallus f. domestica) Goose (Anser anser f. domestica) Duck (Anas platyrhynchos f. domestica) Pigeon (Columba livia f. domestica) Turkey (Meleagris gallopavo f. domestica) 1-1 Total Table 2. Bird remains from the site Zavel, grouped into two chronological phases. centuries AD Wild birds Grey heron (Ardea cinerea) 1 2 Swan (Cygnus sp.) 1 - Grey partridge (Perdix perdix) 3 - Eurasian woodcock (Scolopax rusticola) 1 - Fowl (Gallus gallus f. domestica) Goose (Anser anser f. domestica) 2 - Pigeon (Columba livia f. domestica) - 3 Turkey (Meleagris gallopavo f. domestica) - 1 Total Table 3. Bird remains from the site Zuidstraat, belonging to a single chronological phase (14th-16th c. AD; layer 18 in Gautier, 2001a). centuries AD Wild birds Grey partridge (Perdix perdix) 1 Eurasian woodcock (Scolopax rusticola) 1 Shanks & Sandpipers (Tringa sp.) 1 Carrion crow/rook (Corvus corone/corvus frugilegus) 1 Magpie (Pica pica) 1 Fowl (Gallus gallus f. domestica) 53 Goose (Anser anser f. domestica) 29 Duck (Anas platyrhynchos f. domestica) 4 Turkey (Meleagris gallopavo f. domestica) 1? Total 92 Table 4 The bird remains from the Rijke Klarenwijk (Gautier, 1995a) grouped into an early phase (second half of the 13th first half of the 14th century; phases IB, IC and IIA of De Poorter, 1995) and a later phase (16th first quarter of the 17th century; phases IIb, III, IV and V of De Poorter, 1995). centuries AD 13B-14A 16-17a Wild birds Common teal (Anas crecca) - 1 Eurasian wigeon (Anas penelope)/ Northern shoveler (Anas clypeata) - 2 Swan (Cygnus sp.) - 4 Common goldeneye (Bucephala clangula) - 4 Western capercaillie (Tetrao urogallus) - 1 Grey partridge (Perdix perdix) - 2 Common coot (Fulica atra) - 1 Little owl (Athene noctua) - 1 Tawny owl (Strix aluco) - 1 Carrion crow (Corvus corone) - 1 Fowl (Gallus gallus f. domestica) Goose (Anser anser f. domestica) Duck (Anas platyrhynchos f. domestica) 9 31 Pigeon (Columba livia f. domestica) 1 4 Turkey (Meleagris gallopavo f. domestica) 1 1 Total

4 74 Part I Birds and the provision of food Table 5. The bird remains from the Arme Klarenwijk. The material has been grouped into six chronological units. centuries AD 14 14B-15A 15B-16A 16B-17a 17b-18A 18B-19a Wild birds Little grebe (Tachybaptus ruficollis) White stork (Ciconia sp.) Common teal (Anas crecca) Ducks (Anatinae sp.) Swan (Cygnus sp.) Black or red kite (Milvus migrans / M.milvus) Eurasian woodcock (Scolopax rusticola) cf. European turtle-dove (Streptopelia cf. turtur) Carrion crow/rook (Corvus corone/corvus frugilegus) Songbirds (Passeriformes) Fowl (Gallus gallus f. domestica) Goose (Anser anser f. domestica) Duck (Anas platyrhynchos f. domestica) Pigeon (Columba livia f. domestica) Total Table 6. Bird remains from the site Eenmansstraat (Pigière, 1997), belonging to a single chronological phase (end 13thbeginning 15th c. AD). centuries AD Wild birds 13d-15a Swan (Cygnus sp.) 1? Fowl (Gallus gallus f. domestica) 131 Goose (Anser anser f. domestica) 50 Duck (Anas platyrhynchos f. domestica) 3 Total 185 The site of the Arme Klarenwijk encloses a part of the former Poor Clares convent, namely the whole chapel and a few cloister-buildings, as well as ten small private houses adjacent to the chapel (Claes, 2006). Most of the fauna was found in closed contexts, such as cesspits, that belonged exclusively to the private houses. This fauna should therefore be considered as secular. The archaeological material allowed retention of six chronological units between the 14th and 19th century (table 5). Although sieved material is available, only the hand-collected bones are considered here. This is done because all the other bird bone assemblages brought together in this contribution were exclusively sampled by hand. It is worth noting that the 4 and 2 mm sieve fractions yielded only a small quantity of identifiable bird bones. These include a few bones of thrushes (Turdidae), other songbirds (Passeriformes) and a few elements of small waders. This observation seems to be in agreement with the findings of Ervynck (1993) who stated that the bird species spectrum on historic sites does not change drastically when sieving is practised. The remaining four sites are clearly secular. The site Eenmansstraat t yielded archaeological traces of habitation between the 13th and 20th century (Diekmann, 1997), but the only bird bone sample of sufficiently large size (Pigière, 1997) comes from contexts dated between the end of the 13th and the beginning of the 15th century (phase II in Diekmann, 1997) (table 6). From the site Sint-Katelijnewijk faunal remains are available dating between the 14th and 17th century. Of the ten contexts analysed by Gautier (1995b: 191) eight have been retained for the present study. They can be grouped into four chronological units (table 7). The oldest material represents a rural phase dating to the second and third quarter of the 14th century (Degré, 1995). Another assemblage dates to the second half of the 14th first quarter of the 15th century and corresponds to the onset of urbanisation in this area. There is also material from a subsequent phase, dating to the 15th century, during which building activities took place and craftsmen s workshops were installed. The fourth assemblage dates to the 16th-17th century, a period during which the area prospered as a result of brewery activities. The faunal remains of Treurenberg date between the 14th and 20th century, but here only the largest sample is retained that can be more narrowly dated to the late 16th early 17th century AD (Degraeve, pers. comm.) (table 8). The Dinantstraat, finally, yielded material from the 13th to 20th century (De Poorter, 2001). In the initial publication (Gautier, 2001b: pp ) nineteen assemblages with bird bones

