The Photon Trap

Gravitational Confinement from First Principles,
with Implications for Cosmogenesis and Sustained Fusion

Anni Abigail Casella, March 2026

________________________________________

A few years ago I tried to design a camera that could photograph the shape of light. I failed. But the failure derived real physics — and by the end of it, I was staring at a fusion reactor I hadn't meant to build.

This is the story of that thought experiment, what it accidentally proved, and why it matters for two very different problems: how the universe began, and how we might power civilization without burning it down.

The Impossible Camera

I was studying a simple relationship: light equals mass times the shape of motion. I'd been drawing geometric patterns: reflection signatures, propagation shapes, the structure that motion takes when you stop assuming it's linear and start looking at what it actually does. And I had an intuition I couldn't shake: the universe is the interior of a black hole.

I didn't know at the time that other people had been postulating this. I was following geometries, and the geometries kept pointing inward.

So I asked a practical question: could you build an instrument to observe the shape of propagation directly? Could you watch a single photon move through space and trace the geometry it makes?

Here's what I came up with. Imagine a perfect sphere made of one-way glass — and not the kind you see in interrogation rooms on television, because that's actually two-way glass. Whichever side is brighter sees a mirror; the darker side sees through. That's not what I needed. I needed a material that only passes light in one direction. Truly one-way. Light can enter but not exit. That material doesn't exist. So the experiment started, before anything else, with the need for something impossible. But I kept going. You manufacture this sphere in total darkness, like photographic film — it's never been exposed to a single photon. Then, in a lab, you fire one photon from a laser into the sphere.

The idea was simple: the photon enters, bounces around inside, and leaves a record of its geometric path on the interior surface. A photograph of the shape of light.

It didn't work, of course. But the way it didn't work turned out to be the whole point.

The Constraint Cascade

The first problem was the glass itself. A perfect sphere made of glass would diffract the photon at the entry point. The light bends as it crosses the boundary, which means you don't get an undistorted photon inside — you get an artifact of the lens. If you shape the entry to compensate, it's no longer a perfect sphere. And if it's a perfect sphere, it distorts what you're trying to observe.

So it couldn't be glass. It had to be a field.

The second problem: what kind of field can trap light? An electromagnetic field can bend a photon's path, but to actually trap it — to bend it enough that it curves back into the sphere rather than escaping — you'd need an extraordinarily specific field geometry. Not impossible, but the configuration requirements kept escalating. At some point the constraint became clear: the field has to be strong enough to bend light into a closed path. It has to curve spacetime itself.

It has to be gravity.

So now the thought experiment had evolved. The one-way glass sphere was a gravitational sphere — a region where the gravitational field is strong enough that light can enter but cannot leave. I had been trying to design a camera and had accidentally derived an event horizon.

Nothing, Then a Glowing Sphere

Then came the problem that ended the experiment and started the physics.

The speed of light.

Even inside this gravitational sphere, the photon doesn't slowly trace out a pattern you could capture frame by frame. Along its own path, the photon accumulates no time — not because it moves infinitely fast, but because light doesn't experience duration along its trajectory. It traverses the sphere at c, which in our frame is finite, but the interior is small and the geometry returns every path to the interior. There is no unfolding sequence to photograph. No frame-by-frame record of a shape being drawn. Just a system that crosses from empty to full with nothing useful in between.

Which means that even if you recorded the experiment on video, you would see: nothing — total darkness, the sphere sitting there unexposed — and then, a glowing sphere.

I remember the moment I realized this. It was almost funny. All that design work, all those constraints, all that careful reasoning about one-way glass and field geometries, and the result of the experiment would be: dark, dark, dark, glow.

And then I stopped laughing. Because I knew what that was.

Nothing, then a glowing sphere. That is the description of stellar ignition. That is what happens when a star is born.

The Geometry Nobody Is Naming

The thought experiment was over, but the physics had just started. Because what I had derived, from scratch, starting from nothing but the desire to photograph a photon, was this: a gravitationally confined sphere where energy enters and cannot leave, energy density rises without limit, and the system transitions from dark to luminous in a single moment.

