This handsome two-volume collection joins a number of other recent or forthcoming reference works that cover ancient religious matters.Footnote ¹ Salzman is the general editor with Sweeney joining her for the first volume and Adler for the second. Together they have assembled a very distinguished group of contributors. Classicists will be delighted to see Kearns, Rüpke, van Andringa, Shaw, Kaizer and Salzman herself among the contributors. The collection has been actively and carefully edited, so that the chapters are very broadly similar in scope. Each chapter has a full bibliography, which is up to date and includes scholarship in the major scholarly languages. In short, this is not just another companion. There are c. 60 illustrations and 20 maps. More images might have made it even more useful as a teaching aid. Its real strength is in the high quality of almost all the chapters, which offer clear and authoritative introductions to their subjects, setting religious practices and institutions in wider social and historical contexts. Most also discuss the kind of evidence available, and the majority have taken notice of the editors' injunction ‘to consider how modern theories have shaped and sometimes distorted perceptions of specific religions and their origins’ (vol. 1, p. 6). The essays work well as standalone pieces, but are sufficiently similar in design that they could be used together especially within each volume.

A few words on scope. Geographically the collection covers the Mediterranean World and the Near East (Iran is included but not India). It begins with Sumer and ends with Romulus Augustus. Iconographic and architectural evidence is considered where appropriate, but this is fundamentally a historical not an archaeological enterprise, hence a start date of the Bronze Age in most regions. Fixing limits for collections like this is always a bit arbitrary but a.d. 452 seems a slightly odd choice for the final terminus, making much more sense for the West than for the East, and excluding not just Germans but also and more importantly the emergence of Islam. The Arab Invasions certainly changed the political map in fundamental ways, yet Islam did not appear out of the blue. Its emergence and nature are best understood in the context of late-antique Christianities and Judaisms, and perhaps Sasanian era Zoroastrianism as well. Volume 1 contains fourteen chapters almost all of which consider the religion of particular peoples – Minoans, Hittites, Etruscans and so on – culminating in the Roman Republic and the Celts. The only chapter without an ethnic descriptor is Skjaervø on Zoroastrianism. Volume 2 is very largely a tour of the Roman Empire in nineteen chapters, although De Jong opens with Parthian and Sasanian Iran. The organisation is geographical, Asia, Africa, the Greek East and then Italy and the West. Five chapters deal regionally with the biggest new arrival: ‘Christianity in … Syria, North Africa, Asia Minor, Italy and Gaul’. Judaism has one essay (by Sweeney) in volume 1, ‘Israelite and Judean Religions’, and in volume 2 has another five essays, one each on the Near East before and after the destruction of the Temple, and then ‘Judaism in … Egypt, Asia Minor, Italy and the West’. Some authors are more successful than others in exploring the interconnections between different traditions within a given region, but mostly conventional disciplinary boundaries are respected.

Salzman provides a very good general introduction which among other things notes the limitations of this design:

This is quite correct, and one might add a larger range of thematic issues discussion of which is necessarily dispersed across many chapters: iconography, magic, astrology, initiation, temple architecture, oracles, priesthoods and sacred texts. A general index, an index locorum and the introductions (each editor provides one) to some extent make up for this. The editors have also tried to make sure some themes – gender and identity for instance – are treated in each essay. And of course a thematically organised collection might well have produced a much more uneven geographical coverage.

A different kind of limitation is more problematic given the nature of religious change over the period covered by this collection. For the earlier periods covered it makes a good deal of sense to treat each set of rituals, beliefs, institutions and identities in the local geographical and cultural context of their creation and performance. Doxey's account of Egyptian religion is a wonderful example of the power of this approach. But for later periods this works less well.

The editors treat Alexander's conquest of the East as a game changer. Rather amazingly – in an essay that cites E. Said in the footnotes – Sweeney's introduction to volume 1 states that ‘Alexander's conquest marked the first time that a European culture was able to gain political ascendancy over the Near East’, and treats Hellenisation as an instrument through which the Ancient Near East was dominated and its nations and cultures subverted. Many would prefer to envisage a much longer period of intensifying contacts and movements and connections: Egyptian gods were worshipped in the Aegean world long before Alexander, and we can document a range of syncretisms, translations and borrowings across the Mediterranean from at least the archaic period when Heracles and Melqart or Demeter and Isis began their long associations. It is particularly the new religious phenomena of the last centuries b.c. and first a.d. that fit least easily into a geographical-cum-ethnic frame. The creation of Mithraism is a case in point, drawing on Iranian and Greek inspirations, remodelled to suit Roman social locales, and appropriating new technologies like mysteries and astrology that were shared by many traditions. Any tradition that is not geographically based is difficult to present in this format. And it is not just that the account of Christianity is fragmented into five separate essays: none of them actually tackles either its origins or the reasons for its spread except in passing. The same applies to Orphism or Manichaeism. So paradoxically this History of Religions is unable to discuss some of the major themes of the religious history of the period. Things really did change between 3600 b.c. and a.d. 452. Treating Zeus and Jupiter in separate chapters makes much more sense than separating the treatment of Gallic Christianity from that of Asia Minor: Irenaeus, who divided the religious world so differently, would have been amazed at the importance assigned to geography.

