Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Categories
In this Discussion
[PROMPT] Queer Bodies, Embodying Code (Main Thread)
by Edmond Y. Chang & Jarah Moesch
EDMOND CHANG: Let’s start with a short exercise: as you are reading this (or perhaps even listening to it), for a minute, think about what you are doing, feel what you are doing and thinking, and take in your circumstances and surroundings as you think, feel, and do. Are you seated, standing, laying down? Where are your hands? On a keyboard, mouse, mousepad, perhaps cradling your phone? Is the computer warm on your lap or the screen glowing brightly against your skin? What is going on around you, what catches your senses, is there movement, sound, smells? Are you distracted, multitasking, waiting, rushing, irritated, amused, engrossed, excited?
I wanted to open the forum (and the week) with this little meditation as a way to think about our relationship to the digital, to code, to our bodies, and to others’ bodies. How might code be queer(ed) is one of our questions this week, and how might code be embodied is another?
Jason Boyd and Bo Ruberg in “Queer Digital Humanities” remind us that digital technology is “not only tangible but sensual, luxuriant, and pleasurable. Queerness is invested in identity, representation, and history but also in erotics” (70). They ask us to develop modalities and practices to understand how “information might be conveyed and interpreted through a sense of touch and closeness” (71).
JARAH MOESCH: As we sit with our bodies, our embodiments, our phenomenologies, we can orient ourselves (a la Sara Ahmed’s Queer Phenomenologies) within spacetime to think about how realities are formed and created both for and by us. Cultures and matter are entangled- neither exists without the other.
For example, when I woke up this morning, I found my way into the kitchen to pour myself a cup of coffee. On the way, I stopped for a moment to reach over and hit the ‘power’ button on my laptop, letting the computer ‘boot up’ while I stirred a bit of sugar into my mug, and then settled-in in front of my screen, ready to begin my day researching, writing, drawing, and corresponding. With these few actions I set into motion entire histories and re-inscribed racism, colonization, and genderism by the mere touch of a button.
No system is neutral. The hardware, the code, and user interactions are framed by how humans understand themselves and each other. And not just any humans, but specific humans, who are, amongst other things, aged, gendered, classed, raced, sexualized, nationalized, disabled, and educated in particular ways.
The materiality then, is in the interactions- of executing code (execute!? I mutter to myself as I type this) that is already inscribed with these (violent) histories.
How do these culturally inscribed categorizations get replicated, translated, mutated, opposed, restructured, dissolved within code? What happens to these categories as they shape the 'web' or software? How do top-down hierarchical frameworks reify and normalize social constructions and embodiment? What does it mean to be queer within these structures? What does it mean to be a queer who codes?
EDMOND: Yes, what does it mean to “queer code?” Or perhaps “code queerly?” I guess I would start with Kara Keeling’s delightfully provocative “Queer OS,” which argues:
Queer OS names a way of thinking and acting with, about, through, among, and at times even in spite of new media technologies and other phenomena of mediation. It insists upon forging and facilitating uncommon, irrational, imaginative, and/or unpredictable relationships between and among what currently are perceptible as living beings and the environment in the interest of creating value(s) that facilitate just relations. (154)
Keeling goes on to think through the ways that “Queer OS ideally functions to transform material relations” (154) from race, gender, and sexuality to code, platform, tool, and practice. How might we extend these initial questions, connections, and possibilities to other domains, other transformations?
JARAH: Following Keeling, the authors of QueerOS: A User’s Manual created an intervention that “seeks to address what we perceive as a lack of queer, trans, and racial analysis in the digital humanities, as well as the challenges of imbricating queer/trans/racialized lives and building digital/technical architectures that do not replicate existing systems of oppression.” While this is a user’s guide, and not executable code, it also claims “iterative failure, with no permanent solutions.”
This failure is ultimately queer.
EDMOND: Of course, this brings to mind J. Halberstam’s The Queer Art of Failure that asks and argues,
What kinds of reward can failure offer us? Perhaps most obviously, failure allows us to escape the punishing norms that discipline behavior and manage human development with the goal of delivering us from unruly childhoods to orderly and predictable adulthoods…And while failure certainly comes accompanied by a host of negative affects, such as disappointment, disillusionment, and despair, it also provides the opportunity to use these negative affects to poke holes in the toxic positivity of contemporary life. (3)
For me, my favorite definition of queer comes from Eve Kosofsky Segwick’s “Queer and Now.” They argue that queer is
the open mesh of possibilities, gaps, overlaps, dissonances and resonances, lapses and excesses of meaning when the constituent elements of anyone’s gender, of anyone’s sexuality aren’t made (or can’t be made) to signify monolithically. The experimental linguistic, epistemological, representational, political adventures… (7)
Both definitions invoke contingency, messiness, and polyvocality resisting clear, coherent, perfectly legible order and communication, which Donna Haraway names “the informatics of domination.”
JARAH: Queer has multiple meanings - it can be queer bodies/people, it can be odd, a theory, a practice. One useful way to think about queer is that queer tries to subvert, to go against, and to be outside of normative ways of life. This can look like outright refusal, and it can be a quiet inflection (or something else entirely). As I mentioned before, queer can be failure, though I’d argue that J. Halberstam’s failure is only the beginning.
As I’ve written about before, in The Queer Art of Failure, J. Halberstam frames the argument of the queer politics of failure through the “desire of living life otherwise,” of “getting outside the ‘conventional understandings of success,” “…into a more chaotic realm of knowing and unknowing” (Halberstam, 2011, 3). This situates the queer politics of failure as being against that of U.S. white heteronormative success, placing failure in opposition to, yet still within the framework of success. Additionally, Halberstam claims: “if success requires so much effort, then maybe failure is easier in the long run, and offers different rewards” (Halberstam, 2011, 3).
How is failure easier than success? How do we know when we have failed successfully and when we haven’t?
