How Computing Became Masculine in the Cold War

How Computing Became Masculine in the Cold War

 

Programming was not born male, but rather had to be made masculine.Nathan Ensmenger

Introduction

The computing industry offers a unique perspective on Cold War gender roles. Unlike factory or shipyard work, for which there was a masculine precedent prior to the Rosie the Riveter-era of feminism, there was no set computing culture prior to the beginning of the Cold War because the industry did not exist. As the Cold War progressed and traditional gender roles (particularly, concerns about masculinity) were emphasized, this changed.1  The shift from a gender-neutral to masculine culture of computing can be seen in Cosmopolitan magazine’s 1967 article “The Computer Girls,” which was written just before significant masculinization of computing started and offers insight into the gender dynamics of the Cold War in an industry that did not exist prior to the start of the geopolitical conflict. We see evidence of this gendering through computing recruitment and culture, as well as through the identity of the male programmer.

Postwar Computing Culture and the ENIAC Girls: The Origins of Programming

America’s first fully electronic programmable digital computer, The Electronic Numerical Integrator and Computer (ENIAC), was completed in November 1945, two months after the end of World War II. The ENIAC Girls were the six female operators who worked on this machine. During the wartime labor shortage, computing became a promising alternative for women because of broader career options, especially compared to traditional fields.2 Labor in computer science was still highly gendered, however. For instance, the ENIAC’s hardware was exclusively designed by men,3 whereas the software was female territory. As Janet Abbate explains in her history of women in computing, “operating and programming computers could more easily be constructed as women’s work,” especially given women’s prewar employment in calculations (as “human computers”).4

Programming occupations were not stereotypically masculine because they did not exist before the war. Digital programming began in an American society with a significant deficit of an available male workforce. When men came home from World War II, they returned to the same fields they had left. Because few men had worked in computing prior to the war, women were not pushed out upon their return, and the demand for computers increased postwar. Computer work was uncharacteristically feminine for the time period. Though it was delegated to them because of the perceived clerical nature of the job,5  it was still women’s work.6

Still, 1940s computing culture was far from a feminist ideal. The women who worked on the ENIAC were often left out of the news coverage and general narrative about computers. The very title “ENIAC Girls” omits the names of the women who worked on the machine from the narrative, even when their contributions are discussed in the collective. They were cropped out of photos and not included on panels relating to the machine, despite their intimate knowledge of its workings.7 This lack of coverage combined with the Cold War emphasis on traditional gender roles contributed to the dominant narrative that women should return to pink-collar jobs.8 This was perpetuated by the government and media alike. However, “a fair number [of women] did not leave the [computing] workforce, a fact that the Department of Labor acknowledged even as it urged women toward teaching.”9 In fact, the number of women entering the computer science field continued to steadily rise until 1987, when the proportion of women in professional computing reached its all-time high at 38 percent before sharply declining into the twenty-first century.10 Even though computing had a feminine history, as Cold War assertions of American masculinity rose, computing became relentlessly masculine, kicking women out in the process.

From Computer Girls to Computer Boys: How Computing Became Masculine

There were a variety of practices that employers used to hire programmers. The majority of these practices were often proxy measures, because employers generally had little knowledge of programming and used social categories or other markers to make decisions. Three of the most prominent measures were college degrees, experience, and aptitude tests. Both college degrees and experience measurements had benefits and detriments for women, but aptitude tests provided the easiest entry for women and were often the preferred measure.11 Women could sometimes avoid gender bias or discrimination and instead be hired based off of their “innate” ability to program; the system was not entirely equitable but was ultimately more approachable for women. “The Computer Girls,” Lois Mandel’s 1967 Cosmopolitan article, incorporates all of these different entry mechanisms in a way that gives practical advice for women of all backgrounds looking to become programmers. The tests and other standards for employment are explained through examples of women in the industry, and women are given measures for their likelihood of success through examples of certain tasks that were likely to be within any person’s frame of reference (e.g., school aptitude tests).12 The article even includes a sample problem from the National Cash Register Company aptitude test with helpful hints about how to approach the problem and an encouragement to not overthink. The programmer in the example is a “she.” There is a clear intent for women to see themselves as part of the computing culture and as having the ability to become a “computer girl.”

“The Computer Girls” quotes the (male) director of education for the Association for Computing Machinery (ACM), who said: “Many women still don’t think it’s feminine to be mathematical. In our culture, girls aren’t encouraged along these lines.”13 He went on to suggest that more women should go into computer work in college. Computer science was treated as something that could be and, in the eyes of this director, perhaps was feminine. Computing professions could encompass women and feminine labor; the article notes that there are “some twenty thousand computer girls and one hundred sixty thousand computer men!”14 This lack of women was perceived as an issue, both by the article and men in organizations like ACM. This encouragement, though laced with gendered rhetoric and an emphasis on continuing domesticity at work and at home, implied that computing was a good fit for women.15 The article remarked that “if a girl is qualified, she’s got the job. There’s no sex discrimination in hiring.”16 Though we know this to be untrue, the insistence that computing was more open than other scientific fields and offered the same kind of opportunity for women as teaching painted a narrative that programming could be a women-friendly job.17

