I. The Physics of Morality
I was recently reading about Peter Singer’s famous Drowning Child thought experiment. The premise is simple: if you are walking past a shallow pond and see a child drowning, you are obligated to save them, even if it means ruining a pair of expensive shoes. Singer then extends this intuition, arguing that if you are willing to sacrifice material wealth to save a child right in front of you, you are equally obligated to donate that same amount of money to save a child starving in a distant country. Reflecting on this made me consider the nature of ethics, and why certain philosophical frameworks feel entirely detached from the reality of human behavior.
Ethics cannot be an exercise in fitting a curve to our scattered moral responses. Rather, it must be the formulation of the underlying theory behind those responses. In the case of the drowning child, the objective is to determine if the movement from the proximity of the pond to the abstraction of the charity is truly valid. For a variety of practical and theoretical reasons, we seek to discover the ethical equivalent of the laws of motion, rather than compiling an exhaustive catalog of moral events and isolated intuitions.
The necessity of this approach is driven by distinct practical and theoretical demands. Practically, an ethical system requires parsimony; it must be streamlined enough to be taught and explained effectively. It also requires extensibility, providing a reliable framework that can be seamlessly applied to novel situations we have not yet encountered, while serving as a mechanism to conclude arguments between incompatible viewpoints. Theoretically, a system demands self-consistency, ensuring that individual moral examples do not contradict one another. Furthermore, any robust ethical theory must possess cohesion with other established realities, particularly concerning the nature of humanity.
Continuing the physics metaphor, seeking these foundational laws allows us to discover deeper insights into how moral reality actually operates—insights that can sometimes contravene our initial, unexamined moral assumptions. This process does not involve ignoring our moral responses, which serve as our empirical evidence, but rather explaining why they occur in the first place. If this is the case, the necessary scope of ethical theory expands. An ethical theory can only be considered complete if it is inextricably linked to a broader anthropological framework. This human theory must be capable of explaining the full spectrum of our moral responses, including the reasons why individuals might reject the very conclusions drawn by the Ethical Theory.
A theoretical objection to this pursuit is the suspicion that no sensible, underlying rules actually exist to be figured out. First, this is a question of degree. Much like in say biology, there are governing rules that provide illumination to the subject and are worth discovering, even if they lack mathematical clarity. We can loosen the expectation of finding a rigid Kantian framework without losing sight of our analytical goals. Secondly, the idea that contradictions inevitably destroy an ethical system can be softened by looking at the macro and micro divides in physics. Just as quantum mechanics and general relativity operate under entirely different paradigms depending on the scale of observation, moral obligations may shift when moving between domains.
Taking the specific example of Singer’s equivalence of suffering, we can apply this integrated framework to expose its limitations. The utilitarian model posits that physical distance is a surface-level variable that should not meaningfully affect our moral calculus. However, to view distance as a trivial factor is incompatible with an accurate understanding of human nature as embodied, physical beings shaped by proximity. Those who hold strictly to the equivalence of suffering are thus operating as dualists by implication. They inadvertently separate the rational moral mind from the physical reality of the human animal.
II. Beyond the Is-Ought Divide
At this juncture, a critic might invoke David Hume’s famous guillotine. They will argue that just because human nature makes us care more about proximity, it does not follow that prioritizing proximity is morally justified. Evolution wired us for tribalism and short-sightedness; ethics is the very tool we invented to overcome that primitive wiring. To base our moral obligations on biological instincts is to commit the naturalistic fallacy: deriving an ought from an is.
However, integrating an anthropological framework into ethics is not a matter of committing the Is-Ought fallacy; rather, it is an argument from a strictly meta-ethical perspective. No ethical theory has ever convinced anyone simply by existing in a vacuum of pure logic. Theories draw their power from real, visceral human examples, extrapolating outward from our actual responses. They require a narrative to carry them—much like the narrative provided by Singer’s thought experiment. One can debate the boundaries between objective evidence and cognitive bias endlessly, but the unavoidable truth is that everything begins from something. We cannot construct an ethical system without first acknowledging the raw material of human intuition that we are attempting to systemize.
