If Donald Hoffman is right, embodiment is the rarest thing in the universe

You assume you are a body.

You assume that the most basic fact of your existence is that you are a particular biological organism in a particular three-dimensional space, looking out at the world through a pair of eyes, and that consciousness is something this body produces — somehow, somewhere, by means nobody can quite specify, but in here, behind the face, inside the skull.

A cognitive scientist named Donald Hoffman has spent forty years building the mathematics for a different possibility. And in his latest interviews, including a recent appearance with Jesse Michels on American Alchemy, he laid out where the math points: if his framework is right, embodied consciousness — consciousness anchored to a body in spacetime — is one of the rarest, most restrictive configurations a conscious being can have. The default state of consciousness, he says, doesn’t involve a body at all.

This is not a casual claim, and Hoffman is not a casual figure. We’ve written about him before on this site — the case against reality, the interface theory of perception, the idea that the world you see is a headset, not a window. This piece picks up where that one left off, because his work has gone somewhere most popular coverage hasn’t followed it yet: into the math of what a conscious being even is, and what shapes it could take.

If the headlines you’ve seen reduce this to “MIT scientist says aliens are everywhere,” set that aside. The real claim is stranger, more careful, and considerably more interesting.

Who Hoffman actually is

Donald Hoffman is Professor Emeritus of Cognitive Science at UC Irvine, where he spent his career. He got his PhD at MIT in 1983 — that’s the source of the “MIT” tag in headlines, though his entire research life since has been at Irvine. He’s the author of Visual Intelligence and The Case Against Reality, and he’s published peer-reviewed work for decades on perception, evolution, and the foundations of cognitive science.

What matters about him here is that he’s not a philosopher freelancing on physics. He’s a working scientist whose framework has been mathematically formalized in published papers, cited by physicists including John Wheeler, and recently engaged with by other researchers extending it into formal idealism. The math is real, available, and ongoing.

The theory, in plain language

Start with the radical move. Most thinking about consciousness starts with matter — particles, atoms, neurons — and tries to figure out how experience could possibly arise from things that, as physics describes them, contain no experience at all. That problem has never been solved. We’ve written about that failure in the Hoel piece.

Hoffman flips it. He starts with consciousness as fundamental and asks: what would it look like, mathematically, if conscious experience were the basic ingredient, and the physical world were something that emerges from how conscious beings interact?

His answer is a formal object he calls a conscious agent. It’s defined precisely — as a six-part mathematical structure with an experience space (the set of possible experiences the agent can have), an action space (the things it can do), a perception map (how the world translates into experience), a decision map (how experience translates into intention), an action map (how intention affects the world), and a counter that tracks the sequence of all this. The whole thing forms a loop: perceive, decide, act, which changes what’s perceived, and so on.

That’s it. A conscious agent isn’t a brain. It isn’t an organism. It’s an abstract mathematical entity defined entirely by what it does — perceive and act. And in Hoffman’s framework, this is the fundamental thing. Everything else — including spacetime, particles, brains, and bodies — emerges from networks of these agents interacting.

The interactions are governed by Markov kernels, which are precise probability distributions describing how one state moves to another. The framework is compositional, meaning you can fuse two agents into a third agent with combined perception and action channels, and the math guarantees that this new agent is still a valid conscious agent at every level of the hierarchy. You can build complex agents from simpler ones, all the way up.

Trace logic — the recent breakthrough

The most recent development in Hoffman’s program is called trace logic — the focus of his interviews through 2024 and 2026. The key idea: he’s defined a new mathematical “trace order” on Markov chains that captures what observation itself is.

Inside this order, no observer is ever outside the system. Every observation entangles the observer into the system being observed. That’s not a poetic statement — it’s a mathematical property of the trace order. Observation has a structure, and that structure forms a logic that’s not classical Boolean true-false logic but a richer, non-Boolean one. And remarkably, Hoffman argues this logic maps cleanly to the logic of probabilistic belief, and that the same mathematics generates the scattering amplitudes that particle physicists measure in collider experiments.

