Passion as Capital: The Cultural Production of “Good Computer Scientists”

Abstract

“Do what you love” has become a hegemonic mantra in large swaths of the labor force, and the expectation for workers to be passionate about their work is intense within the technology sector. When we interviewed eight undergraduates at a large Canadian university about what makes someone a “good computer scientist,” participants stressed the importance of passion. Passion was described as a form of capital: to be accumulated and traded for advancement in the labor market. We examine how passion works as a hegemonic norm within this culture, and untangle passion as requiring two distinct processes: conspicuous production (via extracurriculars and internships), and affective passion (via performing interest and enthusiasm for the subject). We explore the relationships and feedback loops amongst conspicuous production, affective passion, and other forms of Bourdieusian capital. In doing so, we uncover the mechanics of the power structures which underlie post-feminist and meritocratic narratives of self-motivated passion.

Plain language summary

Passion in Computer Science Education: The Harm Behind “Do What You Love”

In North America, passion for your work or studies is seen as a positive value. In computer science, being passionate is an expectation. When we interviewed eight computer science students at a large Canadian university about what makes a “good computer scientist,”, passion was high on the list. For these students, being passionate meant not only being interested in computer science, but demonstrating this interest by coding in their free time. It is expected that students are not only able to learn on their own, but want to do it. In this paper, we discuss what is necessary to be a “good” and “passionate” computer science student. Success in the computing industry is not a matter of individual interest, effort and talent. Rather, being passionate requires access to resources such as time, sufficient income for leisure activities, relevant education, friends and family in computer science, and knowledge on where to learn and do activities outside of school. Presenting passion as a simple question of personal interest hides the ways passion acts as a barrier to participation. We argue that it is necessary to shift what it means to be a “good computer scientist” in order to change who is able to participate in the field.

Introduction

With the rise of neoliberal work culture, “do what you love” (DWYL) has emerged as a hegemonic mantra in large swaths of the labor force (DePalma, 2021; Sandoval, 2018; Tokumitsu, 2014). Increasing demands for productivity have been accompanied by a desire for passion in white collar jobs (DePalma, 2021; Sandoval, 2018). Passion promises liberation from unfulfilling and exploited labor: work becomes not a source of income, but the primary means of self-fulfillment (Sandoval, 2018). In both one’s work and one’s leisure time, individuals are expected to be productive, and find pleasure in their productivity (Reagle, 2019; Tarnoff, 2017).

This expectation to show passion for your work is a prominent characteristic of labor in technology-related fields (Bergstrom, 2019; Harvey & Shepherd, 2017; Margolis & Fisher, 2002). This is particularly true in computer science (CS) education, whose ideology is dominated by “constructionism,” a learning theory developed by Seymour Papert with strong ties to the MIT “hacker” culture of the 70s (Ames, 2018). Constructionism promotes a style of education that assumes that students are precocious young men who, if given machines to learn with and minimal instruction, will both be able and want to learn on their own (Ames, 2018).

As scholars in science education and science and technology studies (STS) have noted, narratives of entrepreneurial passion and independent learning are part of a broader set of gendered norms against which CS students are judged and judge each other (Breslin, 2018; Margolis & Fisher, 2002). These norms delimit what is considered credible participation in CS, and shape how (and whether) students develop recognition (by others and themselves) as a computer scientist (Avraamidou, 2020a; Breslin, 2018). The continued marginalization of women and socio-historically minoritized groups in Canadian CS education (Cukier, 2003; Cukier et al., 2002, 2009; Sturman, 2009) warrants a close examination of these norms. Having a detailed understanding of the discourses and structures which underlie norms like passion in CS is necessary to uncover their role in upholding structures of power. This will allow us to more effectively critique and offer alternatives to harmful narratives in computing culture. In this paper, we build on Samantha Breslin’s work on passion as a gendered hegemonic norm among CS students in Singapore by examining the role of passion in developing a computer science identity in undergraduate CS education at a major Canadian university. We develop an emic definition of passion based on how CS students understood it, which we split into two components: affective passion and conspicuous production. In examining the interplay between these two forms of passion in the formation of a credible computer science identity, we uncover the underlying mechanics which allow the marginalization of students to occur in discreet and vicious ways. In particular, we report on how our participants perceive passion as a form of science capital which helps them persist in CS, and examine how narratives that present passion as an individual and self-motivated asset naturalize the marginalization of minoritized groups.

This study is exploratory, and seeks to contribute new theory to address the mechanics of passion as a hegemonic norm in Canadian CS education. We contribute to the call for research to “examine how passion is understood, perceived, and defined in workplaces” (Rao & Tobias Neely, 2019).

Background and Theory

Our analysis is grounded in a feminist poststructuralist theory of identity, which sees identity not as fixed, but as constituted by language and discourse (Weedon, 1996).

We further draw on two central conceptual lenses—Pierre Bourdieu’s theory of capital, and Antonio Gramsci’s conceptualization of cultural hegemony—in order to produce a rich understanding of how students construct their identities, and experience belonging in CS.

Cultural Hegemony

Cultural hegemony refers to the process by which a dominating class manipulates culture such that dominant-class ideologies become accepted by other groups as natural and inevitable (Gramsci, 2007; Hodges, 1998; Mayhew & Patitsas, 2021). This process allows for the “consent” of unequal power relations, as alternatives to the status quo become unimaginable (Breslin, 2018; Gramsci, 2007).

Hegemonic Computer Science Personhood

Breslin applies the concept of cultural hegemony to examine the formation of what she terms the hegemonic computer science personhood at Temasek University in Singapore. The hegemonic CS personhood represents a normative, gendered, heteronormative, US-centric CS identity against which students are judged and judge themselves. Breslin finds that the values of individualism, entrepreneurialism, passion, and neoliberal competition are central to this hegemonic identity. In particular, she argues that women in CS in Singapore must perform passion in relation to both the hegemonic CS personhood and gender norms, as men become the standard against which we measure women’s success. Thus, students learn to compare and judge their own performances of gender norms, individualism, and entrepreneurial passion against the hegemonic CS personhood.

Computer Science Identity

Students develop their own identity as a computer scientist in relation to this hegemonic personhood. Prior research in Science Education has examined this process of identity formation. In order to develop a “science identity,” students must be recognized by others and themselves as a credible science person (Avraamidou 2020a,b). One way this can happen is through legitimate peripheral participation, by which newcomers are introduced to community norms and practices in order to become productive members (Chan et al., 2005; Cheng & Hill, 2022; Fiesler et al., 2017; Safadi & Berente, 2019). However, another element is necessary in building a credible science identity: access to capital.

Bourdieusian Capital

Capital is a concept coined as part of Bourdieu’s theory of social reproduction (Bourdieu & Passeron, 1970). Broadly, it refers to the accumulated skills and resources which can be drawn on for social advantage, and determine their social position (or level of privilege) within a particular field (or social context, such as education).

