Fourteen Words and Thirteen

• Anchor Bible Dictionary — “Acts of the Apostles,” “Apostolic Council,” “Apostolic Decree,” “Philippi,” “Lystra,” “God-fearers.”
• Dictionary of New Testament Background (IVP) — “Roman Colonies,” “Roman Law and Legal System,” “Divination and Magic,” “Citizenship, Roman.”
• New International Dictionary of NT Theology & Exegesis — περιτέμνω, δόγμα, πνικτός, ὕψιστος, διανοίγω.
• Commentary on the New Testament Use of the Old Testament, ed. G. K. Beale and D. A. Carson (Baker, 2007) — the Acts chapter, which is where the Amos 9 quotation is discussed at length, and where the Jeremiah 12:15 proposal for 15:16a is defended.
• 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 ancient column for the numbers to be offset from. There is no Hebrew Acts and no Septuagint Acts. The two scripture columns here are the Greek New Testament and the ESV, and every chapter-and-verse number on this page means the same verse in both.

But the count is still not identical, and one of the differences is inside this study.

Four verse numbers are absent from the SBLGNT text of Acts altogether — 8:37, 15:34, 24:7 and 28:29. That was not read off a list; it was found by walking the MorphGNT file and asking which numbers never appear. 15:34 is in this passage. In the traditional numbering it reads “Notwithstanding it pleased Silas to abide there still,” and it sits between the delegates being sent back and Paul and Barnabas remaining in Antioch. The ESV prints it in a footnote and not in the text. The fifth difference is not an omission but a division: SBLGNT Acts 19 ends at verse 40 where the English prints 19:41, which is why the book totals differ by more than four.

⚠️ What this page did not check. No manuscript, apparatus or edition was consulted on why any of those four is absent. The claim here is only that the printed SBLGNT does not carry them and the traditional English numbering does. Whether a verse is ‘original’ is a question about manuscripts, and no manuscript was examined for this page.

Measured: Ζεύς occurs twice in the whole Greek New Testament, at Acts 14:12 and 14:13, and nowhere else. Ἑρμῆς occurs twice: here as a god, and at Romans 16:14 as a man Paul greets. Λυκαονιστί occurs once. στέμματα — the garlands the priest brings — occurs once.

Borrowed, and flagged as borrowed: The standard reference works connect this scene to Ovid’s story of Baucis and Philemon (Metamorphoses VIII.611–724), in which Jupiter and Mercury walk Phrygia in human form, are turned away by everyone but one old couple, and drown the district. Ovid is not on this disk. That connection is reported from the literature, not measured here, and a reader should treat it as a commonplace worth checking rather than a result of this page.

What the Greek does show without any help is the order of the names: the crowd calls Barnabas Zeus and Paul Hermes, and the text explains the second assignment and not the first — ἐπειδὴ αὐτὸς ἦν ὁ ἡγούμενος τοῦ λόγου, because he was the chief speaker. Nothing is said about why Barnabas looked like Zeus.

Measured: The phrase πνεῦμα πύθωνα occurs once in the Greek New Testament and not once in the 30,637 verse units of Greek Old Testament held here. μαντεύομαι, the verb for what she does for money, also occurs once. κολωνία (16:12) occurs once. ῥαβδοῦχος occurs twice, both in this chapter (16:35, 16:38). στρατηγός occurs ten times in the New Testament; five of the ten are Philippi’s magistrates in this chapter.

Borrowed, and flagged as borrowed: That Πύθων was the serpent Apollo killed at Delphi and that the word came to be used of ventriloquist diviners; that a colonia was a settlement of Roman citizens governed under Italian law with duoviri at its head and lictors carrying rods; that Philippi was settled with veterans after Actium — all of that is from the reference works and from Roman historians not held on this disk. The page reports it as background and measures none of it.

What the background does explain, if it is right: why the charge at 16:21 is about customs unlawful for us Romans rather than about theology, and why the magistrates panic at 16:38 rather than at 16:23. The text supplies the panic; the historians supply the reason.

This one the page could measure, and the result is the most surprising thing in Stage 2.

The slave girl at Philippi shouts δοῦλοι τοῦ θεοῦ τοῦ ὑψίστου. The CATSS corpus holds two Greek Daniels — the Old Greek and Theodotion — and they are different books at exactly this verse.

Searched across all 30,637 verse units held here, the exact sequence δοῦλοι τοῦ θεοῦ τοῦ ὑψίστου occurs in one verse: Theodotion Daniel 3:93. Nebuchadnezzar calls three men out of a furnace with the words a Philippian slave girl uses about Paul and Silas — and eight verses later Paul and Silas are in a prison whose doors come open in an earthquake.

⚠️ What that is not. It is not evidence that Acts is quoting Daniel. There is no citation formula, the run is five words long, and the phrase is an ordinary way of naming the God of Israel on Gentile lips. What it does show is that a run can be edition-dependent: the same verse in the same book scores 5 in one Greek column and 2 in the other, and a page that named only “the Septuagint” would have hidden that.

