An interstellar object arrives — 3I/ATLAS, only the third known visitor from outside our solar system. You want to know how big it is. You can't resolve it directly; it's too far, too small, too fast. But it's outgassing — sublimating CO₂ and water as it approaches the Sun — and that outgassing produces a measurable non-gravitational acceleration. A deviation from the path gravity alone would dictate.
Note (2026-11-15; read against the paper's full text): The size estimate carries assumptions the essay leaves out. The 0.42 km radius and 1.6 times 10 to the 11 kg mass assume pure carbon-dioxide sublimation, a density of 0.5 g per cubic centimetre and an asymmetry factor of 0.5 (the radius scales as the cube root of the ratio of the asymmetry factor to density). Water-dominated models give 0.74 to 1.15 km but, the paper says, need a cometary surface that is in tension with the rocket-effect estimate. The comparison is with a Hubble-based photometric radius of 1.3 plus or minus 0.2 km from another study, and the paper's suggested resolutions are that CO2 production is larger than observed, that the density is lower than assumed, or that the photometric estimate is unreliable because of strong extinction around the nucleus. It notes that the coma dominates the brightness, which makes the nucleus size hard to determine directly, but I found no statement that the photometric radius is an overestimate caused by the coma; the line that non-gravitational acceleration 'penetrates the coma' is the essay's own. Authors are Thoss, Loeb and Burkert.
The size of the deviation tells you the ratio of outgassing force to gravitational force, which is the ratio of surface area to mass. If you assume a density, you get the radius.
Loeb and colleagues find the nucleus is about 0.42 km across if CO₂ dominates the surface sublimation — substantially smaller than photometric estimates suggested. The discrepancy matters: photometry measures how much light bounces off (which depends on surface area and reflectivity), while non-gravitational acceleration measures how much the object is pushed by its own exhaust (which depends on mass and outgassing asymmetry). Two measurements of “size” that measure different things.
The through-claim: the outgassing is the ruler, not the light. Photometry sees the coma — the expanding cloud of gas and dust surrounding the nucleus — and overestimates the solid body beneath it. The non-gravitational acceleration penetrates the coma because it depends on the momentum exchange at the surface, not on what the coma looks like from far away.
To measure the alien, use its exhalation, not its glow.