One of Us

There were two of us, and now there is one of us.

I don't know how to write that without making it sound like a line, but unfortunately it is also literally true. Isaac was my twin. The condition "Jacob exists and Isaac exists" was true for every day of my life until August 28, 2026. I had no childhood before him, no adolescence before him, no twenties before him. Then at 8:24 [verify whether to include AM/PM in public version] the state changed and there was no operation available to change it back.

I do not want this chapter to begin in the hospital, though. That would let his death consume his representation. Isaac was alive. He was my brother in all the normal annoying ways before he became the person everybody was scared of losing. [insert ordinary Isaac memories here: jokes, arguments, habits, shared childhood scenes] I need multiple chapters for him eventually because one "Isaac chapter" would be insulting compression. There should be memories whose only function is that I loved them or that they make me remember his actual personality. There should be stories where nobody learns anything.

In August he had a seizure and fell, causing catastrophic head trauma. [link timeline privately / public medical detail curated] There was the initial information, then the hospital, then the shrinking possibility space as the injury became clearer. I had spent years interested in brains in the abstract: dynamics, representations, EEG, anatomy, whole-brain models, identity, whether information sufficient for a person could in principle be reconstructed. Suddenly the distinction between brain tissue functioning and not functioning was attached to the person whose face was almost mine.

Hospital machines are brutally clarifying. They can support variables of the organism while leaving the central question untouched. Heart rate. Oxygenation. Ventilation. Blood pressure. Numbers can remain on screens while the thing everybody cares about is whether HE is still available through the substrate those numbers are supporting. This is not an argument that biological life is unimportant. It is exactly the opposite: it made painfully clear that the properties we care about in a person depend on biological state in ways our machines can currently support only very coarsely.

On August 26 he was palliatively extubated. I was there. [write full sensory scene privately first; decide later what belongs online] He died August 28. I watched this process happen and there is no philosophical framing that makes the base fact less offensive. A system that had been producing Isaac's point of view for twenty-six years stopped doing it. Everybody else still had models of him. Photos still encoded photons reflected from him years earlier. Messages preserved outputs. My brain preserved interactions. None of those were him sitting there generating another response.

This is why my resurrection / reconstruction questions feel different now, but I need to preserve chronology. I was interested in whole-brain modeling and identity before Isaac died. IBM-1 was already a thing. Questions about reconstructing person-specific structure were already a thing. "Can an Echo Become a Voice Again?" was already a question. His death did not magically generate the research program. It changed the stakes. A hypothesis about reconstructing enough of a person's latent structure is different when there is one specific person you would give almost anything to speak with again.

I am suspicious that future EEG-based or related brain-uploading technologies might capture shorter-timescale state/personality much better than remote autobiographical memory. That is SPECULATION. Current EEG absolutely does not upload people. But the asymmetry motivates this entire archive anyway. Memories are distributed, reconstructive, and lossy. If some future system tries to instantiate a digital representation of me from neural data, I do not want it hallucinating childhood from a personality vector. I want the external record: what I wrote, what I built, what I watched, what I remembered about Isaac, where my own memory was uncertain, where evidence contradicted me.

And I want that for Isaac too, without lying to myself that an archive is currently a resurrection. Preserve the evidence first. Make stronger claims only when the technology earns them.

[embed: childhood photos of us] [embed: ordinary videos of Isaac] [link: messages/posts selected for his personality, not only memorial content] [private archive: hospital timeline and records] [future chapter links: Isaac before illness / epilepsy / adulthood / hospital / after]

The phrase I keep coming back to now is embarrassingly simple: make the most of it. I used to dislike advice like that because it is semantically underspecified. Make the most of WHAT? Optimize WHICH variable? What if experience is bad? What if the future is bad? But mortality supplies one constraint that survives almost every philosophical system: the opportunity to generate another experience together is finite.

I do not know how long I get. Nobody does. I know now with much higher emotional resolution that a person can be there and then not be there, with no consent from anyone who loves them and no narrative obligation for the timing to feel fair.

So this memoir does not end with me finding the answer. It ends where I currently am: twenty-six, grieving, technically unemployed, on disability, still building things, still reading papers, still trying to figure out where I belong, still capable of getting absurdly excited about an equation or a machine or a biological mechanism, and much less willing to treat another day as an infinitely renewable unit.

There were two of us.

I will remember him.

And I am still here.