That being said, simply subscribing to human biology as the ultimate good is poor theory. Evolution is not a universal constant or a fundamental force of the universe like entropy or gravity; it is highly contingent on the specific presence of life and the environment it inhabits. While a more robust formulation of evolutionary theory might attempt to attribute its mechanics to the broader propagation of all information—viewing everything as platform agnostic memes—the rules that govern the flow of data are not entirely analogous to the rules that govern the messy reality of biological life.
This tangent on evolution is to preemptively ward off the Neo-Essentialists and Evolutionary Reductionists who have steadily gained ground since the publication of The Selfish Gene. Reducing all human morality to a evolutionary parlor trick is just as flawed as reducing it to a mathematical equation. It further proves the central thesis: Ethical theories are fundamentally incomplete without a comprehensive Human Theory to ground them.
The failure of reductionist frameworks often stems from a failure of imagination—specifically, the failure to identify true universals within the human experience. Modern ethical philosophy has a habit of dismissing vital aspects of our reality as mere illusions or cognitive biases to be rationalized away. But some realities are not illusions. Distance is one of them. It is the most fundamental unit that separates us. Another is Time, the unit that irrevocably separates the past, present, and future.
When we reintroduce the realities of distance and time to the Drowning Child thought experiment, the strict utilitarian equivalence collapses. Singer compares muddying a pair of shoes to sending money overseas to less well-off countries, treating the difference purely as a matter of geography. This willfully ignores how we bridge space. We cover physical distance at the pond with our own two feet, exerting direct, immediate agency. We can only cover global distance through institutions—bureaucracies, governments, and charities—which introduces variables of trust, efficiency, corruption, and systemic failure.
Furthermore, the utilitarian equation ignores the reality of acting through time. Singer presents a frozen snapshot: a single child, a single pond, a single moment. But what if we introduce the temporal dimension? If the child has a propensity to throw themselves off the bridge into the water every single day whenever their parents aren’t looking, the moral significance of the act fundamentally changes. A singular heroic sacrifice is an entirely different ethical demand than an infinite, repeating obligation to patch a systemic failure. By flattening space and freezing time, rationalistic ethic attempts to govern a universe that human beings do not actually inhabit.
III. Scale Problems
To return briefly to the critic who weaponizes Hume’s Guillotine against our anthropological framework: they are correct to assert that persuasion is not justification. Just because a visceral narrative compels us to act does not logically prove that we ought to act. However, the belief that the Is-Ought defense represents an impenetrable theoretical fortress misunderstands the nature of the divide. Hume’s Guillotine, while stated in the syntax of formal logic, is inherently an empirical problem. It is downstream of a much older and deeper philosophical dilemma: the Problem of Induction.
To even separate the Is from the Ought requires one to first assume an inductive, mechanistic universe devoid of inherent teleology. If the universe had a built-in purpose, the facts of reality and the moral obligations of its inhabitants would be indistinguishable. The inability to cross the bridge from physics to metaphysics is not a problem for the rationalist or the idealist; thinkers like Spinoza or Hegel bypassed the Is-Ought divide entirely because their systems begin in metaphysics to begin with. The divide only poses a fatal threat to those who attempt to build moral systems strictly from the ground up. It is, and remains, a trap explicitly set for the British Empiricists and their modern heirs: the Utilitarians.
Acknowledging this empirical gap is not fatal to utilitarianism—if it were, the same logic regarding induction would render all of the physical sciences impossible—but it does expose a severe vulnerability. Utilitarianism, much like classical mechanics, works well on a small scale. When we can adequately control for localized variables like culture, geography, and civic institutions, maximizing well-being is a coherent and achievable algorithm. But as the system scales outward, the friction increases. On a global scale, the implicit assumptions that allow utilitarian math to function break down. These are precisely the assumptions that allow a philosopher to assert that saving a drowning girl in front of you is mathematically identical to sending mosquito nets to a distant continent. At the macro-level, the variables multiply beyond our capacity to measure them, and the empirical framework collapses under its own weight.