The association of mathematics with women, however, proved to be problematic for women in the industry.18 The fact that the mathematics aspect of computing was more welcoming to women than other prominent disciplines associated with programming (engineering and business, for instance) “meant that too close an association with mathematical work could threaten the professional status of programmers … Mathematical training could be a route into satisfying programming jobs for women but only—paradoxically—if those jobs were not too closely associated with ‘women’s work.’”19 As Cold War ideology about women’s role in society “caught up” to the computing industry, the desire to disassociate from female labor was heightened further. According to Geoffrey Smith’s study of gender in the Cold-War U.S.,

middle-class women who worked or who aspired to professional careers threatened national security, a reversal of conventional wisdom during the Second World War. Many experts advised that female freedom (a.k.a. ‘work’) would lead to ‘race suicide,’ promiscuity, and the consequent explosion of general disease, and the devastation of American youth.”20

Even in an article that masquerades as (and in all likelihood believed itself to be) feminist, “The Computer Girls” still devalues female labor as it encourages women to join the workforce. For instance, one anonymous male programmer is quoted as saying: “And we like having the girls around. They’re prettier than the rest of us.”21 The emphasis was not on women’s skills or intelligence—even in a women’s magazine, in an article that showed cultural awareness of how women were being pushed away from programming, their work was devalued to assist masculinity. As another example, the article emphasizes the likelihood of finding a husband or a “nice male programmer to take a girl home.”22 The work that women did was overshadowed by traditional gender roles, even prior to the significant masculinization of the 1970s.

The devaluation of female labor followed computer science into the latter decades of the twentieth century as women were forced out of computing, even as “The Computer Girls” paradoxically quoted women who said that they were already “fully accepted as a professional” in the sixties.23 The hiring process was still taking place in a field viewed as masculine in a time when female labor outside of the home was seen as a risk to national security, hence the ideological shift toward male programmers. Nathan Ensmerger’s research into masculine culture in computing finds that the shift, “can be explained not only in terms of the professionalization of the discipline but also by reference to very specific structural mechanisms, such as the use of psychometric testing in corporate hiring processes.”24 The belief that programming was an “innate” ability continued into the seventies, but along with the “objective” aptitude tests came personality profiles. The profiles systemically favored candidates who “work[ed] more with machines than with people,” which was effectively coded to be masculine.25 Ensemenger notes that “the association of masculine personality characteristics with innate and intuitive programming ability helped create an occupational culture in which female programmers were seen as exceptional or marginal … [O]nly by behaving less ‘female’ could they be perceived as being acceptable.”26 Because of this, the selection process systemically favored male candidates. This was radically different from the gender-inclusive computing culture described by “The Computer Girls” in the late sixties, in which there was apparently “no sex discrimination in hiring” and femininity was a part of the ideal and plausible computing future.27

“The Computer Girls” promised women a future in which they could be a part of the computing culture. However, as male programmers became more and more anxious about their association with women in their field, they began to work against that future. As Ensmenger put it: “Seen from the perspective of aspiring computer professionals (primarily male), ‘The Computer Girls’ article represented not a celebration of the openness and opportunity inherent in their industry, but an indictment of everything that was wrong with it.”28 The negativity associated with women in computing was informed by a Cold War culture that suggested that any association to femininity was unmasculine, queer, and thus communist. Geoffry Smith writes that in the 1950s, “a man who did not execute his masculine role, epitomized the following equation: ‘I am a failure = I am castrated = I am not a man = I am a woman = I am a homosexual.’”29 The culture of masculine anxiety sparked by the conflation of queerness with communism (and, thus, the conflation of men exuding feminine traits with communism) persisted into the 1960s and 1970s. The association to femininity was a constant source of career anxiety for programmers, just as it was to government employees during the Lavender Scare (perhaps not with the same level of direct threat of being outed or fired, but rather with the threat of being associated with communism or being anti-American). A similar line of reasoning could be made even into the 1970s, when female programmers were doing the same work as male programmers. In the seventies, the equation might have begun with “I am a programmer” prior to “I am a failure” because of its female association, which led to increased performances of masculinity in the field. As a result, women left or did not try to enter the workforce at all. The male-dominated culture was cited by many women as an obstacle to their continued participation in computing. 30 They left because the culture excluded them and the male-dominant culture thus became self-perpetuating.

Conclusion: Why is Computing the Outlier?

