URBAN SOCIAL AND DEMOGRAPHIC RECONSTITUTION. THE CASE OF EIGHTEENTH CENTURY ODENSE

Abstract

Due to high levels of migration and the need to combine information from several sources with scanty identifiers, work on urban places in periods before the nineteenth century meets with severe record linkage problems. This article describes a method using houses, rather than families as the central unit when combining individual level data. The resulting database has been used for a comprehensive community study of the Danish town Odense during the 18th century. When studying the lives of ordinary citizens in the eighteenth century, there is a scarcity of narrative source material such as newspapers and diaries that give detailed information about individuals and groups. Denmark is no exception. However, many sources provide individual level data indirectly which may disclose aspects of a population's demographic and social history. This was the point of departure when research began on a detailed study of eighteenthcentury urban life in Denmark. Traditional first-generation family reconstitution studies, however, used only parish registers as sources and used the married couple as the uniting unit around which all information was assembled.1 Typical identification problems met with when creating the correct linkages in these studies, include high name frequencies, unstandardized name spellings, and incomplete information, such as missing age at death. Also, the reconstituted families might not be representative of the whole population, because migrants are harder to follow in the sources. Inor out-migrants are more often excluded in family reconstitution studies mainly due to the lack of essential information such as age, date of marriage or baptismal records for some children. A further drawback arises when a reconstituted locality reaches a certain critical size. The number of records will be so large, that even with automatic linkage, it is difficult to reach a high degree of coverage because too many links are weak