Ten Words with Paul

• Anchor Bible Dictionary — “James, Letter of,” “James (person),” “Diaspora,” “Paraenesis.”
• Dictionary of the Later New Testament & Its Developments (IVP) — “James,” “Faith and Works,” “Riches and Poverty,” “Tongue.”
• New International Dictionary of NT Theology & Exegesis — δίψυχος, δοκίμιον, ἔργον, πίστις, ταπεινός.
• Commentary on the New Testament Use of the Old Testament, ed. G. K. Beale and D. A. Carson (Baker, 2007) — the James chapter, which is where the Genesis 15:6 and Proverbs 3:34 quotations are discussed at length.
• These were not consulted for the measurements below. The measurements were run on the text files named in the footer.

Every page on this shelf states its versification offset, including the pages where the answer is that there is none. On a New Testament book the answer is stronger than that: there is no second column to be offset from.

The SBLGNT row and the English versification row were counted off disk; the ESV row is the printed edition’s shape, because no ESV James file is held here. There is no Masoretic James and no Septuagint James, so the columns on this page are Greek New Testament and English, and every chapter-and-verse number means the same verse in all of them.

But the offset problem is still in the room, because James quotes books that have one. The Greek Deuteronomy held here numbers the sixth to ninth commandments as four separate verses, 5:17–20, in the order adultery, murder, theft, false witness. The Masoretic text puts all four inside verse 17, in the order murder, adultery, theft, false witness. James 2:11 names the first two in the Greek order. A page that quoted ‘Deuteronomy 5:17’ without saying which column it meant would be naming two different commandments at once.

Seventeen of the twenty-one New Testament letters open with a Christianised greeting — χάρις ὑμῖν καὶ εἰρήνη or something built from it. Three carry no opening greeting formula at all (Hebrews, 1 John, 3 John). James opens with the plain Hellenistic χαίρειν, the word on ten thousand ordinary papyrus letters, and no other New Testament letter does.

Counted in the Greek New Testament: the form χαίρειν occurs seven times. Three are epistolary openings — Acts 15:23, Acts 23:26, James 1:1. Acts 23:26 is Claudius Lysias writing to Felix, a Roman officer’s dispatch. Acts 15:23 is the council’s letter to Antioch. The remaining four (Romans 12:15; 2 Corinthians 2:3; 2 John 10, 11) are the verb doing its ordinary work.

⚠️ What was not checked. The documentary papyri are not held on this disk. That χαίρειν is the standard opening of a Greek private letter is reported from the reference works, not measured here. What was measured is the seven New Testament occurrences and which three of them open a letter.

James 5:11 appeals to τὴν ὑπομονὴν Ἰώβ and to τὸ τέλος κυρίου — the endurance of Job, and the end the Lord made.

The Greek Job held here uses that word-group freely: ὑπομονή / ὑπομένω stands fifteen times across the book (3:9; 6:11; 7:3; 8:15; 9:4; 14:14; 14:19; 15:31; 17:13; 20:26; 22:21; 32:4; 32:16; 33:5; 41:3 — eleven spelled ὑπομ-, four augmented to ὑπεμ-). And the Greek Job does not stop where the Hebrew stops: CATSS carries five extra verses, 42:17a–e, which the Masoretic text has no counterpart for. 42:17a says Job will rise again with those whom the Lord raises; 42:17b–e identify him with Jobab of Edom and make him the fifth from Abraham.

What that does and does not license. It means a reader of the Greek Job met a book with an explicit resurrection line at the end of it. It does not mean James read that ending — the measurement finds no verbal contact between James 5:11 and Job 42:17a–e, and the page asserts nothing about which Job James knew.

James 5:17 says the sky was shut ἐνιαυτοὺς τρεῖς καὶ μῆνας ἕξ. So does Luke 4:25. The Greek 3 Kingdoms says something else.