Bourdieu’s capital takes three forms: economic capital (i.e., material assets), social capital (i.e., access to a network of social relations), and cultural capital (e.g., mannerisms, knowledge, and skills that can be drawn on for privilege and acceptance in a particular environment). Access to these various forms of capital is reflective of social class, as well as other intersecting identity factors such as gender, race, and ability. Bourdieu found that cultural capital plays a vital role in education (Bourdieu & Passeron, 1970).

Science Capital

Science capital—as conceptualized by Archer et al. (2015)—refers to the science-related forms of capital which influence students’ ability to build a credible science identity. This includes, for example, scientific literacy, access to scientific knowledge outside of school, consuming science-related media (cultural capital), and knowing people (such as family members) in science or having people to talk about science with (social capital) (Archer et al., 2015; Gonsalves et al., 2021). Archer et al. (2017) demonstrate how young women in physics draw on an accumulation of science capital to develop a “resilience” against factors which work to exclude them from the field. In the face of uninteresting courses, exclusionary practices, and dominant discourses tying physics to masculinity, young women with strong STEM capital persisted with the expectation of more exciting physics in the future (Archer et al., 2017). As Archer et al. have demonstrated, levels of science capital correlate to factors like gender, ethnicity, and other acquired cultural capital. Margolis and Fisher have similarly examined how women’s persistence in CS is influenced by their ability to gain access to supportive communities and find alternatives to the dominant culture (Margolis & Fisher, 2002).

Conspicuous Production

Our analysis also connects to an emerging area of study focusing on increasing pressures to perform productivity within institutions pervaded by neoliberal ideology based on meritocracy, which has us believe that success is the result of individual effort and talent (Au, 2016; Littler, 2013; Rojek, 2000). Of particular interest here is the term conspicuous production, which refers to the public display of productivity in exchange for social status (Tarnoff, 2017), and is a contemporary update to Thorstein Veblen’s coined “conspicuous consumption” and “conspicuous leisure” (Veblen, 1899).

In his 1899 thesis, The Theory of the Leisure Class, Veblen argued that society was ruled by a leisure class that asserted its power through demonstrations of excess time and money. However, postwar North America saw rising inequality and then a shift toward neoliberalism, requiring the extremely rich to demonstrate that they are deserving of their wealth (Rojek, 2000; Tarnoff, 2017). In contrast to Veblen’s “conspicuous consumption,” which refers to shows of power by the rich through wasteful spending, and “conspicuous leisure,” whereby individuals gain status by demonstrating their ability to spend their time in nonproductive ways (e.g., going on vacation) (Veblen, 1899), conspicuous production highlights a shift toward productivity as a symbol of class power (Tarnoff, 2017). Performing conspicuous production thus involves spending leisure time on something “productive”—even while presenting it as a personal interest. Consider recent trends venerating ultra-productivity, such as the proliferation of productivity apps, and the hagiographies of tech CEOs that depict them as working from dawn-to-dusk (Reagle, 2019; Tarnoff, 2017).

Constructionism

Personal interest is also critical to constructionism, the dominant ideology in computing education (Ames, 2018). Constructionist learning theory assumes that personal interest and access to a programmable computing machine alone is sufficient for learners to learn to program. The learning theory is based around an assumption of the learner as a “technically precocious boy” that is hamstrung by learning in a traditional classroom (Ames, 2019).

Despite constructionism’s failure to sustain long-term success in larger education circles, it remains popular in the technology world, where it has become integral to peoples’ identities as technologists (Ames, 2018). As STS scholar Morgan Ames notes, constructionism originally found its roots in the MIT “hacker” culture of the 1970s (Ames, 2018). As such, many of the tenets of the “hacker ethic” had an influence on the development of the theory, and continue to influence the culture of computing today (Ames, 2018; Coleman & Golub, 2008).

Structures of Power in CS Education

These gendered ideologies help explain the well-documented marginalization of minoritized groups in Canadian CS education (Cukier, 2003; Cukier et al., 2002, 2009; Sturman, 2009) (see also [W. DuBow & Gonzalez, 2020; S. J. Ferguson, 2016; Frank, 2019; Hango, 2013; Wall, 2019] for statistics on the continuing underrepresentation of women in computing in North America). CS is relatively unusual for a STEM field in that the percentage of women has decreased since the 1980s (Patitsas, 2019), and women presently make up about 20% of undergraduates in North America (National Center for Women and Information Technology, 2022) . Efforts within computing education to address these inequalities tend to focus on issues of numerical exclusion and diversity (Mellström, 2009) (see e.g., W. M. DuBow, 2013). However, science education and STS scholars have called for more culturally contextualized research focusing on the experience of marginalized groups already in CS (Breslin, 2018; D. Ferguson & Martin-Dunlop, 2021; Mellström, 2009). This paper responds to this call by investigating the formation of CS identities at a major Canadian university.

Methods

Our research methods fall under the umbrella of feminist methodology rooted in feminist standpoint epistemology (DeVault, 1996; Haraway, 1988; Harding, 1995; Suchman, 1993). In particular, we embrace Donna Haraway’s doctrine of embodied objectivity, or situated knowledges, which sees knowledge as necessarily partial and limited by one’s social location (Haraway, 1988). Standpoint epistemology appreciates how one’s personal standpoint affects the way that one does research, and that pretending otherwise provides a false sense of “objectivity.” Rather than a “Voice From Nowhere” we write in first person to make clear the research process was done by humans, and we aim to be transparent about our standpoints.

We follow in a lineage of feminist STS work aimed at investigating the power structures and breakdowns in technological cultures, and offering ways of imagining these cultures differently (Haraway, 1988; Rosner, 2018; Suchman, 1993).

Context: McGill University

This study took place at McGill University, a large, research intensive university located in the city of Montréal in Québec, Canada. Like most Canadian universities, McGill is a public institution established under a provincial government charter. The School of Computer Science at McGill is located in the Faculty of Science, but also offers Bachelors of Arts. Students who are admitted to the faculties of Science or Arts do not need to apply to CS in order to enter a CS program. The curriculum follows a fairly standard path (CS1, CS2, etc.) similar to the curricula recommended by the Association for Computing Machinery (ACM). Students who complete secondary education in Québec are likely to have attended a CÉGEP (Collège d’Enseignement Général et Professionnel), which covers the final year of secondary education and the first year of post-secondary education (exempting them from their first year of courses at McGill). These students were also likely to have had access to CS courses prior to attending McGill.

Recruitment and Interviews

Ethics approval for this project was granted by McGill’s Research Ethics Board 2 (File 20-11-013). We recruited participants via an email advertisement directed to students in all undergraduate CS programs, as per the approved protocol. The advertisement included a description of the project as investigating passion in CS education, and dealing with issues of gender, race, and class. Twenty-five students responded to the call. I (first author) created an initial list of participants with priority to students who responded first, and then adjusted for greater gender diversity (the majority of respondents were women). We originally contacted ten participants, however two did not respond.

I (first author) conducted interviews with eight students. Written consent was obtained from the participants, and additional time was taken at the start of the interview to review the consent document and address any questions. Participants could opt-out at any time. The interviews were semi-structured with open-ended questions and lasted approximately 1 hr each. I conducted the interviews via the virtual meeting platform of the participant’s choice. The interview protocol is available in the appendix, and included questions about motivation and identity in CS.

