Existence Layers Do Not Imply One Another
Abstract
Finite countermodels separate type, interface, causal, and record existence.
Theorem 1.1 (Type existence does not imply distinguishable existence).
Proof. Machine-checked in Lean as D5/S3/Observer/Existence/ExistenceLayersDoNotImply.type_existence_does_not_imply_distinguishable_existence (✓ std3). ∎
Source. Repository-derived.
Commentary.
Two Boolean states are genuinely different, so their distinction exists at the type level. A constant readout into the one-point type sends both states to the same output, so the interface cannot distinguish them.
Theorem 1.2 (Distinguishable existence does not imply causal existence).
Proof. Machine-checked in Lean as D5/S3/Observer/Existence/ExistenceLayersDoNotImply.distinguishable_existence_does_not_imply_causal_existence (✓ std3). ∎
Source. Repository-derived.
Commentary.
The identity readout separates the two Boolean states at the present time. A constant update maps both states to false after one step, so every positive-time readout agrees and the distinction has no causal existence.
Theorem 1.3 (Causal existence does not imply record existence).
Proof. Machine-checked in Lean as D5/S3/Observer/Existence/ExistenceLayersDoNotImply.causal_existence_does_not_imply_record_existence (✓ std3). ∎
Source. Repository-derived.
Commentary.
Identity dynamics preserve two distinct Boolean states, and the identity readout separates them after one positive-time step. A constant record into the one-point type is stable under those dynamics but assigns the same record to both states, so record existence fails.
References
- Truth anchor:
D5/S3/Observer/Existence/ExistenceLayersDoNotImply.causal_existence_does_not_imply_record_existence - Truth anchor:
D5/S3/Observer/Existence/ExistenceLayersDoNotImply.distinguishable_existence_does_not_imply_causal_existence - Truth anchor:
D5/S3/Observer/Existence/ExistenceLayersDoNotImply.type_existence_does_not_imply_distinguishable_existence