If we have achieved success with our failing, Halberstam offers us many rewards:
[F]ailure “allows us to escape the punishing norms that discipline behavior and manage human development with the goal of delivering us from unruly childhoods to orderly and predictable adulthoods” …“[F]ailure preserves some of the wondrous anarchy from childhood and disturbs the supposedly clean boundaries between adults and children, winners and losers” (Halberstam, 2011, 3).
What does it mean to “escape punishing norms”? If we choose to fail at heternonormativity (and homonormativity), do we escape from the pressures themselves, or just the norms? Once applied, do we ever gain relief from the pressure? Do the norms just simply ‘give up’ on trying to make us change? Is “disappointment, disillusionment and despair” along with the ability to somehow ‘use’ these to ‘poke holes’ in contemporary life as our rewards enough? While investigating future possibilities, Halberstam doesn’t take it far enough, as failing is not failure. You cannot fail purposefully if what you need is a job, or health insurance, or a home.
Conversely, if you don’t need or want those things, then ‘failure’ is no longer failure, but becomes something else entirely. Unfortunately, though The Queer Art of Failure discusses alternative ways of being, or not being, as a way to refuse, circumvent, avoid, rebel, push back against, and move outside of normative modes of being (frameworks), it does not take us outside normative frameworks simply by situating failure as possibility. Failures always lie within and against successes. If you live within a particular normative society, in this case the white heterosexist, patriarchal U.S. society, you cannot be outside of it.
As we turn to code (since we are in a critical code forum), is code written by queer people automatically queer? Is it always inflected with queerness just by being written by queer bodies? Here I mean in the sense of ‘born digital’- objects, artwork, literature that were created with digital output as only form, is there a born queer code? If not, then what is queer code? What can it do, beyond failure? If queer code can’t escape punishing norms, (or pressures, or norms), how might we work within to subvert and rewrite code bases? Is it a bot? a worm? a virus? Here the though of virus brings up interesting correlations with bug chasers. So many interesting theoretical ties- so what could the actual, functional code look like?
EDMOND: I am not sure if anything is “born” anything, but I know what you mean. One of the ways I have tried to think about queer(ing) code is through trying to grapple with writing it. My essay “Why are the Digital Humanities So Straight?,” which riffs on Tara McPherson’s essay “What are the Digital Humanities So White?,” is an academic essay-as-BASIC-program-as-text-adventure-game. I picked BASIC because it was more readable on the page (and because it is the only language I know well enough to pull off an executable program). I wanted to demonstrate how algorithm and code are usually hidden from readers, users, and players, and that code itself is often normative, hierarchical, and embedded with racist, sexist, phobic constructions, mechanics, and meanings. In the essay-as-code-as-game, the reader/player can navigate through three sections on Alan Turing, Ada Lovelace, and Purna Jackson (who is a video game character).
In the Turing mini-game, you are sitting in a small room before a teletype machine, recalling of course Turing’s “Computing Machinery and Intelligence” essay and the oft cited “Turing Test.” The reader/player must navigate how to get out of the room or move on to the next section by exploring what they can or cannot do. Those constraints, those limitations, even as reader/players can have fun with the experience, are part of the text’s argument about the technonormativity of code–you cannot do anything that isn’t already in the accepted and acceptable actions. Below is the code for the very limited command parser for Turing’s mini-game:
5100 REM Alan's Only Choices
5102 REM The correct commands are predetermined but give the illusion of choice.
5103 REM Unable to see the code, the player's commands are arbitrary and contained.
5105 IF Action$="look" THEN GOTO 360
5110 IF Action$="sit down" THEN GOTO 5160
5111 IF Action$="sit chair" THEN GOTO 5160
5112 IF Action$="sit" THEN GOTO 5160
5114 IF Action$="stand up" THEN GOTO 5180
5115 IF Action$="stand" THEN GOTO 5180
5116 IF Action$="get up" THEN GOTO 5180
5118 IF Action$="read" THEN GOTO 5200
5120 IF Action$="read text" THEN GOTO 5260
5121 IF Action$="read message" THEN GOTO 5260
5122 IF Action$="read teletype" THEN GOTO 5260
5124 IF Action$="read paper" THEN GOTO 5260
5126 IF Action$="look teletype" THEN GOTO 5260
5128 IF Action$="look paper" THEN GOTO 5260
5130 IF Action$="open door" THEN GOTO 5210
5132 IF Action$="look console" THEN GOTO 5260
5134 IF Action$="type" THEN GOTO 5220
5136 IF Action$="use teletype" THEN GOTO 5220
5138 IF Action$="remove name tag" THEN GOTO 5230
5140 IF Action$="remove tag" THEN GOTO 5230
5142 IF Action$="hit switch" THEN GOTO 5275
5144 IF Action$="stop being gay" THEN GOTO 5300
5145 IF Action$="come out" THEN GOTO 5300
5146 PRINT "You are constrained by the limits of the room and its design. Try again."
5148 PRINT "You cannot "; Action$; " here."
5150 GOTO 368
The main idea here is that unless you can see the code there are missed opportunities: first, to talk about the ways that the code deeply constrains the player, and second, the missed queer opportunities like the commands “stop being gay” or “come out.” Now, whether this code is queer(er) because I as a queer person of color wrote it is something to think on.
JARAH: Here, I think it is also worthwhile to bring up Zach Blas’ transCoder, a Queer Programming Anti-Language, a coding language that does not work / execute the way it ‘should’ for it do ‘work’ within our current world of code. (Mark Marino has written about this anti-language in relationship to critical code studies). Because it is an anti-language, and does not work, it very much exists outside the normative values of our coded worlds. This queer code has its place as subversion, outside of normativities.
Both this, and the QueerOS mentioned earlier are examples that point towards subversion, towards refusal. And these have a place within queer code, and create a space outside of normative code. Being outside is good. However, when we go to use our everyday normative systems, the code is heternormative still, and we are back inside those normativities, straining against them as we are constrained by them.
