Chapter 8
Animal utilization at the medieval Islamic site of Nora
p. 153-162
Texte intégral
1This chapter reports on the zooarchaeological assessment of animal remains recovered from the ancient Islamic site of Nora in central Ethiopia. Faunal material from Nora was collected over the course of one excavation season, in May 2008. Most faunal remains were collected by hand and no sediment was screened; as a result, no microfauna was present in the assemblage. The faunal assemblage from Nora consists of materials collected from three different areas: the residential areas A (henceforward: “House 1”) and B (“House 2”) of Sector 2, and Zone D of Sector 3 (“Refuse heap”), to the east of the Great Mosque (Fig. 4.7). For House 1 and House 2, the faunal assemblage comes from the stratigraphic horizons dated between the late 13th and the early 16th centuries.1 The faunal assemblage from the Refuse Heap comes from a stratigraphic unit dated to the first half on the 15th century.2 The remains consist of NSP (number of specimens) of 2,505. Of these, 2,374 (95%) were identifiable and 131 (5%) were unidentifiable. Many of the unidentifiable specimens are attributed to the mammalian taxonomic class. For the purpose of this analysis, I have included only specimens identifiable to order, family and subfamily level. Sample size data for each excavation area is summarized in Fig. 8.1. The total number of bones recovered from the three contexts includes 441 from House 1, 571 from House 2, and 1493 from the Refuse Heap. Because there is variation in sample size between the three different areas, one to one correlation was not possible.
Fig. 8.1. NISP (Number of Identifiable Specimens) by Area (N = 2505).
Area | NSP | % Representation |
House 1 | 441 | 18 |
House 2 | 571 | 23 |
Refuse heap | 1,493 | 60 |
Total | 2,505 | 100 |
2The faunal assemblage from Nora is dominated by domestic animals (Figs. 8.2, 8.3). The vast majority of the identifiable faunal remains are from cattle (Bos taurus/Bos indicus or hybrids) (249) followed by sheep/goat (Caprini) (61). It is usually difficult to differentiate sheep (Ovis aries) from goat (Capra hircus) because of their morphological similarity. As a result, bones belonging to sheep and goat are referred at the tribal level Caprini in this text. Other animals present are chicken (Gallus gallus f. domestica) (1), bird (Aves) (11), and rodent (1). Shell (cf. Cypraea annulus) and egg shell were also discovered.
Fig. 8.2. NISP species representation for all Nora fauna (N = 2,505).
Taxa | NISP | % of Assemblage |
Caprini (sheep/goat) | 61 | 2.4 |
Bos taurus/indicus (cattle) | 249 | 9.9 |
Bovidae | 92 | 3.7 |
Gallus gallus (f. domestica) (chicken) | 1 | 0.0 |
Large bovid | 132 | 5.3 |
Small bovid | 1 | 0.0 |
Medium bovid | 57 | 2.3 |
Large mammal | 11 | 0.4 |
Mammal | 1,758 | 70.2 |
Aves (bird) | 11 | 0.4 |
Rodent | 1 | 0.0 |
UID | 131 | 5.2 |
Total | 2505 | 100.0 |
Fig. 8.3. NISP species representation for all Nora fauna (N = 2,505).

