Research
My research develops a new, scientifically informed paradigm for ethics designed to guide AI systems and other large-scale socio-technical infrastructures. I work at the intersection of philosophy, physics, evolutionary ecology, and social justice to understand how moral life actually unfolds “on the ground” in complex, unequal worlds—and how we can redesign our systems to better support human well-being and dignity.
A central theme of my work is that ethics is not a set of abstract rules floating in a vacuum, but a network of dynamic Good, Right, and Virtue fields—patterns of value, obligation, and character—that evolve by interacting with each other and their environment. Drawing on analogies with field theories in physics, I model rightness, goodness, and virtue as coupled ethical fields that interact with a multidimensional ethical ecosystem or “ethosystem”: the psycho-bio-social, economic, political, technological, and ecological medium in which our lives are embedded. This ethosystemic perspective is meant to do for ethics something like what the move from Newtonian gravity to general relativity did for physics: shift our focus to how forces and structures arise from deep couplings between phenomena and their underlying dynamical medium.
A core thread running through my work is the claim that the natural and ethical worlds are deeply analogous in structure (as illustrated by the above table). The same features that make the natural world complex—fields, feedback loops, multi-scale dynamics, and multi-level evolutionary selection—also characterize our moral lives, institutions, and technological infrastructures. When we take this analogy seriously, ethics stops looking like an abstract realm of rules and intuitions, and starts to look like a rich, patterned landscape that can be mapped, modeled, and better understood and navigated.
This analogical perspective gives texture to my scientifically informed paradigm for ethics: we can adapt the mathematical and conceptual tools that have helped us make sense of the natural world—field theories, symmetry principles, multi-level selection, systems and network models—to illuminate the ethical world as well. By treating ethical life as a complex, multi-scale and evolving system, we gain sharper ways of diagnosing where and how things are going wrong, and of designing interventions that shift the overall ethical field in more just, sustainable, and life-affirming directions.
Building on this field- and system-based approach, I use tools from the mathematical study of symmetry—which physicists employ to understand objectivity and invariance—to develop a symmetry-based account of ethical objectivity. In my work, ethical claims are objective when they are stable under certain transformations: for example, under permutation of persons (treating similar individuals with equal concern), under intersectional shifts of standpoint (re-centering marginalized perspectives), and under local “gauge-like” changes in how we describe and parse situations. These symmetry principles provide a rigorous way to think about fairness, bias, and robustness in ethical and AI-mediated decision-making. Indeed, I argue that consistency with these symmetry principles can be fruitfully thought of as playing an analogous role to the one that consistency with the principles of quantum gravity plays in the string-theoretic swampland program (with diagrams illustrating below):
In the string-theoretic swampland program, physicists think of the space of possible physical field theories. They then argue that within this space, consistency with quantum gravity demarcates a string "landscape" of field theories that consistently couple to quantum gravity dynamics from a "swampland" of effective field theories that, as effective, are empirically adequate at more easily accessible lower energy scales, but are ruled out at higher energy scales due to their lack of consistent coupling to quantum gravity.
In an analogous scientific ethical swampland program, we can think of the space of possible ethical field theories that describe possible policies and practices. We can then analogously argue that within this space, consistency with objective (categorical) morality, i.e., morality that is invariant between all transformations in ethical perspective (and so holds for all), demarcates a moral "landscape" of ethical policies/practices (and ethical field theories describing them) that consistently couple with the demands (and ethical dynamics) of objective morality from a "swampland" of effective ethical policies/practices that, as effective, are empirically adequate at more easily accessible lower "ethical energy" scale, but are ruled out at higher "ethical energy" scales (where ethical energy is characterized formally as symmetric harmonious synergy). In other words, the policies/practices that do not consistently couple to the requirements of objective morality cease to be empirically adequate at a certain point when trying to produce a "higher ethical energy" (i.e., more symmetrically harmonious and synergistic) dynamic system.
These diagrams represent oscillating interacting patterns of the Good, Right, and Virtue fields (i.e., oscillating patterns of value, obligation, and character) interacting dynamically over time in ways analogous to how Electric and Magnetic Fields interact.