I posit that since we live in this heteronormative (& homonormative) world, there is a need to break these systems open, to write anew, to queer the code.
I imagine that queer code, just like queer bodies, is functional within its space(s). As queer body is always in spacetime, so too the code: it is written and executed in functional spaces. Sometimes queerness (and queer code) needs to be hidden, sometimes it is outright, subversive, deviant.
- What would it look like to write this kind of functional queer code? Is that even desirable?
- How might creative, artistic, even everyday interventions allow for queer(ing) code?
- How might we create code that changes the very structures of normativity?
- And what does that mean for queerness if it changes normativity?
Works Cited
Ahmed, Sara. Queer Phenomenology, Duke University Press, 2006. https://doi.org/10.2307/j.ctv125jk6w
Barnett Fiona, Zach Blas, Micha Cárdenas, Jacob Gaboury, Jessica Marie Johnson, and
Margaret Rhee, “QueerOS: A User’s Manual,” Debates in the Digital Humanities 2016, edited by Matthew K. Gold and Lauren F. Klein, University of Minnesota Press, 2016, https://dhdebates.gc.cuny.edu/read/untitled/section/e246e073-9e27-4bb2-88b2-af1676cb4a94.
Blas, Zach. “Queer Technologies,” https://zachblas.info/works/queer-technologies/
Boyd, Jason and Bo Ruberg. “Queer Digital Humanities,” The Bloomsbury Handbook to the
Digital Humanities, edited by James O’Sullivan, Bloomsbury, 2023, pp. 63-73.
Chang, Edmond Y. “Why are the Digital Humanities So Straight?” Alternative
Historiographies of the Digital Humanities, edited by Dorothy Kim and Adeline Koh, Punctum Press, 2021, pp. 203-241.
Halberstam, Judith. “Introduction: Low Theory,” The Queer Art of Failure, Duke University
Press, 2011, pp. 1-26.
Haraway, Donna. “A Cyborg Manifesto,” Simians, Cyborgs, and Women, Routledge, 1991, pp.
149-181.
Keeling, Kara. “Queer OS,” Cinema Journal, Vol. 53, No. 2 (Winter 2014), pp. 152-157.
Marino, Mark C. “Disrupting Heteronormative Codes: When Cylons in Slash Goggles Ogle
AnnaKournikova,” UC Irvine: Digital Arts and Culture 2009, retrieved from https://escholarship.org/uc/item/09q9m0kn.
Moesch, Jarah. Designing the Sick Body: Structuring Illness in the Techno Material Age. Diss.
University of Maryland, College Park, 2016.
Comments
Thank you for this beautiful opening piece.
I was wondering how the above might help current debates on AI "bias" (a concept I personally dispute in as much as it leads to technical solutions, i.e. the idea that providing a "diverse enough" training datasets would solve AI sexism). For example, AI-powered facial recognition struggles to recognise (and often misgenders) non-binary people.
I totally support the idea that AI (as other technologies) is a sociotechnical compound that performatively constitutes the observer and the observed (for example, gender recognition software per formatively constitutes the gender of the person it observes). The problem is that AI tends to constitute gender according to a binary.
When a non-binary person fails to authenticate (or even to enrol in a biometric system), is that a productive failure (cf. Halberstam)? Or is it (as it seems at the moment) a disempowering failure that leaves the un-recognised, invisibilised, non-conforming individual powerless?
How can practices of queering code contribute to solving this?
Your point on Gramsci's hegemony as common-sense and the potential of queer practices for constituting new commons by making the invisible visible is really helpful. I guess my question is: how do we proceed from here?
Works of relevance:
Os Keyes (2019) The Body Instrumental, https://logicmag.io/nature/the-body-instrumental/
I love Os' idea that "inferring gender from facial features is complete bullshit". His concept of "bullshit technology" might help here.
Joseph Pugliese (2010) Biometrics, Technologies, Biopolitics. New York: Routledge.
Joseph Pugliese (2005) In Silico Race and the Heteronomy of Biometric Proxies: Biometrics in the Context of Civilian Life, Border Security and Counter-Terrorism Laws, Australian Feminist Law Journal, 23:1, 1-32, DOI: 10.1080/13200968.2005.10854342https://doi.org/10.1080/13200968.2005.10854342. (Pugliese concentrate on race here but, needless to say, gender and race are deeply intertwined).
(With apologies for self-promotion): Eleanor Drage & Federica Frabetti (2023) Copies without an original: the performativity of biometric bordering technologies, Communication and Critical/Cultural Studies, DOI: 10.1080/14791420.2023.2292493
Thank you Federica for your insightful thoughts & questions.
I am not embedded in the AI world, so I am likely missing enough that prevents me from adequately responding. However, I’ll give it a go.
I think that this ‘failure to authenticate’ is exactly why Halberstam’s work didn’t go far enough - if you need to be able to use the system, you cannot fail. Failure to authenticate as the gender you identify with, in order to unlock the bathroom you need to use, means you don’t get to use the bathroom safely, if at all. The AI, as you called it, is a sociotechnical compound, one that is based on particular norms that also exist outside of the code base - we see these norms out in the world: people surveil each other in and around bathrooms - intensely, and dangerously. It is difficult to “escape punishing norms” with this kind of failure (Halberstam, 3).
The releasing / assigning of gender, or queerness, (and race, class, and ableism) to an AI that is built on already heteronormative, binary gender systems, that are also racist, classist, and ableist, makes targets out of some people rather than others. Becoming a target for AI is a punishing continuation of people’s lived experiences.
If you speak with non-binary or non-conforming folks you will get a different response than you will if you talk with the presumably hetero/homo normative AI decision makers. Lived experiences are a form of knowledge that it seems AI doesn’t explicitly use. (perhaps because knowledge, information, and data are all different, but I digress).
