Thomas Smyth, Jill Dimond
In recent years the HCI community has expressed interest in the social impact of technology. Various research projects have explored how technology can be designed to enhance civic participation and public discourse, provide succor during crises and natural disasters, extend services to the homeless, support national reconciliation after civil conflicts, improve healthcare delivery in developing countries, and so on.
Underlying this kind of work is a value judgment that it represents an earnest contribution to the collective social good. Indeed, the noun good has made several appearances. One Georgia Tech initiative styles itself Computing for Good. An upcoming conference dubbed Social Good Tech advertises that "social good organizations deserve the most innovative technologies" . But this discourse then raises the question: What is good? As technologists, we find this question is harder than ever to answer in the age of such ethically confounding enterprises as Facebook and Google.
Here, we explore an analytical tool called the anti-oppression framework as a means of judging the "goodness" of a technological project or artifact. We review the implications of the framework for the field of interaction design, both for the things we design and for the environment in which we design them.
Oppression theory has its roots in women's studies and such ideas as Patricia Hill Collins's matrix of domination, Elisabeth Schüssler Fiorenza's kyriarchy, and Kimberlé Crenshaw's theory of intersectionality. The anti-oppression framework itself was synthesized in the field of social work and is described as "a tool to understand and respond to the complexity of the experience of oppression" . In other words, the anti-oppression framework understands good as that which strives to end one or more forms of oppression.
In contemporary society, oppression unfortunately takes myriad forms. Common among them are racism/white supremacy, colonialism, sexism, transphobia, heterosexism, classism, ableism, religious oppression, and ageism.
Furthermore, each of these types of oppression can exist on individual, institutional, and cultural levels (Figure 1). For example, transphobia might manifest individually in a parent forbidding their boy from dressing in feminine clothes, institutionally in the lack of coverage for transgender healthcare in most insurance plans, and culturally in the commonly held myth that gender and physical anatomy are the same thing.
Further complicating the picture, oppressions can be conscious or unconscious, and often intersect, meaning that a combination of several forms of oppression (e.g., the sexism and racism encountered by a black woman) form a distinct experience that must be understood separately rather than as a sum of its parts.
The concept of privilege is also central to the anti-oppression framework. Privilege refers to unearned advantages enjoyed simply as a consequence of one's membership in a given group. A helpful metaphor compares privilege with the difficulty setting in a video game—being privileged means playing on the easiest setting. Privilege should not be construed as an insult or indictment, nor does it mean that a privileged individual cannot be the target of oppression. Rather, it refers to structural barriers—advantages such as access to a quality education, bank loans, fulfilling employment, and adequate healthcare are all demonstrably tied to privilege.
A final important entry in the anti-oppressive lexicon is the notion of social justice, which represents the opposite of oppression: The freedom of individuals to determine their own paths without having to confront structural barriers. The term also generally refers to the pursuit of such a reality, through which unjust barriers are eroded and abolished.
The anti-oppression framework can serve as a guide for how best to expend resources, be it the choice of a research topic, the focus of a new social enterprise, or the selection of clients and projects as an industry consultant. In discussing the anti-oppression framework in the context of social work, Beverley Burke and Philomena Harrison write that "[a]n understanding of [anti-oppressive] principles brings with it a fundamental transformation in the relationship that exists between the assessment of a situation and the nature of the action that is required to change the existing state of affairs." We suggest that this notion is equally relevant in our field of interaction design.
If presented with the choice to work on an existing or nascent project, one can ask: What oppression would this work strive to eliminate? On what level? At which intersections? On the other hand, if one is in a position to define a new project or enterprise, one can choose an oppression or an intersection of special interests and search for a solution.
Obviously problematic endeavors (e.g., military contracts) will be quickly weeded out by this analysis, and more nuanced shortcomings may also be. A common example is the dichotomy between social service and social change. Some projects may be focused on providing assistance to those affected by oppression without much interest in disrupting the structures that make it possible. While this kind of work is important and praiseworthy, it is at odds with the definition of anti-oppression given above.
Burke and Harrison also write "the anti-oppressive principle of reflexivity demands that workers continually consider the ways in which their own social identity and values affect [their work]." This concept is also applicable in the interaction design context. We must recognize that we may not have a good understanding of the oppressions we seek to work against and must therefore educate ourselves as a first order of business. We must also recognize our technical biases and be prepared to acknowledge that the best solution may end up being a non-technical one.