There is also a wider question here about the extent to which the comparative study of religion serves the end of a historical approach. When religious studies in that sense was devised in the nineteenth century it was conceived very much as a humanistic counterpart to the natural sciences. Rejecting pre-Enlightenment classifications such as the division of the subject into Christianity, Judaism, Islam and Paganism the first religious studies specialists were compelled to create some general categories that might correspond to the taxa used by natural scientists. Polytheism, monotheism, animism and of course ‘religion’ itself were the result with ‘religion’ as a master category in some way equivalent to the biologists’ ‘species’. Each religion was deemed to have core beliefs and practices, a membership, a cosmology and so on. For some purposes – as the early chapters of volume 1 indicate – this still works quite well, just as for some purposes – like driving a car – we can behave as if Newtonian mechanics had not been superseded by Einsteinian physics. Cunningham on Sumer and Doxey on Egypt work well because the societies they study were relatively isolated and well bounded. But de Grummond already has to work harder to separate out the core, native elements of Etruscan religion given the intensive connections between the inhabitants of ancient Etruria and other peoples and the extensive exchanges between Etruscans and Greeks, Phoenicians and the many peoples of Italy.

The situation is actually more serious. Some of the most important research on ancient religion has developed from a critique of that paradigm, beginning from a recognition that the scientific category ‘religion’ was in many respects calqued on Christianity, a generalised version of local norms from which specific content had been evacuated. For many – W. Cantwell Smith, J.Z. Smith, T. Asad, B. Nongbri marking different stages and different answers – even to write of ancient religions is a problematic enterprise. Meanwhile we have learned to write of ritualisation rather than rituals, to be cautious of monotheism, polytheism and heresy as terms freighted with ancient as well as modern polemic, and to distrust sharp distinctions between religion and politics (a move which makes it difficult to distinguish ‘religion’ from its context). Religion now seems to be a concept with its own history, and one best suited to describe very modern forms of Christianity and a few other phenomena (like Neo-Hinduism) that have been calqued on it.

Salzman knows all this very well, and tends to write of religious traditions rather than religions and to emphasise geographical contexts rather than ethnic frames. Her introduction gives some orientation towards the most recent literature and the current state of several questions under debate. But only a few of the contributors have either faced the full implications of these debates, or explained why they are not convinced that they matter. Instead they have mostly used conventional boundary maintaining mechanisms to preserve the fiction that each people has in their own place their own religious tradition. So personal mobility is minimised, because otherwise an account of all the rituals practised in cosmopolitan Antioch or Ephesus would become too complex and confusing. Treating the Christians and pagans from each region separately is another one of these mechanisms. The cost is to make it difficult to appreciate transfers of religious knowledge, or to recapture the lived experience of religion in the larger more cosmopolitan centres of the ancient world or in any of its overlapping diasporas.

And ideas widely shared – such as that gods might inhabit images of them – are marginalised. Syncretism – ubiquitous in dedications and amulets and magic – falls between the cracks in an account organised around less hybrid and mobile religions and peoples. What are we to make of the gradually emerging idea that certain texts had a sacred authority of their own? This is a central theme in the replacement of ancient with medieval religious knowledges across the whole area covered by these volumes. Much of the recent work of G. Stroumsa, C. Ando, J. North, J. Kindt, G. Fowden, E. Eidinow and R. Gordon is hardly visible through these chapters. A comparative history of religions approach simply cannot easily treat phenomena of this kind.

These issues do not detract from the utility of almost every chapter in this collection as a record of what we know of religious traditions in each region (and mostly of how we know it). Just as it is perfectly safe to drive a car as if its length was not altered by the speed you drive it at, this reference work remains an invaluable aide to those wanting to get a sense of the diversity of religious forms across this vast geographical and chronological range. But researchers and students wanting to explore the cutting edge of ancient religious history will need to supplement it with some additional reading.