So perhaps if the decision makers were non-binary or gender non-conforming, disabled, queer, people of color, prior to the creation of AI, some of these ways of being in the world might be included in the base code. I think Donna Haraway’s concept of situated knowledges - that knowledges come from multiple perspectives, and there is no neutral standpoint - is useful for thinking about how queer, non-binary, or non-conforming people in decision making positions might help expand how AI is conceived.
(this reminds me of the 1992 poem “I want a President” by Zoe Leonard. It begins: “I want a dyke for president. I want a person with aids for president and I want a fag for vice president and I want someone with no health insurance and I want someone who grew up in a place where the earth is so saturated with toxic waste that they didn’t have a choice about getting leukemia.”)
Also, though, is there a way to confound the AI? to rework the AI? To queerly rewrite the code base and teach it to recognize multiple genders? or to teach it to refuse to consider gender at all?
(edited to add:) what would it look like to queerly code all AIs to make them not ‘recognize’ any category? to turn them into something else entirely?
—
works cited:
Haraway, Donna, “Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective” Feminist Studies, Vol 14, 3 (autumn, 1988), 575-599.
Halberstam, Judith. “Introduction: Low Theory,” The Queer Art of Failure, Duke University Press,
2011, pp. 1-26.
Leonard, Zoe. I want a president, 1992. https://www.lgbtqnation.com/2016/03/i-want-a-dyke-for-president/
Thank you for all these stimulating thoughts and references. Jarah mentioned "Zach Blas’ transCoder, a Queer Programming Anti-Language, a coding language that does not work or execute the way it ‘should’ within our current world of code". It made me thought of the queer and transgender programmer Jamie Faye Fenton, who created a piece of experimental video glitch art, Digital TV Dinner, in 1978. I discovered it through an article by W. Pow. Would you have any other examples that you would recommend taking a closer look at, and especially for the 70s, 80s, and 90s? I can easily imagine that Queer Code has a long history.
--
work cited:
Pow, Whitney (Whit), "A Trans Historiography of Glitches and Errors", Feminist Media Histories, 2021, Vol. 7, Number 1, pps. 197–230.
Here lies a grinding edge for me regarding queer failure: there is a difference between failing with intent, agency, knowledge, resistance and failing because of the crushing weight of normativity, institutionality, and structural violence. Not everyone can "fail" in the same way or with the same consequences. Given that queerness is often a horizon (to quote Jose Esteban Munoz), what this looks like or feels like or lives like in the "real world" is challenging. Perhaps that's part of the magic--that uncertainty, which current computers (and structures) are very bad at, might be leveraged in some way? I think of face scanning tech that can be confounded by a hat or eye wear or makeup. Perhaps the failure of a computer system to recognize gender (as a binary) is an opening into dispensing with the need to categorize in the first place (a la https://www.theverge.com/2019/1/25/18197137/amazon-rekognition-facial-recognition-bias-race-gender)?
Nothing comes to mind, but I will root around. I am sure there are folks in this hivemind that are way better versed in this history. Obviously there's Turing on up. But a lot of the history of queer programmers and designers is still being uncovered. So much of what people have done was ephemera. I just think about all of the physical media (e.g. disks and printouts) that are just not around anymore.
Thank you very much edmondchang for your answer (and for your thoughts on ephemerality and availability of sources, which are key issues too).
I just came through a few figures like Christopher Strachey here: https://www.thecodingspace.com/blog/2022-05-27-10-queer-pioneers-of-computer-science/
but it would need further investigations on how to retrieve their history, contributions, approach, and notably through CCS and "Queer code". This was a brief historical aside, but let's go back to the present and this exciting thread, that opens new stimulating avenues to rethink past histories !
First of all, thank you for your reflections and references!
This topic of queerness in relation to coding brought to my mind the project Queering Code by Winnie Soon.
The reason why I find this piece of code particularly interesting (apart from its critical perspective on software) is because it works as code poetry and at the same time, the artist shows how to execute it.
I'm intrigued to know what you think about these projects/manifestations which are more valuable on the aesthetic/conceptual realm that on their application or impact in apps that we use on a daily basis...
Such a great discussion already started here, @jarahmo thank you for invoking the Zoe Leonard piece. I have been thinking the same, as I work on a piece about Broken Machines (Sarah Sharma), Glitch Feminism, and GPTs. Forgive the personal share here, but why not say it plain:
Also to say, I love these questions and hope we continue reckoning with them through code:
To create AI/code systems that work outside of categorization would mean returning to the roots (radicality) of computation, where decisions were made about boolean algebra and voltages becoming binary rather than spectra. These are entirely possible and within queer imaginaries, if we don't treat the foundations of computing as fixed and natural (as we have done with gender and sexuality for so long).
Thank you @jarahmo and @edmondchang for such an important thread. Let me encourage everyone to start code critique threads on these objects of inquiry, so we can investigate them individually. (@martapcampos perhaps one for Queering Code, @valerieschafer let's start one for Love Letter Generator, though I don't believe we have the actual code for that one -- not sure.) Perhaps we can draw enough of the algorithm from this article.
I'd also point attention to a few pieces from Poetic Operations by micha cardenas, including net.walkingtools.Transformer and micha's two poems femmeDisturbance.keeling.blackFemmeFunction and femmeDisturbance.janelleMonae.flickerBetweenRealities.
micha has been an important part of our CCS community, and Poetic Operations has much to teach us. (Again, we should set up code critique threads for these. micha's scalar book of her dissertation has code in it.
Also, take a look at this article by cardenas and Blas.
I am interested in seeing queer readings of code not specifically written in the context of identity or written outside or seemingly unaware of the force of norms, sexual or otherwise. What might a queer reading of ELIZA or ALEXA (or code written for ALEXA devices) look like? Can we offer a queer reading of code bases we have looked at in the past, like ADVENTURE, or the Apollo Lunar Lander, the Uroboros quine? What might queer(ing) this code look like and where might it lead us? I'm eager to find out.