σῴζω occurs thirteen times in Acts. Five of the thirteen are in these three chapters, and they are the spine of the narrative.

The passage opens with a man who has faith to be saved and closes with a man who asks how, and in between a council decides that the answer does not include circumcision. That is a reading. What is measured is the five occurrences and where they fall.

Scope. The 108 verse units of SBLGNT Acts 14–16, run against (a) the other 7,819 verses of the Greek New Testament and (b) 30,637 verse units across 64 CATSS Greek files. Exhaustive and deterministic.

Caveat. This finds verbatim contact. It does not find allusion. A passage can be steeped in a book it shares no five words with, and a run of six can be an accident of common vocabulary.

Acts 15:17 and Septuagint Amos 9:12 share fourteen consecutive words: καὶ πάντα τὰ ἔθνη ἐφ’ οὓς ἐπικέκληται τὸ ὄνομά μου ἐπ’ αὐτούς, λέγει κύριος. Tie band 1. The next-deepest verse in the entire Greek Old Testament held here reaches 5.

The argument James makes is that God has taken a people from the Gentiles, and the proof is that Amos says the nations will seek. That is what the Greek says. It is not what the Hebrew says.

Two differences are in play at once, and both are consonantal: Edom and mankind differ by one letter in unpointed Hebrew, and possess and seek differ by one. The page reports both columns and does not say which is prior, which is a misreading, or which is ‘the original’. That is not what a word-run measures, and this shelf does not assert direction of influence. The Amos 9:11–12 page works the same divergence from the Amos side.

What is measured, exactly. That Acts 15:17 agrees with the CATSS Greek Amos for fourteen consecutive words, that no other verse in 30,637 reaches that depth with it, and that the Masoretic Amos at that point has a different noun and a different verb. Not measured: which Greek Amos Luke or James had, whether it read exactly as Rahlfs prints it, or whether any of this happened as narrated.

Acts 14:15 ends ὃς ἐποίησεν τὸν οὐρανὸν καὶ τὴν γῆν καὶ τὴν θάλασσαν καὶ πάντα τὰ ἐν αὐτοῖς. From the heaven to in them is thirteen words, and they stand verbatim in Septuagint Exodus 20:11 and in Acts 4:24. Tie band 1 against the Greek Old Testament; tie band 1 against the New Testament.

Why it matters more than the run of fourteen. The fourteen is expected: the text says it is a quotation. The thirteen is not flagged at all. A missionary speaking to farmers who have never read anything reaches for a sentence out of the Greek Decalogue and does not tell them so — and the same sentence is what the church prayed at Acts 4:24 with nobody outside the room. The formula was evidently how these people talked about God, not a proof-text they deployed.

⚠️ And here is the counter-example printed in full, so the runs above are not read as more than they are. Acts 17:24, the Athens speech, is the nearest parallel in the book to the Lystra speech — same subject, same audience, same argument — and it reaches 4 with Acts 14:15 and 4 with Acts 4:24. Two speeches on the identical theme by the identical speaker share four words. If a run of thirteen proved dependence, a run of four would have to prove independence, and neither conclusion follows from counting.

1. Acts 15:16a from Jeremiah 12:15. The proposal that Μετὰ ταῦτα ἀναστρέψω is drawn from Greek Jeremiah 12:15 (καὶ ἔσται μετὰ τὸ ἐκβαλεῖν με αὐτοὺς ἐπιστρέψω) is standard in the commentaries. Measured against the CATSS Jeremiah held here, the run is 1. The idea is similar; the words are not. This page can measure words, so it records the claim and marks it unconfirmed rather than repeating it as a finding.

2. Acts 15:18 from Isaiah 45:21. Same result: γνωστὰ ἀπ’ αἰῶνος against Greek Isaiah 45:21 measures 1. The verse is three words long in the SBLGNT and there is very little there to match with.

3. The decree’s four items as the Leviticus 17–18 ‘alien in your midst’ laws. This is the most influential reading of the decree, and it is an argument about topics in sequence, not about shared wording. πνικτός does not occur in the Greek Old Testament held here at all, so the third item cannot be matched to a Levitical phrase even in principle. A word-run screen is the wrong instrument for that argument, and reporting a zero here would be misleading rather than informative. The page names the reading and declines to score it.

Acts 15:23 and Acts 15:6 share five words (οἱ ἀπόστολοι καὶ οἱ πρεσβύτεροι), tie band 1 — the phrase that names the deciding body twice in one chapter.

Acts 16:29 and Acts 17:4 share five words (τῷ Παύλῳ καὶ τῷ Σιλᾷ), tie band 1 — which is a run about two proper names travelling together and nothing else. It is printed here because it is the kind of run that looks impressive in isolation and means very little.

Zero on this page means “shares no word at all”. Every Greek-side screen here uses the unigram fallback: a row that shares no two-word sequence still scores 1 if any single word matches. So the run of 1 reported for Acts 15:16 against Jeremiah 12:15 means they share one word and no pair, not that they share nothing.