3The aim of the zooarchaeological study was to assess the potential of faunal material for providing information about the socio-economic, husbandry and butchery practices of an ancient Islamic Ethiopian town at the site of Nora. Zooarchaeological remains were recorded following the protocol for recording animal bones by African zooarchaeologists (Gifford and Crader 1977; Marshall 1990; Walker 1985). MacDonald’s (MacDonald 1992) method was used to differentiate chicken and wild African fowl. Methods used to record age or (where possible) sex are those of Grigson and Payne (Wilson, Grigson and Payne 1982). Sample sizes were suitable for general categorization of age profiles only. Laboratory methods included identification and measuring of bones to class, order, family and subfamily or finer taxonomic level if possible. Animal bones that were not identifiable to family or subfamily were categorized to bovid size class following Brain (1981), Bunn and Kroll (1986), and Marshall (1986). These categories included small bovid (size class I: <23 kg), medium bovid (size class II: 23-84 kg), and large bovid or mammal (size class II: 84-296 kg).
8.1 The faunal assemblage from House 1 (Sector 2 Zone A)
4House 1 yielded 441 fragments of animal specimens. Of these, 333 were identifiable and included: 45 cattle (Bos), 34 sheep/goats (Caprini), 26 Bovidae, 6 large bovids, 221 mammals, 1 rodent. There were 108 UID (unidentifiable) animal remains (Figs. 8.4, 8.5). At this context, bone preservation was affected by weathering, root-etching and burning. Most of the faunal remains from this context came from a possible oven feature (UA 2013, US 2010).
Fig. 8.4. NISP species representation for House 1 (N = 441).
Taxa | NISP | % of Assemblage |
Caprini (sheep/goat) | 61 | 2.4 |
Bos taurus/indicus (cattle) | 249 | 9.9 |
Bovidae | 92 | 3.7 |
Gallus gallus (f. domestica) (chicken) | 1 | 0.0 |
Large bovid | 132 | 5.3 |
Small bovid | 1 | 0.0 |
Medium bovid | 57 | 2.3 |
Large mammal | 11 | 0.4 |
Mammal | 1,758 | 70.2 |
Aves (bird) | 11 | 0.4 |
Rodent | 1 | 0.0 |
UID | 131 | 5.2 |
Total | 2505 | 100.0 |
Fig. 8.5. NISP species representation for House 1 (N = 441).

8.2 The faunal assemblage from House 2 (Sector 2 Zone B)
5There were 571 faunal remains recovered from House 2. Of these, 550 bones were identifiable (Figs. 8.6, 8.7). There is a difference in the ratio of Bos-sized animals to Caprini-sized animals in this context, with more cattle (Bos taurus/indicus) represented. Most of the cattle bones from House 2, like those from other contexts, are aged or very old, as exhibited by the extreme wear stage of the teeth from this site. There is also indication of cut marks on some of the long bone shafts. Body part representation for Bos-sized and Caprini-sized animals is also relatively the same as from the other contexts.
Fig. 8.6. NISP species representation for House 2.
Taxa | NISP | % |
Bos taurus/indicus (cattle) | 62 | 10.9 |
Caprini (sheep/goat) | 2 | 0.4 |
Large bovid | 32 | 5.6 |
Mammal | 450 | 78.8 |
Medium bovid | 4 | 0.7 |
UID | 21 | 3.7 |
Total | 571 | 100.0 |
Fig. 8.7. NISP species representation for House 2 (N = 571).

8.3 The faunal assemblage from the Refuse Heap (Sector 1 Zone D)
6A total of 1,492 bones were recovered from the Refuse heap (Sector 1 Zone D), to the east of the Great Mosque. Species representation for the whole unit shows that more cattle than sheep/goat bones are represented (Figs. 8.8, 8.9). Cut marks and patterns of modification on the long bone shafts indicate processing. Bones from the upper sequence seems to also show burning and poor preservation as opposed to bones from the lower sequence. Body part representation is the same as the other fields, with a large concentration of cattle long bone shafts and cattle and sheep/goat rib shaft fragments. Moreover, all the bird (Aves) bones including the positively identified chicken, are found in this context. None of the other contexts have bird bones.
Fig. 8.8. NISP species representation for the Refuse Heap (N = 1,492).
Taxa | NISP | % |
Bos taurus/indicus (cattle) | 62 | 10.9 |
Caprini (sheep/goat) | 2 | 0.4 |
Large bovid | 32 | 5.6 |
Mammal | 450 | 78.8 |
Medium bovid | 4 | 0.7 |
UID | 21 | 3.7 |
Total | 571 | 100.0 |
Fig. 8.9. NISP species representation for the Refuse Heap (N = 1,492).