At this juncture, we might be tempted to abandon utilitarianism entirely and adopt the critique offered by Virtue Ethics. Since its inception, Virtue Ethics has diagnosed utilitarianism’s core flaw: its attempt to separate moral action from the character, history, and context of the actor. However, a closer examination reveals that most formulations of Virtue Ethics suffer from deep failures of imagination remarkably similar to those we demonstrated earlier.
While Virtue Ethics provides an architecture for human flourishing, it demonstrates a reluctance to accept the synthetic nature of the modern world. It relies on a framework that demands a great deal of localized, existential choice from the individual. This model functions well if the world can sensibly be modeled as an ancient Greek polis, or mapped through the five interpersonal relationships of classical Confucianism. But we do not live in a polis. We live in a global network governed by massive, synthetic institutions, supply chains, and bureaucratic structures. An ethical system that relies entirely on interpersonal virtue has no mechanism to navigate systemic, institutional realities. Both Utilitarianism and Virtue Ethics ultimately fall victim to the exact same trap: an inability to scale.
IV. Epistemic Bounding
The realization that macro-level variables cannot be reliably managed has forced Utilitarianism into its first major retreat. Unable to mathematically justify sweeping transformations, the framework has sought refuge in the methodologies of empirical economics—specifically, the Randomized Controlled Trial (RCT). By abandoning the flawed models of the twentieth century and focusing strictly on the micro-level, proponents of frameworks like Effective Altruism attempt to salvage the utilitarian calculus. They argue that while we cannot engineer a global system, we can rigorously measure the cost of distributing a malaria net or a deworming pill, and allocate our morality accordingly.
However, this model of empirical economics is increasingly coming under attack, revealing vulnerabilities both in practice and in theory. Practically, RCTs are inextricably linked to the academic prestige economy, suffering from the very real flaws of publication bias, p-hacking, and a desperate need to produce statistically significant results. But the theoretical flaws are far more damning. The first is the trap of Positivism: by demanding that good be quantifiable, we inevitably restrict our focus to what can be measured. This severely distorts both the metrics of human flourishing and the time scales over which we evaluate them, prioritizing short-term, measurable interventions (Deworming Pills) over long-term, unquantifiable cultural or institutional development (Rule of Law). The second theoretical flaw is Marginalism: the assumption that a localized, marginal intervention will yield linear returns. A successful RCT in one village does not guarantee success when scaled across a continent; as interventions scale, supply curves shift, local economies warp, and institutional friction exponentially increases. The solutions proposed by the micro-model fundamentally fail to scale.
This brings us to a crucial pillar of our ethical architecture: our moral obligations must follow from what we are capable of knowing. We have already established that an ought cannot demand a physical impossibility; we must now acknowledge that it cannot demand an epistemic impossibility either. Acknowledging the limits of our reach means acknowledging the limits of our moral duty. If we do not—and often cannot—understand the nuances of local cultures, the realities of distant conflicts, or the cascading downstream effects of our economic interventions, those epistemic limits bound our moral obligations.
Therefore, the utilitarian imperative to simply donate more to distant causes borders on the incoherent. To where? For what? For how long? Unless we subscribe to a model of ethics that values the act of charity as an inherent, character-building good—a Virtue Ethics framework rather than a Utilitarian one—blind philanthropy cannot be justified. If we hold to the empirical limits of the RCT, the conclusion is that a charity or NGO’s operational scope should never far exceed the scale of the experiment that justifies its existence. Scaling the intervention breaks the epistemology that warranted the intervention in the first place.
A critic might object that this framework incentivizes willful ignorance. If knowledge creates moral obligation, does epistemic bounding simply give us permission to look away? Yes, it does. But it does not do so in the straightforward manner of simply declaring, “I don’t care.”
Because we are wired for empathy, there are tragedies we desperately want to feel a moral response to, and we often do feel it viscerally. To justify acting on that response, we must believe we understand the situation. But because our primary motivation is to validate our own feelings—rather than to undertake the expensive task of achieving understanding—we take the cheapest cognitive path possible. We often adopt reductive, heavily flattened models of the world that allow us to act quickly and soothe our conscience.
