Where the Quotation Breaks

• Anchor Bible Dictionary — “Peter, First Epistle of,” “Silas/Silvanus,” “Bithynia and Pontus,” “Cappadocia,” “Babylon (place),” “Household Codes.”
• Dictionary of the Later NT & Its Developments (IVP) — “Peter, First Letter of,” “Household Codes,” “Persecution,” “Christian.”
• Dictionary of NT Use of the OT (Beale & Carson) — the 1 Peter chapter, and the Romans chapter on 9:33 for the double stone.
• NIDNTTE — παρεπίδημος, πάροικος, ἀναστροφή, πάσχω, ἐκλεκτός.

1 Peter itself is one of the easy cases. The Greek and the English agree everywhere.

Where the trap is: the Psalter it quotes. Greek psalm numbers run one behind the Hebrew from Psalm 9/10 through 147, because the Greek joins Hebrew 9 and 10 into one psalm. On top of that, the superscription is counted as verse 1 in the Hebrew and the Greek and is not counted in English Bibles — so the verse number inside the psalm shifts as well.

Practical consequence: a commentary that says “1 Peter 3:10–12 = Psalm 33:13–17” and one that says “= Psalm 34:12–16” are saying the same thing. This page gives the Greek number, because the Greek is what is being measured, and prints the English beside it.

The two words 1 Peter uses of its readers were ordinary civic vocabulary in the Greek cities of the empire.

Putting the two together at 2:11 says something sharper than either alone: people who live here permanently and belong here not at all. Whether Peter’s readers held that status legally, or whether he is describing their new social position with a legal word, is not decidable from the letter, and both readings are defended.

The house caution about background material applies here. No inscriptions, papyri or civic law codes were consulted for this page. The definitions above are the standard lexical ones; the social picture behind them is named as the kind of evidence the reading rests on, not as something checked.

1:1 names Pontus, Galatia, Cappadocia, Asia and Bithynia. Pontus and Bithynia were administered together as one province, and this list puts one at the front and the other at the back. Read as a route, it runs clockwise round the interior of northern Asia Minor and comes back to where it started.

That is an observation about the order of five names. It is not evidence of a courier, a schedule, or a first-century postal practice, and none of those was checked for this page. What can be said flatly is that the letter is addressed to a region, not to a congregation, and that it never names one.

5:13 sends greetings from ἡ ἐν Βαβυλῶνι συνεκλεκτή, the co-chosen woman in Babylon.

Three things can be stated without argument. First, the letter does not explain the word. Second, Babylon in Mesopotamia was by this period a much-reduced town, and no early tradition connects Peter with it. Third, Revelation uses Babylon of a city that is plainly not Babylon — and Revelation is on this shelf, so that much is checkable.

The common reading is that Babylon here means Rome. The page reports that as the common reading and does not certify it. What it will say is that the word does work inside this letter regardless of the geography: the letter opens by calling its readers exiles of the Dispersion and closes by locating its sender in the empire that made exiles. Second-Temple texts that use Babylon the same way — 4 Ezra, 2 Baruch, the Sibylline Oracles — were not consulted for this page and are named here only as the literature the argument usually runs through.

A word that occurs exactly once in the whole Greek New Testament is a hapax legomenon. Counting them per book, over the MorphGNT lemma column, gives a blunt but real index of how far a writer reaches outside the common stock.

The line worth stopping on is the last one. 1 John and 1 Peter are exactly the same length — 105 verses each. 1 Peter uses 543 different lemmas and 56 words that occur nowhere else in the New Testament. 1 John uses 233 and one: νίκη, victory, at 1 John 5:4. Two letters of identical length, and one of them has fifty-six times the rare-word count of the other.

What this measures and what it does not. It is a count over the SBLGNT as MorphGNT lemmatises it. A different edition, or a different lemmatiser, would move the numbers a little; the gap between 56 and 1 is not the kind of gap that moves. And a vocabulary count is not an authorship argument. It says how wide a book’s word-stock is. It does not say whose hand held the pen — which matters here, because 5:12 says the letter was written through Silvanus.

All 105 verses of 1 Peter were screened against all 30,637 verse units of the Greek Old Testament held here (64 CATSS files), and separately against the other 7,822 verses of the Greek New Testament. Contiguous runs of whole word forms, never lemmas. Tie band reported beside every run. Re-run 30 August 2026, when Exodus, Leviticus, Numbers, Joshua and the deuterocanonical books came onto the shelf.

