One Word Apart

• Anchor Bible Dictionary — “Luke, Gospel of,” “Synoptic Problem,” “Q,” “Parable,” “Sabbath.”
• Dictionary of Jesus and the Gospels (IVP) — “Travel Narrative,” “Rich and Poor,” “Watchfulness,” “Repentance,” “Herod Antipas.”
• New International Dictionary of NT Theology & Exegesis — ὑπόκρισις, ζύμη, μεριμνάω, ἄφρων, ἀγωνίζομαι, ἄπιστος.
• Commentary on the New Testament Use of the Old Testament, ed. G. K. Beale and D. A. Carson (Baker, 2007) — the Luke chapter, for Micah 7:6 at Luke 12:53 and Psalm 118:26 at Luke 13:35.
• None of these were consulted for any measurement 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 there is none to state. On a New Testament book there is no Masoretic or Septuagint column to be offset from at all.

The SBLGNT row and the English row were counted verse by verse off disk; the ESV row is read from the printed text, since no ESV Luke file is held here. All three run 59 and 35, so every chapter-and-verse reference on this page means the same verse in all of them. The book totals differ only in bookkeeping: the SBLGNT and the English file simply have no Luke 17:36 and no Luke 23:17, and the ESV prints both as footnotes while keeping the traditional numbers. Neither verse is in this passage.

Three numbering problems do bear on results printed here, and none of them is the usual one.

A page that printed “Psalm 118:26” without saying which column it meant, or “a run of 17” without saying what stopped it, would be telling the truth in a way that misleads. Both figures are on this page, with their reasons.

διχοτομέω occurs twice in the whole Greek New Testament — Luke 12:46 and Matthew 24:51 — and its family stands in five verses of the Greek Old Testament held here — seven occurrences, because Exodus 29:17 and Ezekiel 24:4 each carry it twice. Four of the five are butchery at an altar or a covenant; the fifth is meat in a cooking pot.

⚠️ What this is and is not. It is a lexical fact about where the word lives. It is not a measured connection. Luke 12:46 screened against the whole Greek Old Testament reaches a run of 3, tie band 32 — and against Genesis 15:17, Exodus 29:17 and Ezekiel 24:4 its longest shared run is one word. The page prints the word’s Old Testament home and draws no line from it to this parable.

Luke 13:21 and Matthew 13:33 both say ἀλεύρου σάτα τρία. Both words occur exactly twice in the Greek New Testament, and both times in these two verses.

σάτον is the Aramaic seah in Greek letters. Three seahs is the quantity Genesis 18:6 has Abraham tell Sarah to knead for the three visitors — the Greek there reads τρία μέτρα σεμιδάλεως, three measures of fine flour, so the quantity matches and the words do not: measured, Luke 13:21 against Greek Genesis 18:6 shares no two consecutive words at all — a zero on the no-fallback convention, in which single shared words are not counted; the one word the two do share is τρία. Reference works put three seahs somewhere near forty to fifty pounds of flour — bread for a hundred people and more, hidden by one woman under a handful of leaven.

The volume conversion is reported from the reference works, not measured here; the ancient dry measures are not on this disk and the modern equivalents vary by several pounds between authorities.

ἀλώπηξ occurs three times in the Greek New Testament: Matthew 8:20, Luke 9:58 — foxes have holes — and Luke 13:32, where it is a name for Herod Antipas.

What a fox stands for in first-century speech is a question the reference works answer several ways: cunning in the Greek fable tradition, but in some rabbinic usage a small and worthless creature set against a lion. The page cannot settle it from counting and does not try. What it can report is that the word occurs three times in the corpus, that two of the three are the same saying about homelessness, and that this one is the only time it is used of a person.

The reply that follows is a Lukan signature: σήμερον καὶ αὔριον καὶ τῇ τρίτῃ τελειοῦμαι (13:32), then σήμερον καὶ αὔριον καὶ τῇ ἐχομένῃ (13:33) — the same three-day formula twice, with a different third day. τελειοῦμαι in that form occurs once in the Greek New Testament.