As many experts on qualitative methodology have argued, it is not possible to determine the appropriate sample size for a study in advance (Braun & Clarke, 2021; Constantinou et al., 2017; Eynon et al., 2018; Guest et al., 2006; Hesse-Biber, 2007; Quinn Patton, 2002; Sim et al., 2018). Therefore, in order to determine if we had reached saturation, I (first author) conducted a preliminary thematic analysis on my handwritten notes after each interview. We defined saturation as no new major themes in new interviews, and consistency of the interviews with each other and with existing theory. We determined that sufficient saturation was reached after eight interviews, and therefore did not recruit new participants.

Our final participant list included five women, two men, and one queer person (later identifying as a queer male; see Table 1 for participant identity factors). Four students were international, however, Imel and Findiser had lived and attended secondary education in Canada prior to attending university. Four participants were people of color.

Table 1.

Participant Identity Factors.

Participant	Gender	Race	International status
Mithelye	Woman	Black	Domestic
Aninden	Woman	Mixed	Domestic
Mili	Queer	White	Domestic
Golore	Man	White	Domestic
Imel	Woman	Asian	International
Findiser	Woman	Latinx	International
Celothel	Woman	White	International
Indor	Man	White	International

Each interview was digitally video-recorded and/or audio-recorded according to the student’s request. I (first author) then transcribed the interviews verbatim using the automatic transcription feature on Microsoft Stream, followed by a manual correction of each transcript. Video-recordings were used solely for convenience in recording and transcribing, and were not analyzed. Quotes used in this paper were lightly edited for clarity (e.g., removing filler words). We generated pseudonyms for the participants using a randomized name generator for Dungeons and Dragons, Donjon, to generate randomized Elvish names.

Qualitative Analysis

We then performed a qualitative content analysis on the transcript data in order to draw out commonalities across interviews. Each transcript was manually coded by the first author, with the help of the second author and members of our lab. To minimize potential harm to participants, the first author pre-screened transcripts for identifying information, and was the only one who knew the identities of the participants. When the first author noted that a participant discussed taking a course from the second author, the second author recused herself from analysis of that participant. The codes were then placed on a virtual whiteboard and analyzed through affinity mapping with help from members of the lab. The final themes were determined through several rounds of coding and discussion, until a consensus was reached.

Given our methods and our sample size, our goal is to achieve reader generalization and analytical generalization. Reader generalization is the ability of readers to extrapolate findings to other settings, or their own experiences (Polit & Beck, 2010). To support reader generalization, we provide raw data and detailed descriptions in order to allow the reader to draw their own conclusions about the data and how these may apply to other settings. Analytical generalization involves generalizing from particulars to broader constructs or theory (Polit & Beck, 2010). To this end, we place our findings in the context of existing work and theoretical concepts. This study is exploratory, and seeks to build initial theory about the role of passion in how students develop credible CS identities. We do not make claims to statistical generalization, which is generalization from a sample to a larger population.

We believe our data is sufficiently saturated for the purposes of this study. In their work on qualitative methods, Greg Guest et al. found that meta themes were present by the sixth interview, and full saturation occurred between 7 and 12 (they evaluated saturation in increments of six) (Guest et al., 2006). Similar projects have found saturation within eight interviews. Defining saturation as “no new emergent themes,”Eynon et al. (2018) achieved saturation after eight analyzed transcripts. Constantinou et al. (2017) found that “all possible themes” emerged after seven.

Positionality

As part of feminist methodology (DeVault, 1996), our goal was not to avoid or remove our personal experiences and interpretations, but rather to be transparent about how our personal standpoints enrich the analytic process. As such we also wish to clarify our own positions:

Hana Darling-Wolf (First Author)

At the time of writing, I am a recent graduate of McGill university with a Bachelor of Arts and Science in Computer Science and Gender, Sexuality, Feminist, and Social Justice Studies (GSFS). At the time of conducting this research and analysis, I was still completing my undergraduate degree in CS at McGill. My experiences with CS during and prior to completing this degree, as well as my position at the intersection of several identities—including but not limited to a white, queer, nondisabled, woman with relatively little coding experience entering university—motivated me to conduct this research and have influenced both the questions asked, and my interpretation of the data. In particular, my interest in questions of passion and recognition were influenced by my own feelings of disbelonging in CS.

Elizabeth Patitsas (Second Author)

At the times of writing and analysis, I am an assistant professor in the School of Computer Science at McGill, and have taught introductory programming (COMP 202) and upper-level CS courses on STS and CS education. When working with the data, I contributed a standpoint of somebody who has taught introductory CS.

As an undergraduate I demonstrated (with conscious effort) passion for teaching and research. I worked as a TA and RA almost continuously as a student, and held leadership roles in student government. My engagement was genuine, while also being aware that building my CV would be beneficial for a career in academia. I did not participate in hackathons, programming competitions, robotics clubs, etc. as I was not interested in a career in industry. But I did personally experience the cultural push toward demonstrating and performing passion as a CS student.

Analysis: Untangling Passion as Requisite for Studying CS

“I think […] people who are really passionate about [CS] can be good computer scientists.” (Imel)

Consistent with prior literature, our analysis suggested that the performance of passion was vital to developing feelings of belonging and gaining credibility in computing (Breslin, 2018; Harvey & Shepherd, 2017; Margolis & Fisher, 2002).

“[When] moving up to 300 level classes. It just became more dense with—from what I saw—male people who had some type of passion for it. Whereas I didn’t feel the same way, so I felt kind of out of it.” (Mili)

Students in our study were constantly comparing themselves to other students they saw as fitting the normative model of a passionate computer scientist (Golore, Mithelye, Celothel, Imel, Indor), which Breslin calls the hegemonic CS personhood (Breslin 2018) (see section “Hegemonic Computer Science Personhood”). Those that did not fit this model expressed feelings of disbelonging in CS.

In what follows, we offer a close examination of how passion functions as a hegemonic norm at McGill University. As other scholars have noted (Breslin, 2018; DePalma, 2021), the word “passion” simultaneously takes on several meanings for students. We found it useful to distinguish two key definitions of passion, which we will refer to here as affective passion and conspicuous production.

Conspicuous Production

The performance of conspicuous production involves doing extracurricular activities such as creating personal projects, participating in hackathons, being involved in a CS club, self-learning CS, spending time practicing coding outside of courses, or applying for internships. For our participants, the need to perform conspicuous production stemmed from the idea that CS in school was largely theoretical, and not very useful for doing (more interesting) “applied” coding.

