Book Review

David N. Edwards

Books reviewed

PAUL YULE (ed.). Late Antique Arabia. Ẓafār, capital of Ḥimyar: rehabilitation of a ‘decadent’ society. Excavations of the Ruprecht-Karls-Universität Heidelberg 1998–2010 in the highlands of Yemen. xxvii+310 pages, 159 colour and b&w illustrations, 43 tables. 2013. Wiesbaden: Harrassowitz; 978-3-447-06935-9 hardback €98.
Late Antique Arabia. Ẓafār, capital of Ḥimyar: rehabilitation of a ‘decadent’ society. Excavations of the Ruprecht-Karls-Universität Heidelberg 1998–2010 in the highlands of Yemen.

Interacting with the Byzantines and Sasanians, and indeed the Aksumites of north-eastern Africa, the Ḥimyarite polity, centred on Ẓafār (the Safar/Sapphar of Pliny and other Roman sources), was a dominant presence in southern and central Arabia in the early first millennium AD. The research programme reported here, albeit prematurely terminated, represented a bold attempt to establish a firmer basis for understanding the site of Ẓafār and its environs. In a well-watered part of the Yemenite highlands, south-south-east of the modern capital of Sana’a, the remains of the mountaintop site, as now recorded, represent the largest ancient site in the region, after Marib. Dominating much of Arabia a few centuries before the explosive appearance of a new Islamic state from the Hijaz, the importance of the still little known Ḥimyaritic ‘tribal confederacy’ (as it is generally understood), and the site of Ẓafār, cannot be doubted. While raising as many questions as it answers, this volume provides tantalising glimpses of a little-known archaeology, in a region where local South Arabian cultural traditions were entangled with those of its imperial neighbours, as well as local and monotheistic religions of the period.

The survey and excavations are reported in three chapters, with a full report on the most impressive, if still enigmatic, ‘Stone Building’—a substantial ashlar courtyard structure which was the main focus of excavation. A full survey of the city defences, which enclose the mountaintop, was completed although relatively little in the way of internal structures within the circuit could be determined on the basis of surface remains. Test excavations of 19 mainly rock-cut graves and a cluster of larger tombs provide evidence for what may have been quite impressive burials, albeit with little of their original contents surviving. One notable find associated with the latter was a ring intaglio with an Aramaic inscription, a tangible indication of an early Jewish presence in the region.

The artefactual and other finds are presented in varied summary forms, some necessarily incomplete due to the premature termination of fieldwork. A substantial body of pottery (312kg) includes both local and imported material most, if not all, of Late Antique date; glazed Celadon indicates some further medieval activity. Generally undecorated local coarsewares are supplemented by a substantial body (13 per cent by weight) of Late Roman amphorae, sourced to Aqaba, as well as other Roman imports (although very little red-slip). A number of unidentified wares may relate to other external contacts of as yet unknown origin.

Some of the most striking finds relate to the carved stonework of the site and its environs, which has long attracted attention, and which the project sought to record. The sculptural elements, whether in situ in the wall reliefs of the Stone Building, recovered from excavations, or from local inhabitants represent a major body of material; the range of decorative designs include domestic and wild animals, vine/grape motifs, rosettes and a very striking series of bovine heads. The most remarkable find, thought to be a quite late addition to the Stone Building, is a 1.7m-high standing figure, crowned and robed (see Yule 2013); on the basis of other sculptural fragments, it was perhaps one of a number of such figures. Notwithstanding a search for potential Byzantine, Sasanian and indeed Aksumite influences, this would seem to be yet another Late Antique artistic development. Yule would place this in the aftermath of the Aksumite military intervention in Ḥimyar, perhaps the second quarter of the sixth century, although a fourth–fifth century date may be as likely. The function of this building remains open, although it was apparently not a palace.

Small finds include imported vessel glass and beads, copper alloy and iron, a notable copper horse noseband with silver inlays, stone bowls and lamps, seals and intaglios, and a handful of coins. A fragment of a Roman-period inscription in Greek, perhaps dedicatory, is also described along with three Late Sabaic inscriptions (acquired locally) with royal names. These, and 71 smaller fragments, are reproduced in colour plates. Some basic environmental data derived from archaeobotanical and faunal samples indicates much use of hulled barley, as well as oats, some free-threshing and other wheat species, some oil/fibre plants, pulses, grape, date, almond and various wild plants. Most of the animal bones, dominated by cattle (some radiocarbon dated), came from the later phases of the Stone Building, by that stage perhaps used for slaughtering; this material is reported as tabulated metric data. The presence of a few chicken bones may be noted, as well as a few gazelle and ibex. Final sections report on a study of local place names, and a conservation report on the site.

Chapter 13 discusses the various strands of dating evidence relevant to the chronology of the site, making use of 36 radiocarbon dates. Many relate to the third–fourth centuries AD. No definitive statement can be made for the latest use of the Stone Building, although Yule would seem to favour some continued use into the early sixth century, a period when Aksumite intervention is known and with which the sculpture of the crowned figure may be associated. Suggested Christian features of this piece are used to suggest a later rather than earlier date. There is clearly much here to be debated.

The textual report concludes with extensive tabulated data. Attention must also be drawn to an important online resource which complements this book in the form of over 4700 images of sites and finds within the project archive at This may be of interest to many other researchers concerned with the possibilities for providing wider access to archives. For those working elsewhere in the Late Antique world, much seen here will seem familiar, if manifested in less well known South Arabian idioms. The inconclusive nature of much of the work may seem frustrating, although as many will recognise, perhaps no more so than results from more ‘familiar’ regions, such as the Maghreb. Some definitive specialist studies are still awaited and this volume is by no means the final word from this project. The current volume, however, will serve well to establish the very real interest of this archaeology, providing many new insights into a frequently overlooked but crucial part of the Late Antique world.


  • YULE, P. 2013. A Late Antique Christian king from Ẓafār, southern Arabia. Antiquity 87: 1124–35.


  • David N. Edwards
    School of Archaeology & Ancient History, University of Leicester, UK (Email: dne1 [at]