I am an anthropologist of science and technology who conducts ethnographic research on the language and communication sciences, biomedicine (especially mental health care), and computing in the contemporary United States. My research concerns the intersection of language, technoscience, and power, with a focus on the cultural, ethical, and sociopolitical dimensions of listening.
I received my PhD in History, Anthropology, Science, Technology and Society (HASTS) at MIT, my MA in Anthropology at Brandeis University, and my BA in Writing, Literature, and Publishing at Emerson College. I first became interested in studying technoscience and language anthropologically after reading Michel Degraff's article on the fallacy of Creole Exceptionalism in an undergraduate sociolinguistics course.
My research and teaching integrate linguistic and medical anthropology with feminist science and technology studies (STS) and sound studies.
Speech, Signal, Symptom: Machine Listening and the Re-coding of Psychiatric Screening
My book project traces how automation impacts frameworks of language, listening, practices of care, and labor in the contemporary U.S., as the expertise and ethos of Silicon Valley are increasingly called upon to solve psychiatry’s epistemic and bureaucratic ills.
It follows teams of psychiatric and engineering professionals collaborating to build AI-enabled technologies to screen for mental illness based on how a person sounds, rather than what they say. Dominant discourses surrounding these "machine listening" technologies depict them as more objective and efficient than conventional screening, due to their ability to pinpoint subtle, non-semantic signs of mental illness that otherwise exceed the limits of the human sensorium. In contrast, my analysis demonstrates how researchers enact and construct these sonic signatures, alongside notions of gender, race, and ability. I underscore the affective, feminized, and racialized labor that plays a crucial yet often hidden role in the research and development pipeline.
Through a behind-the-screen look at the day-to-day work and hierarchies of value involved in rendering the sounds of mental illness audible, the book articulates the stakes of framing America’s mental health crisis as a technoscientific—and semiotic—problem, contributing to ongoing debates on AI, inequity, and the future of work.
Suspicious and Synthetic Speech
Since the spring of 2020, I have been working on two new projects. One project traces the history of voice stress analysis in criminal-legal applications and the popular imagination from 1956 to the present in the United States. Additionally, I have begun exploratory research on the auditory equivalent of "deep fakes"—also known as vocal deep fakes—and their relationship to other forms of synthetic, imitative speech and disability studies-centered histories of hacking.
I am committed to applied research and organizing work, especially when it involves collaborations that cut across STEM and the humanities and social sciences. I was a participant of a 2019 Alex Trebek Forum for Dialogue on Artificial Intelligence and Mental Health Care at the University of Ottawa and a panelist/moderator on the Ethics of Technology Panel at Tufts University’s Women in Technology Conference. I also co-organized a teach-in at MIT, “#AICantFixThis: MIT, Imperialism, and the Future of AI.”
As a member of the Coalition for Critical Technology, along with over 20 other academic and technology workers, I recently co-wrote and facilitated an open letter condemning the use of criminal legal data to build algorithmic systems to “predict criminality.” The letter underscores the long, multidisciplinary legacy of research debunking predictive policing and related carceral technologies. If you would like to use the letter for teaching purposes, you may download an accessible PDF here. Read more about the letter in the MIT Tech Review, One Zero, and The Verge.
Semel, Beth. n.d. “Listening Like a Computer: Attentional tensions and mechanized care in psychiatric digital phenotyping.” STHV: Science, Technology & Human Values, Special Issue, "Shifting Attention” (editors Rebecca Jablonsky, Tero Karppi, and Nick Seaver). [under review]
Jones, Graham M., Beth Semel, and Audrey Le. 2015. “‘There’s no rules. It’s hackathon.’: Negotiating Commitment in a Context of Volatile Sociality.” Journal of Linguistic Anthropology 25(3):322–345.
A more complete list of speaking engagements and my CV are available upon request. Please feel free to reach out regarding speaking and guest lecturing inquiries.
Upcoming (2021). “Ear from Nowhere: The Listening Subject of Automated Psychiatric Screening.” American Association for Applied Linguistics Annual Meeting, Houston, TX [under consideration] Past 2019. “Listening like a computer: In/attention and the automation of psychiatric assessment.” American Anthropological Association Annual Meeting, Vancouver, BC (Nov. 20) 2019. “Listening/Not Listening: Educations of Inattention in the Automation of Psychiatric Assessment.” Society for Social Studies of Science Annual Meeting, New Orleans, LA. (Sept. 5)2018. “‘I’m not a therapist, but…’ Interactional illusions and the virtually human listener.” American Anthropological Association Annual Meeting, San José, CA. (Nov. 16)2017. “Do androids dream of electric speech? Professional listening and the automation of psychiatric assessment.” American Anthropological Association Annual Meeting, Washington, DC. (Dec. 3)2017. “Sound of mind: the automation of psychiatric listening and assessment.” Society for the Social Studies of Science Annual Meeting, Boston, MA. (Aug. 31)2015. “Good data/good care: speculative psychiatrics after DSM.” American Anthropological Association Annual meeting. Denver, CO. (Nov. 21)2015. “Picturing brains, predicting patients: neuro-sources of evidence in evidence-based treatment research.” Society for Psychological Anthropology Biannual Meeting, Boston, MA. (Apr. 10)2014. “From numerical minds to neurological brains: translating efficacy and producing new evidence in evidence-based psychotherapies.” American Anthropological Association Annual Meeting, Washington, DC. (Dec. 5) 2014. “Tracking the self, installing expertise: Cognitive-Behavioral Therapy and the auto-regulating subject.” Society for Social Studies of Science Annual Meeting, in conjunction with Sociedad Latinoamericana de Estudios Sociales de la Ciencia y la Tecnología, Buenos Aires, Argentina. (Aug. 21)
Upcoming 2021. Title TBD. Michigan Interactive and Social Computing Group, University of Michigan. (April 1)2020. "Digitizing psychiatric assessment and the social implications of the DSM (Diagnostic and Statistical Manual of Mental Disorders)." Guest Lecture, Introduction to Medical Humanities (Instructor Prof. Lan Li) Rice University. (Oct. 20) Past 2019. “Vocal Biomarkers and the Engineering Approach to Mental Illness.” STS @ Tufts, Tufts University. (Dec. 6)2019. “Listening Like a Computer: Computational Psychiatry and the Re-coding of Psychiatric Screening.” STS Circle, Harvard University. (Dec. 2) 2019. “Linguistic Labor and Technologies of Care: Automating Assessment in U.S. Psychiatry.” Department of Anthropology, University of Massachusetts Amherst (Feb. 11)2018. “Do androids dream of electric speech? Listening practices in automated psychiatric assessment.” Scholar Colloquia, School for Advanced Research. (Nov. 7) (video)2013. “Cultural Competence Training and DSM-IV.” Guest Lecture, Psychological Anthropology (Instructor Prof. Janet McIntosh) Brandeis University 2013. “Culture-Bound Syndromes and DSM-IV.” Guest Lecture, Sociology of Mental Health and Illness (Instructor Prof. Peter Conrad) Brandeis University 2013. “Blood Metaphors and Race.” Guest Lecture, Dirt, Disgust, & Contagion (Instructor Prof. Anita Hannig) Brandeis University