Thank you very much @markcmarino for pointing out to this amazing article on the Love Letter Generator. I found two reimplementations of the code, one by Nick Monfort (intended to work in Python 2 (>= 2.5) as well as 3) and one on Github by Gingerbeardman (a PHP Implementation) and I guess we have enough to start a code critique thread with @Titaÿna. More to follow soon / tomorrow (as we are on European time
Thank you for this vital prompt. I just landed here and I'm looking forward to sitting with this and responding more deeply. In the meantime, I want to share an activity I wrote, "Meditation for Our Devices," that touches on some of the sensory aspects invoked in Edmond's opening exercise: https://oilancestors.com/archive/device/
I don't know if it is really within the scope of your question, @valerieschafer , but I think a fun example of codework used in a queer setting is the Natural Bears Classification System detailed by Bob Donahue & Jeff Stoner in Les Wright's The Bear Book. It's very close in idea to the work Marylyn Tan featured this week, and it fits the time period you were looking at (late 80s - 90s). To say it quickly, the "code" was meant to simplify physical descriptions of men, so that a man could be described by a string of letters and symbols. It goes into excruciating details!
In that case, it's less of a queer reading of code, and more of an attempt to (playfully) codify queerness, if that makes sense.
@Lyr Excellent, it makes perfect sense and this is great food for thought (as well as a wonderful example I guess, to introduce students to questions like "embodying code" "self-classification", to Usenet, etc.). Thank you !
Thank you all for the insightful discussion and inspiring work (and amazing references)!
I'll gladly handle the Strachey love letter algorithm thread @valerieschafer @markcmarino. You can expect it in a few hours.
Additionally, the idea of applying queer readings to unexpected code, such as ELIZA or the Apple Lisa, intrigues me. I've already discovered some unexpected comments in the source code. I will continue to explore this perspective, aiming to find fitting examples.
I just wanted to say thank you all for the contributions to my questions on AI. I understand it is not strictly-speaking a CCS question, but I do agree with the idea that more stakeholders (coders) should be involved and that not all failures are equal. Also, the question of how code can be confounded and that could be the starting point for not categorising gender, although I agree with Sarah's cautions on radical computation.
Apologies if we’re meant to be moving the discussion along to concrete analysis straightaway, but I wanted to weigh in on a few of the comments folks have made. I really love the definition of ‘queer’ by Segwick that Edmond brought up, highlighting the:
Others have touched on a lot of why this is difficult to realize through computational systems, the insistence on binaries/gender being one, but I particularly appreciate Sarah’s comment:
Bridging the gaps (no pun intended) from queer imaginaries of possibility, overlap, and gloriously messy excess to queer computations which ultimately do have to reckon with the foundations of computing: Turing’s prosecution, ARPANET and the DoD, binary logics, etc. I agree with Federica that queer developers will help, but The Turing Test is a great example that this isn't enough for queer methodologies (or at least methodologies that remain queer over time) since the test's logic relies completely on a male/female gender binary underpinned by normative cis/Western gender performance. How can we as scholars and artists support specific and radical code systems?
I wonder if there are ways to leverage poetics of refusal/upheaval to trigger literal failures that succeed at more radical goals . I’m thinking of literal poetry like One Art by queer poet Elizabeth Bishop (link), which takes the codified structures of the villanelle and overturns them to succeed at processing rather than documenting situations. I’m also thinking of pursuits like Lucas LaRochelle’s Queering the Map project (link), which uses a normative API (google maps) to embody queerness around the world and hold space for the narratives and dissonances that emerge.
If queerness attempts to avoid monoliths, ( this is not not an Angels in America reference ) then it is inherently opposed to AI systems, which aggregate identities as data in order to extrapolate more valuable information as defined by normative systems (e.g. capitalist, institutional, publishing). I don’t have access to the Queering the Map code itself, but there are a few GitHub repos analyzing and fusing the stories in the project to learn more about variations in queer experiences by nation (link). This is an analysis of a queer project, but is it innately queer? Is sentiment analysis a sufficient tool to analyze queer speech?
I don’t think that queerness is antithetical to quantitative methodologies and the powerful conclusions that they can support, but it seems the gravity of AI towards aggregation will filter away dissonances and excesses (to go back to Segwick's definition again) in datasets in order to convene towards a normative output.
What would an AI look like that seeks the strangest answers, the most radical conclusions; how would the failures of robust conclusions that it would produce sustain nuance and queerness? Are their ways to uphold that dissonance in traditional models?
Would be interested in looking at maybe the Google Maps API to consider embodiment more concretely, and its potential capacity for queer embodiment, if others have interest.
@SarahCiston - I love what you’ve done with this piece - I think the idea of being on the cusp of (normative) existence when already existing feels true to life, both for me personally (queer, sick, disabled) and in terms of this ongoing conversation. The idea of this cusp intrigues me, and I want to think it through some more in terms of what it could mean for confounding the AI, or rethinking the code base.
@valerieschafer - while this is not ‘queer code’ per se, there are trans game designers who have created some very queer and trans work. Merritt K and Anna Anthropy come to mind. Anna Anthropy is well known for Dys4ia a game (done in Twine, if I remember correctly), where the player plays through a simulation of hormone replacement therapy. I had students play through it in my emerging media courses because it changes (& queers) their standpoint, and it plays well, and provides a way in to deeper discussions.
@markcmarino - awhile back (ok, a long time ago…) I had written & presented on queerly reading code comments from a bit of facebook code that ended up in the wild. I am game for doing something similar (but more current) if there’s code to be had. For reference, it’s still on my website here: http://jarahmoesch.com/2011/04/04/critical-code-the_ego_page-queerly-reading-the-computer-code/
Alas, merritt k's games have been on and off the web for a while now, but some are back and I am shocked that LIM is back online, it's so good): https://merrittk.itch.io/
Dys4ia is flash, alas, and sometimes runs with the right emulator: https://freegames.org/dys4ia/
Hi all, thanks for sharing such rich reflections, beautiful ideas, and links to amazing work here.
