Over the last two decades the historical demography of Roman Italy has been at the centre of a very lively debate, featuring a varied array of sources (e.g. documents, archaeological data, comparative evidence) and their widely diverging interpretations. If the traditional reading of such evidence (‘low count’) has held the Italian free population to have declined from 4.5 (225 b.c.) to 4 million (28 b.c.), this view has since been forcefully challenged and many scholars would now favour higher figures and a growth pattern (‘middle count’: 4.6 to 5.9 million; ‘high count’: 7 to 12 million). This is an issue of no little relevance when one considers how much our current narratives of the political, social, economic and cultural transformations taking place between Republic and Empire are in fact heavily rooted on an assessment of coeval demographic conditions in Italy (e.g. the Gracchan reforms, the development of a slave-based ‘villa economy’).
This volume is the latest in a series of important contributions to this debate. However, it clearly stands out from the rest in its attempt to revive the ‘low count’ by moderating/revising some of its most controversial assumptions/readings, like the quantitative importance of slavery in the second century b.c. and the decline of the free rural population since the 160s b.c. (pp. xv–xvi). Its origins lay both in an earlier article (‘Poverty and Demography: the Case of the Gracchan Land Reforms’, Mnemosyne 57.6 [2004], 725–57) and in a wider project that D.L. himself directed at the University of Leiden (Peasants, Citizens and Soldiers: the Effects of Demographic Growth in Roman Republican Italy, 202–88 bc [2004–9]), whose other published outcomes provide a most pertinent complement to his own book: D.L., S. Northwood (edd.), People, Land and Politics (2008); S. Roselaar, Public Land in the Roman Republic (2010); S. Hin, The Demography of Roman Italy (2013).
D.L.'s approach (Chapter 1) follows from a survey of the tenets of the various demographic interpretations underpinning the ‘low/high count’ opposition (little relevance being ascribed to the ‘middle count’), coupled with a comprehensive overview of the wide range of approaches that have since played a role within the debate (from the use of osteological data to estimates of the carrying capacity of Italy, from assessments of past climatic conditions to comparative evidence from other places/periods). D.L.'s verdict is that ‘we can study the problems posed by the demographic history of Republican Italy by starting from these two rival theories and examining the mutually exclusive interpretations of the written and archaeological evidence that they imply’, thus making it possible to assess their explanatory power on the base of ‘the number of “contrived” interpretations required to make them work’ (p. 38).
The first set of evidence to be so examined (Chapter 2) is the Roman and Italian military mobilisation figures for 225 b.c. as preserved by Polybius (2.23–4). These numbers have been routinely employed to derive an estimate of the adult male population, the latter assumed to have constituted about 1/3 of the total population (i.e. including women and children). D.L. is unconvinced by earlier interpretations and his alternative reading would reduce the original ‘low count’ estimate for 225 b.c. to only 3.9 million. Much of his argument rests on the assumption that the Polybian definition of ‘those of military age’ (hoi en tais hêlikiais) might in fact have included men aged over 46 (seniores) – even though this seems to contradict other remarks made by Polybius himself (6.19). Although D.L.'s lower estimates would imply rather high military mobilisation rates, they are considered ‘entirely plausible in light of the highly specific structures of Italian society’ at the time (p. 77).
One would expect census returns (Chapter 3) to provide more secure ground on which to base population estimates. On the contrary, Republican figures have often been called into question in terms of lack of accuracy, as it is usually assumed that until the mid-first century b.c. Roman citizens had to be registered in Rome (resulting in considerable under-registration rates). D.L.'s reading of the evidence calls into question this assumption as he suggests census-taking to have been already decentralised by 169 b.c. (thus requiring little upward adjustment). The crucial issue, however, lies in the Augustan census figures: were they reporting the number of adult males only (‘high count’) or were they including women and children too (‘low count’)? The former hypothesis stands on excellent philological arguments (as no change in census practice is attested), but its demographic implications (i.e. steady growth during a troubled period) seem to D.L. so unlikely as to rule it out in favour of the latter.
D.L.'s ensuing analysis provides a linear narrative of the transformations of the Italian population between 201 and 28 b.c (Chapter 4). Following a first phase of recovery from the losses of the Hannibalic War (such that a low proportion of proletarian citizens was in fact balanced by an input of cheap slave labour), the free population started to grow again in the course of the second century b.c., eventually amounting to 4.2 million in 28 b.c. Even though D.L.'s estimates fall squarely within a ‘low count’ scenario, he explains the background to the Gracchan land reforms as one of relative population pressure, leading to a general impoverishment of the citizen population and a decline in the number of those who met the minimum wealth requirement for military service (i.e. fewer assidui, more proletarii).
Given the high number of towns in Italy at the time of Augustus, low population levels would imply very high urbanisation rates by pre-industrial standards. This claim is openly challenged by D.L. (Chapter 5), who argues that many such settlements were too small to be accounted for. Once a likely density of inhabitants/ha is applied to the area occupied by actual towns, the resulting Italian urban population is shown to be perfectly aligned with patterns found in other pre-industrial societies. Complementary to this analysis is D.L.'s engagement with the evidence for rural population patterns as provided by survey archaeology (Chapter 6). Despite a clear awareness of the problems involved in the (mis)use of survey archaeology, his handling of the actual data is less than systematic and quite impressionistic (in line with the ‘survey of surveys’ genre), two features that make his claim for a closer match with the ‘low count’ somewhat unwarranted.
This is a very important book and essential reading for anyone interested in the demographic development of Roman Italy. Despite his obvious preference for a revised ‘low count’ reading (pp. 284–8), D.L. has managed to produce a remarkably balanced and honest account of the debate. Although not always at ease with the complex subtleties involved in archaeological interpretation, he has presented and discussed the textual evidence with enviable clarity. Furthermore, the four appendices will provide a very useful base for future discussions about the size of the urban population and the number of Roman citizens living outside Italy. Even though many of his claims remain open to question, D.L. has succeeded in showing us that the ‘low count’ cannot be so easily dismissed.