At Sassafras Tech Collective, a worker-cooperative tech consultancy created by the authors, this is the approach we have taken. Our vision document, which we refer to when deciding what work to pursue, lays out our understanding of oppression. We favor projects with a background research component, and several of our main projects grew out of doctoral studies. Below, we review some examples of how the anti-oppression framework has guided our work.
Hollaback is a social movement organization dedicated to ending street harassment, or public harassment motivated by gender, gender expression, or sexual orientation . We work with Hollaback to build technology that enables survivors of street harassment to share their stories. Jill Dimond's research suggests this online storytelling has supported frame transformation in the movement as participants construct shared understandings around the issue.
This work appeals to us as it targets the intersection of sexism, racism, heterosexism, and transphobia—harassment can be directed along any of these axes of oppression. Hollaback also operates primarily on the individual and cultural levels—individual, in that the technology provides a means of recourse for survivors and thus potential dissuasion for would-be harassers, and cultural, in that the movement looks to evolve a cultural shift wherein street harassment would become socially unacceptable.
With the Carter Center, a U.S.-based NGO best known for its election-observation and disease-eradication initiatives, we developed tools to support international election-observation missions. A digitized observation operation ensures more reliable and quickly available data, allowing anomalies to be detected sooner.
We chose to pursue this project because it works against a variety of different oppressions depending on the country holding the election. In a country with a history of religious discrimination, an observation mission might pay special attention to issues of religious oppression. Counteracting sexism is a regular focus of the Center's observations. Broadly stated, the goal of election observation is to ensure a genuine democratic process, and this mission is in line with many contemporary anti-oppression movements. Unlike Hollaback, election-observation missions operate mainly on the institutional level. They are rooted in electoral law and their analyses focus on the operations of the election-management authorities rather than individual actors.
Finally, within Sassafras we are developing an application called LaborShare, which will promote equality and visibility of home labor by computing optimal task assignments according to time and preference. Domestic work is traditionally a woman's responsibility but is usually overlooked as legitimate economic activity. Also, a lack of shared understanding of the work that actually goes into the maintenance of a household can lead to conflict. We believe that technology can help clarify responsibilities in a household and make home labor more visible, equitable, and enjoyable.
Due to the gendered nature of home labor, we view this project as counteracting sexism. It also operates mostly on the individual level, household by household. If the application becomes successful, we envision it as a platform for promoting cultural change as well, but this is not the current focus.
As with any product or project with a stated ethical foundation, the environment in which the product is created or the project is carried out is at least equal in importance to the end result. Unfortunately, in many tech workplaces, a prominent feature of the environment is a startling lack of women and people of color (particularly African Americans, Native Americans, and Latinos). This fact is both a form of oppression (since tech is an ever-expanding field and tech jobs are well paying) and a result of oppression. The invention of anti-oppressive tech also becomes more difficult when some of the very groups for whom the technology might be intended are not represented on the design team.
We believe that the notion of reflexivity as promoted by the anti-oppression framework is central to reversing this disturbing trend. A key first step in constructing an anti-oppressive workplace is to cultivate understanding of, and agreement about, the many forms of workplace oppression. Until all members of the group share a nuanced understanding, any other measures for developing a safe space are destined to fail.
External help in this is available and advisable. Groups such as the AORTA Collective offer anti-oppression education and consulting services. Raising the topic for open and frank discussion with other businesses, labs, or groups can also be helpful in addition to raising awareness. From this basic initial process of reflection will follow specific anti-oppressive workplace policies and measures.
Considering team composition is one such measure. Sassafras has a gender quota of no more than 50 percent plus one of any gender . As our organization grows, this basic policy will prevent over-representation of one gender (typically men) in decision-making processes and thus promote further anti-oppressive measures .
Also important are processes for safe conflict resolution. Raising a complaint about workplace conduct should never be cause for fear, nor should receiving one be. Giving forethought to processes for handling conflicts safely and non-violently  reduces anxiety and reassures prospective members from under-represented groups that the space they are entering is an anti-oppressive one. Such measures are especially important in organizations without a hierarchical chain of command since there may be no clear course of action otherwise.
Just as history has shown that social justice is unlikely to emerge in a non-democratic society, building an anti-oppressive workplace is difficult when decisions are made according to a traditional, non-democratic structure. In such organizations, those who resist entrenched oppressive workplace behaviors often end up as the target of hostility. Recent incidents at Github and Tinder are examples of this in the tech industry.