The rich fool is never accused of fraud, theft, or oppressing anyone. The land produced well — εὐφόρησεν ἡ χώρα, a verb that occurs once in the Greek New Testament.

Read the soliloquy for its pronouns: my crops, my barns, my grain, my goods, my soul. In 12:17–19 the first person singular occurs eleven times in three verses. Then God speaks and the pronouns change hands: your soul is being demanded from you, and the things you prepared — τίνι ἔσται; whose will they be?

The pronoun count is a count of first-person forms including verb endings and is a reading aid, not a screen result. The measured figures for this parable are in Stage 4: six rows, nothing above 4 against anything outside Luke.

1,702 words across 94 verses. Counted off the MorphGNT file: 1,042 words in Luke 12 and 660 in Luke 13.

Eighteen words in these two chapters occur only once in the whole Greek New Testament — among them συγκύπτω (13:11), εὐφορέω (12:16), μεριστής (12:14), σιτομέτριον (12:42) and νοσσιά (13:34) — and the form τελειοῦμαι (13:32) is unique as well, though its verb is not. The page reports these as counts, not as arguments.

All 94 verses of Luke 12:1–13:35 were screened twice: once against the 6,778 Greek New Testament verses outside Luke, once against 30,348 verse units across the 63 CATSS Greek Old Testament files screened (64 files and 30,637 units are held; the Odes, 289 units, are held and not screened — see the correction below). Contiguous runs of whole word forms, never lemmas, with the tie band beside every run.

Against everything outside Luke: all 94 rows reach two, 76 reach three, 42 reach four, 29 reach five, 18 reach six, 12 reach eight, 6 reach ten, 2 reach fourteen, 1 reaches seventeen, and nothing reaches eighteen. Against the Greek Old Testament: all 94 reach two, 78 reach three, 32 reach four, 9 reach five, 2 reach six, and nothing reaches seven.

Every row in this passage that reaches eight or more — all twelve of them — reaches it in Matthew. Not one reaches eight in Mark, in John, in Acts or in any letter. And rows run the other way too: 24 of the 94 measure deeper against the Greek Old Testament than against the New Testament outside Luke, and four of them open a gap of two: Luke 12:53 (4 against 2), Luke 12:56 (5 against 3), Luke 13:4 (5 against 3) and Luke 13:33 (4 against 2). The passage is not uniformly Matthean, and the page is not claiming it is.

On 30 August 2026 the Odes were removed from this shelf’s Greek Old Testament. Four figures on this page moved, and one of them was a hit that had no business existing on a Luke page at all.

⚠️ THE FIRST ROW IS THE ONE TO READ TWICE. Luke 12:52’s deepest contact with the ‘Greek Old Testament’ was four words with a tie band of ONE — a lone, clean, unrepeated partner, which on this shelf is the shape of a real result — and the partner was Luke’s own first chapter.

Ode 9 is the Magnificat and the Benedictus — Luke 1 — and Ode 13 is the Nunc Dimittis, Luke 2, gathered into the liturgical appendix bound after the Greek Psalter. Screened as though they were Old Testament, Ode 9:48 matches Luke 1:48 at eighteen consecutive words, which would be by a wide margin the deepest “Old Testament” contact anywhere in this Gospel. It is Luke matching Luke. On this page the damage was smaller and quieter — a four-word run at 12:52, sitting inside two published aggregates and named in neither — but it is the same error, and a smaller one is easier to publish.

What the reader should take from it. A tie band of one is the shelf’s usual signal that a match is not formulaic, and here a band of one was pointing at the page’s own author. The instrument does not know what is in its corpus. It was caught by asking a question the run engine cannot answer — not ‘how deep is this match’ but ‘what IS Ode 9?’ — and the answer removed a file rather than adjusting a number. With the Odes off the screen, Luke 12:52’s deepest Greek Old Testament contact is three words, tie band 31: an ordinary formula, which is what it always was.