8.4 Bone modification
7Five types of modification were present in the sample: human modification, gnawing, root-etching, trampling and weathering. Modification of the bone by humans before the bone was incorporated in the deposit consisted mainly of impact marks caused from a blow and cut marks resulting from the removal of meat. Very few gnawed bones were present in the faunal remains. Furthermore, it is difficult to identify if they were made before they were deposited or if they are a result of burrowing in the midden by rodents. The degree of weathering involved was not easy to judge. However, some bones showed clear cracking and flaking, as in the weathering stages proposed by Behrensmeyer (1978).
8Bones from Nora bear very few butchery marks: of an NISP of 310 for combined sheep/goat and cattle bones, only 17 have cut marks. The butchery pattern of these few bones, based on the location and orientation of cut marks, is presented in Fig. 8.10. The result shows that butcherying practice involved dismemberment of the distal epiphysial part of long bones for both the hind and fore-limbs.
Fig. 8.10. Butchery marks on cow (Bos tauraus/indicus) remains.
Body part | Portion | Butchery | Function Probable |
Femur | Distal epiphysis | Hack | Dismemberment |
Humerus | Distal & proximal epiphysis | Hack | Dismemberment |
Long bone shaft | Dyphisis | Cut | Filleting |
Mandible | Ramus | Cut, hack | Dismemberment, Filleting |
Metapodial | Distal epiphysis | Cut, hack | Dismemberment, Filleting |
Radius | Distal epiphysis | Hack | Dismemberment |
Rib shaft | Epiphysis | Cut, hack | Filleting |
Tibia | Distal epiphysis | Hack | Dismemberment |
Vertebra | Body | Cut | Filleting |
9Body part utilization was computed for Bos- and Caprini-sized animals (Figs. 8.11, 8.12, 8.13, 8.14). The result shows that, for both Caprini- and Bos-sized animals a large number of rib shaft fragments are represented. In addition, the representation of the following skeletal parts: metapodial, vertebra, and teeth fragments are high for both animal group size. However, long bone shaft fragments, the second largest body part reprented, are only common for Bos-sized animals. Cattle phalanges are also present in large number.
Fig. 8.11. Body part representation for Bos-sized animal remains.
Body part | Count | % of Assemblage | |||
1st phalanx (1PHX) | 12 | 2.70 | |||
2nd phalanx (2PHX) | 11 | 2.47 | |||
3rd phalanx (3PHX) | 7 | 1.57 | |||
Astragalus (AST) | 2 | 0.45 | |||
Calcaneus (CAL) | 5 | 1.12 | |||
Cranial (CRA) | 1 | 0.22 | |||
Femur (FEM) | 3 | 0.67 | |||
Humerus (HUM) | 2 | 0.45 | |||
Innominate (INO) | 4 | 0.90 | |||
Long bone shaft (LBSF) | 96 | 21.57 | |||
Mandible (MAN) | 8 | 1.80 | |||
Maxilla (MAX) | 2 | 0.45 | |||
Metapodium (METP) | 24 | 5.39 | |||
NAVICO-CUBOID | 1 | 0.22 | |||
Paciform (PAC) | 1 | 0.22 | |||
Phalanx (PHX) | 13 | 2.92 | |||
Radius (RAD) | 4 | 0.90 | |||
Radius-Ulna (Rad-Ulna) | 2 | 0.45 | |||
Rib shaft fragment (RSF) | 103 | 23.15 | |||
Scaphoid (SCAF) | 3 | 0.67 | |||
Scapula (SCP) | 8 | 1.80 | |||
Tibia (TIB) | 3 | 0.67 | |||
Teeth (TTH) | 75 | 16.85 | |||
Thoracic vertebra (TVERT) | 5 | 1.12 | |||
Ulna (ULN) | 2 | 0.45 | |||
Unciform (UNC) | 9 | 2.02 | |||
Vertebra (VERT) | 39 | 8.76 | |||
Total | 445 | 100.00 |
Fig. 8.12. Body part representation for Bos-sized animal remains (N = 445).

