The following tables are taken from Richard Saller’s book Patriarchy, Property and Death in the Roman Family; a full explanation as to how the tables were constructed is given in that book (1994:9-70). For our purposes, these life tables provide a guide to the nature of the kinship universe across the life cycle at Rome. The figures to one or two decimal points suggest to the demographically unitiated a high level of accuracy, but it needs to be remembered that the figures are an average figure of all variations across society. Hence, where a reader may think of an example from the ancient world quite different to the average given, there will be another example at the opposite extreme of that experience that would cause the figure given as the average to represent those extremes.
Two tables are given as ‘level 3’ and ‘level 6’ for the senatorial elite, as discussed in the book. The reason two different sets of data are presented is that our estimates of the average life expectancy at birth vary from twenty-five years to thirty-three years. Again we stress here, in response to a general misunderstanding by classicists of average life expectancy, that these averages include those who did not survive the first years of life and those who lived into their eighties and beyond. These figures do not mean that everyone was dead by twenty-five or thirty-three! The tables represent the demographic possibility of a sustainable population into which are placed the paramaeters of the age of marriage, the frequency of remarriage, our understanding of Roman fertility patterns etc. It is a procedure utilised by historical demographers for other periodisations and has been seen as a means to estimate or model: the nature and number of living kin at a specified age; the proportion of people at a specifie age who would have at least one kin relative in each category; and finally the average age of those living kin. In effect, these three tables give us a view into the changing familial structures of the life course.
We stress again, as our final thought on these tables, that the figures are not intended to replicate exactly the Roman life course or all cases; they cannot because there was no single or unified lived experience of birth, marriage, parenting and death. Instead, the model produces a set of probabilities and simplifies the complexities of societal choice to produce a generalised view of