What strikes me is this fundamental tension between generalization (AI) and individualization (at the heart of queer identity and pride), and a similar tension between co-existing needs for both representation and privacy. Large language models like ChatGPT are essentially predicting the "most likely" next word, based on whatever training data, most likely from public sources where the majority of contributors are programmed to follow certain cultural norms. Even a chatbot trained on one therapist's extensive body of work fails to embody her highly attuned capacity for understanding nuance and bringing together threads that aren't obviously, directly connected.
I love the idea of queer(ing) code as a rethinking, reclaiming, retooling that subversively, unapologetically centers the comfort of one's own marginalized identities. Part of the challenge is how algorithms are designed not just to accomplish tasks but to manipulate users' attention–so I wonder if our rethinking, reclaiming, retooling has to match this depth/intensity of underlying intent.
One of my favorite quotes says, “I would not give a fig for the simplicity this side of complexity, but I would give my life for the simplicity on the other side of complexity.”- Oliver Wendell Holmes. I think we're collectively in the dark woods of code complexity, finding ourselves able to reject oversimplified encodings of our lived experiences and still seeking something more authentic and inclusive. To this end, I shared in a separate thread (Drag Queen Data Science) a few thoughts on how data science would benefit from taking several notes from drag culture. The documentary, "Paris Is Burning" interestingly portrays a maximally queer space where categories are explored and performance is judged by its realness. I wonder if it's a fair analogy to say code is a kind of performance, and whether in the future we'll simply be able to judge software by its apparent realness.
Thank you so much for this fantastic prompt @edmondchang & @jarahmo.
When looking at how we Queer code I think one of the most important things is challenging the prevailing assumptions about code and software. As a small example, I was working on developing an interactive narrative and wanted it to be a story that everyone could enjoy. To this end I wanted to have the player be able to select their pronouns and if they would want any romantic themes and the gender of the person that those romantic themes would centre around.
I have not written up any papers on this work but if you want to have a look you can see the work on my website, Blood and Ink, under Interactive Narrative.
If you have spent much time in the games space you would have heard developers claim how difficult developing games with multiple lead genders is (e.g. Ubisoft Blames Lack of Female PCs in Assassin’s Creed on “Reality of Production”). I was perplexed by this so when I started to build my narrative I started with the player's ability to choose their pronouns (initially, see V1, the choices were feminine, masculine and non-binary). What I discovered was that if you start with the default being male or even female it becomes very difficult to build it for non-binary pronouns, as non-binary pronouns are the most grammatically complex. However if you start with the assumption of non-binary when writing and constructing the narrative systems the move to support any gender pronoun choice becomes programmatically simple. I think this can flow on by extension to other aspects of development, it is much simpler to morph models from an androgynous base towards hyper masculine and hyper feminine than to take a prebuilt Masculine and try and morph to androgynous/feminine.
The wonderful thing was while V1 of the project only provided feminine, masculine and non-binary pronoun choices the flexibility by coding he system to assume a non binary default actually made it relatively simple to expand the support to a fully custom pronoun set as well as separating the visual representation of the character and their pronoun set, you can see this in V2.
This process really highlighted for me how cisnormative and heteronormative assumptions in the creation of a game has a massive flow on effect to the code and how the systems that are created. And beyond that how the assumptions of specific normativities in the code flow on to reducing the capacity of a system to actually support diverse options. Inversely it also highlighted that starting with the assumption of diversity actually helps the code and a systems capacity to be flexible.
I felt that the Eve Kosofsky Segwick’s definition of Queer quotes above really articulated this, idea of the possibilities of queerness. I also strongly believe that the active push to queer code is critical in the shifting of the technical and social landscape, in the words of Carolyn Marvin:
[1] Carolyn Marvin. 1988. When Old Technologies Were New: Thinking About Electric Communication in the Late Nineteenth Century (Kindle Edition ed.). Oxford University Press, Oxford.
Thanks so much for the generative prompt, and also to all for the links out to projects (Queering the Map, Oil Ancestors, micha cardenas's scalar, etc.) that have taken me down many pathways. I know little about the technical side of computer code, and nevertheless am interested in the prominence of binary code in computational technologies and even more curious to know about other kinds of examples -- that are not binary -- exist. I believe most computing systems still use binary code to store information/data. [Is this true?] If this is true, what are the implications? I think about the Rena Bivens article, which shows that even when Facebook offered over 50 gender categories for users, on the back end everyone was still coded back into the binary M/F so that FB could sell user info to advertisers.
As I get my bearings on what the relationships are between coding, algorithms, AI/ machine learning, and computational systems more generally, I spend most of my time wondering about data: how it is managed, stored, extracted, and commodified. I spend most of my time thinking about how our systems manipulate data toward capture and extraction.
How/ Can we queer data in ways that potentially subvert their capacities to capture and extract?
This^ question is largely the frame through which I read the opening prompt, and its invitation for us to think about queer failure and refusal.
Two sources/ terms I can offer in relation to these ideas are
1) J. Gaboury's discussion of NULL as "a marker that indicates a state of indeterminacy" (152), since it is a category (in a database system) that does not indicate zero or the lack of something, but rather that the value has not been determined. Though Gaboury warns against understanding NULL as inherently queer (153), they do encourage us to think about the possibilities of "falling out of legibility" of the systems that seek to categorize and surveil us, including racializing and anti-Black modes of surveillance.
2) Lauren Bridges's discussion of "fugitive data," which "slips through the roadblocks of digital captivity to evade its oppressor" and "finds company in the misidentified, misrecognized, and miscategorized" (7). Bridges's example is ImageNet roulette. I don't have ready examples to offer up for discussion, but what strikes me about the ImageNet roulette example is that the work it does is to show how the program/ming is embedded in racist, heteronormative, sexist, classist, and ableist logics. So it is disruptive, and beautifully so, but what are other modes of queering code that are generative without starting from a reactive place, a place of speaking back to systemic violence?
