Words and Power: Computers, Language, and U.S. Cold War Values by Bernadette Longo (review)

Reviewed by: Words and Power: Computers, Language, and U.S. Cold War Values by Bernadette Longo Kendall Giles (bio) Words and Power: Computers, Language, and U.S. Cold War Values By Bernadette Longo. Cham: Springer, 2021. Pp. x + 143. In a world grappling with the role of the human worker amid computer automation, "self-driving" cars, and generative AI, and with technology's elite jockeying over existential risks of AI on the one hand and the dire need to accelerate AI development on the other, in a moment of reflexiveness we might wonder just how we got here. Though there is much more to that story, Bernadette Longo in Words and Power provides a solid foundation, building on her research in writing Edmund Berkeley and the Social Responsibility of Computer Professionals (2015). Instead of a central character and place framing found in a history like George Dyson's Turing's Cathedral (2012), Longo foregrounds words and their importance in the development of a standard, shared terminology to support the growth of early computer development in serving military strategies during the Cold War. Words are also central to her previous books, including Spurious Coin (2000) and, coauthored with David Kmiec, The IEEE Guide to Writing in the Engineering and Technical Fields (2017). Longo is interested in the humanistic study of computer history, and so, as part of Springer's History of Computing series, in Words and Power she provides an externalist and contextual lens, highlighting groups of people working in various institutions and social contexts and showing technological development not as inevitable but contingent. Each of the eight chapters is structured like a standalone academic journal article, with chapter title, abstract, text with embedded citations, and bibliography. Longo argues that the motivation for computing standards derived from early computing device development during World War II. That work was done in isolated labs, resulting in differing computing device designs based on orthogonal technologies, from analog components, servomechanisms, and relays to designs such as John von Neumann's electronic, stored-program End Page 742 architecture. While von Neumann's design eventually became the standard that we still use today, in the 1940s there was no common understanding of what a computer was even supposed to do. In fact, during that time period the original computers were humans, usually female, who performed numeric calculations with calculators, pencils, and paper to support the war effort. But the information needs of the war, such as the creation of ballistic firing tables, quickly overwhelmed the abilities of those human computers, so attention turned to developing more powerful mechanical versions. Once the war ended, tensions with Russia served to highlight the importance of those early computing efforts in supporting a national security strategy based on atomic weapons. Not only were there differing computing technologies to choose from, there were conflicting vocabularies for computing mechanisms that prevented the different labs from being able to communicate effectively, an effort already hindered by national security restrictions. Moreover, the people designing and building those early computers were themselves from different fields, primarily electrical engineering, physics, and mathematics, with their own terminologies and ways of viewing the world. Beyond sharing operating manuals and design proposals, what was needed was a common language just for computing hardware and software, and thus we see in Longo's historical arc through the 1960s the struggle to create such a standard computing terminology, which adapted along with new computing publication venues, curricula, conferences, and a nascent computer industry. Longo thus opens the black box of the creation of the academic field of computer science, deftly illustrating how knowledge is valuable only if it can be communicated. The creation of the computer science profession was neither a straightforward nor obvious process, but instead one of contestation, false starts, coordination, and argument. Longo shows that central to this development was the importance of words, which Longo in part bases on ideas of Michel Foucault's "systematic history of discourses" and Francis Bacon's ideas for the democratization of knowledge through the creation of social institutions. While not only demonstrating the power of language in supporting and even helping accelerate the pace of computer development, Longo's genealogical history of...