Names and texts of Greek authors are abbreviated in accordance with Liddell and Scott, Greek-English Lexicon , those of Latin authors in accordance with the Oxford Classical Dictionary , with only a handful of minor adaptations. For abbreviations of periodicals, the reader is referred to LAnne Philologique .
Preface
The origin of this book may be traced back to my inaugural lecture, in which I argued that the quantitative importance of rural slavery during the second century BC had been greatly exaggerated and that the widely held theory that the free rural population of central-western Italy began to decline in the late 160s BC rested on dubious premises. Even at that time it was clear that, if these basic ideas were correct, it would be necessary to re-examine not only the demographic history of Roman Italy during the last two centuries of the Republic, but a whole range of related issues, such as the economic and social background to the Gracchan land reforms, the gradual proletarianization of the legions, the growth of towns and the multitude of problems posed by the large quantities of archaeological data collected during the field-walking campaigns of the past 50 years.
It was clear from the start that proper treatment of these topics would require a collaborative effort by a team of senior and junior researchers. I therefore owe an enormous debt to the Netherlands Organization for Scientific Research (NWO), whose generous grant made it possible for me to launch an extensive research project involving three PhD students and two post-doctoral researchers.
During the past 5 years, I have incurred great personal and intellectual debts to those who have kept me on my toes by pulling me in completely different directions.
From my friends and colleagues outside the Leiden Institute for History, I should like to single out Peter Garnsey, who volunteered to read the entire manuscript. Only he knows how different some sections of this book would have looked without his perceptive comments. I am also indebted to Bas van Bavel, Professor of Medieval History at Utrecht University, for valuable discussions on historical demography, and to John Bintliff, of the Leiden Faculty of Archaeology, for his acute observations on the methodological difficulties raised by the interpretation of the archaeological evidence. In preparing the final manuscript I have also benefited from the critical observations made by the anonymous readers of the Cambridge University Press.
Of the members of our research group, I should like to thank Saskia Hin for encouraging me not to give up entirely on the fragmentary literary sources. Without this stimulus, the second chapter of this book would never have been written. I also learned a great deal from Saskia Roselaar, whose firm grasp of the legal evidence can only be called frightening, and from Jeremia Pelgrom, who kept reminding me of the many ways in which pre-conceived conceptions of Roman and Italian society continue to shape our readings of the material record. Finally, I cherish happy memories of conversations with Simon Northwood and Paul Erdkamp, who turned out to be inexhaustible sources of information on Roman census procedures, on military matters and on the topic of migration.
On the linguistic front, thanks are due to my long-time friend Rosamund Annetts, who displayed great ingenuity in bringing my English closer to native standards.
As always, my greatest debts are to Olga, for her unwavering and loving support, and to Thijs and Lydian, for constantly drawing me into the fun of family life, travel and sports.
Chapter 1 Evidence, theories and models in Roman population history
1.1 Introduction
The central importance of demographic trends to understanding economic and social developments in pre-modern and modern societies has long been recognized. As early as the eighteenth century, Malthus argued that if humans failed to control their sexual passions the population would inevitably grow faster than its agricultural output. The discrepancy between population and food resources resulting from this would then be resolved primarily by Malthusian checks such as famines and large-scale epidemics.