Fergus Millar examines the centuries following the fall of Rome.

This massive volume, full of varied and interesting material, amounting to more than any single writer or reader could possibly master, represents the culmination of a major enterprise by Cambridge University Press - and is thus very suitably edited by three Oxford graduates, two of whom hold posts at Oxford. Written in the shadow of two great historians, Gibbon and A. H. M. Jones, whose Later Roman Empire (1964) stands out more and more clearly as an unmatched masterpiece, it also reflects a new conception among ancient historians of what "their" field covers: not just the pagan world of classical antiquity, but the Christianised, rapidly changing world - or worlds - of what we have learnt from Peter Brown to call "late antiquity". The volume under review could well be used alongside that produced by Harvard University Press in 1999, Late Antiquity: A Guide to the Postclassical World , edited by G. W. Bowersock, Peter Brown and Oleg Grabar.

Advancing into late antiquity was a welcome new step as regards the new edition of CAH, whose first edition had stopped, in 1939, with volume 12, in the year AD 324, halfway through the reign of Constantine. In the new edition, volume 13 covers AD 337-425, so stopping half way through the reign of Theodosius II, while 14 goes from there to 600. There will never be agreement about starting or terminal dates for the "later Roman empire" or "late antiquity", or about a starting date for "Byzantium", and nor do such questions of terminology really matter. But nonetheless, the division between volumes is clearly misjudged, and an opportunity has been partially thrown away. Volume 13 should have continued to the 450s, to cover the whole of Theodosius's reign (the longest of any Roman emperor), and with that the Theodosian Code, the first and second Councils of Ephesos, and above all the Council of Chalcedon in AD 451, and the defeat and death of Attila in 453. Not only do the 450s mark a real turning-point, in both West and East, but this starting-point for volume 14 would have made it feasible to continue the volume to the most clearly marked turning-point of all, the Islamic conquests of the 630s and 640s that suddenly took Syria, Palestine, Roman Arabia and Egypt out of the Christian empire.

However, volume 14 is still a remarkable achievement, which manages to combine narrative chapters with others looking at structures and patterns, for instance in administration or religious life or education (by the late Robert Browning), and with very informative sections on the arts and buildings and architecture (the latter by Marlia Mango). Noteworthy also is the chapter on the cities by J. H. W. G. Liebeschuetz, significantly titled "Administration and politics in the cities of the fifth to the mid-seventh century, AD 425-640", a foretaste of his major work, just published, Decline and Fall of the Roman City . But the great advance on volume 13 is the inclusion of a series of regional chapters. These serve to bring out with great clarity just how profound the contrasts between the rapidly fragmenting Latin West and the relatively stable Greek East are. Roger Collins, writing on the western kingdoms, or Ian Wood, on the northwestern provinces, or Mark Humphries on Italy AD 425-605, have so many essential events to record that the treatment could hardly go much beyond narrative, even if extensive documentary evidence were available. It is true that, by contrast with the settled empire of the first couple of centuries AD, there is everywhere a relative lack of local documentation in the form of inscriptions or of papyri. But nonetheless the reader who passes from the "western" chapters to the "eastern" ones (Charlotte Roueche on Asia Minor and Cyprus, Hugh Kennedy on Syria, Palestine and Mesopotamia, or James G. Keenan on Egypt) at once enters a different world, which can be known from a multiplicity of sources, and where religious and social change takes place within an essentially stable framework.

ÁñÁ«ÊÓƵ

The East also benefits from three remarkable chapters on societies or regions marginal to empire: the Sasanid monarchy by Ze'ev Rubin, Armenia by R. W. Thomson, and the Arabs by Lawrence Conrad. The last, however, misses the opportunity to bring out the crucial importance of the scraps of writing in pre-Islamic Arabic, which are preserved on inscriptions from the margins of the provincial zone: above all perhaps the bilingual inscription in Greek and Arabic of AD 568 from the Lejja, south of Damascus, recording the foundation of a Christian martyrium by an Arab phylarchos , or tribal leader. Nothing in this period was to have more fateful consequences than the transmission of monotheism and the biblical tradition to Arabic-speaking tribal peoples on the fringes of the empire.

Most readers, or users, of CAH 14 will probably turn to the individual sections that interest them, all equipped with excellent bibliographies. But there is much to be said for reading it cover to cover, which is highly educative and gives a strong impression of the vitality of this area as a field of study in the contemporary world.

ÁñÁ«ÊÓƵ

But, as with the other volumes in the series (which is now complete except for volume 12, on AD 180 to 337), the vividness of the impression left could have been greater, and much is lost by the effect of a general characteristic of style and presentation: the almost complete absence of direct quotations from sources. In this volume we are told with great learning about the world of late antiquity and the mentality of those who lived in it, rather than having them displayed to us. This is all the more of a pity in that, through legal sources above all, we have a mass of testimony, dealing with government, society and religion, and often equipped with elaborate explanations and self-justifications, issuing from - or in the name of - the emperors themselves. This is true, firstly, of the pronouncements included in the Theodosian Code, whose compilation was completed in AD 437. But while the code has been intensively studied, far less attention has been given to the extensive, and sometimes remarkably idiosyncratic pronouncements of later emperors: few are more idiosyncratic than that issued by Justinian in AD 553 on the rules under which Jews were to conduct worship in synagogues. But of this endless stream of imperial pronouncements, not a single one (so far as I can see) is represented in this volume by even a partial quotation. Late antiquity is portrayed here in all its variety; but even the emperors are not quite allowed to speak for themselves.

Fergus Millar is professor of ancient history, University of Oxford.