Modern scholarship

David W. Phillipson, *Archaeology at Aksum, Ethiopia, 1993-7, Volume II* (2000), p. 460

Archaeology at Aksum funerary deposits of the fourth century AD. D and K or blue only. The K site beads were again mainly tursites are associated with occupation and production at a quoise but included a few red, and red over white or lower level of the socio-economic hierarchy and date green cores in the latest phases. between the late fifth and the early eighth centuries. In More change is apparent in the shapes of addition, there is a small group of specimens recovered beads. Dominant in the fourth-century deposits are from Pre-Aksumite horizons at D site; these probably characteristic biconical hexagonally faceted beads (e.g. date between the eighth and the fourth centuries BC, Figs 64a-e, above), followed closely at the Tomb of although admixture with later material cannot be pre- the Brick Arches by drawn beads of a standard barrel eluded. Post-Aksumite and mixed contexts at the Mau- shape with rounded comers and then by other barrel soleum, D and K sites include material as late as the shapes. Oblate beads, perhaps significantly, were abnineteenth and twentieth centuries as well as some sent from these fourth-century funerary contexts. The which is clearly derived from earlier deposits. origin for these drawn beads requires further investiga- The small group from Pre-Aksumite contexts tion, but one contender is northern India (cf. Stem at D site is the only assemblage in which stone speci- 1991: pl. xxxviiia). The gold-in-glass beads from the mens outnumber those of glass. Carnelian, however, Tomb of the Brick Arches are of typical Roman type. seems to have continued in use until later times. Products of the Near East and Nile Valley seem to The predominant glass bead colours from the have been more numerous at Aksum in later periods. Tomb of the Brick Arches are opaque red and red In Late Aksumite times oblates are the most striped (346 beads), opaque and translucent turquoise common at both K and D sites, followed by small (271) and green (315), cobalt blue and blue (48) and often irregular drawn beads and barrels. In the Postyellow beads (51). The much smaller numbers from Aksumite/mixed material small drawn beads continue the primary deposits in the Mausoleum show a similar to be the most numerous, while recent blue ring beads predominance of turquoise (7) and blue/cobalt blue (9) of the kind still used for decorating baby carriers are but only 3 green and 1 red. It is possible that the red also fairly common. It seems that the spherical beads example was intrusive from higher levels in the same with contrasting coloured glass around a coloured core trench (MG 80), where contexts of phases II and III are likewise recent. yielded 208 red opaque drawn beads (Fig. 193a, above). These Aksumite sites demonstrate again how The 23 colourless glass beads from the Tomb of the difficult it is to assign most beads to particular periods Brick Arches have no parallels in the Mausoleum nor, or places of origin. Important exceptions are the bicowith one D-site exception, in later periods. nical hexagonally faceted beads, the gold-in-glass beads During the Late Aksumite period at D site and possibly the relatively large opaque red beads from blue/cobalt blue and turquoise again predominate, fol- the Mausoleum. One or two elaborate recent beads, lowed by green, red and yellow. K site shows a similar such as the large black oblate with contrasting marpattem. There was thus little change in colour-prefer- vered spots from Mausoleum phase 5 in which may be ence between the fourth-century deposits and those of of Roman, perhaps Syrian origin, are also more read- Lake Aksumite date. ily identifiable. Other beads may have been manufac- Far fewer glass beads were recovered from tured locally in Late Aksumite times if not before: post-Aksumite and mixed horizons. Those from the evidence for the working of glass has been noted Mausoleum trenches were mainly blue, cobalt blue above, while the production of stone beads is indicated and turquoise. Those from the D site were turquoise by bars, cubes and partially bored beads. Ivory-working techniques (Laurel Phillipson) Unambiguous literary evidence (e.g. Casson 1989: 53; quantities, but also that it was shaped, turned, carved Wolska-Conus 1968-73) indicates that ivory was a va- and employed with such skill and for such a variety of lued export from parts of ancient northeastem Africa purposes as to indicate that its use was part of a longincluding especially Aksum, but says nothing about standing and accomplished tradition. the forms in which the ivory was exported, whether as Two scraps of lathe-turned waste from K site whole tusks or as finished artefacts. As will be detailed attest to local ivory working in the sixth century, to below, material of fourth- to seventh-century date re- the use of an efficient lathe and fine iron or steel covered in the course of the present research demon- blades as well as to the presence, most probably, of strates not only that ivory was present in considerable skilled specialist turners (Fig. 403). The ivory is 460 …