More specifically, students identified “useful” CS as computer science which would allow them to become marketable and employable in industry. As Indor remarked, “if you don’t apply [what you learn in university] in concrete examples then, I don’t think the theory can lead you really far in the field.” Imel similarly noted the importance of developing “skill” in applied CS in order to become employable:

“there’s some people with a very high GPA but after they graduate they cannot find any job or get any offers ’cause they don’t know the applied stuff […]. So they feel really sad ’cause they have a high GPA but it’s not very useful. And many people told me that companies don’t care about your GPA, they only care about your skill.” (Imel)

Crucially, while poor performance in school did affect some students’ perception of themselves as (not) good computer scientists (Imel), performing well in school was not enough to lay claim to a CS identity. As one student explained:

“I’ve done pretty well in classes […] grade-wise, but it’s mainly just the fact that it comes back to everything is theory, so I have so few projects that I actually coded, so I’m like I can’t really be called a good computer scientist when I actually don’t really code that often.” (Celothel)

As this and previous comments suggest, students tended to align being good at CS (and the meaning of CS itself) with programming and developing skills. Although students often placed self-value in their grades (Aninden, Imel, Indor), and felt pressure from McGill to get a high GPA (Mili, Mithelye, Celothel, Imel), they were made by peers and industry representatives to feel like their performance in school did not matter. For example, Indor experienced pressure to prioritize internships in his interactions with other students online:

“[O]n Reddit, following CS majors […] what a lot of people say is that you have to prioritize job interview searching in comparison to university. So basically, you don’t care much about your grades, but what matters is if you get an internship. […] for me, university is more important, but maybe […] this is why I still don’t have an internship.” (Indor)

Likewise, Imel observed that good computer scientists often have low GPAs because they choose to spend time on personal projects rather than school. This focus on extracurricular skill development makes sense in the context of a technology culture pervaded by constructionist ideals, which draws heavily on tenets of the “hacker ethic.” This includes the belief that “[h]ackers should be judged by their hacking, not bogus criteria such as degrees, age, race, sex, or position.” This prioritization of “hacking” (or in this case, “practical” coding skills) over degrees was reflected in students’ responses. Despite pressures to perform well in school, students are made to feel that their grades are not “worth much”:

“You can have good grades, but if you don’t have any experience, then I don’t feel like it’s worth much.” (Indor)

Although Indor and others claimed to prioritize university, good grades did not afford them credibility as a computer scientist. Rather, being a good computer scientist required “taking the time to go beyond” what they were taught in school through self-learning (Indor) and the creation of projects (Imel). Because courses typically failed to give them the skills they felt were needed to get a job in industry, it was necessary to spend time gaining these skills outside of class to be considered a good computer scientist. Thus, performing a hegemonic CS identity involved conspicuously producing CS outside of school.

Affective Passion

Passion also holds a more obvious affective connotation. We found that it was necessary for students to perform self-motivated interest in doing CS in order to gain credibility as a computer scientist. When a student did not perform their interest in computing, their belonging was questioned by their peers:

“they said if you are not interested in it and if you struggle then you should not stay in the computer science program.” (Imel)

Indeed, affective passion acted as a requisite for credibility in CS. This was made clear in how participants articulated their motivations for doing CS. Despite students’ emphasis on becoming employable, we found that employability was not seen as an acceptable reason for entering CS.

Paradox: You Can’t Be in it (Just) for the Money

An interesting paradox we observed was that while students advocated for job-readying courses, and placed great importance on learning CS skills that would make them employable, it was not acceptable for students to enter CS for that purpose. Indeed, several students commented on the importance of being motivated to do CS by some “pure” form of interest, rather than practical financial reasons (Findiser, Mili, Aninden, Indor):

“I would say that you can kind of differentiate between those who have a passion for technology and those who are just in it for the money.” (Aninden)

Per our participants, in order to be a good computer scientist, it was necessary to have affective passion for CS. This passion, in turn, could not be motivated by factors other than personal interest. Indeed, for many students, doing CS for money or a job was diametrically opposed to having interest in CS, and therefore being a good computer scientist.

“I’ve heard that some people are just interested in CS because you can earn a lot of money compared to other jobs and if you’re doing this, then maybe you’re going to earn a lot of money, but will you be happy? Probably not. […] You should be happy in your job, not earn a lot of money.” (Indor)

Students that did not fit this bill were made to feel like they were doing it for “the wrong reasons:”

“I feel like all my motivations were for the wrong reasons, looking back on it. I don’t think I enjoy it a lot of the time, which I feel should have been my main deciding factor. [The] motivation that I had was to survive in a capitalist world or something” (Mili)

Another student, Celothel, lamented the expectation that students be passionate, arguing that it should be sufficient for students to just want a job, and went on to note she did not identify as a computer scientist because it was not her whole life.

Thus, while students emphasized the necessity of doing work outside of school in order to make them employable, the motivation for doing this work could not be employment. This contradiction reveals how affective passion acts a moral boundary (Lamont, 1992) against which students judged others and themselves. Only those who performed affective passion for CS could become a good computer scientist:

“a good developer is interested in what he’s doing. He may be really good at it, let’s say a person is really good at coding, but it doesn’t interest him, then… Is he a good developer? That’s a complicated question.” (Indor)

Being a Good Computer Scientist Requires Both Conspicuous Production and Affective Passion

Conspicuous production and affective passion require different actions of students—the production of CS or an affective performance. Separating conspicuous production and affective passion allows us to discuss both of these passions(s) without losing sight of how they work in different but intertwined ways.

Per our participants, it was necessary to perform both affective passion and conspicuous production. In particular, in order for other students to acknowledge their affective passion for CS, it was necessary for students to engage in conspicuous production:

“I find the students that I would say are more passionate about technology are ones that take the initiative to do external projects and hackathons.” (Aninden)

Participants argued that students that were passionate did these activities in their free time because they enjoyed doing them (and by extension, CS). That is, students emphasized the importance of both having an interest in CS, and spending the time on projects to back it up:

“[To be a good CS student] I think they have to be highly motivated to do what they’re doing. I think they have to enjoy computer science. I think they have to prioritize it and maybe put some things on the back burner.” (Mili)

Indeed, several students expressed that they felt they were not good computer scientists or CS students because they did not exhibit both affective passion and conspicuous production:

“No, because I don’t work hard enough at it. I have the drive, I have the hopes to change things in the future, but I haven’t been able to do that yet because I just haven’t done it.” (Aninden)

Thus, we see that affective passion and conspicuous production act as hegemonic norms which shape who is considered a credible participant in CS.

Passion(s) is Hegemonic

In her thesis, Breslin highlights how hegemonic norms surrounding the performance of entrepreneurial passion in CS in Singapore force women to perform passion in relation to both the hegemonic CS personhood, and gender norms (Breslin, 2018). Women are held to male standards of confidence, motivation, and interest without questioning whether these are desirable, and for whom this is possible (Breslin, 2018; Margolis & Fisher, 2002).

Performing passion in your free time requires “freedom from social obligations, other interests, and other claims to personal time” (Breslin, 2018; Margolis & Fisher, 2002). This reveals the way that conspicuous production is gendered and classed—labeling other activities such as housework, caring for dependents, part time jobs, or interest in a less hostile subject “distractions” (Breslin, 2018). Conspicuous production is also ableist: not all students have the energy or access to work during their personal time (Dolmage, 2017).