5 7 Bird remains from Late Medieval and Postmedieval sites in Brussels, Belgium 75 Table 7. Bird remains from the Sint-Katelijnewijk (Gautier, 1995b). The material, from eight assemblages ranging in date between the 14th and 16-17th century, has been grouped into four larger units. centuries AD 14b-14c 14B-15a Wild birds Grey heron (Ardea cinerea)? Garganey (Anas querquedula) Northern shoveler (Anas clypeata) Swan (Cygnus sp.) Grey partridge (Perdix perdix) Tawny owl (Strix aluco) Fowl (Gallus gallus f. domestica) Goose (Anser anser f. domestica) Duck (Anas platyrhynchos f. domestica) Pigeon (Columba livia f. domestica) Total Table 8. Bird remains from the site Treurenberg, belonging to a single chronological phase (late 16th early 17th century AD). centuries AD 16d-17a Fowl (Gallus gallus f. domestica) 51 Duck (Anas platyrhynchos f. domestica) 2 Turkey (Meleagris gallopavo f. domestica) 6 Total 59 Table 9. Bird remains from eight well-dated chronological units of the Dinantstraat (Gautier, 2001b). centuries AD 13 13B-14A 14B-15 16d 17A 18A 19A 20B Wild birds Grey heron (Ardea cinerea) Common quail (Coturnix coturnix) Grey partridge (Perdix perdix) Eurasian woodcock (Scolopax rusticola) Black-headed gull (Larus ridibundus) Thrushes (Turdidae) Fowl (Gallus gallus f. domestica) Goose (Anser anser f. domestica) Duck (Anas platyrhynchos f. domestica) Pigeon (Columba livia f. domestica) Turkey (Meleagris gallopavo f. domestica) Total were distinguished. With the exception of two assemblages that were not retained because they only contained one and two bird bones respectively, the material has been grouped into eight chronological units for interpretation (table 9). Discussion Some comments on the identifications The taxa in the tables have been listed exactly as they are published in the literature. It would have been desirable to validate the identifications of some of the published remains, but the material was not available for re-study. The specimen initially identified at the site Rijke Klarenwijk by Gautier (1995a) as Western capercaillie (Tetrao urogallus) ) would represent a species that has not been found on any other archaeological site in Belgium. Probably, it did not occur in the northern half of the country, where Brussels is located. Still today, it is a vagrant species for Belgium (Heinzel et al., 1982). Other galliform species of similar size are peacock (Pavo cristatus s f. domestica) and turkey (Meleagris gallopavo o f. domestica), but without further verification such an alternative

6 76 Part I Birds and the provision of food Figure 2. Proportion of remains of consumed bird versus those of consumed mammals. For each assemblage the total sample size (n) is indicated as well as the period (LM = Late Medieval; EMod = Early Modern Times; Mod = Modern Times). Sint-Katelijnewijk (LM) n=1640 Hoogstraeten (LM) n=949 Zuidstraat (LM) n=609 Eenmansstraat (LM) n=1281 Rijke Klaren (LM) n=1691 Dinantstraat (LM) n=1279 9,1 10,9 14,6 14,4 16,3 28,7 Arme Klaren (LM) n=7212 7,2 Zavel (LM) n=794 3,3 all LM (n=15455) 11,3 Hoogstraeten (EMod) n=960 10,8 Zavel (EMod) n=410 10,5 Sint-Katelijnewijk (EMod) n=701 10,3 Dinantstraat (EMod) n=1596 9,9 Rijke Klaren (EMod) n=5266 6,1 Treurenberg (EMod) n=988 6,0 Arme Klaren (EMod) n=2632 4,7 all EMod (n=12553) 7,0 Dinantstraat (Mod) n=672 8,8 Hoogstraeten (Mod) n=1325 8,8 all Modern (n=1997) 8, identification cannot be substantiated. The black-headed gull (Larus ridibundus) ) found at the Dinantstraat in a late 13th early 14th c. AD context (Gautier, 2001b) is commonly found in inland towns today, but finds of Larus spp. in Medieval towns of western Europe are very scarce, even in coastal areas (O Connor, 1993: p. 160). Therefore this identification must perhaps also be treated with caution. A rare species today is tawny owl (Strix aluco) which has been reported from Rijke Klarenwijk (Gautier, 1995a) and Sint- Katelijnewijk (Gautier, 1995b); here as well, verification of the material would be desirable. Also among the Anatidae finds from Brussels mentioned in the literature some identifications seem, to some extent, questionable in a sense that they have been attributed to species on the basis of a rather restricted reference collection (Gautier, 1995a: p. 183), This could be the case for the finds of garganey (Anas querquedula) and Northern shoveler (Anas clypeata) mentioned from Sint- Katelijnewijk (Gautier, 1995b). Another remark that should be made here is that all the remains of mallard/domestic duck and greylag goose/domestic goose have been considered both in the literature and in our own analyses as representing the domestic form, although no osteological proof for this has been provided. Because of the Late Medieval and Postmedieval date of the material, it is likely that most of these bones are indeed from domestic animals (Albarella, 2005), but it cannot be excluded that a few remains from greylag goose (Anser anser) or mallard (Anas platyrhynchos) are present within the collections. It is unclear whether tame mute swan (Cygnus olor) is present in the studied assemblages from Brussels since none of the Cygnus remains could be confidently identified to species. A final identification that needs some explanation is the Streptopelia cf. turtur r (cf. European turtle- dove) attested in a Late Medieval context from Arme Klarenwijk. The measurements of the ulna