Against the Greek Old Testament: 104 of the 105 verses reach two, 58 reach three, 23 reach four, 12 reach five, 9 reach six, 6 reach seven, 5 reach eight, 3 reach nine, 1 reaches ten, and nothing reaches eleven.

⚠️ What the corpus growth moved, and what it did not. Over the previous 44 files and 23,061 units this same screen read 46 reach three, 20 reach four, 11 reach five, 7 reach six. Twelve more rows now reach three, three more reach four, one more reaches five, two more reach six. The top of the table did not move at all. 3:12 is still 10 with a band of 1; 2:7 and 4:18 are still 9; 2:22 and 2:6 are still 8; 1:24 is still 7; and no row in the letter reaches eleven. The letter’s deepest contact survived a corpus a third larger. Sixteen rows changed their maximum. Two of them matter: 1:16 (2 → 5, band 87 → 1, against Greek Leviticus 19:2) and the two doxologies at 4:11 and 5:11 (5 → 6, band 1), whose deepest Greek Old Testament partner is now 4 Maccabees 18:24 — a witness, not Scripture — the last verse of that book, which closes ᾧ ἡ δόξα εἰς τοὺς αἰῶνας τῶν αἰώνων ἀμήν. 1 Peter closes both of its doxologies on the same six words. Two Greek-speaking Jewish writers reaching for one fixed formula is what that is, and the page reports it as a formula and nothing more.

1 Peter 3:10–12 is one continuous quotation of Greek Psalm 33:13–17 (ESV Psalm 34:12–16). It is unbroken in the source. Measured verse by verse in Peter it reads 4, 3, 10.

The screen is verse-bounded, so it reports ten. Run past the psalm’s own verse boundary by hand and the figure is sixteen. 1 Peter 3:12 is ὅτι plus sixteen consecutive words of Greek Psalm 33 — all ten of 33:16 and the first six of 33:17 — and then it stops, immediately before τοῦ ἐξολεθρεῦσαι ἐκ γῆς τὸ μνημόσυνον αὐτῶν, to cut off the memory of them from the earth. That last clause is the only part of the psalm block Peter leaves behind.

The same pattern holds two verses apart in 1 Peter, at 2:22 and 2:24, both drawing on Isaiah 53.

Stated carefully: across the letter’s quotations, the runs that stay deep are the ones whose source has no first- or second-person forms to move, and the runs that collapse are the ones where Peter has re-addressed the sentence to his readers. This is a description of a pattern in measured data. It is not a claim about which Greek text Peter had in front of him, whether he was quoting from memory, or whether he used a collection of proof-texts. A run screen cannot see any of that, and this page does not pretend otherwise.

Four places on this page where the measurement gives a small number and the small number does not mean what a careless reader would take it to mean.

And one in the other direction, which is the more useful warning. 1 Peter 1:25 is quoting Isaiah 40:8 — the whole surrounding sentence is Isaiah 40:6–8, and 1:24 measures a clean 7 against Isaiah 40:7. But the screen’s deepest hit for 1:25 is not Isaiah at all: it is Greek Psalm 116:2 (ESV 117:2), at a run of 5 with a tie band of 1, on the words κυρίου μένει εἰς τὸν αἰῶνα. Peter wrote the word of the Lord where Isaiah has the word of our God, and that single substitution handed the top slot to a psalm about the enduring truth of the Lord which has nothing whatever to do with the case. The deepest run is not always the source, and a tie band of one does not make it one.

1 Peter 2:6–8 sets Isaiah 28:16, Psalm 117:22 and Isaiah 8:14 in a row. Romans 9:33 fuses the first and the third into a single sentence. Measured against the same Greek columns:

Two writers pair the same two stones in the same order and use the same word σκανδάλου — and that word does not stand once in the whole of Greek Isaiah. Searched: σκανδαλ- occurs 0 times in CATSS Isaiah; προσκομμ- occurs twice (8:14 and 29:21); the phrases λίθος προσκόμματος and πέτρα σκανδάλου occur nowhere in any of the 64 Greek Old Testament files held here. That negative was re-run on 30 August 2026 over all 30,637 verse units — Exodus, Leviticus, Numbers, Joshua and the whole deuterocanon included — and it did not move. It is a stronger negative than it was, over the same two phrases. (For the record, σκανδαλ- is not absent from the new books: it stands in Leviticus, Joshua B, Wisdom, the Psalms of Solomon and three times in Judith. It is the two phrases that are absent.) The page reports that both writers have something the Greek column in front of us does not, and asserts nothing about where either of them got it — not a shared source, not a testimonia collection, not one copying the other.