Every number on this page is an agreement between two printed texts. The New Testament figures come from the SBL Greek New Testament as digitised by MorphGNT; the Greek Old Testament figures from Rahlfs’ Septuaginta as digitised by CATSS. Neither has a critical apparatus on this disk, and no manuscript was examined.

That bears on this passage harder than on most. The Western text of Acts — Codex Bezae and its allies — is famously longer and differs at several points in exactly these chapters, including the form of the decree at 15:20 and 15:29, where some witnesses drop what is strangled and add a negative golden rule. None of that was consulted. The run of 14 at Acts 15:17 is a fact about the SBLGNT and Rahlfs, and a different edition on either side could shorten or lengthen it. Acts 16:12 is a textual crux in its own right — whether Philippi is called a leading city of the district or a city of the first district depends on a conjecture, and the page quotes the SBLGNT and takes no view.

The count is solid; the inference is not the count. The 37 first-person-plural finite verbs in Acts 1–15 were pulled from the MorphGNT parsing column (person field 1, number field P, mood field I / S / D / O), and each was then read in context to check whether it stands inside quoted speech. All 37 do. That last step was done by reading, not by a program, because “inside quoted speech” is not a field in the file, and a reader who disagrees about any one of them should say so.

And the interpretation is wide open. Whether the we-sections mark an eyewitness, a source document, a sea-voyage convention, or an authorial device is an argument this page contributes nothing to except the location of the seam. 16:10 to 16:17, then silence until 20:5. Anyone who wants to argue from that has to account for why it stops when Paul and Silas are arrested and resumes years later at the same city.

No Greek critical apparatus. No papyri — P⁴⁵, P⁷⁴ and P¹²⁷ all carry parts of Acts and none was consulted. No Codex Bezae. No patristic citations. No Latin, Syriac or Coptic versions. No inscriptions from Pisidian Antioch or Philippi, no Roman legal sources, no Ovid, no Strabo, no papyrus letters for the χαίρειν / ἔρρωσθε frame. No commentary on Acts was read for any figure on this page.

And the historical questions stay open, because the measurement is silent on them. Whether Acts 15 and Galatians 2 describe the same meeting; whether the letter to Galatia was written before or after the council; whether the decree was a theological compromise or a table-fellowship rule; whether Paul knew of it when he wrote 1 Corinthians 8–10. Counting shared word-runs cannot decide any of that, and this page does not pretend otherwise.

The place this page is thinnest. It leans on two rows. Acts 15:17 and Acts 14:15 carry the whole thesis, and they are two verses out of 108. They were not cherry-picked — they are the top two of an exhaustive screen of every verse in the passage against both corpora — but a pattern established by the top two results of one screen is a pattern of two, and the page says so instead of calling it a rule.

One claim that is easy to overstate, and is deliberately not made. It is tempting to write that Acts 14:15 shows “the apostles thought in Septuagint.” The page will not, because Acts 17:24 — the same speaker, the same argument, a Gentile audience again — shares four words with it and not thirteen. One deep run and one shallow one, from the same mouth on the same subject, is not a habit. It is two data points.

• C. K. Barrett, A Critical and Exegetical Commentary on the Acts of the Apostles (ICC, 2 vols., 1994–1998).
• Joseph A. Fitzmyer, The Acts of the Apostles (Anchor Bible 31, 1998).
• Craig S. Keener, Acts: An Exegetical Commentary (4 vols., 2012–2015) — the standard collection of the Greco-Roman background this page had to borrow.
• Richard Bauckham, “James and the Jerusalem Church,” in The Book of Acts in Its Palestinian Setting (1995) — the classic treatment of the Amos citation.
• W. Edward Glenny, Finding Meaning in the Text: Translation Technique and Theology in the Septuagint of Amos (2009) — on the Greek Amos 9:12 specifically.
• None of these were consulted for the measurements on this page, which were run on the text files named in the footer.

Three chapters. A man healed at Lystra and a man stoned at Lystra, in the same paragraph. A council that decides the largest question the church would ever face and writes it down in a letter that opens greetings and closes farewell, like every other letter in the Greek world. Then two men in a Roman colony, singing at midnight with their feet in the stocks.

And underneath it, a measurement. The one sentence in these chapters that the text explicitly calls Scripture agrees with the Greek Amos for fourteen consecutive words. The one sentence addressed to people who had never heard of Amos, spoken at the top of Paul’s voice to stop a sacrifice, agrees with the Greek Exodus for thirteen — and with a prayer the church had already prayed, in Jerusalem, before any of this started.

What the page will not do is tell you what that proves. It does not say Paul was quoting Exodus, because he does not say so and a run is not a claim about intent. It does not say the Greek Amos is right and the Hebrew Amos is wrong, or the reverse; it prints both and names the two letters that differ. It does not say the shift to we at 16:10 means an eyewitness. It says where the words are the same and how unusual that is, and then it stops.

The chapter that begins with a decree about what Gentiles need not do ends with a jailer washing the wounds he was ordered to guard, and setting a table in his own house at three in the morning. The council’s document is being carried through Asia Minor in the same satchel as the men who are living the answer to it.