THE MEDIEVAL HOUSEHOLD: LARGE OR SMALL?

Abstract

N medieval population history where the basic data refer so often to holdings and households, the size of the 'multiplier', or number of persons per household, is obviously crucial. Consequently, ProfessorJ. C. Russell's substitution of the 3.5 multiplier for the traditional five has modified greatly the estimates of medieval population densities.2 Needing a multiplier to use in conjunction with the number of landholders listed in Domesday Book and in the manorial extents, Russell used old and new methods to solve the problem. First, he argued that the group represented by the landholder was essentially the nuclear family, i.e. husband, wife, and children. Secondly, having the figures from the I377 poll tax of the number of persons over age I3 per house, he used his life tables to estimate the number of those who were 13 and under. Thirdly, by analysing the relationships between age distribution, age of marriage, and the percentage of families headed by women, he arrived at the hypothesis that 3-5 was the average size of the family. The construction of life tables and the use of such variables as age distribution and age of marriage marks an important innovation in medieval demographic studies. Professor Russell's work, if valid, thus possesses empirical and theoretical importance of the first order. The validity of his methods and conclusions will be examined in the following pages.