7 7 Bird remains from Late Medieval and Postmedieval sites in Brussels, Belgium 77 (Bp 5.5 mm; SC 2.5 mm) fall within the variation of Streptopelia turtur r and outside that of the ringneck dove (S. risoria) documented by Fick (1974). However, of the latter species, which now should be called Streptopelia roseogrisea (BirdLife International, 2008a), only five individuals were measured, versus 15 for S. turtur. Because the ringneck dove has been domesticated, it is likely that its size variation is wider than shown in Fick (1974). The only other possible record of European turtle-dove ever mentioned for Belgium is from a 17th-18th century context from St. Peter s Abbey in Ghent (Ballmann, 1978). The measurements of the Brussels ulna are just below those provided by Fick (1974) for the Eurasian collared-dove (Streptopelia decaocto), a species that only expanded its territory into western Europe during the mid-20th century (Pascal et al., 2006). Wild and domestic birds in the diet Although the majority of the bird remains presented in tables 1-9 is from animals that were consumed, there is also material that should not be considered as food refuse. When trying to quantify the proportion of birds versus mammals in the consumption refuse, or when investigating the culinary importance of wild versus domestic birds, it is vital to exclude animals that were not eaten. Of the taxa listed in the tables the following have not been regarded as consumption refuse: all raptors, i.e. the black or red kite (Milvus migrans / M. milvus), the little owl (Athene noctua), the tawny owl (Strix aluco), and all corvids, i.e. the magpie (Pica pica), the carrion crow (Corvus corone), the Eurasian jackdaw (Corvus monedula) and those that could belong to either carrion crow or rook (Corvus frugilegus). The black-headed gull (Larus ridibundus) is not considered as a consumed species either. None of the birds mentioned are traditionally eaten in our region during Late Medieval and Postmedieval times (Serjeantson, 2006: pp ), although there are some exceptional cases with clear evidence for consumption of corvids in northern France (Clavel, 2001: pp ). However, when found in an archaeological context, the taxa mentioned above are usually considered as carcass remains of animals that died naturally or that were killed on purpose and discarded without being consumed. This interpretation is often supported by the presence of multiple bones belonging to the same individual. At the Brussels sites this is the case, for instance, for the 19 bones of jackdaw found at Hoogstraeten, which represent two clusters found in a pit fill and a refuse layer, respectively. On many Belgian sites of Medieval and Postmedieval date, remains of Passeriformes are relatively rare, probably because of the lack of sieving. As a result of their rarity they did not receive much attention and may have been neglected as a food resource despite the fact that historical written and pictorial evidence suggests the contrary (Serjeantson, 2001). Although songbirds like small rodents can sometimes be considered as prey of raptorial birds or as intrusive animals (sensu u Gautier, 1987) that were trapped accidently and died naturally in structures such as wells or latrines, they are retained here as food animals. In an attempt to compare the dietary importance of birds at the various sites, the frequency of their remains is used as a measure. Although it is clear that this does not reflect the actual nutritional value of birds in the diet, compared to other animal (and plant) food, the relative abundance observed within and among sites can show trends that may be significant in terms of access to food and dietary preferences. Such an approach is justified here because all the faunal remains were sampled in a similar manner, through hand-collecting in the trench. The number of phases retained in the tables mentioned above are numerous and variable in duration. To facilitate inter- and intrasite comparisons the material has been grouped below in wider chronological units as follows: Late Middle Ages: 13th-15th c. AD, Early Modern Times (16th-late 18th c. AD), and Modern Times (19th-20th c. AD). When the consumed birds and mammals from the nine sites are considered, it appears that the proportion of birds is variable with percentages ranging between 3.3 and 28.7 %. A remarkably high proportion is found at Sint- Katelijnewijk where birds make up more than a quarter of the consumption refuse, but on most sites the contribution of birds is 10% or less (fig. 2). The relatively low numbers are in agreement with earlier observations on other urban sites in the northern part of Belgium, such as Gent (Ervynck, 1993). When the data are averaged for the wider chronological units, it appears that a decrease occurs in the contribution of birds from 11% in Late Medieval times to 7% in Early Modern Times. The two assemblages of Modern Times each have about 9% of birds. In northern France a similar chronological trend is seen with a (much higher) proportion of 25% birds in the 14th century, decreasing to a value of less than 6% comparable to the Brussels example in the 17th century (Clavel, 2001: p. 109). The French data, however, are combined assemblages including bone counts from rural, urban, religious and noble sites.

