Women in 1900: Gateway to the Political Economy of the Twentieth Century (review)

In 1900 American women could not vote, serve on juries, run for elective office. Women’s economic situation fundamentally depended on their marital status. The overwhelming proportion of adult women married. In 1900 85% of women over the age of 25 in the U.S. were married or widowed. Women’s legal, economic and social identity was subsumed into their husbands’ under the traditional legal theory of marriage known as coverture. If single, women were potential wives and mothers, primarily young women who would come under the economic protection and “cover” of their husbands in the near future. Widows were at the end of their lives, again without the protection of a spouse, but presumably provided for by their dead husbands or grown children. The small number of women outside such family relationships were truly “women adrift” in Joanne Meyerowitz’ term, ¹ and the subject of much worry about the failure of society to provide for them properly.

A century later, these formal legal dependency relationships for adult women have by and large disappeared, though not the vestiges of coverture and discrimination in labor markets and social institutions. Women still marry, bear children, and confront the life course stages based upon their marital status and biological clocks. Today they do so, however, in an equal opportunity society. Americans live longer, have smaller families, and thus the rhythms of the life course are fundamentally different from those a century ago.

Christine Bose asks what the world of women in their economic and family roles was like in 1900 by way of making an implicit comparison with the political economy of women’s situation today. She uses the public use microdata sample (PUMS) from the 1900 U.S. population census. ² Bose systematically examines women’s situation in terms of work and family and household systems, and provides a nationally representative analysis of both the fundamental patterns of all women’s lives, and the diversity of experiences based upon women’s ethnic, racial, regional situation, and urban or rural residence. The book is a model of clear exposition of the national patterns with their local or regional exceptions, and thus will be very useful in evaluating both the temporal comparisons from the beginning and end of the twentieth century, and for placing the historical case study literature into a national context.

Needless to say, the published results of the 1900 census already provide much of the gross patterning that Bose interrogates. So why bother with a reanalysis? She provides two answers. First, she has a twenty first century vantage point, and is particularly interested in seeing if work patterns for adult women might look different (and similar to those today) if the unmeasured work of women in family enterprises, particularly farms, was added to work performed in the market economy. Second, she wishes to read through the 1900 census assumptions about women’s “proper” roles based upon the then conventional notions of dependency and coverture to see if the census data themselves need to be restructured to reveal different patterns. Accordingly, she reorganizes and reinterprets the way the census reported “families” to identify “hidden” female headed households and the “unemployed” housewives who worked at home. She employs multivariate statistical analysis to evaluate the determinants of particular patterns. For example she compares being a head of household as recognized in 1900, vs. being a “hidden head” and asks if the determinants of the different statuses were “cultural,” that is a function of one’s racial or ethnic status, or a function of the strong geographic concentration of particular groups of women. She argues that once one takes the geographic concentration of African American and non immigrant Euro American women in rural areas into account, many of the reputed “ethnic” differences disappear.

These reinterpretations and analyses need to be performed on individual level data, like the census public use samples. Bose estimates that in 1900...