The Testaments of the Twelve Patriarchs give each son of Jacob a deathbed speech. Naphtali, Gad and Asher — the eighth, ninth and tenth — speak of the order of nature, of hatred, and of the two ways. No clean Greek of these books exists in any free source, so for this page 28 verses were read by eye off the page images of R. H. Charles’s Greek edition of 1908, and the other 138 were screened through the scan’s OCR, which can only understate. In 1913 Charles wrote that the Gospel sayings on forgiving a brother depend on Gad 6:3–7 — “That the N.T. passages are here dependent cannot be reasonably denied.” — and, in a list of places where he said Paul drew on the book, named three more verses of these testaments. Measured on his own Greek, read off his own pages, every pair he named in those 1913 pages from these three testaments shares between one and four consecutive words. The clause of Asher 6:2 that he said was taken over “bodily” into Romans 1:32 reaches four only when one movable letter is set aside — a letter his 1913 quotation prints and his 1908 page does not. And Asher’s two ways share one word with each of the two-ways texts measured (Deuteronomy 30:15, Jeremiah 21:8, Matthew 7:13–14), while their deepest Greek contacts are with Ecclesiastes and Sirach. These books are not Scripture for us: they are read here as witnesses from the world between the testaments, and tested by the canon. See the method →
The Testaments of the Twelve Patriarchs are a book of twelve farewell speeches, surviving in Greek and in other versions, one for each son of Jacob, each telling a life, drawing a lesson and foreseeing the tribe’s future. Charles held them to be a Hebrew work of the second century BC with later additions, Jewish and Christian, some of which he bracketed; the date and the extent of the additions are argued, and this page reports his view without testing it. They are not Scripture for us, and not deuterocanonical either. We read them as we read the other Second-Temple books — as a window on how some Jews, and later some Christians, taught about hatred, forgiveness and the inner life — critically, gratefully for the light they throw, and testing everything by the canon.
⚠️ What this page can and cannot measure — read this first.
HELD: Charles’s English of all twelve testaments (1,000 verses); the page images, OCR and word boxes of his Greek edition (1908); the page images and OCR of his English (1917) and of APOT vol. 2 (1913), with the late Hebrew Testament of Naphtali in his English; the Greek Old Testament (30,348 CATSS units screened) and the SBLGNT (7,927 verses); and the Abegg/ETCBC transcription of the Dead Sea Scrolls, which includes 4Q215 (measured for names only, never printed).
NOT HELD: any clean Unicode Greek of the Testaments; de Jonge’s critical edition; the Armenian and Slavonic versions except as Charles reports them; the Hebrew of the late Testament of Naphtali.
CONSEQUENCE: headline runs are measured only on the 28 verses read off the page, on the shelf’s normaliser (normalize.grk_seq_cmp), against the whole SBLGNT and the whole CATSS corpus, with tie bands both ways. The other 138 verses are screened through the OCR, and an OCR run can only be shorter than the true one. No English is measured against any other English as evidence: Charles translated both the Greek Testaments and the late Hebrew one, so English runs between them would measure Charles.
Testaments of Naphtali, Gad and Asher — ten places, Greek read off the page
Greek: R. H. Charles, The Greek Versions (1908), read off the page images · English: Charles (1908; SPCK 1917)α column Καινοποιός μου ἡ θυγάτηρ·
bef, (A), S¹ column Καινόσπουδός μου ἡ θυγάτηρ· εὐθὺς γὰρ τεχθεῖσα ἐπιλαβομένη τοῦ μαζοῦ ἔσπευδεν θηλάζειν.Naphtali 1:11–12Charles 1917 · N 1:11 “And he was taken captive and was bought by Laban; and he gave him Euna his handmaid to wife, and she bore a daughter, and called her name Zilpah, after the name of the village in which he had been taken captive.”
Charles 1917 · N 1:12 “And next she bore Bilhah, saying: My daughter hasteneth after what is new, for immediately that she was born she seized the breast and hastened to suck it.”
Bilhah and Zilpah are sisters here — both daughters of Rotheus by Euna — which is what a note on this shelf’s Jubilees 24–29 page reported and could not then check. The English of 1:12 renders the right-hand column, not Charles’s α text, which has only the short line he marks corrupt; and the English village in 1:11 is the reading his apparatus gives for β–f, A, S¹ (κώμης), where his text has πόλεως. Measured: 1:11 shares six consecutive words with Genesis 30:21 (Dinah’s naming), tie band 1 — a naming formula.