8 78 Part I Birds and the provision of food Table 10. Proportions of the domestic birds, ordered by period, and within each period sorted in descending order according to the values for goose. The proportions have also been calculated for all the material from each chronological phase. LM = Late Medieval; EM = Early Modern; Mod = Modern. Fowl Goose Duck Pigeon Turkey Dinantstraat (LM) n= Rijke Klaren (LM) n= Arme Klaren (LM) n= Zuidstraat (LM) n= Eenmansstraat (LM) n= Sint-Katelijnewijk (LM) n= Hoogstraeten (LM) n= Zavel (LM) n= all Late Medieval (n=1684) Rijke Klaren (EM) n= Dinantstraat (EM) n= Sint-Katelijnewijk (EM) n= Arme Klaren (EM) n= Hoogstraeten (EM) n= Zavel (EM) n= Treurenberg (EM) n= all Early Modern (n=832) Dinantstraat (Mod) n= Hoogstraeten (Mod) n= all Modern (166) The species spectrum illustrated by tables 1-9 shows the expected general pattern for urban assemblages, with a majority of domestic birds. The proportions of all domestic species, by period and site (table 10), show that domestic fowl is always the best represented species, followed by goose and duck. Pigeon and turkey are poorly represented in general. Trends that appear are that the average (small) contribution of duck seems more or less constant through time, and that the decrease in importance of goose from Late Medieval times onwards is compensated by an increase in the number of domestic fowl (table 10). The aforementioned results are in agreement with findings from England (Serjeantson, 2006) and northern France (Clavel, 2001: pp ) where, in urban sites, the order of importance of the domestic species is the same, and where goose numbers also decrease from the Late Medieval period onwards. Remains of turkey generally occur in relatively small numbers, except at the site of Treurenberg where, in a late 16th early 17th century AD context, six out of 59 identifiable bird bones belong to this species. Single finds of turkey include a carpometacarpus at the site Zavel in a refuse layer dating to the second half of the 16th beginning of the 18th century, another carpometacarpal at the site of Dinantstraat t in a context dated to the first half of the 17th century and a tibiotarsus from the same site in a context dated to the 20th century. Also at the Rijke Klaren site a single turkey bone a tibiotarsus was found in a refuse layer that contained pottery dated mainly to the last quarter of the 16th and the first quarter of the 17th century. A final record comes from a context at Hoogstraeten, dated to Modern Times, in which a coracoid was found. Besides the aforementioned finds that are in agreement with the present historical and archaeozoological knowledge, which situates the introduction of turkey in the 16th century (Crawford, 1984), there are two bones from contexts that are apparently too early. This is the case for a coracoid from Hoogstraeten that would be pre-16th century, and for an unspecified, but securely identified bone element (Gautier, 1995a: 183), found at the Rijke Klaren site in a Late Medieval context (2nd half of the 13th 1st half of the 14th century). These finds will be ignored here, as well as the sacrum fragment from a 14th-16th c. AD layer at the Zuidstraat that Gautier (2001a) hesitated to attribute with certainty to turkey. The proportions of bones from wild versus domestic birds calculated for the various sites from Brussels show that there are two localities with a relatively high amount of bones from wild taxa, namely Zavel (Late Medieval phase) and Hoogstraeten (all three phases) (fig. 3). When the proportions of wild versus domestic mammals are plotted against the bird data, it appears that the three assemblages from Hoogstraeten also yield the highest percentages of wild mammal. The species found at this site include hare, roe deer, red deer and wild boar. The Hoogstraeten bird and mammal bone assemblages clearly reflect

9 7 Bird remains from Late Medieval and Postmedieval sites in Brussels, Belgium 79 Zavel (LM) n=792 Hoogstraeten (EMod) n=955 Hoogstraeten (LM) n=934 Hoogstraeten (Mod) n=1309 Sint-Katelijnewijk (EMod) n=684 Sint-Katelijnewijk (LM) n=1619 Dinantstraat (EMod) n=1589 Rijke Klaren (EMod) n=5229 Zavel (EMod) n=409 Dinantstraat (Mod) n=672 Zuidstraat (LM) n=605 Arme Klaren (LM) n=7195 Arme Klaren (EMod) n=2570 Dinantstraat (LM) n=1274 Eenmansstraat (LM) n=1281 Treurenberg (EMod) n=987 Rijke Klaren (LM) n=1689 0,5 0,33 0,26 0,7 0,3 0,19 0,35 0,58 0,86 0,43 0,54 0,09 o 0,65 o 0,13 1,92 2,07 1,66 3,04 2,76 2,52 2,44 3,39 3,37 5,56 5,31 5,06 4,66 4,7 6,84 9,03 10,58 23,1 Figure 3. Proportion of wild bird remains within the consumed domestic birds, and proportion of wild mammals within the consumed mammals. For each assemblage the total sample size (n) is indicated as well as the period (LM = Late Medieval; EMod = Early Modern Times; Mod = Modern Times). The sample sizes differ slightly in some cases, compared to those given in Figure 2, because for the calculations only the hare bones have been retained within the leporids. The status (wild