Companion Stars

A star that shares a system with another star has a name, and the name is a companion star. Astronomers needed the word for their own reasons, long before it was any use to us.

It is not a borrowed term or a softened one. It is the working vocabulary of stellar astronomy — the second body in a bound system, the one whose gravity you have to account for if you want the first one's motion to make any sense at all.

Helen Edgar's galaxy already holds it. In My Monotropic Galaxy (More Realms, June 2026) she maps twenty constellations of her own Autistic self, and the thirteenth is Companion Stars.

Note the plural. Not the person — the people. This zine follows that plural outward, from one bond to a whole community, and the astronomy does the same thing on its own: two bodies, then three, and at three something changes that changes everything after it.

But first there is a fact to get straight, because the popular version of it is wrong — and this collection exists partly to take science back from the smoothed retelling.

You have almost certainly met the line most stars are in binaries — the Sun is the odd one out. It is repeated constantly, and it is not what the measurements say.

In 2006 Charles Lada put the binary statistics next to the stellar mass function and published the result as his title: most stars are single. Two-thirds of main-sequence stellar systems in the Galactic disk are composed of a single star.

The reason is arithmetic rather than romance. Multiplicity climbs steeply with mass, and the galaxy is overwhelmingly made of the small end:

M dwarfs are the majority of stars by a wide margin, and about three-quarters of them are alone. The big ones almost all have someone; there are hardly any big ones.

Which leaves a smaller claim, and it holds all the way down: where the binding is real, it is real physics — measurable, long-lasting, and structured in a way worth looking at closely. That is the zine. Not you will be found. Just: here is exactly what being bound is, and what it isn't.

Two bound stars do not have one sitting still while the other goes round it. Both move, about a point between them called the barycentre, which is not located in either star.

Sirius is the worked example, and it is the brightest star in our sky. Howard Bond and colleagues combined two decades of Hubble imaging with photographic plates and nearly 2,300 measurements going back to the nineteenth century, and pinned the orbit down: Sirius A and its white-dwarf companion Sirius B trace their paths in 50.13 years, with dynamical masses of 2.063 ± 0.023 and 1.018 ± 0.011 solar masses.

Now the part that keeps this honest. Equal binding is not equal motion. The barycentre sits nearer the heavier star, so Sirius A moves in a small ellipse and Sirius B in one about twice as wide. The heavier body is displaced less.

That distinction is worth keeping. A relation can be genuinely mutual and still ask more travel of one party than the other, and pretending otherwise is how the person doing more of the moving ends up told they are imagining it. The pull is equal. The displacement is not. Both are facts about the same system.

Here is where a warmer piece would say and neither star has to change shape to stay in relation. That is not true, and the way it fails is the most useful thing in this zine.

Bring two stars close and they stop being spheres. Tidal forces stretch each one along the line joining them — the same effect that raises Earth's ocean tides, scaled up until it is the shape of the star itself. Astronomers see it directly: the brightness of a close pair rises and falls as those stretched shapes turn, which is why the effect has its own name, ellipsoidal variability.

Push closer still and the deformation runs to its conclusion:

At the last stage the star can no longer hold its own outer material: it spills across to its companion, and what each one becomes is rewritten by the transfer. Proximity is not automatically good for a star.

And the far end of the range is enormous. Gaia has resolved over a million bound pairs out to separations of a parsec, with orbital periods running to a hundred million years. Those stars are genuinely companions — held by mutual gravity, one orbit longer than the entire history of flowering plants — and each of them keeps its own shape completely. Distance does not weaken the binding. It only stops the binding from being the thing that reshapes you.

This is the argumentative centre, and it is not an interpretation. It is a set of theorems.

With two gravitating bodies you can write the solution down. Newton did. Give the positions and velocities now, and there is a formula that returns them for any time you like, exactly, forever.

Add a third body and that stops being available. Not because it is hard — because it was proved not to be there:

So astronomers integrate these systems numerically: step the whole configuration forward a little, recompute every force, step again. There is no shortcut, and there is no way to do one body at a time.

The same work is where chaos was found. Poincaré discovered that arbitrarily small differences in the starting conditions can lead to completely different outcomes later — and the crucial part, the part the popular word destroys: nothing is random. Every rule is exact and fully known. The system is deterministic and still unpredictable far out. Those are compatible, and holding them together is the whole idea.

This community named both halves of that spread before we went near the astronomy, and the naming is David Gray-Hammond's.

Writing with Katie Munday, he describes the Chaotic Self as “our sense of Self being a fluid and moving entity, constantly changing and reshaping as we receive new information and interact with the environment” — and, on why the change does not run backwards: “You can't unqueer a queer mind.”

That is a claim from lived experience, not from physics, and it should stay that way. But it names the same shape the theorems name: a self that is rule-governed and still not walked back, still not predicted far out.

His Autistic Rhizome is the community version — “communities that exist of networks with no single point of origin. They are interlinked but not dependent on one another for their existence” — and the two ideas meet exactly where this zine does: as one of us changes, “the relational change with our immediate environment transfers the process onwards to the rest of our community in somewhat of an unpredictable manner.”

Helen's galaxy holds the matching constellations, the ninth and the sixth:

Now the honest part, which we are not smoothing. A rhizome and a bound system are describing different things, and both are true:

A network can be resilient in exactly Gray-Hammond's sense and any cluster inside it can still be behaving exactly as spread six describes, at the same moment. Neither claim needs the other to be weakened.

Strip a bound system down to what actually holds it, and there is one item on the list.

No agreement. No shared origin — many wide pairs formed apart and were captured, or drifted from a cluster that has long since dissolved. No requirement of similarity: Sirius A is a bright main-sequence star and Sirius B is the burnt core of one that died first. No maintenance, no renegotiation, no performance. Mutual gravity, and that is the whole of it.

And it does not fall off with familiarity. It falls off with distance, and even at a parsec it is still there, still doing the one thing it does.

That is the claim, kept small enough to be true. Not that you will be found — spread three forbids it. Not that closeness is safe — spread five forbids that too. Only this: where the binding exists, it is made of nothing that can be revoked for underperformance, and where three or more of us are in it, no one of our paths was ever the kind of thing you could work out alone.