Screened against the other 7,822 Greek New Testament verses: 104 of the 105 reach two, 60 reach three, 21 reach four, 9 reach five, 6 reach six, 5 reach seven, 5 reach eight, 3 reach nine, 2 reach ten, 2 reach eleven, and nothing reaches twelve.

The two deepest are a blessing formula and a doxology — set pieces, not citations. Nobody thinks Ephesians is quoting 1 Peter at 1:3, or that Revelation is quoting it at 1:6. They are the fixed openings and closings of Greek-speaking Christian prayer, and a run screen finds them because they are fixed. Compare that with 3:12, where a run of ten sits on a verse the letter is visibly quoting. Depth alone does not tell you which is which; the tie band and the eye do.

The 5:5 line is the one worth staring at. Greek Proverbs 3:34 begins κύριος. 1 Peter 5:5 and James 4:6 both begin ὁ θεός, and both then run identically for six more words. Each agrees with the Greek Proverbs at 6 and with the other at 8. This page states that and stops. It does not say James used Peter, or Peter used James, or that both used a common source, or that either of them was reading a Greek Proverbs different from ours.

Until 30 August 2026 this section carried the opposite heading. It read “the biggest gap: three books of the Pentateuch are not on this shelf,” and it printed 1 Peter’s two most famous Old Testament borrowings as NOT MEASURABLE. Both are measurable now. The page prints the whole history rather than the new numbers alone.

The prediction this page made, and could not test, was right. The finding in stage 4 is that quotation depth tracks grammatical person. This section used to say, in terms: Leviticus 19:2 would have been an excellent test of it. The Hebrew of that verse is second person plural — קְדֹשִׁים תִּהְיוּ, you shall be holy — and Peter keeps the second person, so the prediction would be a deep run. What the Greek Leviticus says this page cannot tell you, because it is not on this shelf. The Greek Leviticus is on the shelf now, and it reads ἅγιοι ἔσεσθε — second person plural, as the Hebrew is. Peter keeps it, changes nothing, and the run is five words with a tie band of one. The thesis no longer rests only on the psalm block, Isaiah 53 and the proverbs. It has been tested against the Pentateuch, on a book that was not on this disk when the prediction was written, and it held.

⚠️ Read that in the right order, because it is the only honest way to read it. The prediction was published first, with its reason and with the admission that it could not be checked. The book arrived afterwards. Nothing was adjusted to fit. Had Greek Leviticus 19:2 read ἅγιός εἰμι ἐγώ as Lev 11:44 and 11:45 do, the run would have been 3 and this section would be printing a failed prediction instead. That is exactly what the other four holiness verses in the book do, and they are printed beside it in the codex line at 1:16.

The Greek Old Testament held here is 64 CATSS files, 30,637 verse units — Exodus, Leviticus, Numbers, Joshua in both the A and B recensions, and the deuterocanonical books all arrived on 30 August 2026, and every figure on this page has been re-run over them. It is still not the whole Septuagint tradition: the CATSS text is a Rahlfs-based diplomatic transcription rather than the Göttingen critical editions, and its apparatus was not consulted. A negative reported here is a negative over these files.

Where a hit lands in Sirach, 4 Maccabees, 1 Esdras, Tobit, Judith, Wisdom, Baruch or the Psalms of Solomon, this page names the book and says what it is: a witness, not Scripture — the same label this shelf puts on Psalm 151 and on Jubilees. Those books are measurement partners here and nothing more.

It finds verbatim contact. It does not find allusion, and it is blind to vocabulary that has been reordered.

1 Peter 2:9 is the clearest case on this page. It carries γένος ἐκλεκτόν, λαὸς εἰς περιποίησιν and τὰς ἀρετάς, and Isaiah 43:20–21 carries τὸ γένος μου τὸ ἐκλεκτόν, λαόν μου ὃν περιεποιησάμην and τὰς ἀρετάς μου. Three matched ideas, and the screen returns 1 and 2, because Peter has re-cased and re-ordered every one of them. A page that reported only the run numbers here would be reporting a falsehood by omission, so this page says plainly that the instrument is wrong for that verse.

The screen is also verse-bounded. The sixteen-word figure at 3:12 was obtained by concatenating Greek Psalm 33:16 and 33:17 by hand and re-running the pair; the automated figure is ten. Both are printed.