or domestic) of the rabbit and rabbit/hare remains is unclear. wild bird wild game the high status of the inhabitants that lived here and that had the rights to exploit hunting grounds outside town (cf. Galesloot, 1854). Similarly, the Late Medieval material from Zavel reflects the presence of noble inhabitants that were installed near the Court of Brussels. Hare, roe deer and red deer were also found here, but their percentage contribution to the total amount of mammals is less pronounced. The Early Modern avifauna from Zavel shows a dramatic decrease in wild birds (4.7% versus 23.1% in the Late Medieval contexts) and there is no game at all, despite the fact that a high class population would have lived here at the time. Another characteristic of the avifauna from Hoogstraeten and Late Medieval Zavel is that goose contributes less than 10% to the total amount of domestic birds in both the Late Medieval and Early Modern periods, which are the lowest values observed (table 10). It has been suggested that low proportions of geese might be considered as a typical signature of a noble site (cf. Serjeantson, 2006: p. 137), but why this is so seems unclear. Hunted species, both mammals and birds, are seen as high status nutrition during the Late Medieval and Postmedieval periods, because hunting was restricted to the nobility (Ervynck et al., 2003: p. 432). The link between the status of the consumers and the type of wild birds and their proportions found at the sites from Brussels has been demonstrated elsewhere in Belgium (e.g., Ervynck, 1992) and abroad (e.g. Clavel, 2001; Albarella & Thomas, 2002; Serjeantson, 2006). The highest proportions of wild birds, and the widest species spectrum, are found in castles: for instance, in northern France, Clavel (2001: pp ) found an average of 8% of wild birds in noble sites, versus about 2% in religious, rural and urban sites. He also noticed a steady increase of wild birds in noble contexts dating to the 13/14th, 15th and 16th centuries AD. At Hoogstraeten the proportion of wild birds is slightly higher in Early Modern Times, compared to the Late Middle Ages, but the difference is small and probably not significant. Among the wild bird species found at the Brussels sites, Anseriformes, Galliformes and Charadriiformes are the most common taxa and it appears that the most abundant species are also the ones found on high status sites elsewhere in Europe. The archaeozoological findings also seem to accord well with historical sources (Cosman,

10 80 Part I Birds and the provision of food Table 11. Proportion of immature bones among the domestic fowl remains. % immature total sample size Hoogstraeten Late Medieval Early Modern Modern Zavel Late Medieval Early Modern Arme Klaren Late Medieval Early Modern 4 97 Treurenberg Early Modern all Late Medieval all Early Modern Modern ). The Anseriformes found in Brussels are represented by swan and several duck species, among which common teal is frequent. Woodcock is well represented within the order of the Charadriiformes and among the Galliformes, grey partridge is the most common species. Both species were probably a staple element of wild bird dishes (Albarella & Thomas, 2002: p. 34). The grey partridge is here retained as a wild species, but it is not excluded that it was kept confined in warrens together with rabbits, a practice that is known in Flanders from written sources dating back to the early 14th century (Smit, 1911). Grey heron which, in northern France, is almost exclusively found in castle sites (Clavel, 2001: pp ), occurs in four assemblages at Brussels and in two cases this is from a site with noble inhabitants. Maybe these birds were placed in their feathers on the banqueting tables of the upper-class in a similar way to other large birds, such as peafowl and cranes, that were used to decorate tables and impress guests (Ervynck, 1992: p. 153; Clavel, 2001: pp ; Serjeantson, 2006: p. 142)? Provenance of the birds It is known from pictorial evidence, such as paintings and drawings, that the past urban environment of Brussels differed to a great extent from the present-day situation by the existence of more open space within the town (Smolar-Meynart & Stengers, 1989). Numerous ordinances forbidding garbage disposal in rivers and canals also demonstrate that accumulation of refuse in Brussels was often problematic (Deligne, 2003: pp ). Both factors must have had an influence on bird life within town. The bird remains from the Brussels sites include several scavenging species that are known to find suitable habitats and food in towns (O Connor, 1993). Black kite was found at Late Medieval Hoogstraeten, and remains labelled as black or red kite (Milvus migrans / M. milvus) occur at Late Medieval Arme Klaren. Kites were historically recorded as scavengers inside towns (O Connor, 1993: p.157). Nowadays the red kite, for instance, still occurs at the periphery of certain towns where they feed on carrion, but populations are in decline almost everywhere in Europe. One of