Bivens, Rena. “The Gender Binary Will Not Be Deprogrammed: Ten Years of Coding Gender on Facebook.” New Media and Society 19.6 (2017): 880-898.
Bridges, Lauren. Digital failure: Unbecoming the “good” data subject through entropic, fugitive, and queer data. Big Data & Society 8.1 (2021): 1-17.
Gaboury, Jacob. “Becoming NULL: Queer Relations in the Excluded Middle.” Women & performance 28, no. 2 (2018): 143–158.
Thank you for such a wonderful opening piece. I was reflecting on how code reflects representation not only as a language but also in the codification of the real world into virtual immersive spaces. Projects like “Queer Skins” that used photogrammetry and volumetric imaging capture to have representation through a collection of artifacts to tell a story. As we are constructing spaces bringing representation into code, how does that fundamentally change the immersive experiences in the stories we tell and how can we queer extended reality environments?
Loving the discussion here. I've posted another Code Critique that I think you will all enjoy, it's The Gay Science by Capricorn van Knapp (pseudonym), a beautiful homage to the tragedy of love, whose title speaks to both its queer love stories and to Nietsche's book Die fröhliche Wissenschaft. It is a sweet and sad piece that draws you into its code for a wistful love letter to the reader.
Such an amazing amount of material generated in this thread to delve into!
Two somewhat random thoughts:
I wonder if there is value in thinking of queer coding / coding queerly through a resistance to the utilitarian imperatives that seem to dominate coding/programming practices. Zach Blas' transCoder resists the imperative of executability. Ed's BASIC program/essay resists the idea that code is exclusively something that is for a computer application to 'read.' It relates to Donald E. Knuth's concept of 'literate programming' which is focused on pleasure: “One of life’s greatest pleasures can be the composition of a computer program that you know will be a pleasure for other people to read, and for yourself to read” (Literate Programming, 1992, p. ix). Knuth explicitly advocates for useless, non-productive programming: “[W]e shouldn’t shy away from ‘art for art’s sake’; we shouldn’t feel guilty about programs that are just for fun…. Some years ago the students at Stanford were excited about finding the shortest FORTRAN program that prints itself out, in the sense that the program’s output is identical to its own source text” (p.11). Is this a queer coding praxis? And there is much code out there in the world that is idiosyncratic, messy and baffling, while still being (mostly) functional: at the other end of the spectrum from literate programming, I wonder if the stock character for programming that works but is not elegant or beautiful or clear (Mort) could be considered a model for queer coding.
Where do markup languages fit into this discussion? Like programming languages, these too can be used in 'improper' ways. And markup languages like the Text Encoding Initiative's XML-based schema have a subversive potential to entangle a primary text with another and perhaps resistant text: a rewriting or writing against or overwriting. I imagine there could be some interesting queer deployments of markup.
Yes! @JasonBoyd, thanks for reintroducing the pleasure of queering reading code ("students at Stanford were excited about finding the shortest FORTRAN program that prints itself out"). Among other things though not explicitly, the book 10 PRINT is about a short program that offers many exquisite code reading pleasures. Reading about Knuth as well as about "failure" earlier in this rich conversation reminds me of Segwick's contemporary Barbara Johnson's wonderful essay "Nothing Fails Like Success," arguably a precursor of queering ways with language and code (1980, reprinted in The Barbara Johnson Reader, 2014).
Thank you @edmondchang and @jarahmo for this fantastic prompt. I've been coming back to it over the week, reading it differently each time. I love the invitation to reflect on our embodied experiences. Today in the middle of writing this, I was feeling a bit blocked so I decided to try moving my work to a cafe, only to discover that my bicycle had become stuck to a post because the lock had become hopelessly jammed. This has happened to me before. Years ago I ended up having to abandon a beloved bicycle in a moment of precarity because the lock became jammed and I didn't have the resources to get the supplies I needed to fix it. And now here I was, earlier today, facing a similar sort of situation, with a stuck bicycle lock. In that moment, fruitlessly wiggling the key in the gummed up lock, I found myself thinking about queer failure, about how sometimes, as Dr. Chang points out, we don't have the privilege to perform failure with intentionality. Sometimes queer failure is simply experienced as things breaking, as setting out to do something, but then stuff broke, things got derailed, and now all that's left is the work of fixing things (if you even have the means to do so).
In an NPR interview last year, Christian Cooper, the comic book artist credited with creating Marvel's first openly lesbian character, and who also helped create Marvel's first openly gay male character, when discussing his experience of being gay while growing up in the 1970s, he used a number of metaphors about locks. He talked about keeping his true self "locked in," about having to keep parts of himself "under lock and key."
A kind of enclosure of the self comes up in a slightly different way in Jacob Gaboury's "On Uncomputable Numbers: The Origins of a Queer Computing," in which Gaboury discusses the dangers of having our interactions mediated by technologies that "function in a state of willful indifference" to individual complexity, in which we are compelled to "define ourselves within a given set of legible identity parameters—race, gender, sexuality, age, taste, preference, etc" but in which any further individual distinctions become lost. "In effect," Gaboury writes, "the self is black-boxed, reducing it to limited set of legible input and output signals."
@edmondchang's interactive work, "Why Are the Digital Humanities so Straight?", responds to this type of peril both through an interactive demonstration, and theoretical offerings. In the demonstration, Dr. Chang gives his player-readers a sense of being boxed in as they navigate a set of legible, prefabricated actions available to them, first roleplaying as Alan Turning, then Ada Lovelace, then as video game character Purna Jackson. Looking at the snippet of code for this game @edmondchang has shared in the prompt, this specific type of danger is underscored: the more our lives become mediated by reductive forms of automation, the more our range of actions will become constrained by the limits of others' imaginations.
