[deanberris]

What this changes

A new argument, TestimonyArguments.SolaGratia, for #24 (part of #3). It covers the research, preparation, implementation and publishing sub-issues (#32–#35) in one pull request, because this session can push to only one branch. The bibliography entries are therefore not a separate PR. No existing result's verdict moved.

What it claims, and what it does not

"Salvation is by grace" divides no one: Trent (Session VI, ch. 5 and 8) and the Remonstrants (art. 4) both affirm it. So the conclusion has three parts:

1. salvation is by grace;
2. the grace that saves is effectual of itself, that is, it brings about the consent it seeks;
3. the will is passive in being made alive (Dort III/IV 10–12, Westminster X.2).

What the solver computes:

• Each position:

  • The Reformed package establishes the conclusion. Neither rival's does.
  • Both rivals reach part 1 (rivals_hold_salvation_by_grace).
  • Trent's premises deny parts 2 and 3 (trent_denies_effectual_grace, trent_denies_passivity). The Remonstrants' premises deny part 2 (remonstrants_deny_effectual_grace).

• The two strands: Ephesians 2 (τοῦτο read as pointing to the whole of being saved by grace through faith) and Romans 9:16 (read as being about salvation). Neither reading is load-bearing alone, and the two together are (touto_not_load_bearing, romans_reading_not_load_bearing, readings_jointly_load_bearing). Each strand also establishes the conclusion alone.

• Finding not designed for: the readings are widely granted, and the dissent is at the steps after them:

  • Robertson reads τοῦτο as Hoehner does, and calls faith "ours".
  • Calvin, himself Reformed, denies that this verse teaches faith as a gift.
  • Arminius grants that faith is God's gift, and that Romans 9 is about salvation, and denies that grace cannot be refused (gift_of_faith_does_not_settle_resistibility).
  • Aquinas grants each half of Trent's canon 4 and still holds grace infallibly effectual.

  So five steps are rated disputed, after a rating-auditor pass. The step that unpacks "God's work alone" into parts 2 and 3 carries no rating, and the docstrings say so.

What it does not do:

• It does not encode a Thomist position, which might reach the sharpened conclusion too.
• It does not encode John 6:44 as a third strand.
• It does not weigh the positions as a Dispute.
• "Faith is God's gift" is one claim here, although Dort and Arminius mean different things by it. The page says so, and a Dilemma would be the next step.

For the reader

• A new page, docs/src/arguments/sola-gratia.md, plus the PDF section, the roadmap table rows, a findings paragraph and the Near-term entry.
• A lay reader finds the verses quoted in English, Trent and Dort and Arminius in their own words, and every rating explained by the reader it rests on.
• The page has had a reader-review pass, and its findings are addressed in the second commit.

Library change: a new Tradition.arminian. Filing Arminius under Reformed or a generic heading would misattribute him.

Checks run

• lake build, lake build TestimonyTools, lake lint, lake exe axiom-audit (library)
• python3 scripts/testimony_lint.py, python3 scripts/test_testimony_lint.py
• the same in arguments/ (pinned counts in Checks.lean updated: 8 claim types, 208/105 postulate checks, 40 refuted packages, 210 entries), and scripts/kernel-replay.sh for both
• bibgen, statusgen, argdoc, argtex, each with --check
• examples/: lake build && lake exe exampletex --check
• mdbook build docs, tectonic arguments.tex
• read the generated page, not only the docstring

Sources

Six new entries. Every identifier and quotation was checked by source-verifier against a public scan or catalogue; unverifiable fields are left out.

• Schaff, Creeds of Christendom vol. 3 (Baker 1977 reprint of the 4th ed., CCEL).
  • Pages verified: Remonstrance 545–549, Dort English 581–597.
  • The printed ISBN catalogues as the three-volume set, so it is in the note, not an identifier.
• Arminius, Works:
  • Vol. 1 (London: Longman, 1825): pp. 600 and 681, read on the page images.
  • Vol. 3 (Auburn and Buffalo, 1853, trans. Bagnall): Analysis of Romans 9, pp. 527–565.
• Robertson, Word Pictures vol. 4. Cited by verse; the Broadman ISBN was verified at Open Library.
• Calvin, Galatians and Ephesians (Pringle, 1854): pp. 228–229.
• Abasciano, JETS 49/2 (2006) 351–371. No DOI exists.

Trent is cited from Tanner by canon and quoted in Waterworth at his page. Lincoln, and Moo on Romans 9:16, could not be verified, so they are not cited. Hoehner p. 343 is known only from a quotation of it.

---

Generated by Claude Code