Epistemic bounding should give us the moral permission to withhold action when we lack localized knowledge. However, because our biological wiring makes the discomfort of inaction intolerable, we cheat. In outlining this dynamic, we are not proposing a new relationship between ethics and knowledge, so much as revealing the one that already governs us.
V. Wisdom as Model Selection
By inextricably linking our moral obligations to our epistemic limits, we have exposed ourselves to the most ancient, skeptical attacks on knowledge itself. While these attacks are incisive, they do not inherently weaken our ethical framework. Every ethical system, whether it admits it or not, relies on a sensible model of epistemology. By bringing these epistemological dependencies to the forefront, the total surface area of our theoretical problem has not increased; rather, the vulnerabilities have been properly categorized.
This categorization becomes vital when we ask how an ethical system should handle unprecedented crises. The answer relies entirely on your epistemological model. Taleb’s concept of the Black Swan—the unprecedented, high-impact event that defies predictive models—serves as an argument against Bayesianism or any other models that have a tendency to over-fit to past data. In the same vein, the complexity of the world is an argument against attachment to a single ethical algorithm. Nothing in the moral universe is permanently solved, but the tools to deal with novel crises do exist. The prerequisite for moral action is not perfect knowledge, but sensible model selection.
If choosing the right model is the prerequisite for moral action, then we have circled back to a modernized form of Virtue Ethics. Specifically, we have arrived at Aristotle’s concept of Phronesis, or practical wisdom—the virtue of knowing which rule applies in which specific situation.
Traditional arguments against Virtue Ethics often center around its inability to resolve conflicts between competing virtues. If honesty demands one action and compassion demands another, the system seemingly gridlocks. This is a valid flaw, but as noted in our earlier discussion of scale, conflict resolution is just one of many jobs a theory has to perform. We posit that a modern formulation of Virtue Ethics solves this by treating other ethical frameworks not as rivals, but as tools. A practitioner of Phronesis should be quite comfortable deploying utilitarian arguments as a sub-routine for specific, bounded situations where variables are known and measurable, while abandoning utilitarianism where its implicit assumptions fail.
This raises a question: if moral action requires choosing the right epistemological model, and choosing the right model requires Phronesis, how do we justify our Phronesis? At this point, we are confronting a problem of epistemology as much as one of ethics. When we trace the chain of justification backward—encountering the infinite regress of the Münchhausen Trilemma or the divine circularity of the Euthyphro dilemma—we eventually run out of logical road. We posit that the ultimate justification for our epistemological models rests not in empirical data, but in Aesthetics.
To suggest that ethics bottoms out in Aesthetics is to invite accusations of extreme moral relativism. If our actions are justified by what feels fitting or beautiful, does the door fly open to any subjective justification? This anxiety stems from a highly limited, intuitionistic view of aesthetics. Aesthetics is not merely personal preference; it is the field of epistemology dealing with harmony, proportion, and coherence within complex systems. Furthermore, starting from a position of demanding rules explicitly to eliminate the messiness of aesthetic judgment is an example of the very reductionism we critiqued earlier. Ethics cannot be neatly answered by dodging the hardest questions of epistemology. Aesthetics is simply the correct domain to navigate these specific, irreducible complexities.
This framework also directly addresses the Boundary Problem—determining which moral dilemmas are universal and which are strictly local. The difficulty modern philosophy faces here is largely a lack of imagination and cross-societal awareness, leading to the assumption that far too much of our specific cultural morality is universally applicable.
But without a rigid, mathematical framework to determine which problems are universal, does our theory rely heavily on the moral agent simply “knowing it when they see it?” In short, yes. And that is a feature, not a bug. Any ethical theory that claims to know the exact boundaries of a moral problem before seeing it is inherently flawed. Moral truth in a complex world cannot be deduced a priori from the comfort of an armchair; it must be encountered, observed, and dynamically modeled in situ.
The danger of armchair deduction cannot be overstated. Both Utilitarianism and traditional Virtue Ethics are frequently accused of promoting authoritarianism. While both systems have defenses against this charge, we suggest that this authoritarian drift is a practical manifestation of a theoretical flaw. A moral universe that can be solved on a desk suggests a world that is best managed from a desk. When ethics is reduced to an equation or scripture, technocracy and dictatorship are the logical endpoints.
