ABSTRACT
Any increase in the infant mortality rate (IMR) is a serious problem not only for the couples directly involved but for society itself. 1 Moreover, changes in infant mortality (IM) have both a selection effect and a scarring effect simultaneously. The former means that, from a eugenics point of view, the lower the IMR, the more babies survive, while the latter sees a lower IMR as having resulted from an improved disease environment for babies. 2
Clear rise and falls in the IMR can be observed over time, but the duration of falls has tended to be relatively short in Japan, as well in European industrialized countries such as England and Wales, the Netherlands and Greece. 3 This indicates that the coexistence of both long-term and short-term determinants of declines in the IMR should be seriously considered in studies taking a demographic historical perspective. 4 In Japan, IM fi rst drastically declined in the late 1920s and the trend has largely continued until today. But why was Japan’s IMR so high until the early twentieth century and what were the short-term, yet effective, determinants of its decline thereafter? When addressing these questions, we must consider that epidemiologic transitions usually occur alongside demographic transitions. 5 Until now, very few studies have considered the relationship between changes in IM and fl uctuations in disease incidence from the perspective of epidemiologic transitions. 6 This chapter aims to shed new light on the existence of the disease beriberi that not only survived during an epidemiologic transition in modern Japan but contributed to the fl uctuations in IM, especially in the city of Osaka, the center of Japanese economy between the early modern period until the early twentieth century and where some marked socioeconomic changes occurred during urbancentered industrialization.