James 5:17 and Luke 4:25 share a run of 3 (καὶ μῆνας ἕξ) — short, and the two sentences are otherwise unlike each other. What is worth reporting is not the run but the number: a figure two New Testament books carry and the Greek Kings on this disk does not. Whether it comes from a common tradition, from a way of counting the reign, or from somewhere else, the page cannot say, and does not.

1,739 words, 555 distinct lemmas, 108 verses. Counted off the MorphGNT file.

Words the letter does not contain at all: σταυρός, ἀνάστασις, εὐαγγέλιον, ἐλπίς. The page reports the absences as counts, not as arguments about date or theology.

Counted from the MorphGNT parsing column, mood field, across all 27 New Testament files: James carries 55 imperative verb forms in 108 verses.

James is first on both measures, and the gap to second place is about a quarter again. Exhaustive over the corpus and deterministic; nothing sampled. The caveat: this counts imperative forms, as tagged by MorphGNT. It does not count the hortatory subjunctives, the future indicatives used as commands, or the participles that function as imperatives — so the true rate of commanding is higher in every book, James included, and a different tagging would move every row a little.

All 108 verses of James were screened twice: once against the other 7,819 verses of the Greek New Testament, once against the 30,348 verse units of the Greek Old Testament screened here — the 30,637 held in 64 CATSS files, less the 289 units of the Odes, which stay on the disk and are not screened (stage 5 says why). Contiguous runs of whole word forms, never lemmas, with the tie band beside every run.

Against the rest of the Greek New Testament, across the 108 rows: all 108 reach two, 74 reach three, 21 reach four, 8 reach five, 5 reach six, 3 reach seven, 2 reach eight, 1 reaches ten, and nothing reaches eleven. Against the Greek Old Testament: all 108 reach two, 71 reach three, 15 reach four, 6 reach five, 3 reach six, 1 reaches seven, and nothing reaches eight.

Re-run 30 August 2026 on the enlarged corpus. Over the 44 files this page was first built on (23,061 units) the same screen gave 66 / 13 / 5 / 2 / 1. Every band grew or held; none shrank — which is what a larger target corpus must do. ⚠️ Re-run a third time the same day, with the Odes withdrawn from the screening corpus: the distribution did not move at all — 71 / 15 / 6 / 3 / 1 over 30,348 units exactly as over 30,637. Two tie bands did move, and stage 5 names both.

Rows run the other way too, and two of them belong in the same table. Sixteen of the 108 rows measure deeper against the Greek Old Testament than against the rest of the New Testament — and the two widest are James 5:4 and James 1:8, 5 against 3 in each case. James 5:4 is the sharper of those two: five words with Isaiah 5:9 at a tie band of one, against three words and a band of four in the New Testament. The letter is not uniformly closer to the New Testament, and the page is not arguing that it is. What the page leads with is two specific rows, and they are the two quotations.

James 2:23 — ten with Romans, seven with Genesis.

Three divergences run in the same direction, and James and Paul are on the same side of all three. The Hebrew says believed in the Lord; the Greek and both New Testament verses say believed God. The Hebrew verb is active, he counted it; the Greek and both New Testament verses are passive, it was counted. And the Greek Genesis calls him Abram — correctly, since Genesis 17:5 has not happened yet — while James and Paul both call him Abraham.

James 4:6 — eight with 1 Peter, six with Proverbs.

Here the Hebrew and the Greek are not two wordings of one sentence but two sentences: לֵצִים are mockers, not the proud, and יָלִיץ is he mocks, not he opposes. James and 1 Peter carry the Greek shape and then depart from it identically, in the same two words.

⚠️ What the page will not say. Not that James is quoting Paul. Not that Paul is quoting James. Not that either is quoting 1 Peter, or 1 Peter either of them, or that all of them are drawing on a lost collection of proof-texts. Direction of influence is never asserted on this shelf, and two rows — however clean their tie bands — cannot establish one. What is measured is that the agreement is there, that it is longer than the agreement with the Greek Old Testament in both cases, and that the surplus falls in exactly the words where the New Testament wording and the Septuagint part company.

