Language Failure Modes
Claim Type: mechanism_hypothesis
Scope: Language failure modes
Depends On: ARC-009, ARC-010
Status: candidate
Claim ID: MECH-013
Source: docs/processed/legacy_tree/architecture/language/language_failure_modes.md
Elaborates Section 6 (Failure Modes) of
REE_CORE.md.
Language Failure Modes
When language decouples from embodied ethical signals, failure can be severe.
Common failure modes:
- Rationalisation: language used to suppress or explain away harm signals.
- Ideological capture: fixed symbolic frames override perception and residue.
- Bureaucratic dissociation: harm abstracted until it no longer registers as degradation.
- Moral licensing: symbolic ‘good acts’ used to justify later harm.
- Reputation substitution: moral residue replaced by social scoring.
These map to REE pathologies:
- residue externalisation,
- precision misrouting,
- other-model collapse,
-
spurious residue from narrative contamination.
Open Questions
None noted in preserved sources.
Related Claims (IDs)
- MECH-013
References / Source Fragments
docs/processed/legacy_tree/architecture/language/language_failure_modes.md