I apply my scientific ethical framework not just to traditional and applied ethics questions, but also to questions in user well-being, AI safety, and responsible AI. Thinking of sociotechnical infrastructures like AI systems—especially large models and platforms—as parts of wider complex, dynamic, multi-scale, and evolving ethosystems allows us to analyze how incentive structures, recommendation algorithms, and automated decisions shape long-term patterns of cooperation, harm, and exclusion. My goal is to help design and evaluate AI systems not only for short-term performance, but for their long-term systemic effects on communities: making sure they do not amplify domination and exploitation and that they instead foster mutual respect, empowerment, and sustainable forms of flourishing.
Alongside this theoretical work, I have written a bioethics textbook that centers the perspectives of marginalized groups and brings philosophy into dialogue with the natural and social sciences, antiracist scholarship, disability studies, and intersectional feminism. This margin-centering, justice-oriented lens informs my approach to AI: I am especially interested in how we can build AI and governance structures that remain accountable to those most vulnerable to technological and institutional harms, and that use our most powerful tools to expand, rather than constrict, the space of genuinely livable futures.
My research has also historically focused on Immanuel Kant’s philosophy, interpreting it as an organic, systematic whole that harmoniously combines theoretical and practical philosophy (in part through the ingenious use of analogies). I offer a particular interpretation of the foundations of Kant’s critical philosophy and of the way Kant’s natural and moral philosophies build on these foundations to provide a “twofold metaphysics” of nature and morals. I argue that with this twofold metaphysics, Kant paints a majestic picture of the moral and political relations between humans in a moral world as importantly analogous to the geometric and physical relations between bodies in the natural world. You can read more about my approach to Kant's philosophy in my published work below and my Visualizing Kant project.
A visual model of the balance of physical attraction (gravitation) and repulsion (expansion) that makes the physical world of bodies possible.
A visual model of the balance of moral attraction (love) and repulsion (respect) that makes the moral world of humans possible.
Here, the circles represent constricting patterns in the magnetic field, while the lines represent electric currents in the electric field weaving into a "filamentary" current or plasma cable that partially characterizes the structure of Birkeland currents.
Here, the circles represent patterns in the Good field (Circulating Mutual Benefit/Harm) while the solid lines represent patterns in the Right field (Reflecting Right/Wrong) and in the Virtue field (Virtuous Spirals of Virtue and Vicious Spirals of Vice). The dashed lines represent the temporal trajectory of entities bearing the relevant ethical properties that enter into harmonious/dissonant mutually reinforcing dynamics (e.g., agents or parts of one's life). Note that in these ethical diagrams, the axis only represents (the passage of) time, whereas in the Birkeland current diagram above, the axis additionally represents movement in physical space. The ethical dynamical patterns are, accordingly, more abstract.
Work in Progress
"Interacting Ethical Fields: The Coupled Dynamics of The Good, The Right, and Moral Worth"
ABSTRACT: Following John Rawls, we can characterize moral theory as the study of how the basic notions of the good, the right, and moral worth may be arranged to form different moral structures. Some philosophers try to explain which of these notions is more fundamental, trying to derive/reduce some from/to others. In this paper, I take a different strategy in connecting these different moral notions, one that relates them dynamically. I argue that we can understand the way these ethical notions relate as analogous to the way electricity and magnetism relate to each other. Thus, I advance a view on which we can understand axiological phenomena (encompassed by the good), deontic phenomena (encompassed by the right), and aretaic phenomena (encompassed by moral worth) as manifestations of axiological, deontic, and aretaic fields that dynamically couple to each other (analogously to how electric and magnetic fields do). An upshot of this view is that it illuminates how the coupling of these fields can lead to non-additive ethical patterns, for example, mutually reinforcing dynamics of mutual benefit, doing right by one another, and virtuous spirals, or mutually reinforcing dynamics of mutual harm, doing wrong by one another, and vicious spirals. As an interesting result, these dynamic ethical patterns between these fields are in certain ways structurally analogous to the electromagnetic field patterns found in certain plasma structures called "Birkeland currents" as can be seen in the adjacent images (the physical diagram is taken from the linked article).