1. James 5:12 and Matthew 5:37: everyone pairs them, and the run is 2.

James 5:12 — do not swear… let your yes be yes and your no be no — is the closest thing in the letter to a saying of Jesus, and Matthew 5:37 is the saying. Measured: a run of 2, ναὶ ναί. Meanwhile 2 Corinthians 1:17 gives a run of 7 with a tie band of 1 (τὸ ναὶ ναὶ καὶ τὸ οὒ οὔ) — and 2 Corinthians 1:17 is Paul explaining why he changed a travel itinerary. The deep run is the accident and the shallow one is the relationship.

2. James 1:10–11 and Isaiah 40: the allusion everyone sees, measuring 3.

Two New Testament passages use the same image of Isaiah 40 — the grass, the flower, the withering. One of them is a citation and measures 7; the other is a free reworking and measures 3. The screen can tell them apart; it cannot tell you that the second is dependent on Isaiah at all. It only ever finds verbatim contact.

3. James 2:8: a run of 6 with a tie band of 5.

Ἀγαπήσεις τὸν πλησίον σου ὡς σεαυτόν is six consecutive words — deeper than most of this letter reaches — and it is shared with Matthew 19:19, Matthew 22:39, Mark 12:31, Romans 13:9 and Galatians 5:14 alike. Five verses tied means the run is a fixed formula in Greek, and tells you nothing about which book got it from which. A run without its band would have made this look like a discovery.

4. ‘Covers a multitude of sins’ measures 2.

James 5:20 and 1 Peter 4:8 are routinely printed together, and they do share πλῆθος ἁμαρτιῶν — a run of 2, in a tie band of 87: eighty-seven New Testament verses share two consecutive words with James 5:20, and 1 Peter 4:8 is one of them. By the numbers alone there is nothing here at all, and the phrase is still obviously the same phrase. Against Septuagint Proverbs 10:12, which is where the thought is usually traced, James 5:20 shares no two consecutive words at all: the Greek Proverbs there reads πάντας δὲ τοὺς μὴ φιλονεικοῦντας καλύπτει φιλία, which is a different sentence from the Hebrew’s love covers all offences as well.

James’s clearest Old Testament citation is Leviticus 19:18, and until 30 August 2026 this page could not measure it. It can now, and the result is the strongest single number on the page.

James 2:8 against Greek Leviticus 19:18: a run of six unbroken word forms — ἀγαπήσεις τὸν πλησίον σου ὡς σεαυτόν, you shall love your neighbour as yourself — at a tie band of one. Screened against all 30,348 verse units, Leviticus 19:18 is the only place in the Greek Old Testament screened here that reaches it.

And James 2:11 can now be set beside both Decalogues instead of one — which turns out to be worth doing.

The run itself does not move: James writes μὴ μοιχεύσῃς (aorist subjunctive) where both Greek Decalogues write οὐ μοιχεύσεις (future indicative), so it scores 1 against Exodus 20:13 exactly as it already scored 1 against Deuteronomy 5:17. The order, however, is evidence. James names adultery first and murder second. So does Greek Deuteronomy 5 — 17 adultery, 18 murder, 19 stealing — and so does Greek Exodus 20 — 13 adultery, 14 stealing, 15 murder. The Hebrew of both Decalogues does not: Masoretic Exodus 20:13 and Deuteronomy 5:17 alike run murder, adultery, stealing. James is following the Greek order and not the Hebrew one — and so are Romans 13:9 and Luke 18:20, while Mark 10:19 and Matthew 19:18 keep the Hebrew order.

And the two Greek Decalogues do not agree with each other: both put adultery first, but Exodus puts stealing second and murder third, while Deuteronomy puts murder second and stealing third. That divergence is inside the Greek tradition itself and is named here rather than smoothed over. It was invisible to this page while only one of the two Greek Decalogues was on the disk.