For this reason, we have chosen to build Sassafras as a worker cooperative. Worker co-ops are similar to the more familiar consumer and agricultural co-ops in that all patrons own a share of the business, except that the patrons are the workers themselves as opposed to the consumers or producers.
This arrangement makes for a flat ownership structure, in which each employee of the business (except trial members) owns exactly one share, an obvious prerequisite for genuine democracy. But note that democracy does not imply majority rule. Many worker co-ops have chosen decision by consensus as a more just standard. Consensus can be difficult to achieve on some issues, but it eliminates the threat that a decision unfavorable to a minority is made despite their objections.
The worker co-op is an old organizational form that is enjoying a resurgence in the U.S. (especially in tech; see Figure 2) and has been a long-term fixture in other areas. The Mondragon Corporation of Spain's Basque region is the world's largest federation of worker co-ops, employing some 80,000 people. Worker co-ops tend to create long-term, stable jobs and a concern for community benefit. Many espouse a "multiple bottom line," wherein the business's objectives are not limited to financial returns and include other values such as environmental sustainability, community impact, and worker happiness. This philosophy implies a dedication to worker-friendly policies such as full benefits for all employees and liberal paid time off, including maternity leave.
Reasonable work hours are also an important measure, especially in the tech industry, where long workweeks are common and sometimes even worn as a badge of honor. The expectation of employees to work long hours is not only unsustainable and inhumane  but also embeds several oppressions. Parents, especially mothers, are often unable to work long hours due to responsibilities to their children. Single and/or working-class mothers are especially vulnerable here—a policy of long work hours, even if only during certain "critical periods," virtually guarantees their exclusion. Ableism and ageism also come into play—some people may not be physically capable of performing 10-or 12-hour days.
One challenge in working on anti-oppressive projects is that such projects are frequently underfunded. To us, this is a manifestation of classist/capitalist oppression and a product of the vast wealth disparities in the U.S. and elsewhere. It is no coincidence that endeavors aimed at building equality and counteracting oppression and privilege struggle to gain access to capital. Such is the self-reinforcing nature of our predominant economic and social systems.
This fact suggests several measures for an anti-oppressive business model. First is a commitment to leanness, since the lower costs are kept, the lower and more accessible rates can be. The somewhat lavish features expected of a modern tech workplace (catered lunches, high-value real estate, expensive swag, and the like) are at odds with this. We suggest that these features, which seem designed to attract employees in a highly competitive labor market, would actually prove less attractive than a genuine ownership stake, democratic role, and social purpose if given the chance.
Second is a willingness to adjust rates on a sliding scale whereby clients pay what they are able to afford. This policy combined with leanness measures can allow better-funded clients to subsidize less well-funded ones, with neither paying more than fair market value. The scope of work can also be creatively adjusted so that a viable, high-quality artifact can be delivered on a limited budget.
To return to the article's title, we have hopefully conveyed that what we call "anti-oppressive design" is concerned with not only the theory and method of design itself but also the domain and environment in which design is carried out. It is an evolving collection of tools and approaches.
The anti-oppression framework and related notions of privilege and social justice provide a structure judging a design project's "goodness." Practitioners and researchers alike can use the framework to pose incisive questions about a project's background, vision, and objectives rather than relying on vague intentions or received wisdom about what constitutes good.
Anti-oppressive design workplaces can be constructed through careful and continual reflection, requests for outside help, and the implementation of genuine democracy. These measures are especially needed in the tech industry, where stark gender and race disparities prevail.
The worker cooperative is an excellent model for implementing these ideas. Employee ownership and workplace democracy go hand in hand, and stepping outside the traditional corporate model also makes way for innovative anti-oppressive business models. The worker co-op is an increasingly popular choice for both new and existing social enterprises, especially in the tech industry.
A new movement of anti-oppressive tech is growing.
7. A helpful commentary on this issue can be found at http://wearemammoth.com/2013/11/long-hours.
Thomas Smyth is a worker-owner at Sassafras Tech Collective, a worker-owned cooperative focused on building technologies for social change. He holds a Ph.D. in computer science from Georgia Tech and lives in a co-housing community in Ann Arbor, Michigan. firstname.lastname@example.org
Jill Dimond is a worker-owner at Sassafras Tech Collective. She holds a Ph.D. in human-centered computing from Georgia Tech. She has worked for Google and as an interaction designer. Her work has appeared in the New York Times, NPR, BBC, and the Washington Post, among others. email@example.com
Copyright held by authors. Publication rights licensed to ACM.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2014 ACM, Inc.