Recent research has considerably increased our understanding of the factors associated with the American epidemiological transition in the late nineteenth century. However, uncertainty remains regarding the impact on mortality of specific changes ancillary to urbanization and industrialization in American cities and towns. The broad objective of the Connecticut Valley Historical Demography Project is to examine changing relationships between socioeconomic status, the rise of new urban-industrial communities, and cause-specific mortality trends during the rapid development of New England manufacturing. To address these issues, the present analysis examines two emergent urban centers in Massachusetts, adopting a micro-demographic approach to explore late-nineteenth-century and turn-of-the-century determinants of mortality.

By our late-twentieth-century experience, survival of an economy seems critically dependent on well-established rights to private property and a return to labor that rewards greater effort. But that need not be so. History provides examples of microsocialist economies that, internally at least, did not use prices to allocate resources, allowed for little private property for participants, and provided a constant return to labor independent of effort. Some of these economies may even be termed “successful,” if success is taken to mean survival over several generations (Roehl 1972). If these communities survived despite their lacking the conditions generally thought necessary for success, a question worth asking is how this occurred, for we can then shed some light on what really is necessary for economic survival. The role of time is critical here, for even if the microsocialist economies examined in this article eventually became the merest shadow of their former selves, the fact that they did flourish for so long makes them a valuable counterexample and, hence, a phenomenon in need of explanation.

The Civil War was the bloodiest event in American history. For every 10,000 persons, 182 died during the war. Wartime mortality was especially severe for young men; about 8% of all white males between the ages 13 and 43 died (Vinovskis 1990). Among those who survived the war, many were disabled due to wartime injuries. According to the estimate of Claudia Goldin and Frank Lewis (1975), human losses account for 37% and 28% of the direct costs of the Civil War for the North and the South, respectively.

A question of perennial interest to social historians is the degree to which mortality during the war varied across economic classes (Murdock 1964; Hess 1981; Lavine 1981; Riggs 1982; Rorabaugh 1986; Kemp 1990; Geary 1991). Starkly put, the question reduces to this: Was the Civil War a poor man’s fight? In a recent study, Maris Vinovskis (1990) suggested that the answer to the question is yes. Vinovskis found that recruits from disadvantaged backgrounds were more likely to die or to be wounded than servicemen who had higher-ranking jobs or greater wealth, presumably because they were more susceptible to disease or were sent on more dangerous missions. However, he didn’t provide evidence of how the socioeconomic backgrounds of recruits affected their susceptibility to disease and military missions. Moreover, since his study was based on a relatively small sample composed of recruits who resided in a single town, it is unclear whether the results could be generalized for Union army recruits at large.

Sex preference has long been an issue in both mortality and fertility studies. In mortality studies, higher mortality rates for girls than boys have been documented in many regions of the world, especially in South Asia (Waldron 1983). These studies suggest that the patriarchal family system puts more value on male offspring than female offspring and that this valuation determines such mortality patterns (Das Gupta 1987; Levine 1987). Recent studies examining this process of discrimination have found that family composition is a major intervening mechanism (Simmons et al. 1982; Muhuri and Preston 1991). The Japanese family system is especially useful for examining these patterns. It is characterized as a stem family system, similar to other Asian countries in its patriarchal lineage but different in its possibility of female inheritance. Under a stem family system, a son typically inherits the household, but a daughter can also inherit when there are no sons. Despite the uniqueness of the Japanese family system, little attention has been paid to the mortality pattern of children in Japan relative to those in other Asian countries. Evidence is mixed regarding sex-selective practice in early modern Japan (Smith 1977; Hayami 1992; Saito 1992b).