diff --git a/docs/ADR/ADR-0008-adaptive-query-routing-intent-history.md b/docs/ADR/ADR-0008-adaptive-query-routing-intent-history.md
index bf70fd4..31225a8 100644
--- a/docs/ADR/ADR-0008-adaptive-query-routing-intent-history.md
+++ b/docs/ADR/ADR-0008-adaptive-query-routing-intent-history.md
@@ -114,6 +114,8 @@ classify(q, history) ≠ f(dominant_type(history))
 classify(q, history) = f(intent(q), resolve_references(q, history))
 ```
 
+**Rationale:** Small LLMs implicitly compute `P(type | history)` instead of `P(type | message_content)`. The distribution of previous intents must not influence the prior probability of the current classification. Each message is an independent classification event — a session with 10 `RETRIEVAL` turns does not make the next message more likely to be `RETRIEVAL`. The `<history_rule>` in the classifier prompt enforces this explicitly.
+
 ### RC-04 — RAG bypass (priority: medium)
 
 Query types that bypass Elasticsearch retrieval: