Do not panic! If this 800-page tome arrives on your desk, you are not obliged to read it cover to cover. Of those 817 (to be precise) pages, 623 form an enormous, exhaustive and laudable catalogue of all 1388 cities for the period meeting the criteria to be included in the book. At this level, this is a ‘big data’ project, as not only are there the best part of 1400 sites, but for each, there are standard fields of information such as size, monumentality and civic status, as well as brief overviews and bibliographies. The monumentality entry can vary enormously, from the 236 monuments listed for Rome to the zero of Pausanias’ famously under-monumentalised Panopeus (which therefore cannot make it into this catalogue). This leaves 115 pages of text and general bibliography, 11 pages reproducing 31 tables, then 54 pages with 22 graphs (rank-size distributions) and 90 maps and diagrams. So the ‘operative’ part of the book comprises some 180 pages.

The catalogue is at once a major strength and a terrible weakness. It is the product of a huge labour for which Hanson can only be applauded with wonderment. All 1388 sites are recorded using the same criteria, making them comparable and thus potentially the basis for enormously informative interrogations of the data. But to do this effectively it would all need to be online and searchable. We are informed that Hanson has assembled such a legacy database and that “This could be put online . . .” (p. 8) with provision for searching or augmenting it. Let us hope a reliable platform for this can be found as soon as possible. It is a great frustration that this extraordinary resource is for the moment so laborious to use; indeed, the current format will probably put many off the attempt. A digital version of the catalogue would also have meant that this book could have come in at under 200 pages.

On the basis of his catalogue, Hanson has constructed maps depicting a huge range of locational, chronological, monumental, spatial and other variabilities in the patterning of Roman urbanism. These also constitute a hugely valuable resource and it can be confidently predicted that they will soon become familiar on a thousand PowerPoints (with due acknowledgement).

The text seeks to use all this information to analyse the phenomenon of Roman urbanism under the late Republic and high Empire, both in its own terms and in relation to other pre-industrial urbanisms (particularly in western Eurasia), and ultimately the phenomenon of urbanism over the longue durée. Given the largely spatial nature of the data, it is perhaps not surprising that underlying much of the analysis is an essentially economic model of urbanism, although perhaps more use could have been made of network theory. Hanson's approach fits into long-running debates about the nature of pre-industrial urbanism and of Roman urbanism specifically, especially the ‘consumer/producer city’ debate. The literature on this and other debates is clearly laid out and the present study situated in relation to it. But the Roman city was in important respects an ideological construct. It remains significant that in Europe such places only appeared within the empire, suggesting some cultural/ideological cleavage along the edge of Roman power. Within that line there was an impulse to invest in this form of settlement, even in provinces where it had to be created more or less ex nihilo. Places such as Italica, Lepcis or Constantinople were created or enhanced by emperors for personal and dynastic reasons. Likewise, so much of the monumentality recorded here was paid for as a result of the ideology of euergetism with a calculus of social rather than financial cost-benefit. And still other cultural factors lay behind the skewed distributions of monument classes such as the amphitheatres so clearly mapped here.

Overall, a case is made that the rank-size distribution of Roman cities was less ‘primate’ than often thought and the curve only slightly concave, suggesting a better-integrated urban system (in largely economic terms) than frequently envisaged. There were more cities of the ‘middling’ sort, particularly in the provinces of temperate Europe, which therefore appear less ‘underdeveloped’ than the common perception. The degree of integration is argued to show that though Roman-period urbanism may to an extent be an agglomeration of pre-existing traditions of urbanism, the existence of the imperial system promoted economic and cultural integration over and above what Rome found as she acquired her territories.

The rapid increase in urban places in the first century BC is explained largely in terms of the short-term but huge resources were made available by the acquisition of empire, with the first and second centuries AD as expressive of the resources of a longer-term period of increased productivity (although the latter needs more demonstration), before the system began to break down in the third century under the impact of external factors such as wars and plagues, possibly springing the Malthusian trap. But for all the quantity of Roman cities and the quality of their monuments, Hanson argues that they did not or could not make the economic difference that would have allowed them to break out of the constraints of the agrarian resources and energy sources or the mercantile and financial systems of the ancient world, and to initiate developments analogous to those that took place in the early modern period. In this, as in so much else, the Roman Empire was enormously impressive but ultimately a dead end. Researchers on both the Graeco-Roman world and on world urbanism have much to thank Hanson for in his data and analyses, as well as occasion to grind their teeth in frustration at the rebarbative format in which they are currently available.