Charles 1908 · α Ὁμοίως δὲ καὶ οἱ Ἐγρήγοροι ἐνήλλαξαν τάξιν φύσεως αὐτῶν, οὓς κατηράσατο Κύριος ἐπὶ τοῦ κατακλυσμοῦ, δι᾽ οὓς ἀπὸ κατοικησίας καὶ ἄκαρπον τάξας τὴν γῆν [ἀοίκητον].Naphtali 3:4–5Charles 1917 · N 3:4 “But ye shall not be so, my children, recognising in the firmament, in the earth, and in the sea, and in all created things, the Lord Who made all things, that ye become not as Sodom, which changed the order of nature.”
Charles 1917 · N 3:5 “In like manner the Watchers also changed the order of their nature, whom the Lord cursed at the flood, on whose account He made the earth without inhabitant and fruitless.”
Sodom and the Watchers side by side, both changing the order of their nature — the pairing Jude 6–7 also makes. Measured: Naphtali 3:4 shares two words with Jude 7 (as Sodom), 3:5 one with Jude 6. From Jude’s side, Jude 7’s deepest Greek Old Testament run is five words, tie band 1: Jeremiah 30:12 in the Greek numbering (49:18 in English Bibles), Σοδομα καὶ Γομορρα καὶ αἱ. And 3:4 shares no word at all with Romans 1:26 (εἰς τὴν παρὰ φύσιν, for what is against nature). The idea is shared; the words are not. The phrase τάξιν φύσεως stands in none of the 7,927 SBLGNT verses and none of the 30,348 Greek Old Testament units.
margin, bdg, A, S¹ ἀνέγνωνNaphtali 4:1Charles 1917 · N 4:1 “These things I say unto you, my children, for I have read in the writing of Enoch that ye yourselves also shall depart from the Lord, walking according to all the lawlessness of the Gentiles, and ye shall do according to all the wickedness of Sodom.”
A testament citing another writing by name. The English I have read is the margin’s ἀνέγνων; Charles’s text has ἔγνων, I knew. Jude 14 names Enoch too, and introduces his words with Προεφήτευσεν δὲ καὶ τούτοις … λέγων, prophesied … saying. Naphtali 4:1 and Jude 14 share one word, ἐν. The page does not look for the cited words in 1 Enoch. The English file prints this line with a stray stop (for I. have read); it is printed here as the 1917 page has it.
“The devil shall flee from you” — ὁ διάβολος φεύξεται ἀφ᾽ ὑμῶν. James 4:7 has φεύξεται ἀφ’ ὑμῶν: a run of three, and those three words stand in one New Testament verse and no Greek Old Testament unit. From James’s side the deepest Greek Old Testament run is three, tie band 2 — Esther 8:1 and Tobit (S) 6:8 — the first on Haman’s epithet, τῷ διαβόλῳ, the slanderer, a word match only; the second where, for one afflicted by a demon or an evil spirit, every affliction is said to flee: καὶ φεύξεται ἀπ' αὐτοῦ πᾶν ἀπάντημα. And 8:4’s own deepest New Testament run is not James: it is four words, tie band 3 (Luke 6:22, Acts 15:12 and Acts 21:19). The English file has a full stop after through you; the page has a comma, and so does this line.
Five words with David, tie band 1. καὶ ὅταν ἤρχετο ὁ λέων, and whenever the lion came, is the verse’s deepest Greek Old Testament run: 1 Kingdoms 17:34 (1 Samuel 17:34 in English Bibles), where David tells Saul how he kept his father’s sheep, καὶ ὅταν ἤρχετο ὁ λέων. The five-word sequence stands in one of the 30,348 units. And the same verse touches Gethsemane: ὡσεὶ λίθου βολήν, about a stone’s throw, stands in one New Testament verse, Luke 22:41, and in no Greek Old Testament unit. A shepherd’s boast and a measure of distance: no dependence is asserted either way.
