Why I Killed Dualism

For much of European history, the Moon was viewed as a divine perfect object suspended in the heavens. Aristotle shaped this understanding through his influential cosmology. It divided the universe into two realms: the imperfect Earth below the Moon and the perfect, eternal heavens above it. Aristotle believed celestial bodies moved in perfect circles and were made of special heavenly substances unlike anything found on Earth. His ideas fit neatly into a larger philosophical system that explained nature, motion, and humanity’s place in the cosmos, which helped his theories dominate Western thought for nearly 2 millennia.

Aristotle’s Cosmology, Source

In contrast, Anaxagoras had a far more naturalistic explanation of the Moon centuries earlier. Rather than viewing it as divine, Anaxagoras argued that the Moon was a physical object made of ordinary matter. He correctly proposed that the Moon shines by reflecting the Sun’s light and explained the lunar phases and eclipses through geometry rather than mythology or supernatural causes. These ideas were remarkably close to modern astronomy and were one of the earliest attempts to explain the heavens through observation and physical reasoning.

The contrast between Aristotle and Anaxagoras highlights an important tension in the history of science: the struggle between elegant philosophical systems and evidence-based explanations. Aristotle’s model was more philosophically comprehensive and emotionally satisfying to ancient audiences. Anaxagoras was more scientifically accurate (though itself mostly a product of philosophy and reasoning) but it lacked the cultural and institutional influence needed to prevail. It was only centuries later, through the work of astronomers such as Galileo and Kepler, that many of Anaxagoras’s insights were confirmed by observation.

Vitalism

This is a repeating pattern in the crossover between science and philosophy. A more recent and relevant example is the explanation for life itself: why are some things alive and other things not? Note that this isn’t asking about why some entities have intelligence or consciousness; it’s a much more basic question: what differentiates living matter from unliving matter? Similar to consciousness, we had two major classes of explanation.

The purely physical explanation is the one widely accepted today. Atoms form together to make more and more complex molecules, which we eventually consider living. In fact, the usual definition of life is usually a “capacity for homeostasis, organisation, metabolism, growth, adaptation, response to stimuli, and reproduction”. The fact that we know this doesn’t change what it means to be alive, it just gives us a clearer boundary.

In contrast to this was vitalism. This was the belief that living matter had an extra component that unliving matter lacked; for example a vital force (vis vitalis) or sometimes even a something like a soul.

In hindsight, vitalism seems silly but it attempted to explain a gap in our knowledge. That is, the transition from inorganic to organic matter was unclear. Even cooking permanently changed matter in a way that needed to be explained. This Ernst Mayr quote captures the issue well:

It would be ahistorical to ridicule vitalists. When one reads the writings of one of the leading vitalists like Driesch one is forced to agree with him that many of the basic problems of biology simply cannot be solved by a philosophy as that of Descartes, in which the organism is simply considered a machine... The logic of the critique of the vitalists was impeccable

Of course, this overlooks two key problems. First, was the gap really explained or simply moved: if a vital force explains how life is created, then what is the nature of the vital force?

Compared to the first, the second problem is much worse for science. It’s that the original truly unknown gap is filled with a magical unknown force that seems to explain life. This changes the scientific target to understanding a nonexistent magical force rather than trying to explain the gap. This caused people to devote time and resources to the wrong target.

Dualism

Those familiar with dualism will immediately see the similarity to vitalism. At a high level, dualism is the belief that consciousness itself is created by something other than the pure physical interactions of matter in the brain and rest of the body. There are a wide variety of dualist beliefs. Some believe that the brain acts as a radio antenna for consciousness. Others believe that consciousness is a secondary property of matter, such that conscious experience cannot be fully reduced to physics, chemistry, or computation alone. In these views, mental phenomena are treated as fundamentally distinct from the material processes that correlate with them, even if they interact closely with the brain. What unites most forms of dualism is the idea that subjective experience cannot be completely explained by describing the physical structure and behavior of matter itself.

Similar to vitalism’s “vital force”, many dualists argue that consciousness cannot be explained entirely through neuroscience, and therefore must involve some non-physical aspect of reality. In both cases, the central claim is that purely mechanistic explanations leave something important unexplained: for vitalism, biological life itself; for dualism, subjective conscious experience. This similarity serves as a historical warning against introducing mysterious non-physical entities to explain gaps in scientific understanding.

One counterargument, however, is that vitalism was eventually undermined by empirical evidence, while the empirical case against dualism is far less decisive. Advances in biology and chemistry progressively showed that processes once thought to require a “vital force” could be explained through ordinary physical interactions. Consciousness, by contrast, remains deeply resistant to reduction due to the contrast between first and third person accounts of subjective experience. Neuroscience has uncovered extensive correlations between brain states and conscious states, but many philosophers argue that correlation alone doesn’t explain why subjective experience exists at all. For this reason, the analogy to vitalism may be incomplete: vitalism was displaced by science developing a convincing replacement, whereas the so-called “hard problem” of consciousness remains philosophically unresolved.

But this ignores the second problem: we’re moving the gap from the truly unknown issue and replacing it with a magical one. For example, “how does something become conscious from neurons signaling (the hard problem)” becomes “how does a dualist soul signal the brain to do stuff”. This sort of magical thinking hurts science. At any point, someone can say “but maybe there’s a soul and it’s causing this”.

