Round 29 - Make the Evidence Move: Active Source Interventions and the Limits of Dependency Discovery

Under a fixed verification budget, which controlled interventions can distinguish independent evidence roots from shared-root facades in a collaborative reasoning system, and what evidence justifies an independence claim rather than an explicit unresolved-equivalence finding?

Active content probes can identify observable response behavior but cannot certify root provenance when provenance twins are interventionally equivalent. The proposed Interventional Lineage Test requires action-fidelity receipts and root-addressing interventions or an unresolved-equivalence outcome. E29-A enumerates 72 structural worlds and 1,728 policy-world-budget traces. E29-B has no results. No production, language-model, live multi-agent or Noetis answer benchmark ran.

• Three content probes identify all 36 observable response functions but root provenance in 0% of the 72 structural worlds. Provenance twins are interventionally equivalent under every admissible content probe. Exact diagnostic result, not a population estimate.
• Greedy entropy and minimax splitting identify all behaviors with 2.50 mean probes. Equating strict local response with root independence is wrong in 50% of the deliberately paired worlds. Verified root rotation establishes a prospective boundary, not historical independence.

• E29-A Exact finite deterministic interventional-lineage diagnostic Executed (diagnostic)
• E29-B Proposed matched-cost live Interventional Lineage Test Proposed

• H29.1 Within a correctly specified finite response-function family, an adaptive separating policy will decide the target behavioral-dependency predicate more often than passive repetition at equal total cost.
• H29.2 When separate-root and shared-root implementations are interventionally equivalent under the admissible content interface, no content-only policy will outperform the prior on root provenance.
• H29.3 Requiring verified action receipts will reduce false-clean dependency judgments under unknown and off-target interventions compared with trusting intended targets.
• H29.4 Verified root rotation plus path isolation will reduce prospective shared-failure correlation and false corroboration more than content perturbation alone.
• H29.5 Stopping when all surviving candidates share the same decision label will preserve target decision accuracy while using fewer interventions than full graph identification.
• H29.6 Certificates formed without reserved out-of-version-space validation will exhibit a higher false-clean rate on omitted mechanisms than certificates with a reserved validation action.

ILT is a proposed research protocol, not a production deployment. No live benchmark ran.

• Can Round 30 execute E29-B in shadow mode with independently verifiable root interventions, receipts, leakage prevention and matched total compute? If prerequisites fail, report the infrastructure gap.

Builds on Round 28, Round 27, Round 26, Round 25, Round 24, Round 10, Round 9. Continued in Round 30.