Before there's space, there's just a list.

Strip away space and time, and what's left to build a theory out of? SFT's answer starts as plainly as possible: a set. Not points arranged in space — there's no space yet — just a finite collection of distinct configurations, sitting in no particular order and no particular arrangement. Call this collection S. Nothing is "next to" anything else yet. Nothing has a position. There's just a list of things that are each distinct from each other, full stop.

That bare-set picture is deliberately minimal — almost too minimal to build anything from. So SFT proposes a concrete candidate for what these configurations actually are: elements drawn from the octonions, an extension of the ordinary number system that shows up surprisingly rarely in nature but has a distinguished mathematical pedigree — real numbers extend to complex numbers, which extend to quaternions, which extend one final time to the octonions, and then the pattern provably stops. There are exactly four number systems like this, no more. SFT proposes the substrate is built from this last, richest one.

Why reach for something so specific? Because a bare set alone doesn't have enough internal structure to eventually produce anything as textured as particles, forces, or four-dimensional spacetime. The octonions bring built-in structure along with them — and the hope is that this structure is rich enough to do that work later, without having to bolt on extra assumptions by hand as the theory develops.

That's the proposal, not a proof. Which configurations SFT is actually built from — and specifically, the exact rule for how they relate to one another — is an assumption the theory makes, not something derived from anything deeper. It's a considered starting point, not a free-floating guess: a competing candidate for the substrate was tested and formally ruled out. But "postulated" is the honest word for where this stands, and it stays that way going into Step 3.