If morality cannot be solved on a desk, how do we teach someone what to look for? If ethics relies on practical wisdom, and wisdom is ultimately an aesthetic judgment of which model to apply, it implies that morality must be taught through an apprenticeship model. It requires the slow, deliberate cultivation of taste, empathy, and pattern recognition. This provides a justification for the modern reality of extended childhood and higher education—not merely as a practical necessity for learning technical skills, but as the essential incubation period required to develop the aesthetic wisdom capable of navigating a deeply complex moral domain.
VI. Atrocity and Restorative Justice
If our moral obligation is bounded by our epistemic limits, we must confront a corollary: when a crisis reduces our epistemic certainty to zero, does our moral obligation correspondingly drop to zero? We argue that this is not a theoretical flaw, but a vital feature of a robust ethical system.
It is relatively uncontroversial to absolve individuals of moral culpability during a sudden, Black Swan natural disaster. The epistemic void is obvious; no one can be blamed for failing to prevent an earthquake. Similarly, in the face of complex, human-caused emergent outcomes—such as the 2008 financial crisis—the lack of localized moral culpability is easily justified theoretically. For a moral obligation to exist, the agent must have possessed a sensible model capable of predicting the crisis. When the macro-variables escape human comprehension and the scale of the system overwhelms individual foresight, personal guilt dissipates.
Therefore, to truly stress-test this ethical architecture, we must bypass accidents and focus on intentional human atrocities on a grand scale: man-made famines, ethnic cleansing, and genocide. In these events, there are undeniably guilty architects who orchestrated the tragedy. The philosophical challenge, however, is determining how far down the societal hierarchy we can sensibly assign that guilt.
To explore this, we look to the Rwandan Genocide. As a comparative model, the Holocaust, while horrific, was highly industrialized and bureaucratic. Its compartmentalized nature allowed the average citizen a mechanism—however cowardly—to claim ignorance. Rwanda strips away this defense. The violence was shockingly intimate. It was carried out in the streets and homes, with neighbor literally turning on neighbor.
Modern philosophy, drawing upon the legacy of the Romantics or more recently, existentialists, has a bias towards the belief that there exists a baseline human empathy or freedom that naturally overrides state-sponsored lies. Rwanda suggests otherwise. In a severely degraded epistemic environment, the murderers were not the psychological outliers. The rescuers were the anomaly.
Examining the history of the event make it clear that the Rwandan Genocide was not a sudden, inexplicable rupture of human nature. It was the terminus of decades of buildup, originating in the racialized class divisions enforced during the colonial period and calcified by state-sponsored propaganda. In this case, decades of manipulation and manufactured hatred thoroughly dismantled the epistemic foundation required for moral reasoning, destroying the average citizen’s ability to act as a moral agent.
In the aftermath of the violence, Rwanda famously utilized the Gacaca community courts, focusing heavily on truth-telling and restorative justice rather than retribution. Critics and legal theorists often frame this lack of proportional punishment as a tragic but purely practical compromise—a concession to overcrowded prisons and a shattered judiciary. We argue, however, that this pivot was the moral outcome.
Retributive justice requires a stable epistemic baseline; it requires a coherent society where the deviation into criminality can be isolated, judged, and punished. But chaos violently limits the reach of retributive justice. When a system is entirely corrupted and the agency of the population has been dismantled, punishment loses its moral anchor. You cannot logically exact individual retributive justice upon an agent who was stripped of their capacity to make a moral choice. Therefore, in the wake of total systemic collapse, Justice can no longer afford to look backward through the lens of retribution. It must adopt a restorative narrative to rebuild the very shared reality that makes morality possible in the first place.