Second-wave American feminism was at its height as the masculinization of the computing industry occurred, creating an ironic contradiction. Why was computing not subject to this feminism? Furthermore, why was the “long-term drop in the proportion of women [in computing] counter to the trends in all other STEM disciplines”?31 There are a variety of reasons for this, but I will focus here on the impact of domestic metaphors for computing at the beginning of second-wave feminism. In the 1950s and 1960s, some women offered domestic metaphors (e.g., dress making or cooking) to encourage women to join the workforce and effectively soften the blow of female labor; but as Abbate asks: “[W]ere they promoting gender equality or merely demoting programming to the kind of routine clerical work to which women had traditionally been consigned? … [Ultimately, d]omestic metaphors took advantage of this ambiguity [between skilled work and women’s work] to invite, explain, or justify women’s participation.”32 By domesticating the work women did, it became acceptable in the Cold War era. Considering the patriotism associated with domestic work “for the nation” (by teaching, child rearing, or, in this case, perhaps programming), this narrative echoed common Cold War sentiments but was ultimately overpowered by the desire for masculine traits present at the same time.33 Feminism in computing was co-opted by the reification of gender and masculinity at the beginning of the second-wave movement, ultimately creating an industry uniquely dominated by men. The repeated challenges for and denial of women to do this work were a win for the sociocultural battles of the Cold War. Women have been largely erased from the computing narrative as a result. Ensmenger writes that: “[M]ost computer work—again, particularly computer programming—began as women’s work. It had to be made masculine.”34 This points to the larger sociocultural trend in the Cold War of relegating women to the home and making men “manly,” as seen in the gendered rhetoric of The Computer Girls. Given the role of computers in national security and surveillance today, this points to a continued desire to maintain masculine power and patriarchy.35 The legacy of the Cold War in the computing industry continues to have significant ramifications for women.

  1. Geoffrey Smith, “National Security and Personal Isolation: Sex, Gender, and Disease in the Cold-War United States,” The International History Review 14, no. 2 (May 1992): 306- 337.
  2. Mathematical fields often had long histories of bias toward men, hence why computing was significantly more appealing to women with backgrounds in mathematics. (Abbate, Recoding Gender, 18-19.)
  3. Women were excluded from engineering and were very minimally involved in the hardware as a result.
  4. Janet Abbate, Recoding Gender: Women’s Changing Participation in Computing (MIT Press, 2012), 20.
  5. As the ENIAC girls’ intimate understanding of programming shows, however, this work was by no means trivial.
  6. Nathan Ensmenger, “‘Beards, Sandals, and Other Signs of Rugged Individualism’: Masculine Culture within the Computing Professions,” Osiris 30, no. 1 (2015): 38-65.
  7. Jennifer S. Light, “When Computers Were Women,” Technology and Culture 40, no. 3 (July 1999): 455-483.
  8. Smith, “National Security and Personal Isolation.”
  9. Light, “When Computers Were Women,” 482.
  10. Clakre Hayes quoted in Thomas J. Misa, Gender Codes: Why Women Are Leaving Computing (IEEE Computer Society, 2010); “Women in Computer Science: Getting Involved in STEM,” Computerscience.org.
  11. Abbate, Recoding Gender, 47-50.
  12. Lois Mandel, “The Computer Girls,” Cosmopolitan, April 1967, 54.
  13. Mandel, “The Computer Girls,” 56.
  14. Mandel, “The Computer Girls,” 56.
  15. For instance: “[Programming is] just like planning a dinner”; “The girls may even work at home— while the children are napping”; “What about the chances of meeting men in computer work?” (Mandel, “The Computer Girls,”, 52-56.)
  16. Mandel, “The Computer Girls,” 54.
  17. Take for example this passage from “The Computer Girls”: “After earning a master’s degree at Harvard in astrophysics, Helene Carlson suddenly found that there wasn’t much a woman could do in astronomy.” The article described how Carlson’s newfound career in computing is—according to her—“much better than looking at stars, anyway.”(Mandel, “The Computer Girls,” 54.)
  18. Specifically, the mathematical aspects of computer science; scholars treat mathematics and computer science in a somewhat synonymous manner in discussions of computer science and mathematical trends.
  19. Abbate, Recoding Gender, 56-57.
  20. Smith, “National Security and Personal Isolation,” 328.
  21. Mandel, “The Computer Girls,” 56.
  22. Mandel, “The Computer Girls,” 54.
  23. Mandel, “The Computer Girls,” 54.
  24. Ensmenger, “‘Beards, Sandals, and Other Signs of Rugged Individualism,’” 43.
  25. Ensmenger, “‘Beards, Sandals, and Other Signs of Rugged Individualism,’” 43.
  26. Ensmenger, “‘Beards, Sandals, and Other Signs of Rugged Individualism,’” 51.
  27. Mandel, “The Computer Girls,” 54.
  28. Ensmenger, “Making Programming Masculine,” quoted  in Misa, Gender Codes, 121.
  29. Smith, “National Security and Personal Isolation,” 328.
  30. Ensmenger, “Beards, Sandals, and Other Signs,” 60.
  31. Misa, Gender Codes, 3.
  32. Abbate, Recoding Gender, 68.
  33. Smith, “National Security and Personal Isolation” 330.
  34. Ensmenger, “Making Programming Masculine,” quoted in Misa, Gender Codes, 121.
  35. See “The Male Gazed: Surveillance, Power, and Gender” by Kate Losse (2014).
 
Back to Top
CONFLUENCE