We can also start to see how the moral boundary surrounding affective passion discussed in section “Affective Passion” works to present doing CS for money as unclassy. While upper class people can afford to chose their degree based purely on interest, the hegemonic norm surrounding affective passion in CS works to marginalize people who might have more practical financial reasons for entering a CS program.

The hegemonic nature of passion also works to marginalize minoritized groups in a more subtle way over time. A surprising finding we noticed in our data was that students presented passion as a form of capital which helped them succeed in CS. The following sections will address how students framed affective and conspicuous production as capital on which they could draw to build their legitimacy as computer scientists.

Passion as Science Capital

In addition to acting as a hegemonic norm, we noticed that students described passion (and the performance thereof) as a resource: one accumulates passion, spends passion, exchanges passion for other resources (good grades, a degree, job interviews, etc) and credibility in the field. As previously discussed, the notion of science capital has been termed (Archer et al., 2015) and used in the science education literature to refer to cultural and social capital that, when accumulated, allows students to persist in science even in the face of a hostile environment. In other words, passion was an emic term for a form of science capital students drew on for persistence and legitimacy in their education. In the following sections, we demonstrate how affective passion and conspicuous production worked both separately and together as different forms of science capital.

Affective Passion as Science Capital

Students expressed that passion helped them persevere through their studies. When asked what was necessary to be a good CS student, several participants commented on the importance of passion and interest in being able to “push through:”

“I feel like I’m not the most book-smart person, but my passion for technology has driven me through.” (Aninden)

Affective passion provided the motivation for students to continue in CS despite the “less fun parts” of their CS education. Interest thus helped motivate students to look beyond the “boring” material they learned in school, and also helped them succeed in their courses:

“I’d say [you need] an interest, definitely. ’Cause for me the interest [in a specific course] was less there and that made it harder for some parts of the class to get myself through” (Golore)

In particular, affective passion could motivate students to continue their studies with the expectation of more fun CS ahead:

“if you don’t see the big picture, then you can get demotivated pretty quickly. […] if you’re not passionate about CS, then I don’t think you’ll be able to see the bigger picture, and you’ll just get frustrated over the courses because it doesn’t interest you.” (Indor)

Having affective passion for CS that extended beyond (and around) the less fun CS they saw in the classroom protected them from being demotivated by their lack of interest in course material, contributing to their resilience and perseverance in CS.

Conspicuous Production as Science Capital

Performing conspicuous production helped students gain the skills needed to succeed in interviews and make them employable. Since “applied” CS outside of school was considered “useful,” gaining this kind of experience led to credibility as a computer scientist. In other words, students perceived conspicuous production as a form of cultural capital that they could draw on for acceptance and privilege in CS.

Affective Passion and Conspicuous Production Can be Exchanged

Transferring Affective Passion into Conspicuous Production

Affective passion was also a resource which gave students the necessary motivation to perform conspicuous production (Figure 1). Passionate students were good students because they enjoyed spending the necessary time working on CS outside of class:

“I think that people who are good are the ones that ask OK, but what does this other thing do? And they go try it. Like they don’t mind spending more time on an assignment because, […] they will try different things just for the sake of trying different commands and in whatever programming language. And I guess that advantages them on the final exam” (Golore)

“I think you have to be willing to devote a lot of time toward what you’re doing. Which should be easy, I think, if you really enjoy it, but it’s hard if it’s not one of your favorite things, or not something you want to prioritize all the time.” (Mili)

According to our participants, students with affective passion for CS would necessarily spend more time performing conspicuous production for CS. Performing conspicuous production, in turn, would advantage them both in their studies, and in their future CS careers.

Figure 1.

Affective passion motivates conspicuous production.

Conspicuous Production Can Transmutate into Affective Passion

Just as affective passion gave students the motivation to “put in the time,” a couple participants described conspicuous production as key to increasing students’ interest and motivation to do CS:

“it’s important to be passionate. I think that passion can grow. I feel like the more exposure you have, and […] maybe you don’t like what you’re doing in your classes, and you do stuff outside of McGill and you’re like, oh, I like this, I like that.” (Mithelye)

“I could finally see that I was able to make small projects and I was getting creative about what I could program and it really motivated me.” (Indor)

By creating their own projects and trying new extracurriculars outside of school, students could build their affective passion (Figure 2).

Figure 2.

Conspicuous production increases affective passion.

Passion Capital Builds on Itself

As we have previously seen, participants saw affective passion as the motivation for performing conspicuous production (i.e., students perform conspicuous production because they enjoy it). On the flip side, conspicuous production—in the form of clubs, personal projects, or other CS related activities performed outside of school—could act as the source of (new or sustained) affective passion for CS. Together, affective passion and conspicuous production act as a form of capital which builds on itself (Figure 3). Students that had affective passion would create projects and participate in hackathons which helped them build the skills necessary to create even larger projects:

“Most people in computer science—but me—would make projects for fun, or they would work for a computer science company. […] So obviously they learned so much. Like they’re still students, and they still learn at school, but they found a job. Also I didn’t join any clubs, but many people were joining coding based clubs. And many people would do hackathons. That does not attract me, but if they start doing that they just get a way higher level ’cause it means they do so much for fun in their free time.” (Celothel)

“before learning CS I couldn’t create much because I didn’t know much. But after learning a few languages, some front end some back end, I could create more projects and it’s so rewarding […] that it pushes me to make a new project.” (Indor)

These projects, in turn, would boost their affective passion by introducing them to new tools and concepts, and providing a feeling of accomplishment. Thus, affective passion and conspicuous production acted as a positive feedback loop which could increase the passion capital of students. As Celothel points to, however, increasing one’s passion capital meant having affective passion and conspicuous production to begin with. Moreover, accessing one without the other was not simple. As the following section addresses, gaining access to affective passion and conspicuous production required students to draw on other forms of social, economic, and cultural capital.

Figure 3.

The positive feedback loop of affective passion and conspicuous production.

Passion Requires Other Capital

Because affective passion and conspicuous production act as a form of capital, they can be exchanged for other forms of social, economic, and cultural capital. As we illustrate in the previous section, once affective passion and conspicuous production have been acquired, they build on each other to form a positive feedback loop of passion capital. In addition, both forms of passion can be leveraged to access economic and cultural capital in the form of perseverance, belonging, credibility, and employment. Students could also grow their social connections through participation in clubs and hackathons, or sharing their affective passion for CS with a group of “like minded people” (Aninden).

Inversely, passion capital was acquired through the exchange of other forms of capital. In this section, we start by discussing a myth we observed that works to hide the capital needed: the illusion of passion as natural and innate. We then debunk this myth by outlining the social, economic, and cultural capital necessary to develop passion capital.

The Illusion of Passion as Natural

Understanding how conspicuous production and affective passion depend on other forms of capital is complicated by a myth we observed in the data: that passion is a natural state of being rather than a form of capital. Across our interviews, there was a common assumption that being passionate would automatically lead to success in CS.