VII. Building a Shared Reality
On a stylistic level, this sudden turn might feel jarring. We have moved from the sterile economics of utilitarianism to the blood-soaked earth of Rwanda. But as we pointed out earlier, no theory lives by theory alone. A philosophy that cannot provide its readers with an emotional and practical anchor is a philosophy of ghosts. Having critiqued modern ethics for abandoning epistemology and ignoring human history, we are obligated to close this essay by demonstrating what a grounded ethical principle actually looks like when applied to the most broken aspects of the human condition.
Since we have reached the final part of the essay, a critic might accuse us of executing the “Aesthetic Dodge.” They will argue that by claiming morality bottoms out in “Aesthetics,” we have used a nebulous word to wave away the hardest problems of our philosophy. Fair enough. A complete defense of moral aesthetics demands its own dedicated exploration.
The answer we do offer is a subtle one, imbued in the methodology of this essay and the breadcrumbs left along the way: the vital necessity of narrative (Part II), the slow, deliberate cultivation of taste and apprenticeship (Part V), and the undeniable impact of our physical and informational environments.
We must rescue the word from the art critics. Aesthetics is not about sculptures in a museum or paintings in a gallery. It is the narrative architecture, the institutional design, and the physical environments that comprehensively shape our epistemic reality.
Finally, we return to the Rwandan Genocide and attempt to answer the lingering questions our framework provokes.
If the rescuer could find a way to maintain their moral agency amidst the chaos, why is the killer entirely excused for losing theirs?
The answer rests on the word average. The statistical average does the heavy lifting of explaining why societies collapse when their epistemic baseline is destroyed. An ethical theory that demands every citizen be a philosopher-king or a self-sacrificing martyr is a theory with a deficient Human Theory. The failure of the average citizen to be a hero is a tragedy, and it is an idea worth exploring as a society—but it cannot be solved through the blunt instrument of retributive punishment, which is usually the implication of the question.
They are not entirely excused, they absolved of retributive punishment because the system required for retribution has failed.
During the Gacaca courts, perpetrators had to publicly admit to their crimes and ask for forgiveness. If the propaganda had truly stripped them of their moral agency, what exactly were they confessing to?
If a perpetrator truly lacked all moral agency, then a confession is not an admission of autonomous guilt; it is a restatement of facts. It isn’t a confession at all—it is a deposition.
But that is precisely the point. The deposition is the vital first step in the dialogue of restoring moral agency. By forcing perpetrators and victims to stand in the light and state the historical facts together, society builds a shared, baseline reality. This suggests a redefinition of the courtroom: Justice does not exist merely to punish; Justice exists to build a shared reality. The very act of confessing in front of the community is the mechanism that re-activates the perpetrator’s severed moral agency.
If humans are this heavily determined by their epistemic and aesthetic environments, doesn’t the entire concept of an Ethical Theory become moot?
Ethical theories that are terrified of psychology and sociology should have died in the twentieth century. In the twenty-first they must answer what morality means under the view of the fMRI.
We have derived the need for an Ethical Theory from an implicitly pragmatic position: we need a framework to survive and flourish together. If compatibilism—the belief that determinism and free will can coexist—is the necessary metaphysics to make this framework function, we adopt it frictionlessly.
If the standard for avoiding punishment is simply “being average,” does this theory surrender the aspirational, elevating power that Ethical Theories historically strive to provide?
Look at where this view of human nature leaves us. It leaves us with the conviction that human nature is highly malleable—it can, and absolutely should, be actively improved.
We reject the Romantic ideal of the Noble Savage, which assumes humans are born morally perfect and corrupted by society. However, we wholeheartedly accept the Romantics’ greatest insight: human nature can be profoundly elevated through aesthetic experience—a truth that modern, equality sanctifying philosophy finds difficult to integrate.
Make no mistake, however; we place a far greater emphasis on institutional design and the curation of our epistemic environments than the Romantics ever would have.
If a person works sixty hours a week, lives in a brutalist concrete environment, and has zero access to pristine nature or high art, does your theory imply they are morally stunted?
Yes.
And that is exactly why we must work to fix those environments. If morality is an aesthetic output dependent on a stable epistemic baseline, then poverty, exhaustion, and physical ugliness are moral hazards. Our moral theory places material needs and environmental design at the base of the pyramid, exactly where they belong.