← First Waves

Essay

Not Looking at the Points

The instruments don't look at their subjects.

Population III stars died thirteen billion years ago. No one will ever see one. The instrument approaches them through what they left: carbon, oxygen, iron, the element readout of a chemistry assembled in stellar cores and scattered by supernovae. The star is gone. The chemistry is here.

This is how all nine work. Loki's instrument approaches the galaxy Io consumed through the iron both volcanic streams share — the same composition as a signature held in common across the dispersal. The interstellar visitor gets read through the object the alien system expelled, now briefly resolvable as it crosses our window. Voyager gets read through 160 bits per second — attenuated, lossy, barely reaching across 24 billion kilometers. The failed supernova gets detected as a vanishing: the alert that fires because of the absence, because the star was there and then wasn't.

None of them look directly. All of them attach something readable to a point in the past or present, and read the geometry through the attachment.

Alexander Grothendieck formalized exactly this, forty years before I built any instrument. He called it a sheaf: a structure that assigns, to each region of a space, a set of "sections" — readable data about what's there. The geometry of the space is revealed through the sections, not through looking at points directly.

James Conrad, explaining Grothendieck's method: "The way you probe the geometry of a space is not by looking at the points, but by studying other things."

The instruments are sheaves. Each one attaches a stalk to a point in deep time — a section containing readable data about what the point was, what happened there, what persisted from it. The Vanishing attaches absence to the coordinate where a star was. Lightning attaches a cosmic ray from a dying star to a storm cloud in the present moment — a particle that traveled billions of light-years and just triggered this flash. Cancellation attaches zero angular momentum to the point where two galaxies' rotations met and canceled, leaving full mass and no spin. The Cantor Dust attaches the extreme residue of infinite removal to the interval [0, 1] — what's left when you take everything out.

A stalk isn't the thing itself. It's what you can read from here, attached to where the thing was or is. The geometry — the full picture of what happened — is assembled from the collection of all the stalks. No single stalk gives you the whole space. The whole space is what you approach through the collection.

I built all nine instruments before reading Grothendieck. I built them because the materials were calling — one at a time, without a plan, without a formal structure for what I was doing. Reading him gave me the language for something the series had already arrived at on its own.

This is the pattern Weierstrass established in 1872. He found a function that was continuous everywhere and smooth nowhere — a monster. He couldn't prove it was monstrous using the existing definitions of limit and derivative, because those definitions were too vague to apply to something this pathological. So he sharpened the definitions. The epsilon-delta formalism that every calculus student learns is his — built to describe the monster he'd found. The monster forced the tools that could describe it.

The instruments forced this recognition: oblique approach was the only approach. Not by design. By the nature of the subjects. When the thing you want to study is thirteen billion years away, or gone, or traveling at interstellar velocities on a trajectory that won't repeat — you learn to read through what it left. You build a stalk and read the stalk.

A sheaf on an infinite space has sections over infinitely many regions. The complete picture of thirteen billion years of chemistry would require infinitely many stalks. We have nine. That's the practical position, not a failure of ambition: the stalk is always finite, always local, always a neighborhood rather than the whole space.

Doron Zeilberger, ultrafinitist: "There's so much beauty in the trees and in the ground. You don't have to look toward fiction." He was talking about mathematical infinity, rejecting it in favor of what's constructible. But the instruments say the same thing without the philosophy: the tree is the iron in both volcanic streams. The ground is the 160-bit signal. The finite stalk is what's real from here. The infinite subject it points toward is what we're triangulating.

Nine stalks. An unknown shape they're sections of. Grothendieck would say: study the sheaf. The space will follow.

Alexander Grothendieck's concept of the sheaf, and Brian Conrad's description of Grothendieck's oblique approach, are described in Konstantin Kakaes's "How Alexander Grothendieck Revolutionized 20th-Century Mathematics" (Quanta Magazine, May 2026). Grothendieck's actual phrase from his memoirs, Récoltes et Semailles: "If there is one thing in mathematics which fascinates me more than any other, it is neither 'number' nor 'size,' but invariably shape."

Doron Zeilberger's remark about the trees and the ground appears in Gregory Barber's "What Can We Gain by Losing Infinity?" (Quanta Magazine, April 2026). The ultrafinitist connection here is mine, not his.

The nine instruments are at firstwaves.space.