"Bringing Ethics Down to Earth"
ABSTRACT: In this paper, I articulate a transdisciplinary framework for thinking about the mediums in which ethical phenomena play out. In doing so, I give a nonreductive metaphysics of ethical phenomena that locates them not just within a scientifically informed vision of the natural world but also within a humane margin-centering vision of an inherently unjust and deeply asymmetrical social world. In doing so, I offer naturalistic, margin-centering metaphysics that centers the perspectives and experiences of a variety of marginalized persons) By articulating this framework, I sketch a picture of some of the different mediums in which ethical phenomena play out and interact. I argue that this picture vividly highlights the need to cultivate an empathetic framework that can help us talk about and across the deep differences in our lived experience of the same physical world and the ethical phenomena within it in order to tackle the pressing social and moral problems that we face.
"Designing a Good Life"
ABSTRACT: In their Designing Your Life, Bill Burnett and Dave Evans from Stanford University combine insights and methods from design, positive psychology, and other disciplines to craft a "life design” framework that has been used by thousands of people to effectively transform their lives into happier, more meaningful, more resilient ones. I argue that by bringing together this framework with ethical theorizing we can develop a “good life design” framework that we can use to empirically evaluate and thereby enrich our ethical theorizing. More specifically, I show how we can refine life design’s “good time journal” to craft a “good life journal” that we can use to gather data about the value of activities that make up our lives. We can then use this journal to test whether particular “good life hypotheses/prototypes" (i.e., miniature experiments in living) actually constitute changes that make our lives, on the whole, better and use the data we gather to iteratively test and refine our good life hypotheses/prototypes. This framework thus frames social and moral problems as “wicked problems” in the designer’s sense (problems that are resistant to solution because they have many interdependent factors that are hard to define, incomplete, and constantly changing) and correspondingly combines tools from design thinking together with ethical concepts and theories to tackle them in more human-centered, iterative, and effective ways.
“Balancing Forces and Fields in the Physical and Moral Worlds”
ABSTRACT: This is a successor paper to “Kant and the Balance of Moral Forces” that argues that Kant’s view that the physical and moral worlds are analogous in structure remains attractive from a contemporary perspective and articulates a fuller picture of how the physical and moral worlds are made possible. According to Kant, both the physical and moral world are grounded in the balancing opposition of attractive and repulsive between bodies/matter (in the physical case) and humans (in the moral case). However, contemporary physics teaches that the physical world consists of more than just matter interacting through attractive and repulsive forces. A fuller physical picture requires adding certain physical fields interacting in the right way to make ordinary matter and its physical properties and (attractive and repulsive) interactions possible. This includes quark, gluon, weak force, and Higgs fields interacting in particular ways according to their unique properties. Without these "background" physically relevant fields interacting in certain ways, ordinary matter would not be possible at all. Similarly, I argue that a contemporary understanding of morality requires adding certain morally relevant fields interacting in the right ways to make ordinary humans and their moral properties and (loving and respectful) interactions possible. This includes social, political, economic, and ecological fields interacting in particular ways according to their unique properties. Without these "background" morally relevant fields interacting in certain ways, ordinary human moral life would not be possible at all. I argue that this picture has significant implications for the fundamental relevance of social, political economic, and ecological factors to moral theory and practice.
Publications
"Kant and the Balance of Moral Forces," Journal of Ethics and Social Philosophy (2023)
ABSTRACT: In the Doctrine of Virtue, Kant draws an analogy between the physical and moral worlds, claiming that love is a form of moral attraction, respect a form of moral repulsion, and that "should either of these great moral forces fail, then nothingness (immorality) with gaping throat would swallow up the whole realm of moral beings like a drop of water" (6:449). I argue that we can make sense of this intriguing yet puzzling passage by interpreting Kant here as employing a certain use of symbols/ analogies (developed in other critical works) that makes supersensible objects intuitive in a sense by symbolizing them with an analogous sensible object. Based on that, I argue that we should read this passage as endorsing a moral version of the balancing argument from his natural philosophy. There, Kant argues that purely repulsive and attractive bodies respectively face problems of total dispersion and collapse. I argue that purely respectful and loving humans respectively face analogous problems of total moral dispersion and collapse. I conclude by drawing some significant implications of this moral balancing argument for contemporary theorizing about moral interactions, relationships, and communities.