Held and screened are now different numbers. The shelf HOLDS 30,637 CATSS verse units in 64 files and SCREENS 30,348 in 63; the difference is exactly the 289 units of the Odes, which stay on disk. Ten of the fourteen canticles are second copies of passages already in the corpus — Ode 1 is Exodus 15, Ode 2 Deuteronomy 32, Ode 3 1 Samuel 2, Ode 4 Habakkuk 3, Ode 5 Isaiah 26, Ode 6 Jonah 2, Odes 7 and 8 Daniel (Theodotion) 3, Ode 10 Isaiah 5, Ode 11 Isaiah 38 — which is why the Luke 13:28 and Luke 12:27 bands were inflated: a verse standing once appeared to stand twice. The two remaining canticles that are neither duplicates nor Luke are Ode 12, the Prayer of Manasseh, and Ode 14, the morning hymn; the whole appendix goes out together because it is one book. Nothing else on this page moved: the 17 at Luke 12:46, the 14 at 13:34, the 6 with a band of 1 at 13:35 against Psalm 117:26, the 4 with a band of 1 at 12:53 against Micah 7:6 and at 13:27 against Psalm 6:9, and the whole New Testament column are unchanged.

Luke 12:46 has 28 words. Twenty-six of them stand in Matthew, consecutively, in the same order. The twenty-eighth is identical as well. One word differs.

⚠️ The 26 is a hand-made comparison and the page says so. The screen that produced every other number here is verse-based; it cannot join two target verses on its own, and it did not. The 26 was obtained by concatenating Matthew 24:50 and 24:51 and running the same engine on the result. It is reported separately from the 17 for that reason, and the 17 is the figure that belongs in the distribution table above.

What the page will not say about it. Not that Luke is using Matthew. Not that Matthew is using Luke. Not that both are drawing on a common written source, and not that they are not. Direction of influence is never asserted on this shelf, and a single sentence — however clean — settles nothing. What is measured is that the agreement runs 26 words, that the one divergence is the final noun, and that hypocrisy is a word Luke had already spent, once, at the head of this very block.

And here is the honest complication, printed rather than hidden. Luke is not avoiding the word. ὑποκριταί stands twice inside these two chapters — at 12:56 and 13:15 — in Luke’s own mouth. So “Luke replaced the hypocrites with the unbelievers” is a story, not a measurement, and it is not one this page will tell. The measurement is the 26 and the one; the reader may do with it what the evidence allows.

Across all 94 verses, the deepest contact with the Gospel of Mark anywhere is six consecutive words — and those six words are a quotation of the Psalter.

The deepest of the five is the one that is not Markan wording at all. Septuagint Psalm 117:26 reads εὐλογημένος ὁ ἐρχόμενος ἐν ὀνόματι κυρίου, and Mark 11:9, Matthew 21:9 and Luke 13:35 all carry those six words unchanged. Mark’s single deepest measured contact with these two chapters is a psalm all three Gospels quote.

⚠️ The obvious caution, and it is a large one. Mark does not contain most of this material. A screen cannot find agreement in text that is not there. “Luke 12–13 measures shallow against Mark” is therefore in large part a statement about what Mark contains, not about how Luke wrote. The one place where Mark does carry the same parable — the mustard seed, Mark 4:30–32 against Luke 13:18–19 — the measurement against those three verses is 2 and 4. Luke 13:18 reaches 5 elsewhere in Mark, at 4:26 and 10:14, on the formula ἐστὶν ἡ βασιλεία τοῦ θεοῦ, not on the parable. Having the same parable is not the same as having the same sentence, and that is the genuinely informative result in this table.

1. Luke 13:28: nine consecutive words, and six verses tied for them.

ἐκεῖ ἔσται ὁ κλαυθμὸς καὶ ὁ βρυγμὸς τῶν ὀδόντων is nine words — deeper than all but six rows in the passage — and it stands identically at Matthew 8:12, 13:42, 13:50, 22:13, 24:51 and 25:30. A run of nine with a tie band of six is a fixed formula and tells you nothing about which text got it from which. Printed here so that the run of 17 at Luke 12:46, whose band is one, is read for what distinguishes it. A run without its band would have made these two look like the same kind of thing.