The SBL Greek New Testament marks its variation units in the text itself, and this page checked where they fall relative to every run it prints. The answer cuts both ways and is given in full.

Read honestly this both supports and threatens the page’s own thesis, and both halves belong here.

It supports it where the number is deep. The verse carrying the headline figure — 3:12, ten words and then sixteen — carries no marked variation unit at all, and neither does 2:22, the eight against Isaiah 53:9. Those two numbers do not rest on an editorial choice.

It threatens it where the number is shallow. The 4 at 3:10 and the 3 at 3:11 sit on γλῶσσαν, χείλη and δέ, and all three are marked. If the units at 3:10 restore a pronoun after those two nouns, the shape of the collapse changes. The 1 at 4:8 sits on καλύπτει, the only word it shares with the Greek Proverbs. The 9 at 4:18 has a unit inside it, and so does the 7 at 1:24.

The apparatus itself was NOT consulted. What is reported above is only where the SBL editors marked a variation unit, read off the sigla in the MorphGNT text column — not what the competing readings are, not how well attested they are, and not which is likely original. Every number on this page is a number over the SBL text as printed. A page built on NA28, or on the Byzantine text, could return different figures at the rows marked above and the same figures at the rows that are not.

No critical apparatus. The Greek is the SBL edition as digitised by MorphGNT, with its editorial sigla stripped for display. No variants were adjudicated. Three that bear on verses printed here: 3:18 ἔπαθεν / ἀπέθανεν and ὑμᾶς / ἡμᾶς; 5:2 whether ἐπισκοποῦντες stands at all; 1:22 whether καθαρᾶς stands before καρδίας. None of this page’s measurements sits on those words, but a reader should know they are contested.

No Qumran, no Targum, no Syriac, no Vulgate, no Coptic, no patristic citation evidence. For the double-stone problem at 2:6–8, that is a real gap: the standard discussions turn partly on whether an early collection of stone texts circulated, and none of that literature was examined here.

Authorship is not settled on this page and no attempt was made. The vocabulary table in stage 3 is a count, not a verdict, and 5:12’s through Silvanus is quoted rather than interpreted.

3:19–20 is left standing. He went and proclaimed to the spirits in prison is among the most disputed sentences in the New Testament. This page prints no reading of it. 1 Enoch, which the main line of the discussion runs through, was not consulted and is not on this shelf.

And the letter’s household material (2:18–3:7) is described, not resolved. It stands inside the structure diagram as part of the second block and is not discussed further here. That is a limit of this page, not a judgement about the text.

• Paul J. Achtemeier, 1 Peter (Hermeneia).
• John H. Elliott, 1 Peter (Anchor Yale Bible 37B) — and his A Home for the Homeless on πάροικος.
• Karen H. Jobes, 1 Peter (Baker Exegetical Commentary).
• Peter H. Davids, The First Epistle of Peter (NICNT).
• D. A. Carson, “1 Peter,” in Beale & Carson, Commentary on the New Testament Use of the Old Testament.
• None of these were consulted for the measurements on this page, which were run on the text files named in the footer.

Read 1 Peter 3:10–12 as a piece of editing and it is a busy passage. Peter takes a psalm that is addressed to you — keep your tongue from evil, turn from evil, seek peace — and turns every one of those commands into the third person: let him keep, let him turn, let him seek. Five imperatives re-pointed and two possessives deleted in two verses. The run screen sees the wreckage: 4, then 3.

Then the psalm stops giving orders and starts describing God, and Peter stops editing. The eyes of the Lord are on the righteous, and his ears are open to their prayer. Ten words out of ten, unchanged; and then six more from the next verse, unchanged. Sixteen consecutive words of a Greek psalm, dropped into a letter to five provinces of frightened people, with a single ὅτι welded on the front.

He edits the commands because the commands are now being given to somebody else. He does not edit the description, because what God is like did not need re-addressing to arrive.

And then there is what he left out. The psalm’s sentence continues: the face of the Lord is against those who do evil — to cut off the memory of them from the earth. Peter takes the first half of that and puts down his pen before the second. It is the only part of the whole psalm block he does not carry over, and the page reports the omission without explaining it. Six verses later he writes that Christ suffered the righteous for the unrighteous, and ten verses after that, that the gospel was preached even to the dead. What a reader makes of the sequence is left to the reader.

First page on this shelf for 1 Peter; the companion 2 Peter study is already on it. All five chapters are represented in the codex above.