“if you’re passionate, you will be a good student. But if you’re not, then [you have to be a] hard worker, organized.” (Celothel)

“if you do have that interest, that luckily for me I have, you’ll learn fast. […] I don’t think you need to be a really good student, but if you have the interest in computer science, I’m sure anybody could manage to get good grades and thrive in CS.” (Indor)

This comes back to the idea that if you have affective passion for CS, it is easy to perform the conspicuous production needed to be a good computer scientist (“you have to be willing to devote a lot of time toward what you’re doing. Which should be easy, I think, if you really enjoy it” [Mili]). Although our participants often upheld narratives suggesting that having passion makes CS easy, or that passionate students were good computer scientists, the interviews also contained contradictions which complicated these myths. As one student illustrated, it is not actually as easy to do CS as these myths suggest:

“I like computer science because of the fact that it’s so accessible. It is, but it’s not because if you don’t know how to access it, it feels very hard.” (Mithelye)

In other words, there must be other forms of capital necessary to participate in CS, which become obscured by narratives that treat personal interest as access to CS.

Conspicuous Production Requires Social, Economic, and Cultural Capital

Several students expressed difficulty performing conspicuous production despite their interest and willingness to learn. Imel noted that many students interested in CS—including themselves—still struggle to persevere in their education:

“So many people that I know are really interested in computer science, but [find it] so difficult to continue to study [CS].” (Imel)

Mithelye also expressed how, despite her excitement about entering coding clubs in university, she was not able to participate:

“I tried to sign up for clubs and none of them went through. And I was so excited. I was like oh my gosh I’m gonna go and do robotics—living out my childhood dream. I was like oh maybe I’ll do [club name] but then I didn’t even know where to start with their coding assignment. I didn’t know where to start with like… everything so I just didn’t end up doing any clubs. And then it’s nearing November and I’m like, uh, did I do the right thing with coding? I like coding but I’m not good at it. ” (Mithelye)

Some students experienced this feeling of “not knowing where to start” prior to university. In spite of his assertion that if you have interest, you will learn fast (see section “The Illusion of Passion as Natural”), Indor described the difficulty he had self-learning CS before attending McGill:

“before university, I was interested in computer science but I didn’t really know what to start with, like there was obviously a lot of programming languages and I didn’t really know which one to choose […] I tried getting information here and there, but it wasn’t effective at all.” (Indor)

This feeling of being interested, but struggling to understand CS prior to university was also echoed by Findiser. These responses make clear that simply having interest in CS is not actually enough to succeed and find belonging in CS. Rather, prior knowledge—or cultural capital—was necessary to perform conspicuous production, even with affective passion.

Time, a resource shaped by economic capital and social position (requiring freedom from other interests and responsibilities [Margolis & Fisher, 2002]), was also necessary to perform conspicuous production. Although our participants emphasized the importance of creating personal projects, they also highlighted the difficulty of finding time to work on them (Golore, Mithelye, Indor). Several students informed us that they planned to (or already did) take time over vacation or after exams to work on projects they didn’t have time for over the semester (Golore, Mithelye, Indor).

In addition, students described the importance of having a community, connections, friends, or other social capital in performing conspicuous production (Aninden, Golore, Mithelye, Celothel, Indor). For example, Indor described a friend he could always contact for help when creating projects and learning independently. Not all students had this kind of support, however. Celothel mentioned that part of the reason she does not participate in hackathons is that she does not have friends who do, and does not know how to join the “closed group of people that do hackathons” (Celothel).

Even when students had access to affective passion, additional social, economic, and cultural capital were necessary to do conspicuous production. As we address in the following subsection, the reverse was also true for affective passion.

Affective Passion Requires Social, Economic, and Cultural Capital

Like conspicuous production, social capital was necessary for garnering and maintaining affective passion. For several students, having family members with experience or an interest in computing was a major motivation for entering the field (Findiser, Mili, Aninden, Indor). Having connections with other students that shared their affective passion for technology also motivated Aninden and Findiser to do CS, and helped sustain their affective passion. For Indor, access to a structured education in university, as well as a friend who pointed him to resources for learning independently (i.e., cultural capital) were necessary to motivate his continued interest in the field.

There was also a lot of variance in the amount of experience with which students entered the CS program (some took courses before university [Findiser, Mili, Imel, Indor], one already had CS-related work experience [Golore], one did a short online course and tech-related projects prior to university [Mithelye], and two entered university without any prior experience [Aninden, Celothel]). Depending on their prior experience and access to social capital, students were more or less likely to have developed affective passion prior to university. For example, Celothel—who did not have prior experience or friends and family in the field—did not describe herself as passionate for CS. According to Imel, however, even students with high school experience required additional knowledge to keep up. Economic and cultural capital shaped students’ access to computers and prior knowledge of the field, and therefore their ability to access and maintain affective passion.

Passion Capital is Shaped by Habitus

In addition to social, cultural, and economic capital, students’ ability to develop passion capital was shaped by their habitus. Habitus is a learned set of dispositions that shapes one’s sense of agency and possibility in a particular social context (or, as Bourdieu terms it, a particular field) (Bourdieu, 2005; Hillier & Rooksby, 2005). In other words, habitus is a “sense of one’s (and others’) place and role” in a field (Hillier & Rooksby, 2005).

An emic term we saw used to describe students’ perceived sense of agency in the field of computer science was “confidence.” We found that students’ self-described confidence (or their sense of what activities were possible or appropriate for them to do, i.e., their habitus) shaped their ability to access passion capital. As we discuss below, we observed a gendered pattern in the discussion of confidence among our participants.

Confidence is Gendered

Four of the five women we interviewed (and none of the participants that did not identify as women) brought up the topic of confidence during their interview, and three mentioned confidence when discussing the traits necessary to be a good computer scientist or CS student. Students understood their performance of confidence in relation to that of their peers, and noted how these disparities in confidence are gendered:

“I’m trying to describe my friend in words. ‘Cause my friend is definitely […] confident. Maybe as a girl, I just never make it to office hours. […] I’m not the most organized person. And confidence. I just never use the help that they try to offer us.” (Celothel)

In order to become a better computer scientist, women felt it was necessary to increase their confidence (“I could be better and have better self confidence in myself when it comes to computer science problems” [Mithelye]).

This was possible by performing conspicuous production:

“I would say I feel more comfortable now, but probably because I’m older and I have [experience in a CS club], which makes me feel more confident in my abilities. It’s definitely helped with my confidence in computer science.” (Aninden)

“I have no experience making websites and making applications but this year I was [. . . ] like you’re going to join the club because you really want to and you’re going to sign up for the tech things, and you’re also gonna do [an extracurricular activity] because the only way you’re going to feel more confident in coding is if you do more” (Mithelye, emphasis added)

Passion Capital Requires Confident Habitus

One of the most common activities mentioned for increasing confidence was hackathons, which are a popular form of conspicuous production. However, for some students, participating in activities like hackathons was not so simple:

“maybe [a hackathon is] super fun, but since I’ve never done it […] it’s very hard for me to picture how I could actually manage a project in 24 hours, or 48 hours. It just seems so hard. I definitely don’t have the level to do that.” (Celothel)

“also for the hackathon, those kinds of things, I didn’t participate ’cause I think I’m not good enough” (Imel)

Although extracurricular activities could increase students’ confidence in their abilities, not all students felt like they had the “level” to participate in these activities in the first place. Students thus found themselves in a vicious cycle where the key to gaining confidence (that is, conspicuous production) required having confidence (or a high enough perceived “level”) in the first place. Moreover, since conspicuous production increases affective passion, confidence affected not only students’ ability to perform conspicuous production, but also their ability to gain affective passion for CS.

