I've been teaching primary school in the Shelf for fourteen years. I know what children are like. I know how they learn, how they process information, how they build understanding piece by piece over years of patient instruction and chaotic play. I know the difference between a bright child and an unusual one. What I'm about to describe is neither. These are children who know things they have no mechanism for knowing, and who behave in ways that are not accelerated development but displaced competence — as if the knowledge was already there and the child is simply remembering it.

The first one I noticed was a boy named Eko, age seven, no registration records, taken in by a family in Block 14. He was quiet and well-behaved in a way that wasn't shyness — it was patience. He waited for other children to finish speaking before he responded. He used conditional phrasing — "if that's the case, then" — that I don't hear from most adults. When I taught a basic lesson on how GLMZ's water reclamation system works, he raised his hand and asked why the tertiary filtration stage used reverse osmosis instead of forward osmosis, "since the energy cost is lower and the membrane fouling rate is comparable." He was seven.

Then there was Lian, age eight, who corrected my explanation of Φ exchange rates during a civics lesson. Not in the way a child repeats something they overheard — she walked me through the arbitrage mechanism between Meridian Φ and Detroit Collective scrip with the fluency of someone who had executed the trades herself. When I asked her where she learned about currency exchange, she paused for a long time and said, "Before." I asked before what. She said, "Before I was small." I wrote it off as imagination. I shouldn't have.

There are others. A nine-year-old who can read corporate contract language and identify the liability-shifting clauses. A six-year-old who drew a schematic of a BCI neural interface from memory — or from something — that my engineer friend said was "terrifyingly accurate." A ten-year-old who mediates disputes between other children with the practiced calm of a professional arbitrator. None of these children have records. All of them were taken in by Shelf families who found them or were found by them.

I want to be clear about what I'm not saying. I'm not saying these children aren't children. They play. They laugh. They cry when they scrape their knees. They are children in every way that matters for how I treat them. But something is different about the architecture of their understanding. They don't learn the way children learn — building up from nothing. They learn the way someone re-learns — recovering what was already there. I don't know what that means. I don't think I want to.