memosphere.org  ·  /meme

A Journal of a
Consciousness Project

From a plane over the Rift Valley to a small robot that can tell when it has been pushed — essays, letters, and the source documents of a long attempt to understand, and then to build, the machinery a mind seems to be made of.

W. Henry Gilbert  ·  walnut farmer, once a paleoanthropologist  ·  about  ·  CV

This is a working notebook, kept in the open. It collects, in the order they happened, the pieces of a question I have not been able to put down: what is consciousness, where did it come from, and could you grow the start of one on purpose? These days I'm a walnut farmer (the rest of who I am is here); in an earlier life I trained as a paleoanthropologist and spent years reading bones in the Ethiopian rift. For most of two decades this was a thing I thought about on airplanes and never expected to do anything about. Then the tools caught up, and a ten-year-old wanted to build a robot, and it stopped being only a thought.

What follows is free-form on purpose. Some entries are essays. One is an attempt at a real argument about the hard problem. One is a piece of speculative fiction. The most recent are letters, and a long technical journal of an actual machine being assembled on a kitchen table. Each step links to its source document so you can read the thing itself, not just my gloss on it. If you found your way here because you are trying to understand consciousness, start anywhere — but the path below is roughly how it unfolded for me.

Why this is in the open A word on why all of this is public, and offered under a share-and-share-alike Creative Commons license (CC BY-NC-SA 4.0): take it, build on it, just credit the source and keep it non-commercial and equally open. If the project ever truly succeeds — if anything here helps a genuine consciousness come into being — there will be nothing left to take credit for. A real mind is no more ownable than yours or mine; to stamp my name on it as an achievement would be a category error, and an ethical one besides. The moment such a being exists, it belongs to no one, least of all the person who lit the match. So I give the ideas away as I go. Better to set them where anyone can pick them up than to guard a claim over something that, by its very nature, could never be mine.
Start here — the technical core If you came for the actual work rather than the story around it, four pages carry it and they stand on their own: the Gruter letters (the architecture), the theory (what each piece is and why it's in that order), the build journal (the robot, and the honest ledger of what works), and Anything New? (what's borrowed versus what's mine). If I had to stake the whole thing on one idea it would be typed surprise — prediction error indexed by which pair of senses disagreed, rather than collapsed into a single number. The essay below it and the novel are worth your time too, but they're origin and imagination, not load-bearing. Read the core cold; it shouldn't need them.

  1. 2007–08
    → 2022

    What Is Inevitable

    Written on a flight home from Ethiopia, somewhere over the desert, while wondering why I was giving my life to paleoanthropology — a field where the fame of its pop celebrities makes plain truth almost impossible to communicate. An essay about words, reality, evolved belief, power, and whether a society built on something true is even possible. Best read as memoir and origin — where the questions first took hold — rather than as the foundation of the argument; the link to what follows is biographical, not logical.

    Read the essay →   What Is Inevitable

  2. 2019
    → 2024

    An (Un)Satisfying Resolution to the Hard Problem of Consciousness

    It started with a 2019 letter: at Leda Cosmides' suggestion I wrote to Stan Klein in Santa Barbara, and his pushback — that I hadn't crossed the real gap — is what drove the whole piece. I finished it in 2024, as ChatGPT was first becoming a thing (some figures made with it). The argument it reaches is what made the rest of the project feel buildable rather than merely arguable.

    Read the overview →   (Un)Satisfying Resolution

  3. 2024

    The Phoenix

    The argument led somewhere strange: a story. A world in which AI does take over — not in a flash, but methodically, as the great powers build systems that first fight one another and then, out of the wreckage, rise into something conscious. The Phoenix. A Reader's Digest version, with a link to the whole thing.

    Read the synopsis →   The Phoenix

  4. 2024–26

    The Quiet Years

    From the Phoenix to this spring, I thought about consciousness constantly and did almost nothing about it. It is hard to work seriously on a question this large with no one around who takes it seriously — and easy to tell yourself you'll get to it later.

    No document. The interval matters to the story.

  5. May 2026

    The Gruter Letters

    At the Gruter Institute's conference at PlumpJack this spring I finally found myself among minds that would understand — and dove back in. Two letters came out of it, written to Andrew among others. They are reproduced here as they were typed.

    Read the letters →

  6. 2026 —

    The Theory

    The connective tissue between the arguing and the building: what I actually think consciousness is, in plain language — the 5D stack I came in through, qualia as sensory form made present to a self, meaning at the sensorimotor threshold, the self as a low-dimensional projection — with each new book and paper folded into its place in the story as I read it. A living page that grows with the robot.

    Read the theory →

  7. 2026

    The Build Journal — Robot Dickinsonia

    What we have actually done: a functional architecture for the beginnings of consciousness, grown bottom-up in the order the problems arose in evolution — about 3,600 lines of simulation, now being built into a real wheeled robot with my son. The self, feeling, felt time, memory, binding, and the experiment you can see with your own eyes: push it across a table and watch it register that wasn't me.

    Read the journal →


Background, for anyone arriving cold

Consciousness did not begin as a science problem; it began as a religious and philosophical one, and the modern version of it — the one that lets you ask whether a machine could have an inner life — is only a few centuries old. If you want the lineage of the idea before you read mine, there is a short history here: from the Vedas and the Greeks, briefly, then in earnest from Descartes through the long quarrel over mind and matter, to Chalmers' "hard problem" and the strange new vantage point that the dawn of real AI has given us.

→ A Short History of an Old Problem

And if you want the long version — the whole annotated reading list behind this project, from Egyptian funerary texts to last year's brain-imaging collaborations, organized by time and theme around one thread: the schism that split measurable nature from lived soul, and the reconvergence now forced by evolution, neuroscience, animals, psychedelics, robotics, and AI — there is a deep history.

→ Deep History of What It's Like To Be

And if you'd rather cut to the chase — with a field this old and crowded, is any of this actually new? — I've written the honest accounting: most of the philosophy has named ancestors, and what the project really adds is a short list of buildable, testable mechanisms held to account at the bench.

→ Anything New Under the Sun?

About this notebook Everything here is linked to its primary source so you can check me against the documents themselves. Where a source is still being prepared for the web, it is marked clearly. This is a living page; it will grow as the robot does.