When Computers Were Human (review)

We take for granted the easy computational power of contemporary electronic devices. Yet, even in the recent past, scientific calculations were, of necessity, done by hand, and the laborious job of producing complex scientific calculations was that of human "computers." David Grier's thoroughly researched and detailed book charts the rise of the phenomenon of the human computer. With a nice personal touch, of the sort that is so often sanitized from our accounts of how we come to our research questions, Grier reveals that his interest was kindled by his discovery of his grandmother's education in mathematics at the University of Michigan in the early 1920s. Although she did not subsequently undertake a mathematical 875career, the other five women in her class became human computers.

Grier characterizes the job of the human scientific computer as "blue-collar science." This is perhaps a slightly misleading term, in that these workers were usually highly educated professionals. But it does capture the "back room" nature of their work, and this partially explains why they have been neglected in histories of science and technology. A second reason for such neglect may lie in the recognition that so many of the human computers were women; indeed, computing provided one of the very few scientific career possibilities for women from the 1860s through the middle years of the twentieth century. Despite this, as Grier notes, computing was by no means an easy career choice for women, as they struggled with problems "working in an environment that would constrain their role or even deny that they were part of a scientific endeavor" (p. 84).

Given the immense importance of astronomy for timekeeping, navigation, and surveying, it is not surprising to find the earliest examples of organized scientific computation in the science of astronomy. The first section of the book deals with astronomy over a 200-year period from 1680 through calculations of the return of Halley's Comet and the American prime meridian. Grier uses the successive calculations of the return of the comet to illustrate the changing role of the human computer.

In part 1, Grier identifies three themes of organized scientific computation, themes which are shared with labor history and the history of factory organization: the division of labor, the concept of mass production, and the development of the role of the professional manager. Illustrating the first, he notes that work on the English Astronomer Royal Maskelyne's Nautical Almanac in 1765 was organized as a cottage industry, in contrast to the more radical and bureaucratized division of labor at the French Bureau du Cadre when it became involved in the production of the metric system. Here, a wider appreciation of context might have been employed to good effect, as the British approach toward funding scientific research was notoriously laissez-faire throughout the eighteenth and nineteenth centuries, while the French system favored a centralized bureaucratic system across the range of sciences.

Part 2 extends the theme of mass production, in particular mass production of data. In the late nineteenth century, this was facilitated by a new generation of computational tools, most notably the slide rule, the adding machine, and the punched-card tabulator. The spotlight begins to turn to the social sciences with a focus on census and biometric data. The First World War brought a need for ballistics calculations incorporating a more sophisticated understanding of air resistance. As is so often the case, war spurred technological developments. After the armistice, the vast array of computational machinery and expertise developed during the war in the United States was put to use in peacetime projects, particularly agricultural economics and statistics.

Part 3 looks at the professionalization of the computer. One of the major projects of the interwar years is represented by the publication of comprehensive sets of mathematical tables for scientific and technological purposes. Responding to...