Male, Female, Nonbinary: The Ethical Responsibility of the IT Community

Several US states are advancing the recognition, equality, and dignity of nonbinary and genderqueer people in our society through the addition of a new nonbinary gender identification checkbox on legal documents, such as birth certificates and driver's licenses. We're hopeful that these changes will serve as a catalyst for conversations about the collective ethical responsibility of the IT community to safeguard this data and support its appropriate use once collected. This feels especially important considering that classifications of gender—as well as race and ethnicity—and their codification in systems of record have been used to reinforce discriminatory practices in society for generations.

In the summer of 2017, the state of Oregon was the first to announce that it would recognize a new gender choice, nonbinary, on its driver's licenses. Washington DC quickly followed suit. Currently, Arkansas, California, Colorado, Maine, and Minnesota offer nonbinary driver's licenses, while other states have bills in the legislature.¹

In our state, Governor Jerry Brown signed California State Senate Bill 179 (SB-179) into law on October 15, 2017, enabling "intersex people, like transgender and nonbinary people, to be able to use state-issued identification documents that accurately recognize their gender identification as female, male, or nonbinary." SB-179 requires that a third, nonbinary gender marker be made available not only on driver's licenses but also on birth certificates, identity cards, and gender-change court orders. It also improves the process a person must follow to change their gender marker and/or name on these identifying documents.²

And so, a little over a year after the passage of this bill, the two of us were sitting with IT leaders from across campus listening to a presentation from the working group that had been created to support implementation of SB-179. At UC Berkeley, this has meant answering the following questions: Which IT systems and/or administrative processes collect and store gender identification information? What changes are required to support an individual's nonbinary choice? What downstream systems do they feed? What are the costs and technical implications of the changes? How do we translate a nonbinary gender selection in our systems to best comply with federal reporting requirements where only binary options are allowed? Are their special considerations for international students hailing from countries that are hostile to LGBTQ individuals?

As we reviewed the technical issues, the conversation quickly evolved to the myriad ways that online systems currently reinforce gender norms. How many of us have filled out an online form that requires us to add a title or honorific that denotes gender? While the honorific Mx. has been established, it is rarely used in online forms or systems. And what about marital status? Some women probably still feel like a bit of a rebel when they select "Ms." instead of "Mrs."

The implementation of IT systems updates will need to address these issues as well as the implied promise for how gender identification information is used by the institution after collection. For example, will it be used to generate salutations in donor letters or admittance correspondence? Will it be used for creating student housing assignments or "balanced freshman cohorts" for orientation? Will it be presented to faculty via class rosters?

As we have reflected on these questions, a number of additional questions have surfaced: How do we ensure that we are representing the designation in the most inclusive manner, without "otherizing" it? Does the order or presentation for gender identity options say something about the value that we place on the person inhabiting that gender? Even our existing binary choice has a standard pattern: "male" and then "female" (M/F). That certainly isn't a decision based on alphabetical order. What does that reinforce about gender and position in our society? Will we offer nonbinary as the first choice, the last, or one in between?

At UC Berkeley, those of us in the IT organization will often be the ones answering these questions and making these design decisions. To be truly inclusive, and to honor both the letter and the spirit of SB-179, we will need to get input from the genderqueer community on how best to represent the nonbinary identity.

In their 2000 book Sorting Things Out: Classification and Its Consequences Geoffrey C. Bowker and Susan Leigh Star note: "Classification systems are often sites of political and social struggles."³ Many of their examples focus on race and ethnicity classification systems developed by people in power to suppress and demoralize entire populations. The same holds true for gender.

While the classification of "nonbinary" is embraced and supported by most of the genderqueer community as empowering, with a potentially positive impact on social justice, IT service providers must consider and guard against misuse—or abuse—of this information. We must be mindful of how the data is stored, how it is used, and with whom it is shared. One can't help but think about current events and the unintended consequences of data being used in harmful ways. Analysis of FBI hate crime statistics for 2016 by the Human Rights Campaign shows that of the 6,121 crimes classified as hate crimes in the United States that year, 1,076 were predicated on sexual orientation bias, and 124 were based on gender identity bias.⁴

Student records follow students as they attend multiple institutions throughout their lives. Thus, even if a higher education institution is not located in a state that is considering or has passed legislation around this issue, its IT leaders need to think about several considerations:

1. Data Governance: Who stewards this data at the institution? What, if any, additional considerations, protections, and/or controls should be put in place for access and usage? Who participates in those conversations? Does the institution have student representatives serving in its data-governance model?
2. Ethics: Does the institution have a statement on ethical behavior, an honor code, and/or an IT code of ethics? Can these be used as a framework for conversations?⁵ Will they need to be updated to include this new use case?
3. Implied Promise or Experience: What promises are implied through the collection of this data? For example, when users are allowed to specify their preferred pronouns in the institutional systems, can they expect that formal communications from the institution will address them using their preferred pronouns?

The world of data, classification, and systems will grow only more complex. With the utilization of machine learning and artificial intelligence, the kinds of seemingly simple decisions we make about the design of our systems will have an impact for years to come. We should be thinking about these issues and our role not just in designing data systems but also in supporting social justice. Ensuring that we are treating the data with the same dignity and respect with which we treat the person would be a good start toward meeting our ethical responsibility.

Notes

1. Casey Parks, "Oregon Becomes First State to Allow Nonbinary on Drivers License," The Oregonian/OregonLive, June 15, 2017 (updated August 3, 2017); "State Laws" [https://www.intersexrecognition.org/resources], Resources: Non-Binary Gender, Intersex, Intersex & Genderqueer Recognition Project (website), page updated on January 15, 2019. ↩
2. "SB-179, Gender Identity: Female, Male, or Nonbinary (2017–2018)," California Legislative Information (website), accessed January 7, 2019; "SB 179: Gender Recognition Act of 2017," fact sheet from Equity California (website), accessed December 15, 2018. ↩
3. Geoffrey C. Bowker and Susan Leigh Star, Sorting Things Out: Classification and Its Consequences (Cambridge: MIT Press, 2000), 196. ↩
4. Jordan Dashow, "New FBI Data Shows Increased Reported Incidents of Anti-LGBTQ Hate Crimes in 2016," Human Rights Campaign (website), November 13, 2017. ↩
5. An example is the "University of California Regents Policy on Statement of Ethical Values and Standards of Ethical Conduct," approved May 2005 and amended March 16, 2017. See also Melissa Woo, "Ethics and the IT Professional," EDUCAUSE Review, March 27, 2017. ↩

Jenn Stringer is Chief Academic Technology Officer and Assistant Vice Chancellor, Teaching and Learning, at the University of California, Berkeley. She is the 2019 Editor of the Viewpoints column for EDUCAUSE Review. Benjamin Hubbard is Director of Service Design at the University of California, Berkeley.