That's not just a star. That's a reactor. And it's a reactor whose confinement mechanism is fundamentally different from anything currently being attempted in fusion engineering.

Every tokamak on Earth confines plasma in a torus — a donut shape. The engineering is extraordinary and the plasmas are beautiful... I could watch the videos over and over like watching a river, but the topology is wrong. A torus has a hole. Mathematically, it's the product of two circles, and that hole isn't just a geometric feature — it's a structural defect that every aspect of tokamak engineering exists to compensate for.

Particles in a toroidal field drift outward because the magnetic field is stronger on the inside edge than the outside edge. This is not an engineering problem to be solved. It's a topological problem inherent to the shape. The entire history of magnetic confinement — rotational transform, magnetic shear, ELM mitigation, disruption avoidance — is the history of compensating for a geometry that doesn't close on itself.

A sphere closes on itself. Specifically, a 3-sphere — the higher-dimensional analog of an ordinary sphere — is simply connected. No hole, no preferred drift direction, every point equivalent to every other. And the 3-sphere already contains the rotational structure that tokamaks have to artificially impose: the Hopf fibration, a mathematical structure native to S³, gives you twisting fibers that wrap the sphere in exactly the way magnetic field lines need to wrap a confined plasma.

The twist that confines is native to the geometry rather than engineered on top of it.

Someone will say: you can't build a 3-sphere in three-dimensional space. True. But you don't need to. The plasma doesn't know what shape the vessel is — it knows what shape the fields are. What matters is not the physical container but the topology of the confinement. And there's already experimental evidence pointing this direction: spherical tokamaks like MAST and NSTX show better confinement at lower aspect ratios, achieving higher plasma pressure relative to magnetic pressure. They are moving toward the sphere without having named the reason why.

The Fourth Dimension Is Not a Place

But here's the thing that matters most, and it's the thing the thought experiment actually derived: the difference between a burst and a sustained reaction is not about stronger magnets or hotter plasma. It's about the topology of the feedback.

NIF — the National Ignition Facility — achieved fusion ignition by compressing a sphere of fuel with lasers. Spherical geometry, enormous energy density, and it worked. For a nanosecond. The geometry was right, but the reaction was a single frame. One pulse. No return path.

A tokamak sustains plasma for much longer, but in the wrong shape. The torus bleeds energy through the hole it cannot close.

What the thought experiment revealed is that the missing coordinate — the one that turns a burst into a self-sustaining reaction — is not spatial. It's temporal. It's the reaction's output becoming its input. The energy released by the fusion process curves back through the geometry and re-enters as fuel. The fourth dimension isn't somewhere you build to. It's the system closing on itself in time.

This means confinement is not a container problem. It's a feedback topology problem. The plasma sustains when the geometry creates a closed return path — when energy loss pathways curve back into the reaction rather than escaping. The "wall" is the curvature of return itself.

And that reframes the entire engineering question. Stop asking how do we hold it in. Start asking what field geometry creates a closed return path for the reaction's own energy.

One Proton, One Universe

Now follow the thought experiment the other direction. Not outward, toward engineering. Inward, toward origin.

Take the gravitational sphere seriously. Not as a reactor to be built, but as a condition that already exists. Anywhere gravity is strong enough that light cannot escape — any event horizon — the photon trap is already operating.

So: one proton falls into a black hole. Or rather, one unit of energy crosses an event horizon and enters a region of spacetime from which nothing returns.

What happens?

The energy is confined. There is no exit path. The geometry guarantees that energy density increases because nothing can leave. As density rises, pair production begins — energy converts to matter. More matter deepens the gravitational well. A deeper well means stronger confinement. Stronger confinement means higher energy density. Higher energy density means more pair production.

This is the self-fueling loop. The fourth dimension again — output becoming input. But now the input is matter, and the output is the gravitational field that confines it, and the confinement produces more matter that deepens the field that increases the confinement.

One proton becomes a universe.