β, A, S¹ column, after the first line, which both columns share εἰπὲ αὐτῷ ἐν εἰρήνῃ, ἐξορίσας τὸν ἰὸν τοῦ μίσους, καὶ ἐν τῇ ψυχῇ σου μὴ κρατήσῃς δόλον· καὶ ἐὰν ὁμολογήσας μετανοήσῃ ἄφες αὐτῷ.Gad 6:3Charles 1917 · G 6:3 “Love ye one another from the heart; and if a man sin against thee, speak peaceably to him, and in thy soul hold not guile; and if he repent and confess, forgive him.”
The verse Charles built his forgiveness argument on, in both of his columns. Against Luke 17:3 the α column measures two (καὶ ἐάν) and the β column three (μετανοήσῃ ἄφες αὐτῷ, against ἐὰν μετανοήσῃ ἄφες αὐτῷ). Charles quoted the β column (APOT 2, p. 292); the English follows α. One verse, two columns, two numbers.
Forgive “from the heart”, and give the vengeance to God. Against Matthew 18:35 (if you do not forgive … from your hearts) this verse measures one. The four-word run with Matthew 18:35 is in another verse: Gad 6:1, ἐξάρατε τὸ μῖσος ἀπὸ τῶν καρδιῶν ὑμῶν, put away hatred from your hearts — four words, and the run stands in one New Testament verse and no Greek Old Testament unit.
Charles 1908 · α Διὰ τοῦτο πάντα δύο δύο εἰσίν, ἓν κατέναντι τοῦ ἑνός.
Charles 1908 · α Ὁδοὶ γάρ εἰσιν δύο, καλοῦ καὶ κακοῦ· ἐν οἷς εἰσι τὰ δύο διαβούλια ἐν στέρνοις ἡμῶν διακρίνοντα αὐτά.Asher 1:3–5Charles 1917 · A 1:3 “Two ways hath God given to the sons of men, and two inclinations, and two kinds of action, and two modes (of action), and two issues.”
Charles 1917 · A 1:4 “Therefore all things are by twos, one over against the other.”
Charles 1917 · A 1:5 “For there are two ways of good and evil, and with these are the two inclinations in our breasts discriminating them.”
The two ways, the two inclinations — and one word with the Bible’s two-ways texts. Against Deuteronomy 30:15 and Jeremiah 21:8 in the CATSS Greek, 1:3 and 1:5 each measure one; against Matthew 7:13–14, one. Their deepest Greek Old Testament runs go elsewhere: 1:3 to Ecclesiastes 1:13 and 3:10, ἔδωκεν ὁ θεὸς τοῖς υἱοῖς (five words, tie band 2), and 1:4 to Sirach 33:15 and 42:24, ἓν κατέναντι τοῦ ἑνός (four, tie band 2).
Charles 1913, APOT p. 292, quoting it καὶ πράσσουσι τὸ κακόν, καὶ συνευδοκοῦσιν τοῖς πράσσουσινAsher 6:2Charles 1917 · A 6:2 “For they that are doublefaced are guilty of a twofold sin; for they both do the evil thing and they have pleasure in them that do it, following the example of the spirits of deceit, and striving against mankind.”
The clause Charles wrote had been taken over “bodily” into Romans 1:32. On the shelf’s normaliser it shares two consecutive words with that verse; with the movable ν folded on both sides, four — καὶ συνευδοκοῦσι τοῖς πράσσουσιν. The one letter is on the page: his 1908 Greek prints συνευδοκοῦσι, his 1913 quotation συνευδοκοῦσιν. The English “guilty of a twofold sin” is Charles’s emendation of the word he marks corrupt (Stage 5).
