**On Historical Logos v0.2: this is a substantial advance.** It now carries an operative ontology, eleven argued bridges, a comparative one-person argument, station-specific erasure findings, and current standing attached to the older moves. I checked that the new rival quotations occur in their saved source texts, the source-page hashes match, and the bridge and current-standing references resolve.

The remaining work is more specific.

1. **Make each bridge’s endpoints match the evidence actually supplied.**

   This is the most important remaining repair.

   - `bridge:key` runs **Philo → Peri Hypsous** and is typed `documented_reception`. Its strongest evidence establishes the co-presence of **Sappho and Genesis** in the treatise. Its own supporting passage says reception through Philo or Caecilius remains unsettled. Preserve the documented conjunction; give the possible conduit its separate standing.
   - `bridge:into-the-record` starts at **Peri Hypsous**, but its stated historical sequence and quoted crux are **Catullus → Josephus**.
   - `bridge:portable` starts at **the Apocalypse**, while its supplied argument concerns **Sappho’s sensory sequence and the generalized socket in Augustine**.

   The new bridge structure makes these discrepancies visible—which is progress. Let the topology branch where the argument branches. An order of exposition can remain a sequence while the evidentiary relations take a different shape.

2. **Carry Philo’s modular historical status into the erasure finding.**

   The bridge carefully distinguishes the structural mapping from Sappho’s unestablished material presence in Philo. The erasure row then states, without that distinction, that attribution “dies at the crossing.”

   Separate the propositions:

   - the name is absent: a documented textual finding;
   - the structural mapping survives: the reported mapping result;
   - this is a crossing at which provenance was lost: the genealogical interpretation.

   Keep the third at the strength the argument gives it. Its modularity needs to travel with it into the erasure layer. Otherwise the bridge preserves a distinction that the next representation removes.

3. **Give the one-maker argument a few of its strongest positive demonstrations.**

   The line is now being held: **one composing person**, explicitly distinguished from a collective, and argued comparatively rather than merely attributed to your belief. That fixes a major problem.

   But several steps still point to conclusions—“explains the set,” “nothing requires two”—rather than exposing the particular configuration that earns them.

   The argument layer already contains useful material: `a1576-four-props` and `a1576-one-word-moves`; `a1658-combine-divide` and `a1658-one-operation`; the catalogue patterns in #1733. Select a few and show the actual comparison: what one maker explains through their conjunction, what the several-maker account must posit, and where those explanations diverge.

   This is a selection-and-connection task within the material already collected. It would make the positive case survive compression alongside the critique of the received count.

4. **Keep the rival’s documented statements distinct from your diagnosis of their combined effect.**

   The source work improves the ledger considerably. But the Catholic Encyclopedia supplies a **1910 theological account**; the Wikipedia material supplies contemporary encyclopedic representations. Together they do not automatically constitute one demonstrated current scholarly position.

   There is also a distinction inside the Catholic account worth carrying: it calls the Logos eternally divine and describes incarnation in a concrete person. Your diagnosis—that the interpretive endpoint collapses into biographical personhood—is an argument about what that account and its later reception *do*. It should have its own explicit inferential step.

   Keep three objects visible: the source’s proposition, its treatment of incarnation, and your critique of that treatment. That gives the central disagreement a precise target without surrendering its force.

5. **Run scope C, and reconcile the changed rung-VI dependency before scoring it.**

   Scope C is now much closer to the instrument this needs: reconstruct arguments, alter premises, distinguish refutation from flattening, and require the central commitments to survive.

   But the fifteen reported structural passes do **not** execute those new consequence tests. The supplied back-projection prompt still runs scope A.

   One consequence also records a substantive revision: **rung VI now survives removal of the Josephus thesis and loses a strengthening**, attributed to your 12:48 ruling. The earlier joined Revelation kernel made VI dependent on V. Preserve the new ruling as a dated amendment, include its actual statement, and reconcile the joined representations. The test should then enforce that revised dependency consistently.

A final technical repair: the new reference-bearing fields—`from`, `to`, `move`, `operation_at_from`, `operation_at_to`, and `supporting`—need appropriate JSON-LD identifier handling. The existing T19 checks the main edge array; it does not establish that these newly added references expand as intended.

**The immediate next pass should align those bridge endpoints, strengthen the selected positive maker argument, and then execute scope C against the frozen package.** Those checks will tell us whether the substantial improvements now survive an actual reconstruction.