Not an explosion outward into pre-existing space. An inward deepening that generates space as it goes. What we call the Big Bang is the interior experience of this recursive self-amplification. From inside the process, it looks like everything expanding outward. From the perspective of the geometry, it's everything collapsing inward into higher density, higher complexity, more structure.

This is not a new speculation. Lee Smolin proposed that black holes spawn new universes — his cosmological natural selection hypothesis. Nikodem Poplawski developed models where torsion inside black holes triggers a bounce that initiates a new expanding region. What they proposed as a possibility, the photon trap provides as a mechanism: recursive gravitational confinement with no loss pathway, where confinement and reaction are the same geometry, and the system's output is the condition for its own continuation.

Confinement and Reaction Are the Same Thing

Here's what the thought experiment actually derived, stripped to its core:

An event horizon is not just a boundary. It is a reactor geometry. The same structure that prevents energy from escaping is the structure that drives energy density to the threshold where matter emerges, stars ignite, and universes begin. Confinement is the reaction. Not a container that holds a reaction — the act of containing, performed by the curvature of spacetime itself, is identical to the process of energy amplification.

This identity — confinement equals reaction — is what the toroidal approach to fusion misses. A tokamak treats confinement as a prerequisite for reaction: build the bottle, then heat the plasma. But in the gravitational case, there is no separate bottle. The curvature that confines is produced by the energy it confines. The system holds itself.

For fusion engineering, this suggests a specific direction — and it requires a sentence about gravity that changes everything downstream.

General relativity established, over a century ago, that gravity is not a force. It is a geometry. Mass and energy curve the shape of spacetime, and what we experience as gravity is that curvature. This is not speculative. It is the standard, experimentally confirmed model of gravity. But its implications are rarely followed all the way through.

The thought experiment derived a geometry: closed return paths, no loss pathway, self-reinforcing curvature. That geometry is gravity. Not a geometry that produces gravitational effects — the geometry itself is what the word "gravity" refers to. General relativity settled this. Gravity is the name we give to the curvature of spacetime. If you create the curvature, you have created the gravity. They are the same thing.

Now: we already use electromagnetic fields to confine plasma. The question the thought experiment answers is not what kind of field but what topology. If you arrange electromagnetic fields in the right configuration — spherical, with the closed feedback structure derived above — then you have created the geometry. And if you have created the geometry, you have created what general relativity calls gravity. Not as a secondary effect. Not as something that shows up later at high enough energy. The geometry is the thing.

This reframes what ignition means. It's not just the point where the plasma is hot enough to fuse. It's the point where the feedback loop closes — where the reaction's energy output begins sustaining the confinement geometry that sustains the reaction. The electromagnetic fields create the initial topology. The reaction, if the topology is right, begins holding itself.

Spherical geometry is the starting point. The Hopf fibration's native twist structure is the confinement architecture. And the engineering problem becomes: design field configurations that create closed return paths, so that energy which would otherwise be lost curves back into the reaction zone.

This is a topology problem, not a materials problem. The answer is in the shape.

What a Camera Taught Me About the Universe

I started with a thought experiment about photography. I wanted to build an impossible instrument — a sphere of one-way glass that could capture the geometric signature of a single photon's propagation.

Every constraint I hit forced the next realization. The glass can't work → it must be a field. The field must bend light → it must be gravitational. Light enters but can't leave → I've derived an event horizon. The photon fills the space instantaneously → the experiment fails. And in the failure: nothing, then a glowing sphere.

I didn't assume any physics. I didn't start from equations or prior theory. I started from "how would you photograph the shape of light" and followed each dead end until it opened into something real.

What I found: the event horizon, stellar ignition, the origin of the universe, and the unsolved problem in fusion energy are not four separate phenomena. They are one geometry operating at different scales. Gravitational confinement with no loss pathway, where the confinement and the reaction are the same thing, where the system's output is the condition for its own continuation.

The photon trap is not a thought experiment that failed. It's a thought experiment that succeeded at something other than what it attempted.

It derived the structure of reality from a camera that couldn't exist.