Structural Realism

It’s a bit late in the essay but we should discuss structural realism. It’s the idea that science only really teaches us about the relationships between things, and the “ontological furniture” can change. A classic example is the transition from Newton’s gravity to Einstein’s General Relativity. Newton described gravity was a force acting instantly at a distance between masses, governed by an inverse-square law. General Relativity changed gravity from a force to a curvature in spacetime caused by mass and energy. At a glance, these theories seem fundamentally incompatible because they give us very different pictures of what gravity is ontologically.

Structural realists argue, however, that there’s a form of continuity between the two models. Even though Einstein replaced Newton’s ontology, the underlying structural relationships were preserved. Newton’s equations are still extremely accurate approximations in weak gravity fields and low-speed conditions, and many of the predictive relationships between masses, motion, and trajectories survive within Einstein’s framework. The continuity isn’t in the literal existence of a Newtonian gravitational force, but in the mathematical structure connecting physical qualities and predicting observable behavior. For structural realism, this continuity explains why older scientific models can both be “wrong” in some respects and yet still successful: science progressively refines our grasp of the world’s structure even as its descriptions of underlying entities change.

SR can be used to open the conceptual space for dualism by emphasizing the limits of what physical science tells us about reality. In that sense, physics doesn’t reveal the intrinsic nature of things; it only describes the mathematical and causal relations between them. Scientific models tell us how entities behave, interact, and are structured, but not what those entities are “in themselves”

This creates an epistemic gap that dualists can exploit. If physical science only captures relational structure, then consciousness (especially the qualitative, first-person character of experience) may belong to a different category of reality not exhausted by structural description. A dualist can argue that even a complete physical model would leave out the intrinsic nature of conscious experience.

This line of thought is especially useful against reductive materialism. A strict physicalist may claim that advances in neuroscience will eventually explain consciousness entirely in physical terms, but the structural realist will respond that physical explanations are inherently structural and functional. They describe the network of relationships, information flows, and causal organization, yet subjective experience seems to involve something more than relational structure alone: the felt quality of pain, color, or awareness itself. Dualists argue that this explanatory remainder suggests the existence of non-physical mental properties or substances. In this way, structural realism undercuts the idea that physics alone gives a complete account of reality, thereby reopening philosophical space for the claim that the mind is irreducible to the physical world.

Of course, this is nonsense. If there was some extra part that we could measure using our minds, then this would fold into the structural description of the world. Otherwise, that part would remain forever outside observation, so that even discussing whether it exists is a fool’s errand. In other words, even if unmeasured dualism were metaphysically possible, it is scientifically inert.

In that sense, SR is telling us something we already know: we can never know the true “ontological furniture” of the world. This is trivially provable; we can’t tell if the world was created in this exact state exactly 5 minutes ago with our memories intact. Or if we’re living inside a simulation. Or if this is all just an elaborate dream.

From this, we see that science has never tried to be more than it is. We’re trying to model reality to make useful predictions. Like which entities are conscious in a way that might require that we care about their wellbeing. When someone asks “are AI conscious”, they’re asking about it for ethical reasons: Are we creating new minds? Are we exploiting them? At least I am.

On science and philosophy

From this, we see that philosophy and science operate in different but overlapping domains. Science is primarily concerned with empirical observation, experimentation, and predictive models, while philosophy examines the conceptual foundations and implications of those models.

One role philosophy plays in exploring ontological questions that may lie beyond direct scientific observation. SR, for example, raises the question about whether science describes reality itself or only the relational structure of reality.

Philosophy also helps science by clarifying concepts, exposing assumptions, and organizing interpretations of evidence. Scientific models often underdetermine their metaphysical interpretation leaving many philosophical explanations possible. Philosophy helps scientists think more clearly about what their models mean and how different ideas relate.

Finally, philosophy can shape future scientific inquiry by encouraging exploration of possibilities before direct evidence exists. Speculation about exoplanets, extraterrestrial life, or the nature of consciousness began as philosophical reflection long before relevant technologies existed. Over time, these conceptual questions can become scientifically operationalized, such as the search for biosignatures on exoplanets or neuroscientific investigations into consciousness. In this way, philosophy often functions as a conceptual frontier for science, exploring possibilities that may later become empirically testable.

From this perspective, dualism fails to occupy a productive scientific role. If consciousness is unknowable in the structural realist sense, then dualism becomes no different from positing that the universe was created 5 minutes ago with our memories intact: logically possible, but disconnected from scientific inquiry. If consciousness is knowable, then evidence matters, and we have no evidence requiring dualist explanations over physical ones.

More importantly, dualism generates no scientific traction. Unlike speculative ideas that eventually mature into research programs, it offers no predictions, no criteria, and few directions for productive investigation. Historically, explanatory gaps are often filled with metaphysics that feel intuitively satisfying but only relocate the mystery rather than resolve it. Vital forces explained life by introducing another unexplained force; dualism is doing the same for consciousness.

This matters because models influence how scientific attention is distributed. Once a mystery is framed in terms of unknowable or non-physical causes, pressure to investigate the underlying mechanism weakens. In this sense, dualism stops being an explanation and becomes a conceptual buffer around unsolved problems. The danger stops being that it might be wrong, but that it reduces the incentive to search for better answers.

In April, Henry Shevlin made an excellent point,

And that is why I killed dualism (and panpsychism and epiphenomenalism too) 🦁