the reasons held responsible for this, in France and Spain, is the decrease in the number of rubbish dumps at the edges of towns (BirdLife International, 2008b). The most common scavengers in the sites studied, however, are the corvids of which the following species have been identified: magpie (Pica pica), carrion crow (Corvus corone) and Eurasian jackdaw (Corvus monedula). Another species that may have found suitable shelter and food in town is the European turtle-dove (Streptopelia cf. turtur) possibly attested in a Late Medieval context from Arme Klarenwijk. Many of the wild birds identified from the sites in Brussels are aquatic, but it is questionable that they were obtained from the Zenne, the local river, or from one of the canals inside town. It has been argued (O Connor, 1993: p. 157) that, with the exception of the mallard (Anas platyrhynchos), Anatidae should not be considered as urban scavengers. Remains of swans occur in at least four Late Medieval and two Early Modern sites from Brussels, but they could not be identified with certainty to species level. If they are mute swan (Cygnus olor), they could be considered as tamed birds kept under controlled conditions such as in parks (Albarella & Thomas, 2002: p. 34). It is tempting to consider the Late Medieval find from Zavel and the Early Modern one from Hoogstraeten, as deriving from animals that were kept in the nearby park of the ducal palace. The provenance of the grey partridge is not very clear either. Possibly they were kept in warrens together with rabbits,

11 7 Bird remains from Late Medieval and Postmedieval sites in Brussels, Belgium 81 a practice that is known since Late Medieval times (Smit, 1911) and that is also suggested by the co-occurrence of both species in a 14th century refuse layer of the castle of Londerzeel in Flanders (Ervynck et al., 1994: pp ) and in a context of the 16th century castle of Eindhoven, The Netherlands (de Jong, 1992). All the Late Medieval and Early Modern contexts from the Brussels sites with grey partridge have also yielded remains of rabbit, but it would be preferable to combine these data in the future with historical information before considering this as evidence for the keeping of the species in confinement. The other wild birds not mentioned above must have come from outside town. These may have been purchased at the market or may have been obtained directly from properties located outside the city walls, in the case of the noble households and other well-to-do inhabitants that owned large estates. Figure 4. Right tarsometatarsal of a cock with sawn of spur, from a 16th century context at the site Zavel, compared to a complete, modern specimen. The domestic birds that have been found on each site may have been bred by peasants living outside town, although it is likely that local husbandry took place as well, indicated by the presence of immature bones. In England, immature chickens are also found in urban sites, with percentages varying between 20 and 40% (Albarella, 1997; Serjeantson, 2006). At the end of the Middle Ages a significant increase is noted in immature domestic fowl (Albarella, 1997). Among the geese, the percentage of immature bones in towns increases dramatically in the 16th century (Serjeantson, 2006: p. 145). In the case of Brussels, proportions of immature bird bone have not been mentioned in the previously published faunal reports. Table 11 therefore only mentions the data that were recorded during our own analyses of four of the sites. The figures refer to domestic fowl only, the number of observations on the other species being too low. The proportion of immatures varies between 4% and 29%, but on average the contribution of young birds seems lower than in England. The sample size for the assemblages from Zavel and Treurenberg is restricted, and it is unclear whether the trends in the remaining, larger, assemblages from Hoogstraeten and Arme Klaren are significant: compared to the values for the Late Medieval period there is a slight decrease in the proportion of immature bone in Early Modern Times at Hoogstraeten and a more pronounced decreasing trend occurs at the site Arme Klaren. When the data for all the sites are averaged by larger periods, a more or less constant contribution of young animals is seen, with a possible slight increase through time. Special finds In Late Medieval times cock-fighting was a favourite pastime (Reeves, 1998), and even from the Classical Period there is written evidence for the breeding and training of cocks for fighting (West, 1982: p. 260). Archaeozoological hints to the use of cocks in animal baiting are rare, however. In a context, dated to the 16th century from the site Zavel at Brussels, a tarsometatarsal of a cock was found of which the spur has been completely sawn off, close to the shaft (fig. 4). It is believed that the removal of this spur can be seen as evidence for cock-fighting, rather than a measure to protect other fowl or to avoid people getting hurt while handling live cocks.

