<Cognitive Axioms 04_2>

The Absence of Transformation

26w391v171_0921 Bs2FoB

1. Reference and transformation

When you pick a room, you hear about a room a friend once lived in. Was the rent about the same? How far was it from the station? Was it cold in winter? That story helps you set up the items and scales for weighing your own room. But you can't carry your friend's room over as it is. The floor is different, the windows are different, and above all your situation is different. The first is reference; carrying the whole room over would be transformation.

Reference and transformation are different things. Reference means taking an earlier case and consulting it while you set up the axes of a new problem. Transformation means carrying a single substance over into another coordinate system and reproducing it exactly. The gap between the two is not small.

For an axis transformation to hold, three conditions must all be met: fixed axes, an assumed substance, and a demand for identical computation. Physical tensors (<Cognitive Axioms 04_4>) need axis transformations because they posit a hypothetical substance and must secure coordinate independence — the substance has to come out the same whichever axes you observe it from. This is transformation at the formal level, where identity is guaranteed by definition. On paper it holds perfectly.

2. Reality has no transformation

Transformation and reference rest on identity and similarity. Transformation demands identity with no gaps; reference gets by on similarity.

In reality, gapless identity never holds. Complex systems block the reproduction of the same conditions through emergence, nonlinearity, and context dependence. Entropy blocks the preservation of the same state. And every problem's goal is its own. Even formal transformation becomes reference the moment it's applied to the real world. The only thing real cognitive computation actually uses is reference.

Building a new wooden footbridge while looking at the stone bridge downstream.
Building a new wooden footbridge while looking at the stone bridge downstream.

3. Two paths for handling a problem

Narrow it down to how people handle problems and it gets clearer. Faced with a new problem, cognitive computation has only two paths. Set up new axes on a blank page, or set up axes by consulting earlier cases.

Each problem forces its own axes, so the condition transformation relies on — the same substance preserved while only the axes change — simply doesn't come up in handling problems. As <Cognitive Axioms 04_1> showed, human problem-handling doesn't posit a substance, so coordinate independence never becomes an issue. And a tool for solving a non-issue — the axis transformation formula — isn't needed either.

4. What fills the gap

What fills the space left by transformation is reference, and setting up new coordinate systems. Reference is borrowing, not moving. The old problem's axes stay with the old problem; the new problem looks at them and then sets up its own.

Keeping the old map book open while drawing a new map.
Keeping the old map book open while drawing a new map.

5. What can't be carried over gets borrowed

Complex systems make axis transformation meaningless and force reference. Optimization forces new coordinate systems, since the objective function and constraints differ from problem to problem. When a new problem comes up, the situation forces the axes, the problem is solved on those axes, and earlier problems are consulted. Axis transformation has no place in this structure.

Then what are the precise references that look like transformation? Reference has layers. <Cognitive Axioms 04_3> deals with them.

What can't be carried over gets borrowed.