Increasing Passion Capital Requires Both Conspicuous Production and Affective Passion

In order to access affective passion and conspicuous production, students required existing economic, social, and cultural capital to “trade in.” Inversely, when students did not have access to other forms of capital—such as time, social connections, and prior knowledge—it was difficult to acquire passion capital. This is because spinning up a positive feedback loop of affective passion and conspicuous production required having both to begin with. Students who started with affective passion but not the capital and habitus to perform conspicuous production found their affective passion decreasing. Having access to conspicuous production but not the affective passion to persevere made it difficult to increase conspicuous production. Thus, developing passion capital required having both the habitus and the social, economic, and cultural capital required to acquire affective passion and conspicuous production.

Discussion

Comparison With Prior Literature

In this article, we demonstrated how the performance of passion acts as a moral boundary which delineates what is considered legitimate participation in CS. Similarly, Harvey and Shepherd examined how affect perpetuates hegemonic notions of credible participation in games, which work to naturalize the marginalization of women in the field (Harvey & Shepherd, 2017). This is also consistent with Breslin’s finding that CS students in Singapore saw passion as a necessity for being a “good” computer scientist (Breslin, 2018). Like Breslin, we found that passion was an important characteristic of the hegemonic CS personhood against which students judged and compared others and themselves (Breslin, 2018).

Breslin discusses how students are compared and judged both on their grades and their performances of passion. She emphasizes grades as particularly significant within a Singaporean context, where they have a serious impact on students’ future opportunities and paths (Breslin, 2018, pp. 237–238). In our interviews, however, we found the performance of passion to be more significant for determining a student’s credibility as a good computer scientist than GPA. A high GPA could even come in contradiction to students’ perception of others as a good computer scientist, as “good computer scientists” performed passion by prioritizing conspicuous production over all else, including coursework.

This finding is interesting when compared to Harvey and Fisher’s description of credible participation in game development (Harvey & Fisher, 2015). They note how the use of accessible do-it-yourself game design tools is delegitimized as “dabbling” in game development, whereas legitimate participation in gaming necessitates a formal game design education (Harvey & Fisher, 2015). Our finding that conspicuous production is necessary for legitimacy as a computer scientist within the context of formal education suggests that measures of legitimacy may change (and contradict) in formal and informal tech-related environments—such that the marginalization of particular groups (self-taught “dabblers” or the formally trained passion-poor) is always naturalized. Future work could further investigate these contradictions.

We found that discourses surrounding passion in CS work to naturalize the individualistic image of the passionate computer scientist, and obscure the barriers that make being passionate more complicated than simply having personal interest. Indeed, the performance of passion required other capital such as confidence, time, social connections, and previous experience. This agrees with other work exploring the barriers to interest. As game studies scholar Kelly Bergstrom demonstrates, participation in leisure activities is more complicated than an autonomous choice motivated by personal interest. Rather, several barriers (such as social privilege and intrapersonal, interpersonal, and structural barriers) shape who is actually able to participate in these activities (Bergstrom, 2019). Breslin also notes how performing passion necessitates “freedom from social obligations, other interests, and other claims to personal time” (Breslin, 2018). Margolis and Fisher similarly argue that women should not have to sacrifice sleep, family responsibilities, or other interests for success in CS (Margolis & Fisher, 2002).

Our findings also support the cycle of confidence and passion identified by Margolis and Fisher. In their book, Unlocking the Clubhouse, they describe a trend of women who enter CS with interest and enthusiasm for the field, but begin to doubt their abilities in comparison to their peers’, and start questioning their belonging in computing (Margolis & Fisher, 2002). If their confidence does not return, their interest in the field also declines. Since the hegemonic image of a computer scientist is passionate, a decrease in interest can lead women to further question their belonging, resulting in a downward spiral of interest and confidence (Margolis & Fisher, 2002). Our findings further explain this poverty trap: students had to perform conspicuous production in order to maintain their affective passion, but performing conspicuous production required having a confident habitus, which, in turn, required performing conspicuous production.

Moreover, the representation of habitus as an issue of “confidence” plays into postfeminist narratives of individual empowerment and choice which place the blame on women for their supposed “lack” of confidence (Favaro, 2017). Both confidence and passion reflect this neoliberal feminist tendency to insert positive affect into structures of discipline and exploitation, shifting the focus from collective empancipatory struggle to “interiorized affective spaces” (Favaro, 2017; Rottenberg, 2014).

These findings further support Bergstrom and Margolis and Fisher’s work complicating the association of non-participation with disinterest. Margolis and Fisher highlight how the promotion of a singular, obsessive interest in computing fails to acknowledge other ways to “be in” and be interested in CS (i.e., just because a student does not stay up all night working on a project does not mean they are not interested). They also note that the notion that women leaving CS are “just not interested” oversimplifies a more complicated process which results in the loss of interest (Margolis & Fisher, 2002). Bergstrom similarly notes that dismissing non-participation as “lack of interest” obscures the barriers which prevent participation (even with interest) (Bergstrom, 2019).

The imperative to perform passion in CS can be situated more broadly in rhetoric of DWYL (Jaffe, 2021). Labor activists and scholars have for years noted how this attitude enables workers’ exploitation by pushing workers to go “above and beyond,” while rendering worker satisfaction an individual rather than structural issue (Breslin, 2018; DePalma, 2021; Dyer-Witheford & De Peuter, 2005; Harvey & Shepherd, 2017; Jaffe, 2021; Sandoval, 2018). However, this work has largely focused on affective passion. In particular, DePalma articulates an emic definition of work passion which encompasses attraction, enjoyment, motivation, and perseverance (DePalma, 2021)—characteristics that we bring together under the umbrella of affective passion. In this paper, we contribute a definition of passion which includes conspicuous production in addition to affective passion. Although this framing has yet to be tested in other professional fields, we propose that investigating the interplay of affective passion and conspicuous production may be useful in uncovering the power dynamics in other/broader cultures of work.

Limitations and Future Work

Our study used only one set of interviews with participants, rather than longitudinal data, and is limited as such. We do note our high agreement with prior, longitudinal analyses (notably [Breslin, 2018] and [Margolis & Fisher, 2002]), which were furthermore conducted in other countries (Singapore and the United States, respectively). The first author is currently doing research to compare how our findings on passion in Canadian CS education compare to the role of passion in other fields. Future work may also build on this study by including the perspective of other stakeholders in the university, such as instructors.

The findings of this project are likely to be influenced by the pool of respondents for this study. The recruitment advertisement mentioned passion, belonging, and a focus on the intersecting identities of CS students at McGill. However, we believe that the standpoint of students who responded to this call for participants out of concern about passion and belonging in CS are particularly useful in examining hegemonic norms in CS.