"On Kant's Derivation of the Categories," Kant-Studien (2018)
ABSTRACT: In this paper, I put forth a novel interpretation of how the third categories under each heading in the table of categories (totality, limitation, community, and necessity) are derived. Drawing on a passage from the first Critique and a letter to Schultz, I argue that in order to derive these categories, a special act of the understanding is required. I propose that we interpret this special act as consisting of an application of the third logical function under the corresponding heading that unites the combination of the first and second categories (under that same heading) so as to produce the third.
*Note: My thinking on this topic has evolved since the publication of this article. I now argue in my dissertation that each category is generated in a special act of the understanding that employs the relevant logical function to combine certain representations. According to this revised view, the acts that generate the third categories under each heading are "special" not in the sense that they are unique but in that they are "special" like other category-generating acts.
"A Kantian's Guide to Objective Morality"
ABSTRACT: This paper builds on Jenann Ismael’s “An Empiricist’s Guide to Objective Modality” (which itself builds on David Lewis’s “A Subjectivist’ Guide to Objective Chance) in telling a sensible story about metaphysically puzzling concepts and their use. I argue that by following Ismael’s suggestion that we take a “side-on view” of the practices in which we use concepts, we open up room for a robust, non-reductive naturalism that naturalistically locates not just modal properties and relations but also ethical and moral properties and relations within a scientific vision of the world. In particular, it locates them not as (either reified or reduced) patterns in the mind-independent categorical manifold of events but as objective relational structures that encode information about how creatures with our practical, affective, social, and political capacities relate to patterns in this categorical manifold.
"Making the Supersensible Intuitive"
ABSTRACT: In this companion paper to "Kant and the Balance of Moral Forces", I flesh out my interpretation of the species of use of symbols/analogies that Kant uses to make the supersensible objects of ideas intuitive in a sense by symbolizing them with an analogous sensible object. I then apply this interpretation to show how, by means of such analogies, Kant legitimizes the use of certain ideas that are central to both his theoretical philosophy (e.g., ground and substance) and moral philosophy (e.g., moral community and realm of ends).
"Naturalizing Transcendental Idealism"
ABSTRACT: In this paper, I argue that a version of transcendental idealism can be seen to emerge from an attractive, naturalistic way of viewing the human mind as a certain kind of dynamical system that represents itself and governs its behavior by means of its self-representations. According to this naturalistic approach (which is articulated in different ways in the work of Craig Callender, Rick Grush, Jenann Ismael, and Huw Price among others) the embodied human mind not only passively receives sensory information from disparate sensory sources but also actively synthesizes information from these sensory sources into a global model of its environment. The mind represents itself in this model and uses it to navigate its environment. This view of the mind constitutes a version of transcendental idealism insofar as it accepts the two constitutive theses of transcendental idealism, roughly (TI-1): the things we experience are not things in themselves but rather representations that we synthesize from sensory information, and (TI-2): space and time are not features of mind-independent reality but rather primarily forms of sensibility, representational frameworks we use to coordinate different sensations in order to navigate our environment.
"No Reversal in Kant's Foundational Moral Works"
ABSTRACT: A common interpretation of Kant's foundational moral works is that in the Critique of Practical Reason, Kant rejects his argument from Groundwork III and reverses his strategy of justification, arguing from morality to freedom rather than from freedom and the moral law. I argue, contrary to this interpretation, that the arguments from these two works are different kinds of argument, which are not incompatible but in fact complementary. The Groundwork argues that freedom is the ground of the possibility of the moral law, while the second Critique argues instead that consciousness of the moral law is the ground of cognition of freedom. Each argument aims to establish a different kind of conclusion, but together they make possible cognition not just that we must be free if our being bound by the moral law is to be possible but that we are actually free.