Alex Thinius

Machine learning offers new possibilities for developing more precise diagnostics and treatments, but the increasing use of sex stratification in precision medicine algorithms raises concerns. Using Alzheimer's disease (AD) research as an example in which machine learning approaches are applied to a heterogenous, socially patterned disease, this paper examines how the move toward sex-specific “pink” and “blue” algorithms reinforces biological sex essentialist assumptions and their attendant harms. We analyze three examples of sex-stratified algorithmic approaches in AD research, and identify three interacting processes-effacing contested knowledge, obscuring social factors, and ossifying binary sex categories-that can occur when binary sex variables are incorporated into predictive models. These case studies demonstrate that even in models intended to be causally agnostic, sex categories are likely to be interpreted as decontextualized, self-evident health determinants in a manner that can imply causality of biological sex. We call for establishing ethical norms and empirical standards for including gender/sex variables in precision medicine algorithms to avoid perpetuating crude ontologies of sex and gender that undermine both scientific validity and health justice.

To explain observed disparities in health outcomes between men and women, sex essentialist approaches assign causal primacy to sex-related biology. In this essay, we present three case studies to illustrate how sex essentialism can distort human biomedical research and distill three maxims for countering this distortion: (1) engage in responsible citation practices; (2) generate and weigh alternative hypotheses for apparent observations of sex differences; (3) take care in constructing the appropriate denominator when making sex comparisons. We offer these maxims as broadly applicable standards of evidence to guide biomedical research that includes analysis of potential sex differences, as well as to support Institutional Review Boards (IRBs), funders, publishers, and peer reviewers in evaluating sex difference findings. If widely applied, these maxims would substantially improve the rigor, precision, and utility of the knowledge base of sex and gender science.

How many sexes are there? How many genders? Answering these questions is not just challenging: It is about as silly as trying to determine how many kinds of stories or how many species there are: Where? And when? And, most importantly, in what sense? Instead, I suggest taking a closer look at the double historicity of phenomena and concepts of sex/gender: Research concepts of sex and gender are context-dependent, plural, and historically dynamic; and so are the phenomena that these concepts refer to. For reasons of both scientific accuracy and justice, understanding the double historicity of sex/gender transforms the “How many are there” questions. Instead of absolutist single-answer approaches of the “There are x number of sexes/genders” style, it demands a research ethos that involves contextual awareness regarding concepts and phenomena; reflective distance from one’s own analytical research concepts and aims; and epistemic humility.

Alex Thinius, in “Sex-Gender in Life-Science Research: Conceptual Renegotiations and an Enactivist Vision,” discusses how researchers are increasingly acknowledging the urgency that the concept of “sex” be redefined. In contrast to concepts of sex-gender differences as stable and dichotomous, in current research on sex-gender, there is a growing consensus that sex is far more nuanced, variable, and interacting with gender in complex ways. The article aims to open up a research horizon for pluralist and dynamic concepts of sex by looking at a family of theories that mediate between gender theories and the life sciences, potentially integrating complex systems theory and critical phenomenology: enactivism. While endorsing the strength of this constructive integration, the author stresses that there is still great unexplored potential for reconceptualizing the sexes beyond grounding it on a sex/gender or male/female binary.

Sex is often thought of as a straightforwardly binary categorical variable. Yet there is considerable variation in would-be sex traits; from genitals and hormones to morphology, neurology and behaviour, there is rarely if ever a categorical binary. We introduce a strategy that researchers use to deal with this variation: Individualising Variation (IV). IV involves treating non-binary and gradual variation as idiosyncratic, as individual differences rather than sex-based differences. Using the contrasting cases of sex identification in field ornithology and the debate about sex differences in neuroscience, we illustrate IV and investigate its epistemic and conceptual consequences. We argue that IV stabilises the ontological picture of sex as categorical and binary. While IV can be an epistemically benign research strategy in some cases, we argue that it can also be epistemically detrimental. This is because of its ability to mask evidence that would otherwise challenge related assumptions about the phenomenon of interest, such as what sexes are and what they look like. We also identify an alternative strategy, De- individualising Variation, which works against IV and helps life scientists recognise variation beyond categorical binaries.

What does it mean to be of a particular gender? I answer this question with an account of genders as dynamic categories, exploring the analogy between what genders are (e.g., men or women) and what genres are (e.g., Novels, Ballads, or Hip-Hop). For instance, due to its relation to other and earlier pieces, we recognize, e.g., a particular song as Hip-Hop. However, the piece will also develop that genre further. Likewise, e.g., the category of men emerges, persists and transforms through a specific sort of response of individuals to earlier supposed men, which emerges in social interactions. Drawing on critical and (trans-)feminist theory, phenomenologist, enactivist, and systems theoretical approaches, I show that gender categories themselves develop in a dynamic between three elements: (1) at any given time, there is an enactment class of individuals ambiguously belonging to the category in question; (2) these individuals are, in an embodied and intersubjective way, enacted as practical reinterpretations of earlier members of that category; (3) this unfolds in a matrix of hermeneutic and material relations, which loop with both the class and the enaction. Thereby, that an individual is gendered, emerges between two levels of enaction, that of bodily people and that of the dynamic between them. As what an individual is gendered in enaction, however, is constituted by relations within and beyond that situation. This responsive realization gives rise to feedbacking histories of acts, people, relations, and enactment classes, thus explaining how gender can be both solid and open to change.

This paper discusses Sosa’s via media between existential relativism and absolutism. We discuss three implications of Sosa’s account which require some further clarification. First, we distinguish three alternative readings of Sosa’s account – the indexicalist, the homonymist and the (proper) relativist reading – and argue that they differ with respect to two crucial points: (a) they lead to different analyses of the lack of disagreement in existential discourse, and (b) they differ with respect to the question of whether conceptual schemes pick out different senses of “exist” or whether they pick out different entities to exist. Second, we ask Sosa to answer on four problematic implications of his final position: (a) Sosa appears to change the topic from ontology to semantics without solving the ontological issue. (b) It is puzzling why Sosa finally accepts the initially implausible explosion of reality. (c) Sosa is forced to accept that disputants really disagree in existential disputes (although faultlessly). (d) We offer an even simpler alternative option to reconcile the realist and the relativist intuitions by clarifying what is meant by “conceptual relativism”, without arguing for existential relativity at all. Third, we argue that Sosa’s argumentative reliance on an appropriate development of conceptual schemes drives him not only to a position of pure conceptual absolutism, but even to a more traditional form of ontological absolutism according to which nature itself manages to cut the cookies. In contrast to his apparent intention, this discharges Sosa’s via media from any relativist intuition.