Urban Demographic Stagnation in Early Modern Germany: A Simulation.(Statistical Data Included)

Although the demographic consequences of the Thirty Years' War (1618-1648) have received considerable scholarly attention, little is known about the population recovery that followed. This research note, with the aid of simple simulations, seeks to specify the limited capacity of small towns in South Germany--a region that boasted a dense and vibrant urban network during the fifteenth and sixteenth centuries--to make up the demographic losses suffered during the war. The findings suggest that these towns could grow rapidly only through substantial migration and that the absence of such growth indicates a fundamental restructuring in patterns of rural-to-urban migration. The methodological difficulties of obtaining reliable information about small-town population growth warrant the use of simulations for exploring the urban demographic stagnation that gripped South Germany. [1]

THE SHARLIN HYPOTHESIS RECONSIDERED At least since the 1741 publication of Sussmilch's Die gottliche Ordnung in den Veranderungen des menschlichen Geschlechts, the first study to document the yawning deficit of births vis-a-vis deaths in early modern cities, historical demographers have generally agreed on two essential and interrelated features of the urban demographic system. First, according to the law of natural decrease, or urban graveyard effect, extremely high mortality rates prevented city populations from reproducing themselves. Population density, inadequate sanitary conditions, poor housing, and related factors made cities breeding grounds for disease and death. Second, city populations could increase only as a result of substantial in-migration that could overcome the population losses caused by the high mortality. Historical demographers have readily postulated these two features for cities with more than 10,000 inhabitants. The parish registers of every early modern city of that size reveal that the annual number of deaths invariably exceeded that of births. Without a stream of migrants to offset the difference, these cities would have gradually declined in population. But because the parish registers of urban communities with fewer than 10,000 inhabitants show that the annual number of births usually equaled or exceeded that of deaths, historical demographers implicitly assume that the law of natural decrease did not apply to small towns and that in-migration did not serve the vital function that it did in large cities. [2]

The size of the surplus of births over deaths in German small towns after the Peace of Westphalia varied considerably (see Table 1). Durlach had a substantial surplus, but Schwabisch Hall had a deficit. Luzern's hefty surplus during the first four decades of the eighteenth century became a deficit for the next six decades. What is the significance of this range and variability? If these towns had closed populations, the difference between births and deaths would indicate the community's ability to reproduce itself. Migration greatly complicates the situation, however; the demographic experiences of two groups--migrants and non-migrants--must be isolated and compared. Sharlin critically examined one manner in which the presence of migrants and non-migrants in cities could dramatically affect the ratio of births to deaths. His analysis, which challenged the conventional wisdom concerning natural decrease in preindustrial cities, merits careful consideration. [3]

Sharlin's challenge rests, in large measure, on the recognition that two distinct groups of city dwellers--permanent residents and temporary migrants--made entirely different contributions to the births and deaths recorded in the vital registers. Sharlin argued that the law of natural decrease prevailed only among the temporary migrants. Be cause they were overwhelmingly single, they contributed few births but more than enough deaths to cancel the surplus of births produced by the permanent residents. Aggregating the two groups of city dwellers results in a greater number of total deaths than births, but this negative balance reflects the population's composition, not the level of urban mortality. [4]

Sharlin's critical reassessment of the law of natural decrease excluded towns with less than 10,000 inhabitants. Since the vital registers from these towns generally showed birth surpluses, he assumed that the law of natural decrease did not apply. But this recorded surplus of births over deaths--which, as Table I indicates, did not occur everywhere--does not by itself conclusively demonstrate that the law of natural decrease did not prevail. [5]

The argument can be rigorously stated with reference to the net reproduction rate (NRR), a demographic measure of the female population's reproductive potential. The NRR is the ratio of the number of daughters that a cohort of newborn female babies will bear to the number of babies originally in the cohort, assuming fixed age-specific mortality and fertility rates; it expresses the cohort's ability to replace itself. In the calculation of the NRR for a female birth cohort, the mortality rates are used to determine the fraction of the cohort that will die in infancy, childhood, and early adulthood without reproducing any children. The mortality rates are also used to trace the continued decrement of the cohort during the childbearing period. Some women will bear at least one child but will die before reaching the age of menopause. Since their reproductive spans are trun cated, they bear fewer children than would otherwise be the case. Finally, once the number of women alive in each of the childbearing years h as been established, the fertility rates determine the number of daughters that they will bear. The ratio of the number of daughters to the size of the initial birth cohort, the NRR, can be less than, equal to, or greater than unity, indicating that the number of daughters born is, respectively, less than, equal to, or greater than the number in the original cohort of newborn females. [6]

The births and deaths recorded in a town's vital registers give no reliable indication of the NRR'S value, because, similar to Sharlin's argument, two distinct populations--natives and migrants--contributed unequally to the recorded vital events. In particular, migrant mothers bore a substantial number of the infants included in the birth totals. Both native and migrant mothers came from much larger birth cohorts that were drastically reduced by infant and childhood mortality. The calculation of the NRR takes this mortality into full account; the town's death registers did not do so. The registers captured the native female cohort's infant and child mortality, since this cohort was actually present in the town when exposed to the risk of dying, but they did not capture this mortality for migrant females, most of whom did not reside in the town during their infancy and childhood. Since migrants generally migrated as adolescents or young adults, the parish registers contain no evidence reflecting migrant infan t and child mortality; only adult mortality left a wake. In other words, the town's death registers did not record the early mortality of the original birth cohort from which non-native mothers came. Because infant and child deaths constituted the bulk of annually recorded deaths, migrants almost certainly contributed disproportionately more to the birth side of the ledger. [7]