Moving Towards Epistemic Autonomy: A Paradigm Shift for Centering Participant Knowledge

Leah Hope Ajmani, Talia Bhatt, Michael Ann Devito · 2025

Read Moving Towards Epistemic Autonomy: A Paradigm Shift for Centering Participant Knowledge by Leah Hope Ajmani, Talia Bhatt, Michael Ann Devito (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-14648.txt.

Introduction

Leah Ajmani and co-authors argue that research with pushed-aside groups must guard each person's control of knowledge about the self. The work dates to 2025. They call this guard epistemic autonomy. Tech systems keep pushing such groups to the edge, the authors say, despite rich past work on justice.

The paper builds the idea from a trans care rule about self rule. Six short tales from two trans women show how harm looks when the guard is breached. The authors then shape the idea into a full research stance. They give two adapted methods that flow from it, and urge their field to lead the shift.

Catalog record

CREATORLeah Hope Ajmani, Talia Bhatt, Michael Ann Devito
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE85 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-14648

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Moving Towards Epistemic Autonomy: A Paradigm Shift for Centering Participant Knowledge Leah Hope Ajmani, Talia Bhatt, Michael Ann Devito Abstract . Justice, epistemology, and marginalization are rich areas of study in HCI. And yet, we repeatedly find platforms and algorithms that push communities further into the margins. In this paper, we propose epistemic autonomy–—one’s ability to govern knowledge about themselves—as a necessary HCI paradigm for working with marginalized communities. We establish epistemic autonomy by applying the transfeminine principle of autonomy to the problem of epistemic injustice. To articulate the harm of violating one’s epistemic autonomy, we present six stories from two trans women: (1) a transfem online administrator and (2) a transfem researcher. We then synthesize our definition of epistemic autonomy in research into a research paradigm. Finally, we present two variants of common HCI methods, autoethnography and asynchronous remote communities, that stem from these beliefs. We discuss how CHI is uniquely situated to champion this paradigm and, thereby, the epistemic autonomy of our research participants. Conference: CHI Conference on Human Factors in Computing Systems; April 26-May 1, 2025; Yokohama, Japan CHI Conference on Human Factors in Computing Systems (CHI ’25), April 26-May 1, 2025, Yokohama, Japan DOI: 10.1145/3706598.3714252 ISBN: 979-8-4007-1394-1/25/04 CCS: Human-centered computing HCI theory, concepts and models 1. Introduction As Human-Computer Interaction (HCI) researchers have increasingly testified about their experiences of being transgender, Black, Indigenous, and disabled, there is a clear theme of fighting against pervasive erasure and disempowerment within computing systems (e.g., ( Haimson et al., 2020 ; Erete et al., 2021 ; Lazem et al., 2022 ; Ymous et al., 2020 ) ). In many cases, both researchers and the users of our systems have expressed a sense of powerlessness. The fundamental changes these communities call for seem more like a distant future rather than a close reality. This feeling of powerlessness serves a structural function; it is crucial to the status quo of patriarchy, racism, and ableism that marginalized communities internalize this message of disempowerment. Therefore, it is essential that HCI, as Computer Science’s most human-facing subdiscipline, foster precisely the opposite. Creating power here is the process of creating knowledge. There is power in knowing oneself and being known by others on one’s own terms ( Fricker, 2007 ; Jackson, 2018 ; Hogeland, 2016 ) . However, the restrictive influences of whiteness, patriarchy, and colonialism are deeply embedded in our current knowledge processes. These theories of knowledge tend to favor a search for one objective reality, provable only through means those already in power deem acceptable ( Cunningham et al., 2022 ; Kannabiran, 2023 ; Rorty, 1985 ) . This dynamic results in a world where it is nearly impossible for those in the margins to know and assert themselves on their terms. Furthermore, we, as researchers, inadvertently spread a message that those in the margins are worthy of being observed but not worthy of being observers. In doing so, we further a prejudiced ideology: epistemic injustice . Epistemic injustice is the stripping of