12 82 Part I Birds and the provision of food Figure 5. Caudal view of the pathological furcula of a jackdaw, found in a Late Medieval refuse layer at the site Hoogstraeten. to the shaft of the tarsometatarsals. However, the natural spurs do not necessarily have to be cut off to allow the setting of artificial spurs as is suggested by an archaeological find from Camber Castle (Sussex, England). Excavations of mid-16th to mid-17th century contexts yielded a cock tarsometatarsal with an intact natural spur. Parallel with the spur, on the anterolateral surface of the shaft, a green stained patch occurred which, according to X-ray fluorescence, consists of mainly copper and zinc. This has been seen as possible evidence for a cock wearing a brass spurred ring for fighting (Connell & Davis, 2001: p. 309). It has been suggested that artificial spurs inflict wounds that heal more easily and get less infected than those from the natural spur (Means, 1911: p. 14). An alternative explanation, that is valid today, is that fights are ended more quickly when the cocks are wearing gaffs, thus permitting a larger series of interesting fights in a given time span (B. Goddeeris, pers. comm.). Numerous types of artificial spurs are known (Finsterbusch, 1980: pp ) whose setting requires the complete or partial removal of the natural spur. A late 18th early 19th century context at Greyfriars, Oxford (West, 1982), yielded a spur sawn off close to the shaft, comparable to the find from Brussels. Two tarsometatarsals with cut off spurs have also been found at Late Medieval contexts dated to 15th-16th centuries in the city of Beja, Alentejo region in Portugal (see Moreno-García & Pimenta, this volume). The specimens differ from the preceding two examples in the sense that the spur has been cut off at a larger distance about 0.8 cm and 1.3 cm from the shaft. The fauna from Dichin in Bulgaria (Johnstone, 2007) yielded a Late Roman (5th-6th c. AD) example of another only partially removed cock spur (Johnstone, pers. comm.). Here, only just the tip has been removed, but there are several cut marks further down. It seems that in this case a major part of the horny spur may have been removed, but the bony part is still rather long, i.e. about 2 cm. The Greyfriars and Brussels specimens seem to be the only examples thus far of almost completely cut off spurs, and may have been set with metal gaffs comparable to the ones seen on the modern, stuffed fighting cock depicted in West (1982: p. 258). The steel gaffs are put over the truncated natural spurs, and attached with leather bandage The Late Medieval contexts from the site of Hoogstraeten yielded skeletal remains of two jackdaws represented by 7 and 12 bones, respectively. The most complete specimen was found in a pit fill dated to the 14th century on the basis of the associated archaeological material. The other individual is broadly contemporaneous and comes from a refuse layer (S. Modrie, pers. comm.). Jackdaws are reputed for their ability to adapt to human presence and are easily tamed (Lorenz, 1952). For that reason, it is worth considering the possibility that skeletons found in urban contexts such as Hoogstraeten, represent tamed birds. This has been suggested for two individuals found in a Postmedieval domestic context at the Bedern Foundry site at York O Connor (2003: p. 90). One of the individuals from Hoogstraeten had a pathological furcula: the right part (i.e., clavicula) of the bone shows a healed fracture at about 1 cm from the proximal end (fig. 5). The articulation of the right clavicula with the coracoid also shows a severe deformation. A slight thickening is seen on the left clavicula, at less than a centimetre distance from the junction with the element of the opposite side, showing that the bone was damaged here too during the animal s life. Healed fractures near the symphysis have been recorded on numerous furculae of ptarmigan (Lagopus mutus and Lagopus lagopus albus) from Late Pleistocene sites in Central Europe (Tasnadi-Kubacska, 1962: pp ). The high incidence of this type of fracture in ptarmigan

13 7 Bird remains from Late Medieval and Postmedieval sites in Brussels, Belgium 83 was explained as a result of the fierce battles that take place between males during the breeding season whereby heads and breasts receive hard knocks, although it has also been suggested that the fractures result from flying against obstacles (Lambrecht, 1933: p. 883). A fractured furcula would not immobilise birds, which should still be able to escape from enemies (Lambrecht, 1933: p. 885). However, because the breakage of the furcula often occurs in association with fracturing of the coracoid or scapula, the wing cannot function normally and this limits its use for flight until healed (Chubb, 1998). It is unclear when the Hoogstraeten specimen became injured and whether the fracture can be seen as a result of handling by people. Perhaps the fracture is the reason why the bird may have been captured and possibly tamed consequently. Conclusions On the basis of about 2800 bird remains from nine Late Medieval and Postmedieval sites within the town of Brussels, it has been possible to document the role that birds played in the food provisioning. With the exception of the raptors, corvids and gulls, all the bird species are considered food items. When the proportion of birds versus mammals is considered within and among sites, it appears that birds usually represent 10% of the remains or less and that, when the data are averaged chronologically, a decrease occurs in their abundance from Late Medieval to Early Modern Times. Relatively high proportions of wild birds have been found at two sites (Hoogstraeten n and Late Medieval Zavel) that were inhabited by a high class population. Those sites also yielded hare, roe deer, red deer and wild boar, the wild mammals that are typically obtained from hunting grounds outside town. Hoogstraeten and Late Medieval Zavel also have very low amounts of