How this site works

Seek is for readers who want to learn Scripture and its teachings. Question is for readers who want to test it. They differ in order and tone, never in facts: a caveat, a dating problem or a weak piece of evidence appears in both.

Text

What the biblical text itself says, checked against the editions below.

Tradition

How Jewish and Christian traditions (and sometimes Islam) have read it, in their own terms.

Evidence

What history and archaeology have actually found. Each claim is judged on its own, never as a verdict on the whole story.

Scholarship

Inference from the evidence (who wrote it, when, what borrowed from what), where scholars hold different positions.

1. A researcher drafts from primary sources and the original languages, citing every claim.
2. An independent fact-checker re-reads every source, re-runs every count against our database, and corrects or removes what doesn’t hold.
3. A historian adds manuscripts, art, liturgy and other links; a scholar checks that every tradition is represented fairly.
4. Pages say plainly when they are still drafts. Nothing quotes Wikipedia or an AI summary as a source.

Know the Text is built from the original texts and published scholarship. AI tools help us search those sources and check every citation. Every claim links to a source you can check, and each page is reviewed before it’s published. Short answers marked ‘AI summary’ are written on the spot from our sources.

Some searches show a short AI summary of our sources. A small AI model (Claude Haiku) is given only material from this site (verses, the reference works, our notes and stories) and must cite it on every sentence. The site then removes any sentence that cites nothing, or contains a name, number or verse reference its sources don’t contain. The sources are always listed with it. It can still be wrong, so it is labelled, and the pages it cites are the real answer. Questions are counted anonymously so the most-asked can become researched pages, which are written and fact-checked like the rest of the site.

Searches, best matches and links are chosen by plain rules, not by AI.

Topic pages draw on Easton's Bible Dictionary (1897), Smith's Bible Dictionary (1884), International Standard Bible Encyclopedia (1915), Nave's Topical Bible (1896) and Torrey's New Topical Textbook (1897): public-domain works of Protestant scholarship. Their history and archaeology can be out of date, and every topic page says so.

Every edition on the site, with its licence. Verse numbers follow the King James scheme so that editions line up; each edition’s own numbering is shown where it differs.

Edition differences come from STEP Bible (TAGNT and TAHOT): “Data created by www.STEPBible.org based on work at Tyndale House Cambridge (CC BY 4.0)”. Source data and documentation · CC BY 4.0. We expand codes and names, remove tagging, and group consecutive words into derived views; we do not distribute copies of the raw data files.

The Greek comparison names Nestle–Aland 27th and 28th editions (NA27/NA28), Tyndale House GNT 2017 (Tyn), SBLGNT 2010 (SBL), Westcott and Hort 1881 (WH), Tregelles 1879 with the 2009 revision (Treg), Scrivener’s 1894 Textus Receptus (TR), and Robinson–Pierpont 2005 Byzantine text (Byz). The source’s comparison reflects those versions, rather than every later revision: the SBLGNT text displayed here is v1.2 (2023), which adds John 7:53–8:11. Occasional manuscript sigla are expanded using the CSNTM manuscript catalogue; asterisks identify the first hand before corrections. Coptic, Latin and Syriac witnesses are not tied to a specific edition by this source. Double brackets flag disputed passages; inclusion does not imply an editor regarded the passage as secure.

Hebrew comparisons name the Leningrad, Aleppo and Cairo (Cairensis) codices, Ben Chaim’s Second Rabbinic Bible and Biblia Hebraica Stuttgartensia (BHS). Written forms (Ketiv), read-aloud forms (Qere), Dead Sea or Judean Desert readings, scribal traditions and scholarly reconstructions are identified separately. BHK means Biblia Hebraica Kittel; LXX means the Greek Septuagint. These sources are witnesses and editorial choices, not a ranking of authority.

SBLGNT apparatus: Michael W. Holmes, SBL Greek New Testament. Copyright 2010 by the Society of Biblical Literature and Logos Bible Software. The SBLGNT is licensed under a Creative Commons Attribution 4.0 International License. We parse the entries and expand witness sigla. WH = Westcott and Hort 1881; Treg = Tregelles; NIV = the Greek text behind the NIV; RP = Robinson–Pierpont. Margin and appendix readings, double brackets, and editorial conjectures retain their labels.

LSJ (TFLSJ): Liddell–Scott–Jones, edited and formatted by Tyndale House and STEP Bible. “Data created by www.STEPBible.org based on work at Tyndale House Cambridge (CC BY 4.0)”. Source · CC BY 4.0. We link entries by extended Strong’s identifiers and convert formatting and hover references to readable text. Entries supplied from Abbott-Smith where LSJ has none keep that source label.

• 1En-Charles 1 Enoch, tr. R. H. Charles (APOT 1913) (1913) · Public Domain
• ASV American Standard Version (1901) · Public Domain
• BSB Berean Standard Bible (2016/2023) · Public Domain (dedicated 2023-04-30)
• LXXE Brenton English Septuagint (1851) · Public Domain
• RP2018 Byzantine Textform (Robinson-Pierpont 2018) (2018) · Public Domain (Unlicense)
• DBY Darby Translation (1890) · Public Domain
• DRC Douay-Rheims (Challoner), 1899 American edition (1752/1899) · Public Domain
• GNV Geneva Bible (1599) (1599) · Public Domain
• JPS JPS Tanakh (1917) (1917) · Public Domain
• Jub-Charles Jubilees, tr. R. H. Charles (APOT 1913) (1913) · Public Domain
• KJV King James Version (1769) with Apocrypha (1611/1769) · Public Domain (outside the UK; Crown letters patent apply to printing in the UK)
• LSV Literal Standard Version (2020) · CC BY-SA 4.0 (Copyright 2020 Covenant Press)
• James-ANT M. R. James, The Apocryphal New Testament (1924) - selections (1924) · Public Domain in the US (published 1924; James d. 1936)
• SBLGNT SBL Greek New Testament (2010 (v1.2 2023)) · CC BY 4.0 (Copyright 2010 Society of Biblical Literature and Logos Bible Software)
• SCR1894 Scrivener's Textus Receptus (1894), unaccented (M. A. Robinson) (1894) · Public Domain
• LXX-Br Septuagint (Greek text of Brenton's 1851 edition) (1851) · Public Domain
• TR-Boyd Textus Receptus - R. A. Boyd 2022 collation (Stephanus 1550 / Elzevir 1624 / Scrivener 1881) (2022) · Public Domain
• WLC Westminster Leningrad Codex (Open Scriptures Hebrew Bible) (OSHB 2024) · CC BY 4.0 (WLC text itself public domain)
• WEB World English Bible (Classic) with Deuterocanon (2020) · Public Domain ("World English Bible" is a trademark)
• YLT Young's Literal Translation (1898) · Public Domain

People, places and events: Theographic (CC BY-SA 4.0). Cross-references: OpenBible.info (CC BY). Original-language words and lexicons: STEPBible (CC BY 4.0). Meaning-based search runs on our own server; no searches are sent elsewhere except the text of questions sent for an AI summary.

Use Report a problem at the bottom of any page. Every report is read, and corrections go through the same fact-check as everything else.

No accounts, no ads, no tracking. We count what people search for, by day, and whether we had a good answer, so we know which pages to write next. The count keeps only the search words: no IP address, time, cookie or anything else about you. The site remembers your lens (Seek or Question) in a cookie on your device. AI summaries send your question, and nothing about you, to the AI model that writes them. The site is delivered through Cloudflare, which carries the connection, and Cloudflare's Turnstile check may run before an AI summary to keep bots from using it.