one’s rights to be a knower, such as the right to be an authority on one’s experience ( Fricker, 2007 ) . Sociotechnical systems are ripe venues for epistemic injustice, as they often have their own rules, norms, and politics ( Reddy and Chandrasekharan, 2023 ) . Epistemic injustice rapidly compounds for those with intersecting marginalized identities. One key example of this phenomenon is how sociotechnical systems act on transgender women. Trans women are subject to egregious forms of epistemic injustice as they are repeatedly questioned as authorities of their own gender identity and health information. While online spaces can provide a venue for sharing information and building community, they have also formed algorithmic traps that actively endanger trans women seeking to use said systems ( DeVito, 2022 ) . More broadly, trans health and cultural information often get unduly flagged as misinformation ( Haimson et al., 2021 ) . These dynamics are concerning given that a systemic lack of online health knowledge contributes to disproportionately high suicide rates in trans populations ( Moody et al., 2015 ; McNeil et al., 2017 ; Tebbe and Moradi, 2016 ) . At its worst, epistemic injustice is a violent act—propagating the gatekeeping and discrediting of life-altering knowledge for trans women. As the amount of information, curation, and subsequent moderation on the internet grows, so does the embeddedness of existing power structures around knowledge, with dire practical consequences. Given that HCI research deeply informs the design of these platforms, we risk propagating epistemic injustice through our practices. At the same time, HCI is uniquely positioned to fight this same injustice, as work on how research interacts with justice, epistemology, and identity has a rich history in our field. In the past decade, we have seen a new wave of HCI research intended to interrogate power structures, evaluate our research practices, and—ultimately—build more just technologies ( Bardzell and Bardzell, 2011 ; Søndergaard and Campo Woytuk, 2023 ; Menking and Rosenberg, 2021 ; O’Neill et al., 2024 ) . And yet, time and time again, we find platforms and algorithms that push communities further into the margins ( Walker and DeVito, 2020 ; DeVito et al., 2018 ) . This phenomenon is unsurprising, given how transphobia, misogyny, and patriarchy are deeply woven into our social fabric. However, recent work suggests that sociotechnical systems can serve as sites of resistance, a way to rip the stitches and weave a new fabric ( Ajmani et al., 2024a ) . Furthermore, sociotechnical research itself can further this mission. As research is considered “canon knowledge,” we all have the opportunity to push against epistemic injustice through our beliefs and practices. To that end, we ask: what does an orientation of HCI research towards fighting epistemic injustice look like? In this paper, we propose a re-orientation of HCI towards prioritizing the epistemic autonomy of our research subjects and their communities. We define epistemic autonomy as one’s ability to govern how they know themselves and how they are known to the world. We specifically build on autonomy as articulated in transfeminism ( Koyama, 2020 ) —drawing heavily on the experiences from the two transfem authors of this paper. After establishing what epistemic autonomy means in research settings, we demonstrate the serious consequences of violating one’s autonomy. Using storytelling methods ( Ogbonnaya-Ogburu et al., 2020 ) , we include personal stories from (1) an online community context and (2) an academic context within HCI. Inspired by previous work on making CHI more inclusive and progressive ( Bardzell and Bardzell, 2011 ; Ogbonnaya-Ogburu et al., 2020 ; Frauenberger, 2020 ) , we frame epistemic autonomy as a research paradigm. While we name an epistemic problem here (i.e., epistemic injustice), we believe the solution also requires a fundamental inquiry into our ontological and methodological beliefs. Finally, we apply our paradigm to two current methods in HCI, autoethnography and asynchronous remote communities (ARCs). We outline variants of these methods, scaffolded autoethnography and member-checked ARCs, to demonstrate how researchers can better prioritize epistemic autonomy in their current research practices. In sum, this paper contributes a theory of epistemic autonomy , a critique of current HCI research beliefs, and two methods for fighting epistemic injustice within our research.

Read the complete text