goose which would be another signature of noble sites. Some of the wild birds found at the Brussels sites are considered urban scavengers, i.e. the corvids and the kites, and some of the Anatidae were possibly obtained from the local river or the canals inside the town. It is unclear yet whether the remains of swans and of grey partridge should be considered as evidence for their keeping in confinement. It is likely, however, that the majority of the wild birds came from outside Brussels. Although the domestic birds may also have been obtained mainly from outside town, it appears from the presence of immature bones that local husbandry took place as well in all periods. The fact that birds played not only a nutritional role, is illustrated by a cock tarsometatarsal with its spur sawn off and by a healed clavicula of a jackdaw, which are believed to be indicative of cock-fighting and of bird taming. Acknowledgements We acknowledge the archaeologists of the Brussels Capital Region for making the material available for study and we thank Ann Degraeve, Sylviane Modrie and Britt Claes for context and dating information on the unpublished sites. Cluny Johnstone shared unpublished information with us. We benefited from discussions with Anton Ervynck (Flemish Heritage Institute) and Boudewijn Goddeeris (Royal Belgian Institute of Natural Sciences) on some of the avian finds. Sheila Hamilton-Dyer (Southampton) kindly made the linguistic corrections. This paper presents research results of a convention with the Brussels Capital Region (P4/IRScNB/2006-3) and of the Interuniversity Attraction Poles Programme Belgian Science Policy (P6/22). References Albarella, U., Size, power, wool and veal: zooarchaeological evidence for late medieval innovations. In: G. De Boe & F. Verhaeghe (eds), Environment and subsistence in medieval Europe. Papers of the Medieval Europe Brugge 1997 Conference Volume 9. Institute for the Archaeological Heritage of Flanders, Zellik, pp Albarella, U., Alternate fortunes? The role of domestic ducks and geese from Roman to Medieval times in Britain. In: G. Grupe & J. Peters (eds), Feathers, Grit and Symbolism. Documenta Archaeobiologiae 3, pp Albarella, U. & R. Thomas, They dined on crane: bird consumption, wild fowling and status in medieval England. Acta zoologica cracoviensia 45 (special issue), pp Ballmann, P Knochenfunde von Vögeln aus der Abtei Sankt Peters zu Gent, Belgien (VII bis XVIII Jahrhundert). Le Gerfaut De Giervalk 68, pp

14 84 Part I Birds and the provision of food BirdLife International, 2008a. Species factsheet: Streptopelia roseogrisea. Downloaded from on 13/2/2009 BirdLife International, 2008b. Species factsheet: Milvus milvus. Downloaded from birdlife.org on 17/2/2009 Chubb, K., Upper body trauma; doing postmortems: a study of 300 Red-tailed Hawks. Beaks, Brains & Bones. Experiences in Wild Bird Trauma, vol. 2. Avian Care and Research Foundation. Verona, Ontario. Claes, B., Archeologisch onderzoek van het voormalige Arme Klarenklooster (Br.), Archaeologia Mediaevalis Kroniek 29, pp Clavel B., L animal dans l alimentation médievale et moderne en France du Nord (XIIe- XVIIe siècles). Revue Archéologique de Picardie, Numéro Spécial 19, pp Connel, B. & S. Davis, The animal bones. In: M. Biddle, J. Hiller, I. Scott & A. Streeten (eds), Henry VIII s coastal artillery fort at Camber Castle, Rye, East Sussex. An archaeological, structural and historical investigation. Oxford Archaeological Unit for English Heritage: pp Cosman, M.P., Fabulous feasts. Medieval cookery and ceremony. George Braziller, New York. Crawford, R.D., Turkey. In: I.L. Mason (ed.), Evolution of domesticated animals. Longman, London & New York, pp Degraeve, A., Archeologisch noodonderzoek aan de Zavel te Brussel (Br.), Archaeologia Mediaevalis Kroniek 24, pp Degré, S., Een akker, een brouwerij, een overdekte markt Analyse van de archeologische structuren. In : S. Degré (ed.), Brouwerijen in de Sint-Katelijnewijk. Archeologie in Brussel 2. Brussels Hoofdstedelijk Gewest, Monumenten en Landschappen, Brussel, pp de Jong, T., Huisdieren, jachtwild, vissen en weekdieren: een weerspiegeling van gevarieerde maaltijden. In: N. Arts (ed.), Het kasteel van Eindhoven, archeologie, ecologie en geschiedenis van een heerlijke woning. Museum Kempenland, Eindhoven, pp Deligne, C., Bruxelles et sa rivière. Genèse d un territoire urbain (12e-18e siècle). Brepols, Turnhout. De Poorter, A., Archeologisch onderzoek. In: A. De Poorter (ed.), De Rijke Klarenwijk: van Priemspoort tot klooster. Archeologie in Brussel 1. Brussels Hoofdstedelijk Gewest, Monumenten en Landschappen, Brussel, pp De Poorter, A., Het archeologisch onderzoek op een terrein in de Dinantstraat. In: P. Blanquart, S. Demeter, A. De Poorter, C. Massart, S. Modrie, I. Nachtergael & M. Siebrand (eds), Rond de eerste stadsomwalling, Archeologie in Brussel 4. Brussels Hoofdstedelijk Gewest, Monumenten en Landschappen, Brussel, pp Diekmann, A., Archeologisch onderzoek in de vindplaats van de Eenmansstraat. In: A. Diekmann (ed.), Middeleeuwse ambachten en stedelijk wonen. Archeologie in Brussel 3. Brussels Hoofdstedelijk Gewest, Monumenten en Landschappen, Brussel, pp Ervynck, A., Medieval castles as top-predators of the feudal system: an archaeozoological approach. Château Gaillard 15, pp Ervynck, A., The role of birds in the economy of medieval and post-medieval Flanders: a diversity of interpretation problems. Archaeofauna 2, pp Ervynck, A., W. Van Neer & P. Van der Plaetsen, Dierlijke resten. In: A. Ervynck (red.), De Burcht te Londerzeel. Archeologie in Vlaanderen, Monografie 1. Instituut voor het Archeologisch Patrimonium, Zellik, pp Ervynck, A., W. Van Neer, H. Hüster-Plogmann & J. Schibler, Beyond affluence: the zooarchaeology of luxury. World Archaeology 34, pp Fick, O., Vergleichend morphologische Untersuchungen an Einzelknochen europäischer Taubenarten. Dissertation, Munich. Finsterbusch, C.A., Cockfighting all over the world. Saiga Publishing, Hindhead, Surrey.