American ethnic groups vary in terms of movement on the economic ladder, position on the political spectrum, and educational attainment. Members of individual groups also congregate in particular occupations; Koreans in Los Angeles are overrepresented in the retail liquor industry (Light and Bonacich, 1988) and Greeks in New England are concentrated in pizza businesses (Lovell-Troy, 1980). While many of these differences may be obvious, an explanation for why they occur poses a serious challenge to social scientists. In response, scholars have proposed two theoretical frameworks, cultural and contextual, for understanding ethnic behavior. A cultural interpretation stresses that American ethnic groups’ behaviors are based on cultural traits brought from their countries of origin. In contrast, a contextual analysis explains ethnic conduct as immigrants’ responses to their new societies.

This paper deals with the distribution of power among Belgian political parties during the interwar period. In the 1930s Belgium, like most European countries, was confronted with the electoral success of extreme right- and left-wing parties that wanted to change the existing political system into an authoritarian one. Usually, historians draw attention to the rapidly growing share of seats in Parliament held by extreme parties as a sign of their increasing influence on Belgian politics. Among game theorists, however, it is widely accepted that the proportion of seats is a poor proxy for power relations (Schotter, 1979). It is indeed possible that a political party acquiring a higher proportion of seats in Parliament loses its capacity to influence the outcome of a vote, and vice versa.

Realignment of the American party system is a major focus for students of voting behavior, congressional policy making, and political history (Key, 1955; Burnham, 1970; Clubb, Flannigan, and Zingale, 1980; Kleppner, 1979; Sundquist, 1973; Campbell, 1977; Trilling and Campbell, 1980; Ladd and Hadley, 1978; Brady and Stewart, 1982; Brady, 1985; Sinclair, 1982). The transformation of the party system encompasses significant shifts in electoral support for the major political parties, changes in congressional policy making, and a redirection of national policy making. The rise of new party eras is intriguing because the constitutional structure seeks to minimize such major institutional and policy change. The separation of powers, federalism, and different recruitment patterns for political leaders are structural factors inhibiting major policy change (Brady, 1985). Realignment overcomes these barriers as highly volatile and cross-cutting issues arise. These issues produce linkages between electoral choice, congressional voting, and the broad shape of national policy making.

During the past two decades, there has been an outpouring of research on the seaboard mortality associated with intercontinental migration during the seventeenth, eighteenth, and nineteenth centuries. The focus of historical interest in this linkage between mortality and migration has been the Atlantic slave trade. We now have mortality rates on voyages from various regions in Africa to various destinations in the Americas, from the late seventeenth century to the mid-nineteenth century (see Curtin, 1968, 1969: 275-286; Klein and Engerman, 1976, 1979; Klein, 1978; Postma, 1979; Miller, 1981; Cohn and Jensen, 1982a, 1982b; Cohn, 1985; Eltis, 1984, 1987; Steckel and Jensen, 1986; Galenson, 1986). These slave studies have spawned renewed interest in the mortality associated with other seaborne populations, and mortality rates have been calculated on Dutch immigrant voyages to the East Indies during the eighteenth century, European convict and immigrant voyages to North America and European immigrant voyages to Australia during the eighteenth and nineteenth centuries, and Indian and Pacific Islander indentured labor voyages to Fiji and Queensland, Australia, during the late nineteenth and early twentieth centuries (see Riley, 1981; Eltis, 1983; Cohn, 1984, 1985, 1987, 1988; Grubb, 1987; Ekirch, 1987; Morgan, 1985; Shlomowitz, 1986, 1987, 1989; McDonald and Shlomowitz, 1988, forthcoming).

The slave trade, death, and misery were inseparable long before abolitionist writers took up the slave trade as a subject in the late eighteenth century. Throughout the historiography there has been widespread recognition that Africans entering the trade died not only during the middle passage but during the process of enslavement and travel in the interior, on the African littoral awaiting shipment, and after arrival in the Americas. Europeans directly involved in the traffic were at risk in the last three of these four phases of transition between life in Africa and life in the Americas, and tended to die at rates comparable to their human cargoes. In the shipboard phase, and probably also in other stages of the journey, mortality in the slave trade normally exceeded that in other long-distance population movements. In the nineteenth century this differential widened as rates on other long-distance routes fell (Cohn, 1984; Eltis, 1984; Grubb, 1987; Klein, 1978; McDonald and Shlomowitz, 1989, forthcoming). To date, most explanations have focused on morbidity and mortality on board ship; data on the preembarkation phases are no more available to us today than to the abolitionists 150 years ago. For shipboard mortality, overcrowding on the ship, psychic shock, and violence have not fared well as explanations in the work of the last two decades, although the interplay between the first two and resistance to disease suggests further consideration. The present study focuses on shipboard mortality, but it is based on a large and complex dataset. It begins with a discussion and preliminary analysis of the nineteenth-century data. This is followed by a review of the various hypotheses on mortality in the slave trade.