Measurement and observation
The Shape You Could Never See
For years it was a sphere. It was never a sphere. Nothing about the object changed — the observers did.
Established ScienceWhat we actually know
Radar gives distance and speed superbly and shape poorly, especially from a target that reflects everything sent at it. Optical, laser-ranging, structured-light and interferometric methods each recover different parts of a surface, and combining them is how a real geometry is reconstructed. A circle's circumference over its diameter is pi; measure thousands of edge points well enough and you can test whether a cross-section is genuinely circular and to how many decimal places.
Where the extrapolation begins
A difference of 14.8957 centimetres across a body 108 metres wide is roughly one part in seven hundred. You could stare at it for a lifetime and never see it. Discoveries of that kind are not made by looking harder; they are made by measuring differently, and by taking the residuals seriously instead of rounding them away.
What the novel invents
The object is a prolate spheroid: two equal short axes at 108.063303958 metres, one longer axis at 108.212260519. It always was. Successive surveys agreed with pi to six, then eight, then ten decimal places, and the elongation only emerged when the uncertainty fell below it. Rourke asks what changed. Maseko's answer is the archive's thesis in five words: nothing — we learned to measure.
Questions to think about
- — How can an object be described correctly as a sphere and then correctly as not a sphere?
- — Why does agreement with pi improving over time say more about the observers than the object?
- — What in your own life would look perfect at low resolution?
Where to read next
- — Geodesy and metrology — surface reconstruction from multi-station ranging
- — Introductory geometry of spheroids; oblate versus prolate forms
Continue in the archive