Dr. Chang offers the theoretical concept of technonormativity as a way to name and problematize this terrifying form of enclosure:
In the face of such peril, I appreciate Dr. Moesch's call to action:
In offering a contribution to this call to queer the code, I can perhaps offer a gentle pushback against a couple myths about binary computers, in the hope of perhaps offering a bit of comfort to those who have encountered distressing lies about the "nature" of computers. Forgive me for descending into a bit of Baradian agential realism here... Where to start?
In early 2020, I had the good fortune to interview Lee Felsenstein, the lead designer of an early iteration of the Internet, and as the interview wound down, Lee and I both divulged into a bit of a playful exploration of the way ones and zeros actually work (when the circuits are even laid out in a binary fashion lol). Lacking any technology to demonstrate the things we were talking about (we had meet at a cafe in Berkeley, and neither of us had brought any tools or equipment), Lee and I began miming things out. Making hand gestures. Lee kept making this excellent gesture like a cat pawing at a piece of string as he described what's actually going on in a binary computer when you get down to it:
“You have a whole voltage level, that’s the zero, and then 1 comes and [cat paw gesture] baps it down. Not all the way down, though. Just a bit. Like this [cat paw gesture].”
All this is to say, there's no separation between the one and the zero, they are hopelessly fused together. The one is just a little less of the very same current that, when whole, is zero. The idea that zero is is an "off" is a myth. One isn't an "off" either. Neither of them is an "off"! If you're talking about an actual, modern fully digital electric circuit there is no "off," and also, zero is always bigger, by necessity, than one. Zero is a wholeness. One is slightly less. There is no physical distinction between zero and one, except that zero shrinks a little in order to become one.
So. Computers, even when laid out in binary, have a deep queerness to them, and zero is always larger than one, as far as the voltage levels are concerned. But also, computers don't have to be binary. Just as Lee Felsenstein was fired from NASA for his family’s communism, nonbinary computers—like the 1958 Soviet ternary (base-3) computer, the Setun, and the Canadian ternary computer, the QTC-1—have been been pushed out of computing discourse.
This strange, pervasive social practice of obfuscating the true nature of computers, of keeping the reality about their essence under lock and key, might be compared to the types of things Karen Barad encountered in her investigations in theoretical physics. While looking at Quantum Field Theory, for example, she found found that a variety of "queer" electron behaviors were being systematically erased in the technical language by (white, straight, male, cis) physicists (Barad 2018).
Anyway, computers. When computers are event doing the whole "binary thing" (which absolutely isn't required, essential, or even the most efficient way to do things), the actual circuitry, the articulators below their bits, are rarely described properly. In fact, so far, I've only rarely been able to get proper description, even from people who've built computers from scratch. Lee Felsenstein is one of a select few who isn't scared of naming the things the sees...
What types of fear and discomfort must engineers and physicists have been experiences in the mid 20th century that they invented myths to obscure the realities of the machines they were working on and the particles they were observing?
TW: The next paragraph contains mentions of anti-queer harm in the workplace. Feel free to skip it.
.
.
.
.
.
.
.
.
.
Perhaps we can look towards the work of Historian David K. Johnson to better understand this bizarre era, espeically his book, The Lavender Scare: The Cold War Persecution of Gays and Lesbians in the Federal Government (University of Chicago Press, 2004). In this work, he traces a moral panic that spread from the State Department and ultimately infected workplaces both within and beyond U.S. government jobs, through which paranoid ideas about queer people--including the bizarre idea that all all sexual deviants people are somehow part of an international spy ring--were used as justification to not only fire suspected queer people from their jobs, but also interrogate, detain, and even "out" folks to their families.
/TW
These horrific workplace practices were protected by the force of law with the signing of Executive Order 10450 in 1953, which remained on the books and continued to be used until 2017. During this 64 year era, STEM workplaces were especially effected, due in part to the way research funding works in those fields, and I think it kinda messed with people's heads in those fields, or at least with people's ability to speak clearly about what they were perceiving.
In trying to understand this strange era with my students, I've found the 2019 short story, “Our Aim Is Not to Die” by A. Merc Rustad to be quite helpful. The story centers the terrifying experience of a neurodivergent, queer person as they navigate state-mandated identity standards amidst a regime of escalating repressive technonormative apparatuses. The story also plants some hopeful seeds, but it really takes you into the headspace of what that regime must have been like.
Anyway, I've fixed my bike lock. At some point while writing this, I got up and walked through the strange enclosures unfolding in Berkeley right now and managed to make it to the hardware store, where I found some plant-based machine lubricant (yay!). My bike will live another day!
The more I think about it, the more the distinction seems to blur between the movement to make queer and the Right-to-Repair Movement. Anti-queer regimes contribute not only to a blackboxing of our selves, but they also obfuscate our machines, making harder to repair, improve, and know them intimately.
Queering the code in this often goes mean building the machine yourself, from scratch. I imagine the Digital Humanities might have a lot to gain if we were to make computers from scratch together more (and I'm talking real scratch, not that pre-fab junk, but building the thing bespoke, from the physics on up). I think the critical code studies discourse especially would benefit.
When you make a machine from scratch, you can really get a sense of how there's nothing essential about the binary. We build down from our assumptions about the code, rather than the other way around. The machine is an extension of the code.
Thank you all for this discussion. I wanted to bring in another question, that of intersectionality. As we have discussed in previous working groups, when we isolate a critical approach, we anatomize our study and also risk separating groups rather than building coalitions. Sarah Elwood makes the argument:
How does this intersectional approach open up the discussion? I should note that
@edmondchang you has thought a lot about the politics of the glitch!
Elwood, S. (2021). Digital geographies, feminist relationality, Black and queer code studies: Thriving otherwise. Progress in Human Geography, 45(2), 209-228. https://doi.org/10.1177/0309132519899733