How this apparatus is made
Generation and review
Each verse's apparatus — word-by-word parsing, textual notes, lexical and syntax discussion, Old Testament substrate, cross-references — is drafted with AI assistance (Claude, via the Anthropic API), working from the Greek or Hebrew source text and a fixed set of cited primary tools: lexicons, grammars, critical editions, and commentaries, listed under "Built on" at the foot of every verse. It does not consult or match the wording of any existing English translation.
Every verse carries a status, shown at the top of its page:
- Draft
- AI-drafted, not yet checked by a human editor. This is the default and current state of most of the corpus — draft content is real, sourced work, not a placeholder, but it hasn't been verified yet.
- Reviewed
- Checked against the sources it cites.
- Final
- Settled.
A verse is never promoted out of draft automatically — that's a human decision, made by reading it, every time.
The approach behind the approach
This apparatus tries to render what a text says with as much precision as the Greek and Hebrew allow, without collapsing genuine ambiguity into false certainty. That is itself a stance on how Scripture asks to be handled — Jesus and the Scriptures traces where that stance comes from: his own first-century relationship to the text, argued from the Gospels themselves rather than assumed.
The foundation is always the original language: Greek for the New Testament, Hebrew for the Old. Early translations, the Septuagint above all, are valuable witnesses and are discussed alongside the original, but a translation never stands in for the text it translates.
Source text and licensing
The Greek text is the SBL Greek New Testament (SBLGNT), © 2010 the Society of Biblical Literature and Logos Bible Software, used under a Creative Commons Attribution 4.0 International License. SBLGNT differs from the Nestle-Aland/UBS text in several hundred places, documented in its own apparatus; where a verse's own textual notes discuss a variant, they say which edition supports which reading.
The Hebrew text is the Westminster Leningrad Codex (WLC) — the text of the Leningrad Codex (AD 1008/1009) — as published by the Open Scriptures Hebrew Bible project. The text of the WLC is in the public domain; the project's lemma and morphology data are licensed under a Creative Commons Attribution 4.0 International License, credited to the Open Scriptures Hebrew Bible Project. Verse numbers follow English Bibles; where the Hebrew numbering differs, the Hebrew number is shown beside it.
Feedback
Spotted an error, or have a question or a request? Use the Feedback button at the top of any page. Your note goes to the people who make this site, along with the address of the page you were on. If you leave an email address it is used only to reply. To keep spam out, a short-lived counter based on a scrambled form of your connection address is kept for a few hours; the address itself is not stored.