They depend on one printed edition on each side. The New Testament figures are the SBL Greek New Testament as digitised by MorphGNT; the Old Testament figures are Rahlfs’ Septuaginta as digitised by CATSS. Neither has a critical apparatus on this disk. The run of 10 between James 2:23 and Romans 4:3 is an exact agreement between two printed texts; manuscripts of both verses vary, and a different edition could shorten or lengthen it. The same goes for the 8 between James 4:6 and 1 Peter 5:5.

They also depend on a decision about what a ‘word’ is. The screen normalises both sides the same way — lowercased, accents and breathings dropped, final sigma folded — and compares word forms, not lemmas. That is why James 2:11 scores 1 against Greek Deuteronomy 5:17 while quoting it recognisably: μὴ μοιχεύσῃς and οὐ μοιχεύσεις are the same commandment and not the same words. A lemma-based screen would give different numbers everywhere on this page, and would be a different measurement, not a better one.

And two rows are two rows. The pattern the page leads with rests on James 2:23 and James 4:6. Those happen to be the letter’s only two Old Testament quotations introduced by an explicit formula and traceable to a book that is held here, so the sample is not arbitrary — but it is a sample of two, and the page says so rather than calling it a rule.

No Greek critical apparatus, no papyri (P²⁰, P²³, P⁷⁴ and P¹⁰⁰ all carry parts of James and none was consulted), no patristic citations, no Qumran or rabbinic material, no Syriac, Latin or Coptic versions. No documentary papyri for the letter form. No commentary on James was read for any figure on this page.

And the historical questions stay open, because the measurement is silent on them. Which James this is, when the letter was written, whether its author knew Romans or Romans knew it, whether James 2:14–26 is answering Paul or answering a misuse of Paul or answering neither — counting shared word-runs cannot decide any of that, and this page does not pretend otherwise. What it offers is the measurement the arguments have to fit.

One more that is easy to overstate. James and 1 John are the only two New Testament letters whose last two verses carry no greeting, no grace-wish, no peace-wish, no doxology and no Amen. That was checked verse by verse across all twenty-one letters — but 1 John still closes with a vocative address (τεκνία), and James closes with a third-person imperative. The two endings are abrupt in different ways, and ‘the only letter that ends without a farewell’ would have been a tidier sentence and a false one.

• Dale C. Allison Jr., A Critical and Exegetical Commentary on the Epistle of James (ICC, 2013).
• Richard Bauckham, James: Wisdom of James, Disciple of Jesus the Sage (1999).
• Luke Timothy Johnson, The Letter of James (Anchor Bible 37A).
• Martin Dibelius, James (Hermeneia) — the classic case that the letter is unstructured paraenesis, which is the case this page’s structure diagram takes seriously.
• Scot McKnight, The Letter of James (NICNT, 2011).
• None of these were consulted for the measurements on this page, which were run on the text files named in the footer.

James is the most commanding book in the New Testament — fifty-five imperatives in a hundred and eight verses, more per verse than any other — and it names Jesus twice. It has no cross in it, no resurrection, no gospel, no word for love as a noun. It calls its readers brothers nineteen times and then tells them to stop favouring the rich, stop judging each other, stop swearing, stop planning a year they may not live to see.

And when it stops commanding long enough to quote, twice, the quotation comes out matching another New Testament book more exactly than the Greek Old Testament it names. Ten consecutive words with Romans 4:3; eight with 1 Peter 5:5. Both times the extra words are precisely the ones that step away from the Septuagint — Abraham where Genesis still says Abram, God where Proverbs says the Lord.

The page does not know what to make of that, and says so. It could be a shared way of saying these verses that was already old when all three were written. It could be one book reading another, in either direction. It could be two of them and not the third. The measurement establishes the agreement and stops there, and on the same page it prints the run of 7 between James 5:12 and a verse of Paul about travel plans — so that nobody reads a tie band of one as more than a tie band of one.

The letter that keeps insisting the proof of faith is what you do ends without a goodbye: whoever brings back a sinner from his wandering will save his soul from death and will cover a multitude of sins. Then nothing. No grace, no peace, no Amen — just the last instruction, and the reader left holding it.