Link contradicting ideas together in your graph — spatial proximity forces the confrontation that compartmentalization prevents

Why This Is a Rule

The mind's default strategy for contradictions is compartmentalization: keeping conflicting beliefs in separate mental contexts so they never confront each other. You believe "people are rational agents" in your economics notes and "people are systematically irrational" in your psychology notes, and neither belief is troubled by the other because they never share a page. This is intellectually comfortable and epistemically dangerous.

Knowledge graphs can either enable or defeat compartmentalization. If contradicting ideas live in separate clusters with no connecting edge, the graph mirrors and reinforces mental compartmentalization. If they're explicitly linked with a "contradicts" relationship, every traversal that touches either note surfaces the contradiction. You can't encounter one without the graph reminding you of the other.

The confrontation is productive. Encountering a contradiction forces one of three resolutions: (1) one belief was wrong — update it, (2) the beliefs apply to different scopes — articulate the boundary conditions, (3) the contradiction is genuine and unresolved — hold it explicitly rather than hiding it. All three outcomes improve your epistemic state. Compartmentalization produces none of them.

When This Fires

• When you notice you hold contradictory beliefs in different domains of your knowledge system
• When adding a new note that contradicts an existing one — link them immediately
• During graph audits when looking for beliefs that should conflict but don't touch each other
• When using typed edges (Typed edges enable AI reasoning; embeddings only find proximity — label relationships explicitly) — "contradicts" is one of the most valuable edge types

Common Failure Mode

Filing contradicting ideas under different categories and never cross-referencing them. Your "management" notes say "give people autonomy" and your "quality assurance" notes say "standardize processes." These ideas genuinely tension each other, but if they're in separate folders with no cross-link, you'll advocate for whichever one the current context primes — never confronting the trade-off.

The Protocol

(1) When you create a note that contradicts an existing note, create an explicit "contradicts" edge between them. (2) Add a brief annotation explaining the nature of the contradiction: scope conflict, value conflict, empirical disagreement, or logical incompatibility. (3) During periodic graph reviews, search for beliefs that should conflict but aren't linked. Same-topic notes making opposite claims are candidates. (4) When traversal surfaces a contradiction, attempt resolution: is one wrong? Do they apply to different scopes? Is the tension genuine? (5) Document the resolution (or the explicit non-resolution) as a note linked to both — this is where some of your deepest thinking lives.