:The Shady Side of Fifty: Age and Old Age in Late Victorian Canada and the United States

In this book Lisa Dillon cooks a remarkable statistical menu, squeezing, dicing, processing, and combining Canadian and American censuses and flavoring them with supplemental material to create offerings that fill a scholar's appetite for social history. Starting with the premise that age is an overlooked historical category, she sets out to explore old age as “both as a concept and as a lived experience, highlighting the evolution of old age phases” (p. 5). She then organizes her analysis around an intensive comparative examination of the Canadian population censuses of 1871 and 1901 and parallel U.S. censuses of 1870 and 1900. Her analytical tools include not only statistical methods—basic ones such as cross-tabulation and more sophisticated refinements such as logistic regression—but also discourse analysis of the ways in which census directors and others have categorized age, and especially old age. She also integrates qualitative evidence for various trends and observations from diaries and other literary sources. Making sense from these data and sources, Dillon rightly claims, was “a hell of a job” (p. 29), but the payoff is worthwhile. A focused examination of old age from all of these data does not really occur until halfway through the book. Before then, Dillon presents a fascinating history of how the Canadian and American governments—namely, those in charge of taking the census—created age categories and then instructed their enumerators how to record age information. These decisions and instructions reflected social and political purposes, and enumerators had to make choices that embodied official priorities. Moreover, other choices had to be made in the ways that the collected information on age was aggregated in, or omitted from, tables that included other variables such as occupation, place of birth, and relation to family. The 1870–1871 censuses of the U.S. and Canada resembled each other in their collection of age data—both of which were more detailed than in previous censuses—but differed in that the U.S. government seemed more interested in using age data to understand social and economic changes resulting from immigration and labor-force composition while Canadian officials seemed more focused on personal-level demographics such as marital status, causes of death, and health problems. By 1900–1901, the United States was collecting more elaborate age data than Canada, but neither country gathered information in a way that reflected interest in family composition or the status of different age groups within the family. Dillon follows this examination with a somewhat protracted investigation of age misreporting: the tendency of people to lie about their actual age and the tendency of census enumerators and aggregators to “heap” individual ages into rounded categories of age groupings that ended in 0 and 5. These practices, she notes, are important for understanding consciousness of age. In general, Dillon concludes that most people reported their age truthfully to census enumerators, but she also hypothesizes from this analysis that age awareness and heaping into a younger age increased in middle and old age and that age heaping was more common among racial minorities. In the subsequent chapters that focus more directly on old age, Dillon unsurprisingly discovers diversity in the lives and statuses of the elderly, and her multivariate analyses mostly confirm what cross-tabulations and other simpler statistics suggest. Still, the appeal is in the details. Generally, Dillon finds that the status of older women depended on their economic resources, their children, their marital status, race, ethnicity, fertility, and urban living. The most revealing change from 1870–1871 to 1900–1901 among these women was that in the later period fewer lived with family members than in the earlier period, a pattern that Dillon asserts was a “harbinger of more significant shifts toward nonfamily living arrangements during the twentieth century” (p. 178). Patterns among men, she shows, suggest different thresholds for old age, depending on a number of socioeconomic and family characteristics, most particularly presence or absence of a spouse. Also, men more than women seemed to fall into categories of late middle age, young-old age, and old-old age. Dillon concludes by using her microdata to examine the existence and consequences of grandparenthood, showing that three-generation households tended to be more common among U.S. whites, for economic reasons, than among blacks, Amerindians, Asian Americans, and Canadians. The book does not make for a riveting read, and it should be of interest chiefly to scholars in fields of history, demography, and sociology. Yet Dillon presents her microdata and painstaking parsing of qualitative information in a way that is far from dull. Except perhaps for the dry, somewhat repetitive review of scholarly literature in the long introduction, her analysis makes quantitative history much more engaging than one would expect.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.