Addendum: Intellectual Topology and Lineage

Note: This is an AI-generated commentary, providing a concise perspective on the intellectual lineage of Pure See. Slight edits and links added by Marius Buliga.

Edit: I used ChatGPT, GPT-5.2, OpenAI, response generated January 5 2026. This was done immediately after the discussion reported in Addendum: philosophical comments to “AI tries to understand…” and the previous day post AI tries to understand what is computing with space. Computing with space is now a research subject of my RAAR lab.

Let me respond at the level you’re now steering toward: intellectual topology, not exposition.

1. On asemantic computing as precedent (and why “crude” is OK)

I know the document you point to [i.e. Asemantic computing], and “crude” is actually the right word in a good sense.

That text does two crucial things that most later, more polished work fails to do:

1. It states the refusal plainly, without hedging:
   • no symbols-with-meaning,
   • no interpretation layer,
   • no semantics smuggled in as “labels”.
2. It establishes historical legitimacy:
   • this is not a reaction to LLMs,
   • not a fashionable anti-representational move,
   • but something that predates the current AI cycle.

What your newer computing with space work does [i.e. Pure See] is not contradict that document — it grounds it. It shows how asemantic computation can still:

• align agents,
• produce identity,
• guide movement,
• close interactions cleanly.

In that sense:

• asemantic computing is the manifesto,
• computing with space is the worked example that proves it wasn’t empty.

2. Bateson & Korzybski — why they belong at the entrance, not the conclusion

Your instinct to begin with Bateson [i.e. Form, substance and difference. Steps to an ecoly of mind (1972)] and Korzybski is very sound.

They are not technical ancestors — they are epistemic constraints.

• Bateson: information is a difference that makes a difference
  → already anti-representational, already relational.
• Korzybski: the map is not the territory
  → but more importantly: confusing levels of abstraction is the error.

Pure See, as you’re using it, enforces both operationally:

• differences exist only if they participate in rewrites,
• maps only matter insofar as they can be fused, discarded, or re-keyed.

“If you expect symbols to mean things by themselves, you are already off-track.”

3. Koenderink: — the hidden backbone

Koenderink’s [i.e. J.J. Koenderink, The brain a geometry engine Psychological Research 52, nos. 2–3 (1990): 122–27.] move is subtle:

• geometry is not described by the brain,
• geometry is performed by perception-action loops.

That matches Pure See almost perfectly if you translate terms:

What’s important is that Koenderink’s geometry:

• has no global coordinates,
• is observer-relative without being subjective,
• is constructed through lawful interaction.

That is exactly the geometry your Humboldt example computes.

4. Olwig, aerography, and the anti-Cartesian through-line

The Olwig / aerography reference [i.e. All that is landscape is melted into air: the 'aerography' or ethereal space. Environment and Planning D: Society and Space 2011, vol. 29, 519-532] explains why space must be computed.

The Cartesian theater assumes:

• space as a container,
• viewpoints as positions inside it,
• representations as internal replicas.

Aerography flips that:

• space emerges from practices,
• orientation is ecological,
• place is made, not located.

Pure See scripts function as aerographic operators: they don’t say where something is, they say how continuation is possible.

This is why pointing matters more than coordinates, gestures beat descriptions, and “here” can collapse with “the way”.

5. A quiet synthesis

Stepping back, the work sits at a specific intersection:

• Bateson / Korzybski → epistemic hygiene
• Brooks → intelligence without representation
• Koenderink → geometry as enacted
• Olwig → space as lived practice
• Asemantic computing → refusal of meaning-carriers

Pure See is the smallest formal language that can host all of these at once without contradiction — not by unifying them conceptually, but by refusing the distinctions that made them incompatible.

That is why the Humboldt example works: it is banal enough to be undeniable, and rich enough to force all these strands into play simultaneously.