2. Luke 12:53 and Micah 7:6 — and Matthew, which is supposed to be the closer one.

Matthew 10:36 says a man’s enemies will be the members of his own household and Micah 7:6 says a man’s enemies are the men of his own house. In sense they could hardly be closer. Measured against the Septuagint their longest shared run is one word — four word forms are shared in all (αὐτοῦ, αὐτῆς, ἐχθροί, οἱ) and no two of them stand together — because Matthew writes οἱ οἰκιακοί where the Greek Micah writes οἱ ἄνδρες οἱ ἐν τῷ οἴκῳ. Luke, whose sentence is a symmetrical list rather than a quotation, happens to hit four of Micah’s words in a row. The screen finds verbatim contact. It does not find allusion, and here it gets the two backwards.

3. Luke 13:27 and Matthew 7:23 quote the same psalm verse, from opposite halves of it.

Luke keeps the psalm’s first four words and rewrites its last four. Matthew rewrites the first and keeps the last four verbatim. Each agrees with the Septuagint more than either agrees with the other. On a page whose headline is a 26-word agreement with Matthew, this is the row that shows the same two books doing the opposite thing eleven verses later.

4. The mustard seed measures shallower than the leaven standing next to it.

The parable with two parallels measures shallower than the parable with one, and they are consecutive sentences. Luke’s mustard seed is planted εἰς κῆπον ἑαυτοῦ, in his own garden; Mark’s is sown ἐπὶ τῆς γῆς, on the ground; Matthew’s ἐν τῷ ἀγρῷ αὐτοῦ, in his field. Luke’s becomes a tree; Mark’s becomes the greatest of shrubs. Three tellings, one image, and very little shared wording. Having the same story is not the same as having the same sentence.

5. The narrow door: three shared words, and the run stops at the noun.

Luke 13:24 and Matthew 7:13 share διὰ τῆς στενῆς — and then Luke says θύρας and Matthew says πύλης. Three words, band 1. The adjective στενός occurs three times in the entire Greek New Testament and this is one of them; the sentences around it are otherwise unalike, and Luke’s continues into a scene Matthew does not have at all, the householder getting up and shutting the door.

Every deep result here is a Luke–Matthew agreement, and there is a two-century literature about what those agreements mean. None of it was consulted, and none of it can be settled by this method.

Two-source theory, Farrer–Goulder–Goodacre, the Griesbach hypothesis, the reconstructions of a sayings source and the arguments that no such source existed — each of them predicts that Luke 12:46 and Matthew 24:50–51 will agree closely, and the 26-word run does not discriminate between them. A page that presented the 26 as evidence for any one of those positions would be misusing its own instrument. This one presents the 26 and stops.

What a determined critic could fairly say against the page. That selecting the deepest runs and calling the pattern “Matthean” is close to circular, since the double-tradition material in this block is by definition the material Matthew shares. That objection is correct, and it is why the counter-examples are in Stage 4 rather than in a footnote: 24 of the 94 rows measure deeper in the Greek Old Testament, Luke 12:53 measures 2 in the whole New Testament outside Luke, and the parable Mark also carries measures shallower than the one he does not.

It depends on one printed edition on each side, with no apparatus. The Greek is the SBL Greek New Testament as digitised by MorphGNT. Luke 12:46 and Matthew 24:50–51 carry variants in the manuscript tradition, and no critical apparatus is held on this disk. The 26 is an exact agreement between two printed texts; a different edition could lengthen or shorten it. The same applies to the 14 at Luke 13:34 and the 11 at Luke 12:27.