Two spans in Charles’s square brackets, in the Greek and the English alike — his sign for an interpolation, and reported here as his judgement. Inside the first, ἐσθίων καὶ πίνων, with the and that follows the bracket, shares four words with Matthew 11:19 and Luke 7:34 — tie band 3 from Asher’s side (Acts 15:17 also reaches four, on another phrase, and all the nations). From the Gospels’ side, Matthew 11:19 and Luke 7:34 each reach four in the Greek Old Testament, tie band 1, on the same words: 3 Kingdoms 8:65 (1 Kings 8:65 in English Bibles), Solomon’s feast, ἐσθίων καὶ πίνων καὶ εὐφραινόμενος. A phrase both corpora have. Outside the brackets the dragon’s head is broken ἐπὶ τοῦ ὕδατος, on the water (Charles’s English has “in the water”), and those three words are shared with Psalm 73:13 in the Greek numbering (Psalm 74:13 in the Hebrew), συνέτριψας τὰς κεφαλὰς τῶν δρακόντων ἐπὶ τοῦ ὕδατος; they stand in four Greek Old Testament units (Deuteronomy 32:51, Joshua (B) 11:5, Numbers 20:24 and Psalm 73:13), and the words for head and dragon are not shared in form. And that preposition is not his α family’s: his apparatus (p. 181, note 26) gives α and β–g reading διά, prints the reading of MS g and the Armenian, and refers the text to Psalm 74:13. With α’s word the run with the Psalm is two.
Versification — a third state, stated because it is not the usual one. There is no Masoretic, no Septuagint, no King James, no Vulgate and no ESV text of the Testaments: the shelf’s King James, Vulgate and CATSS loaders each raise on the name rather than return nothing. The numbering is Charles’s, inline in his Greek of 1908 and his English of 1908/1917.
The English file, counted (verses per chapter): Naphtali 12 · 10 · 5 · 5 · 8 · 10 · 4 · 10 · 3 (67); Gad 9 · 5 · 3 · 7 · 11 · 7 · 7 · 5 (54); Asher 9 · 10 · 2 · 5 · 4 · 6 · 7 · 2 (45) — 166 of the file’s 1,000 verses.
The Greek, counted where it can be: the last verse numeral of every chapter was read off the page images — 25 chapters — and 25 of 25 agree with the English file’s count. Every other numeral the page relies on was read with the verse it begins. Internal numerals not read for this page were not checked one by one.
The offsets that DO exist are between the columns of Charles’s own Greek: he prints parallel columns at Naphtali 1:12, Gad 2:3–5, Gad 6:3, Asher 2:3–4 and Asher 6:6. At Gad 2:3–5 his α column has no verse 4 — it runs 3, 5 — and the English file’s 2:4 (“And thus through covetousness we were bent on slaying him.”) is the β column’s Καὶ οὕτως τῇ πλεονεξίᾳ ἐπληροφορήθημεν τῆς ἀναιρέσεως αὐτοῦ. Naphtali 6:10 is numbered 10 in the Greek (read off the page); the English prints ‘9.’ a second time there, and the file numbers it 10 after the Greek.
The Shape of the Passage
Sequence (counted): the testament form, three times over⚠️ Where this page sits. The Testaments of the Twelve Patriarchs are filed here under Pseudepigrapha & Second Temple. They are in no canon this church knows of and are not deuterocanonical. They are not Scripture for us, and nothing taught at Pleasant Springs rests on them.
Author
Anonymous, writing as the dying sons of Jacob. Each testament opens with the same frame — “The copy of the testament of Asher” — and closes with a death and a burial at Hebron.
Charles’s account, reported and not tested: the “Testaments of the Twelve Patriarchs were written in Hebrew in the latter years of John Hyrcanus” — “in other words, between 109 and 106.” (APOT 2, 1913, p. 282). He also marked with square brackets the words he judged to be interpolations (not every addition he identified: APOT 2, p. 290 lists the whole of Naphtali 4, the chapter that cites Enoch, among the first-century Jewish additions, and his English leaves it unbracketed); in the English of these three testaments a bracket opens in 11 verses (Naphtali 5:5, 5:8, 8:3, 8:4; Gad 5:7, 6:5, 6:6, 8:2; Asher 5:2, 7:3, 7:4). The page reports his brackets as his judgement and dates nothing.
A text that survives in more than one form. Charles printed his Greek in parallel columns where his manuscripts part; in these three testaments he does so at Naphtali 1:12, Gad 2:3–5, Gad 6:3, Asher 2:3–4 and Asher 6:6.
Audience
Sons, and through them a people. The vocative is constant — “And now, my children, I exhort you” — and the predictions look past the sons to scattering and return.
Readers who know Genesis. Bilhah and Zilpah, Joseph sold to the Ishmaelites, the lion and the flock: the stories assume Genesis 29–37 and add to it.
People living with a grudge. Gad’s whole teaching is on hatred, told by a man who admits he wanted his brother dead: “I confess now my sin, my children, that oftentimes I wished to kill him”.
