Published online by Cambridge University Press: 11 February 2009
We have extremely strong reasons for supposing that the exposure of infants, very often resulting in death, was common in many different parts of the Roman Empire, and that it had considerable demographic, economic and psychological effects. The evidence for the first of these propositions has been reviewed or alluded to in several recent publications.1 However, a thorough new study, covering the whole of Greek and Roman antiquity, would be worth while. In the meantime Donald Engels has declared that in the Greek and Roman worlds the exposure of children was ‘of negligible importance’ (‘The problem of female infanticide in the Greco-Roman World’, CPh 75 (1980), 112–20).
1 Harris, W. V., ‘Towards a Study of the Roman Slave Trade’, Memoirs of the American Academy in Rome 36 (1980), esp. pp. 123–4Google Scholar; Duncan-Jones, R. P., ‘Demographic Change and Economic Progress under the Roman Empire’ in Tecnologia, Economia e Società nel mondo romano: Atti del Convegno di Como...1979 (Como, 1980), pp. 69–71Google Scholar; on the fundamental discussion by Brunt, P. A., Italian Manpower, 225 B.C.-A.D. 14 (Oxford, 1971), 148–54, see below, p. 116Google Scholar; see also on the Hellenistic and Roman periods Pomeroy, S. B., Goddesses, Whores, Wives, and Slaves (New York, 1975), pp. 140, 164–5, 228Google Scholar. Etienne, R., ‘La conscience médicale antique et la vie des enfants’, Annales de Démographie Historique (1973), pp. 15–46, does not give a clear account of infanticide, but the comments of J. Bourdon (pp. 51–3) are worth reading.Google Scholar
2 On the problems involved in defining infanticide see especially Dickeman, M., ‘Demographic consequences of infanticide in Man’, Annual Review of Ecology and Systematics 6 (1975), pp. 109–10. I shall not be concerned in this paper either with the infanticide of deformed infants or with infanticide by selective neglect (a topic much written about by recent anthropologists).Google Scholar
3 Engels, pp. 118–19. In fact the death rate would not rise quite to 44 per 1,000, since in any ancient society normal mortality included many infants under one year of age, i.e. some of the victims of infanticide would have died of other causes during their first year.
5 See Hopkins, K., ‘The age of Roman girls at marriage’, Population Studies 18 (1965), 309–27.Google Scholar
6 For another formulation of the relationship between infanticide and early marriage see Brunt, pp. 151–2.
7 The basis of Engels’ argument here seems to be that an increase in birth rate necessarily produces an increase, over both short and long term, in death rate. But the only such increase which would inevitably follow would be a temporary one resulting from a temporary rise in the proportion of infants in the population.
8 ‘That the practice was common in Roman times is universally admitted...’, p. 105; ‘exposure was familiar even in Athens of the classical period’, p. 107.
10 op. cit., p. 153.
12 cf. Harris, op. cit. p. 123. The Loeb translator (W. C. Helmbold) mistakenly gives ‘when poor men do not rear their children, it is because...’.
13 Very many instances are mentioned in Carr-Saunders, A. M., The Population Problem. A study in human evolution (Oxford, 1922), pp. 146–9, 169, 179–80, 190–1, 196, 216–22, 255–62 (Rome: 258–9)Google Scholar. For more recent bibliography see Marshall, J. F., etc., ‘Culture and Natality: a Preliminary Classified Bibliography’, Current Anthropology 13 (1972), p. 274, and above all Dickeman, op. cit. pp. 134–7Google Scholar. On prehistoric infanticide see Polgar, S., ‘Population History and Population Policies from an Anthropological Perspective’, Current Anthropology 13 (1972), p. 206Google Scholar. On female infanticide in modern historical periods see the bibliography provided at the beginning of Knodel, J. and Vos, S. De, ‘Preferences for the Sex of Offspring…’, Journal of Family History 5 (1980), 145–66.Google Scholar
14 Petersen, W., ‘A Demographer's View of Prehistoric Demography’, Current Anthropology 16 (1975), esp. p. 234.Google Scholar
15 Dickeman, p. 130.