It depends on a decision about what a word is. Both sides are normalised the same way — lowercased, accents and breathings dropped, final sigma folded — and compared as word forms, not lemmas. That is why Luke 13:34 stops at 14: ἐπισυνάξαι and ἐπισυναγαγεῖν are two aorist infinitives of one verb and the screen treats them as different words. It is also why Luke 12:43 stops at 10, for a transposition of two words that are both present in both verses. A lemma-based or gap-tolerant screen would give different numbers on every line of this page, and would be a different measurement, not a better one.

And the 26 is one sentence. It is the deepest row in this passage and the seventh-deepest of Luke’s 1,149 verses measured against Matthew — Luke 16:13 reaches 26 against Matthew 6:24 within a single verse pair, and Luke 9:58 and 11:32 both reach 24. So the phenomenon this page leads with is real and it is not unique, and saying so is more useful than letting the reader assume it is a one-off.

No Greek critical apparatus. No papyri — P⁴⁵, P⁶⁹ and P⁷⁵ all carry parts of Luke and none was consulted. No patristic citations, no Diatessaron, no Marcionite Luke, no Syriac, Latin or Coptic versions. No Qumran or rabbinic material, so the rabbinic use of fox at 13:32 is reported from reference works and not measured. No documentary evidence for dry measures, so the size of three σάτα is likewise reported, not counted. No commentary on Luke was read for any figure on this page.

Three specific things this page would like to have measured and could not. (1) Whether the SBLGNT’s text of Luke 12:46 is disputed at ἀπίστων — that is exactly the kind of question an apparatus answers and this disk cannot. (2) Whether Luke 13:35’s omission of ἔρημος against Matthew 23:38 is editorial or textual — same problem. (3) What any of these runs measure against the Greek Sirach or Wisdom material as printed in other editions; the CATSS files here are one digitisation of one printed text.

And the questions the counting is simply silent about. Who wrote this Gospel. When. Whether the travel narrative is a deliberate literary construction or a gathering of loose material. What Jesus said in Aramaic before any of it was Greek. Counting shared word-runs decides none of it, and this page does not pretend otherwise. What it offers is the measurement the arguments have to fit.

• Joseph A. Fitzmyer, The Gospel According to Luke X–XXIV (Anchor Bible 28A).
• John Nolland, Luke 9:21–18:34 (Word Biblical Commentary 35B).
• François Bovon, Luke 2: A Commentary on the Gospel of Luke 9:51–19:27 (Hermeneia, 2013).
• Mark Goodacre, The Case Against Q (2002) — for the position this page’s measurement is most often taken to support and does not.
• John S. Kloppenborg, Excavating Q (2000) — for the position it is second most often taken to support and also does not.
• None of these were consulted for the measurements on this page, which were run on the text files named in the footer.

Luke 12 opens with a crowd so large it is trampling itself, and the first thing said is beware of the leaven of the Pharisees, which is hypocrisy. Ninety-four verses later the block ends outside a city that kills prophets, with a house left empty. Between the two there is a farmer who built bigger barns, a woman bent double for eighteen years, a fig tree given one more year, a door somebody gets up and shuts, and a servant who assumed his master was going to be late.

And in the middle of it stands one sentence that agrees with Matthew for twenty-six consecutive words. Not seventeen — seventeen is only where Matthew’s verse number falls. Twenty-six out of twenty-eight, and the one word that differs is the last noun: Luke puts the servant with the unbelieving, Matthew with the hypocrites.

The page cannot tell you why, and it says so on the same screen where it says the number. It cannot even tell you that Luke was avoiding the word, because he uses it twice in these very chapters, at 12:56 and 13:15. What can be said is what was counted: the run, its band of one, the verse division that truncates it, and the fact that hypocrisy is the word this whole block opens with and occurs nowhere else in Luke.

The two chapters end at a door and a house. The door is narrow, and the one who shuts it says I do not know where you come from. The house is Jerusalem’s, and it is left to them — ἀφίεται ὑμῖν ὁ οἶκος ὑμῶν — until they say the words of a psalm the Septuagint numbers 117 and the ESV numbers 118: Blessed is he who comes in the name of the Lord. Six words. All three Gospels carry them identically, and they are the deepest thing Mark shares with these two chapters.