Roman Life Expectancy: The Pannonian Evidence

IN A RECENTLY PUBLISHED ARTICLE, I used the document commonly called Ulpian's life (Aemilius Macer, Dig. 35.2.68 pr.) in order to develop a hypothetical Life Table for the Roman Empire.' At the end of that article, I argued that, in all likelihood, only new evidence (such as a more complete collection and analysis of skeletal remains) will prove final and decisive in establishing the validity of my hypothetical Life Table. In the present article I will discuss some little-known skeletal evidence from western Hungary, evidence that offers considerable support for my Life Table. This evidence has the additional merit of helping to explain why the epigraphic data on mortality, despite their almost overwhelming quantity, are of marginal use to historical demographers. Table I presents the basic details of my Life Table.2 A life table is a device for describing, through statistics, the mortality experience of a given population; it might be thought of as a series of snapshots, from different angles, at the exact birthdays of the members of the population. By convention, these data are given for birth (age 0), age 1, age 5, and then at five-year intervals up to 80. In my Table I, the first column, q(x), describes the likelihood that a person of age x will die before the next indicated birthday, the probability of death; thus, the Life Table indicates that about 10.6 of every 100 persons reaching their thirtieth birthday will die before their thirty-fifth. The second column, 1(x), provides the corresponding figure for the number of survivors from an original cohort of 100,000 babies; by following these figures down the column, one can observe the toll of mortality upon the model population. In my Life Table, about two-thirds of all live-born children are dead by age 30. The third column, e(x), gives average life expectancy; those persons who reach the birthday of age x can anticipate living this many additional years, on average. A person of age 30 has, in my Life Table, a mean of about 23 years of life remaining. Finally, the fourth column, 1000m(x), gives the age-specific death rate per thousand from age x to the next indicated birthday; this statistic, called the central death rate, gives an idea of the rate of mortality during the interval: from age 30 to 35, about 22 per thousand.