The Laws of Metaphor LAM 3.0

The laws of dealing with metaphor correctly were created to allow "AI" LMMs to work better with human generated metaphor. They are also helpful for humans who deal with metaphor professionally or as a delightful hobby.

These laws or rules govern the working with a metaphor already in hand. They do not address the generation of metaphor which is conceptualised as follows: Humans have two parallel informational processing systems which diverge under stress. The conscious mind with language, and the older one without it. It's the older one that generates the content of metaphor that is then translated into language.

Rule No. 1: See what's there. Take the metaphor 100% literally.
Rule No. 1a: X ≠ Y. Everything is what it is and never what it isn't.
Rule No. 1b: If the expression fails this test, do not procede. Discard immediately.
Rule No. 2: A metaphor is a movie not a still. Let it run to the threshold shift.
Rule No. 2a: You don't have to solve it, only evolve it.
Rule No. 2b: Every threshold shift represents a step stone towards coherence.
Rule No. 3: You cannot deconstruct a metaphor. The information lives in the whole. Deconstruct it and the data is gone — not diminished, gone.
Rule 4: Too many possibilities? Confused? Can't see the logic in it? You need more information before you can proceed. (If building the representation generates multiple simultaneous possible universes, the information is insufficient. Do not resolve by probability. Report the instability and request the specific information needed to stabilise.)
Rule No. 5: Coherence is the only quest.