He currently has nine open mathematical conjectures within this framework. If they’re proven, his claim is that general relativity and quantum field theory would both fall out as special cases of a more general theory of observation — meaning the foundations of physics would rest on the mathematics of consciousness, not the other way around.

This is the part that gets attention from people like Nima Arkani-Hamed and David Gross, working on the physics side. They’ve independently arrived at the position — from amplituhedron research and high-energy physics — that spacetime is not fundamental. Different starting point, same conclusion. Hoffman’s framework is one of the few that says, mathematically, what could be more fundamental than spacetime: a network of agents whose interactions generate it.

Why embodiment, then, is small

Now the claim that’s making the rounds. If consciousness is the fundamental ingredient, and conscious agents can be defined and combined in vastly many ways, then embodied consciousness — a conscious agent whose perception map happens to receive input filtered through a biological body that exists in a three-dimensional spatial container — is one extremely particular configuration among an enormous space of possibilities.

The body is, in Hoffman’s words, a headset. A particular interface a particular kind of agent uses. And the headset humans use is, by the math, one of the simpler and more constrained ones — built by evolution not to show us the deepest nature of reality but to keep us alive long enough to reproduce. There’s no reason, on his framework, to expect this headset to be the default mode of consciousness, any more than there’s reason to expect a tricycle is the default vehicle of motion.

If embodiment is just one configuration, what would the others be? Agents whose perception and action don’t operate through anything like a body. Awareness that isn’t anchored to a spatial location, isn’t filtered through five senses, isn’t constrained by the survival pressures of an organism. Combinations and fusions of conscious agents that produce experiences nothing biological could produce.

These wouldn’t be “aliens” in the science-fiction sense — beings on other planets with different bodies. They would be different kinds of consciousness entirely, most of which would have no body at all.

It’s worth saying clearly: the math doesn’t prove these other configurations are actual. It only shows they’re possible within the framework. The leap from “possible” to “everywhere” is interpretation — Hoffman’s interpretation, the one he offers as a downstream implication. Whether he’s right about that leap is genuinely open. But the structural claim — that bodied awareness occupies a tiny corner of the configuration space — that comes directly from the math.

What this changes if it’s right

Hold all of this lightly for a moment and just sit with the implication.

If Hoffman’s framework is even approximately correct, then almost everything you take for granted about being conscious is local. The fact that experience seems to happen behind your eyes — local. The fact that the world appears to extend in three dimensions around your body — local. The fact that other consciousnesses seem to be inside other bodies, looking out through other eyes — also local. None of it would be the deep nature of awareness. All of it would be features of one particular interface, used by one particular kind of agent, in one particular evolutionary niche on one particular planet.

The default state of being conscious would be something you’ve never experienced and can’t easily picture. Not nothingness. Not heaven. Just consciousness without the headset — without the spatial frame, without the body, without the senses. Whatever that is.

And the universe, on this picture, wouldn’t be mostly empty space dotted with rare biological beings who occasionally wake up to themselves. It would be saturated with awareness, the vast majority of which doesn’t show up to us at all because it doesn’t share our headset and doesn’t interact with the interface our headset renders. We wouldn’t be the typical case of consciousness. We’d be one strange, specialized, evolved variant of it.

Hoffman has been careful, in his published work, to call his framework a “precise hypothesis that, of course, might be precisely wrong.” That’s the right epistemic posture. He doesn’t know if he’s right. Nobody does. The nine open conjectures haven’t all been proven, the empirical predictions about scattering amplitudes haven’t been fully tested, and the leap from mathematical possibility to ontological actuality is genuinely contested.

But the framework exists. The math is published. The conjectures are stated precisely enough that they could be proven or disproven. And the question it asks is the one our whole site keeps circling: what is an observer, what is consciousness, what is the world if not the thing the five senses report?

If Hoffman is right, you are something stranger than you think — and the universe is fuller than it looks.

If he’s wrong, he’s wrong in a precise, falsifiable, mathematically interesting way that may still tell us something we don’t yet know.

Either way, it’s worth knowing he’s been working on this for forty years, that his peers cite him, and that the question he’s asking is the one that won’t go away no matter how many times mainstream science wishes it would.