Fig. 8.13. Body part representation for Caprini-sized animal remains (N = 91).
Body part | Count | % of Assemblage |
Astragalus (AST) | 1 | 0.01 |
Atlas (AVERT) | 1 | 0.01 |
Cervical vertebra (CVERT) | 3 | 0.03 |
Femur (FEM) | 5 | 0.05 |
Innominate (INO) | 1 | 0.01 |
Lumbar vertebra (LBSF) | 5 | 0.05 |
Mandible (MAN) | 4 | 0.04 |
Metatarsal | 2 | 0.02 |
Metapodium (METP) | 14 | 0.15 |
Radius (RAD) | 2 | 0.02 |
Rib shaft fragment (RSF) | 34 | 0.37 |
Tibia (TIB) | 2 | 0.02 |
Teeth (TTH) | 7 | 0.08 |
Vertebra (VERT) | 10 | 0.11 |
Total | 91 | 1.00 |
Fig. 8.14. Body part representation for Caprini-sized animal remains (N = 91).

8.5 Age profile
10A mortality profile was constructed using data from cattle and sheep/goat. The age profile was computed based on tooth wear stage and, if available, epiphyseal fusion (Grigson and Payne 1982). The age profiles show that almost all the cattle were adult or old, while the domestic Caprini (sheep/goat) sized animals, were young or juvenile. However, adult sheep/goat animal remains are also represented (Figs. 8.15, 8.16).
Fig. 8.15. Bos-sized animal remains by age. Y: Young, YA: Young Adult, J: Juvenile, A: Adult, O: Old.

Fig. 8.16. Caprini-sized animal remains by age. Y: Young, YA: Young Adult, J: Juvenile, A: Adult, O: Old.

8.6 Discussion
11Cattle, sheep/goat and chicken are the main domesticates present at Nora. The relative proportion of cattle to sheep/goat suggests preference for cattle utilization. This is relatively usual in African urban sites. However, the growth rate of sheep/goat herds and the smaller size of the animals means that at Nora they were usually slaughtered more frequently for household use (Dahl and Hjort 1976; Marshall 1990). It is noticeable that pigs, wild or domestic, are absent from the Nora fauna. This is consistent with Islamic (as well as Ethiopian Christian) practice (MacLean and Insoll 2003). Chicken have been rarely identified in African archaeological sites, so their presence here is significant. It adds to a growing body of data regarding the timing and distribution of chicken in Africa. Few specimens have so far been identified in Aksumite sites (Cain 2000; Helina Solomon Woldekiros and D’Andrea 2017), and in Islamic contexts at Swahili sites (Marshall 2000).
12The elevated representation of long bone shafts, rib shafts, cranial fragments, and vertebral fragments, is the same in all the three contexts (House 1, House 2, and Refuse Heap) and reflects the effects of butchery, sampling and density-mediated attrition. The breakage pattern shows heavy processing of bones at the site. The age profile for cattle shows that, in addition to meat production, animals might have contributed to several other aspects of the economy like milking and traction. The age profile for sheep/goat shows a variety of age categories consistent with slaughter for meat and other uses: the production of milk, leather, and wool might have been involved in the economic specialization of this Islamic site.
13Future regionally-based zooarchaeological approaches that look at animal body part distribution and butchery patterns will likely shed light on meat distribution and consumption practices, in comparison with other faunal studies from Islamic sites in sub-Saharan Africa (Gaastra and Insoll 2020; Insoll 2003). Moreover, even though the faunal assemblage from Nora provides the first glimpse of animal exploitation at an ancient Islamic town in Ethiopia, it is important to look at several spatially distinct contexts to gain an insight in the household production versus market-based distribution and consumption of domestic animals (Cain 2000; Zeder 1991).
14The faunal remains from Nora reflect, to a limited extent, the economic basis of the community: dependence on domestic animals and preference for cattle as opposed to sheep/goat. However, the faunal specimens and the contexts excavated are too few to allow reconstruction of the means by which the community obtained its food, the existence of contacts outside the community, or class status within the community (Cain 2000; Zeder 1991). They nevertheless reflect the first contribution to understanding the role that animals played in this medieval Islamic community and highlight the importance of cattle.
Notes de bas de page
Auteur

Le texte seul est utilisable sous licence Creative Commons - Attribution - Partage dans les Mêmes Conditions 4.0 International - CC BY-SA 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Nora, ville islamique de l’Éthiopie médiévale (fin du XIIIe-début du XVIe siècle) : les fouilles de 2008
François-Xavier Fauvelle et Romain Mensan (dir.)
2024