One area of research that would be useful to investigate in the future is the intersection of formal and informal CS education. As mentioned in section “Comparison with Prior Literature,” examining hegemonic narratives among informal learners of CS and students receiving a formal education may reveal how these narratives work to exclude certain people from both spaces. It may also be useful to examine how contradictions between narratives surrounding “accessible” online resources for self-learning CS, and the reality of needing capital to participate impact feelings of belonging for CS learners.

Finally, conspicuous production as a concept is relatively new, and we would like to see more work on conspicuous production both within and outside of the technology sector. Although we make no claims to the generalizability of our findings to other fields (or even other institutions), this paper illustrates the value of looking at both affective passion and conspicuous production in order to understand the mechanics of power underlying discourses of “do what you love.” Examining affective passion alone is not sufficient for understanding the positive feedback loop which makes the passion-rich richer, and keeps those without the required capital and habitus passion-poor. We hope that this will provoke research on neoliberal affects in other sectors to consider how these two forms of passion may work together in different contexts.

Reflections and Implications

As we have discussed in section “Passion as Science Capital”, the way our participants described passion suggested that passion works as a form of capital on which they could draw for privilege in the field of CS. This framing points to the ways passion works as a positive feedback loop for gaining more passion and persevering in CS. However, (as noted in section “The Illusion of Passion as Natural”), the presentation of passion as natural and individualistic hides the other kinds of capital necessary to gain passion capital in the first place. Moreover, the suggestion that participation in conspicuous production is a simple question of personal interest contributes to post-feminist narratives that “everyone can do CS” and that “the only thing that is hindering someone from participation is themselves” (Harvey & Fisher, 2015). This meritocratic framework places blame on unsuccessful students for their own passion-poverty.

Thus, while passion does act as a form of capital in CS culture as it is, it is important to also remember its role as a hegemonic norm. Our findings challenge the innocence of passion, illustrating how the cycle of affective passion and conspicuous production limits who can gain credibility as a computer scientist.

The issue is not that these students lack passion, but that students need to perform a particular form of passion in order to credibly participate in CS. In other words—as others have argued (Archer et al., 2017; Harvey & Shepherd, 2017; Margolis & Fisher, 2002)—making CS more inclusive requires a cultural shift in what it means to “be in” CS. Simply inserting socio-historically marginalized groups into CS as it is allows only individuals with enough capital and resilience against exclusionary norms (“those who can tolerate the pain”) to persist in the field (Harvey 2021). Therefore, making CS less exclusionary involves changing what it means to be passionate in CS, as well as challenging the idea that individualistic passion should be valued at all. This, of course, is not a simple task. However, supporting different ways of engaging with and “being” a computer scientist in the classroom is a good start.

It also remains valuable to consider how CS can be made less painful for students already in CS. For one student, the disconnect between “coding” and more theoretical concepts in her courses made it difficult to persist (“many people told me that they feel confused too […] ‘cause they don’t know why they’re learning that, and [that it’s] really difficult to continue study[ing]” [Imel]). Another student placed blame on the university for failing to make her passionate about CS (Celothel). In particular, Celothel noted the disconnect between what she learned in school and the aspects of CS that were “fun” and would make her employable (such as group work). It is crucial for instructors to be explicit about how the concepts they are addressing in class connect to other aspects of computing (“applied,” theoretical, and material [Mayhew & Patitsas, 2021]) in order to promote understanding and prevent students from feeling “lost, unsupported, unconnected, and unable to bolster their own sense of belonging in the field” (Margolis & Fisher, 2002). It is up to instructors to foster a level of understanding and interest that allows students to persist in the field. However, in doing this it is also crucial to promote ways of being a good computer scientist that do not involve performing conspicuous production and overwork.

Conclusions

As a relatively young and newly important economic sector, the IT sector is in many ways a vanguard of labor and professionalization issues in society more broadly (Amrute, 2020; Chang, 2019; Ensmenger, 2010; Eubanks, 2018). Hackathons, once limited to computing circles, have proliferated to other domains such as physics and music (Dubber, 2016; Feder, 2021). Formal education plays a vital role in the instillation and reinforcement of the culture of the field (Abbate, 2012; Chang, 2019; Patitsas, 2019), and the dominant culture of the field in turn produces technologies which reinforce societal inequalities (Benjamin, 2019; Wajcman, 1991).

In this paper, we have explored the role of passion in claiming a CS identity in Canadian undergraduate CS education. We examined how passion works as a hegemonic norm which determines what is considered legitimate participation in CS, and how this credibility is gained.

The performance of passion happened in two important ways: through affective passion (expressions of interest or motivation for doing computer science) and conspicuous production (spending leisure time on CS-related activities). Being a good computer scientist required doing both. Despite students’ emphasis on employability, it was not acceptable to do CS for the job—the hegemonic CS personhood spent their free time on CS, and enjoyed it too. This reflects a larger trend within neoliberalism to fetishize productivity, and present being conspicuously productive as a sign of class. This also fits into larger classed narratives that hide exploitation under the guise of “doing what you love” and mythologies of “work-as-play.”

Interestingly, students also described passion as a form of science capital which could help them persevere and succeed in CS. In particular, affective passion and conspicuous production acted together to generate a positive feedback loop of motivation for continuing in CS. However, students that did not have access to affective passion or conspicuous production found themselves in a vicious cycle requiring them to already have one in order to gain the other. Indeed, contrary to dominant myths which suggest that participation in computing is a question of personal interest, our analysis made clear that other forms of capital were necessary to perform passion in the first place. This brings to light the ways in which dominant narratives surrounding passion in CS obfuscate the larger structures of power which make gaining interest and participating in computing more complicated for members of socio-historically marginalized groups.

In sum, in order to change who participates in CS, it is necessary to change what it means to credibly participate in the field. The takeaway from this paper should not be that socio-historically marginalized groups need to be instilled with more passion for CS (although connecting course content to skills perceived as employable can help students persist). We found that myths surrounding passion in CS maintain structures of power by repressing other ways of being in CS, and obscuring the capital needed to be passionate. Therefore, challenging the exclusion of individuals with minoritized identities and creating a more socially just CS requires a broader acceptance of different ways of legitimately participating in CS—including ways of being without passion, or ways that embrace alternative definitions of passion.

More generally, we call into question the neoliberal and gendered valuing of individualized positive affects (such as passion and confidence). These affects normalize violence by redirecting blame to the individual rather than larger structures of power. We found that understanding the mechanics underlying myths of passion and confidence required looking at the interplay of affective passion and conspicuous production. We hope that this paper will provoke work in other sectors to consider how these two forms of passion work together in upholding neoliberal structures of power.

Supplemental Material

sj-docx-1-sgo-10.1177_21582440241267157 – Supplemental material for Passion as Capital: The Cultural Production of “Good Computer Scientists”

Supplemental material, sj-docx-1-sgo-10.1177_21582440241267157 for Passion as Capital: The Cultural Production of “Good Computer Scientists” by Hana Darling-Wolf and Elizabeth Patitsas in SAGE Open