Orient on the Testaments, on their text, and on the editions that letter it differently.
Search by topic: Testaments of the Twelve Patriarchs; testament (literary form); Beliar; two ways; forgiveness; Watchers; Naphtali; Gad; Asher; 4Q215.
Which articles are relevant?
- Anchor Bible Dictionary — entries on the Testaments of the Twelve Patriarchs and on Beliar. Subjects to look up; the exact entry titles were not checked, and the work is not held on this disk.
- Eerdmans Dictionary of Early Judaism — the Testaments, testaments as a form, and the two-ways teaching. Not held.
- Dictionary of New Testament Background (IVP) — the Testaments and the Pseudepigrapha. Not held.
- On the text — any article that says whose Greek it quotes. Charles (1908) printed an α text; M. de Jonge’s edition (1978) is built on another manuscript family. A Greek quotation of the Testaments that does not name its edition cannot be checked against this page.
Bibliography surfaced:
- R. H. Charles, The Greek Versions of the Testaments of the Twelve Patriarchs (Oxford, 1908). Held as page images and OCR; the Greek of this page.
- R. H. Charles, The Testaments of the Twelve Patriarchs (London, 1908; SPCK, 1917). Held; the English of this page.
- R. H. Charles (ed.), The Apocrypha and Pseudepigrapha of the Old Testament, vol. 2 (Oxford, 1913), with Appendix I, the late Hebrew Testament of Naphtali. Held as page images and OCR.
- M. de Jonge et al., The Testaments of the Twelve Patriarchs: A Critical Edition of the Greek Text (Brill, 1978). In copyright; not held, not consulted.
- H. C. Kee, “Testaments of the Twelve Patriarchs”, in J. H. Charlesworth (ed.), The Old Testament Pseudepigrapha, vol. 1 (1983). Not held.
A deathbed form, a late Hebrew testament, a Qumran fragment, and the Greek Bible beside the shepherd’s boast.
Every Greek line in this stage is Charles’s (read off the page), a CATSS row or an SBLGNT verse, checked by the self-check. The Hebrew sources are reported by name and content only.
Which parallels shed light on the world of the text?
A testament. A father who knows he is dying gathers his sons, tells his life, teaches, predicts and is buried with his fathers. Genesis 49, Deuteronomy 33 and Tobit 4 are farewells of the same kind. Named, not measured.
A late Hebrew Testament of Naphtali (Charles’s title), whose visions differ on Joseph. Charles translated a Hebrew text he called late in an appendix to APOT 2 under the title “TRANSLATION OF A LATE HEBREW TESTAMENT OF NAPHTALI, WHICH CONTAINS FRAGMENTS OF THE ORIGINAL TESTAMENT” — “These fragments are in italics, and their parallels in the original Testament given in the margin.” Read side by side in his two English translations, the visions of Naphtali 5–7 share their furniture and differ in their verdict on Joseph (Charles held that the Hebrew contains fragments of the original Testament):
| Testament of Naphtali 5–7 Charles’s English of his Greek (1917) | late Hebrew Testament of Naphtali 2–7 Charles’s English of the Hebrew (APOT 2, Appendix I) | |
|---|---|---|
| who bids them seize | “Isaac, the father of my father” | “and lo, our father came and said to us” |
| the sun and the moon | “Levi laid hold of the sun, and Judah outstripped the others and seized the moon” | “When Levi heard it he seized a staff in his hand, and jumped upon the sun and sat and rode thereon.” |
| Joseph | “But Joseph came, and seized him, and ascended up with him on high.” | “and Joseph only remained alone upon the earth.” … “And Joseph got up and mounted upon the bull.” … “and Joseph beat him till he had taken from him ten against his will, and only two were left with Judah.” |
| the ship | “there was written upon the ship, The ship of Jacob” | “This ship belongs to the son of Berachel, and all the good therein.” |
| the wreck | “And Joseph fled away upon a little boat” | “and the ship went a wrong course, and the waves of the sea dashed it on a rock, so that the ship was broken up.” |
| what it means | “Then Levi, girt about with sackcloth, prayed for us all unto the Lord.” | “and for the wickedness of my son Joseph you will be sent into captivity”; “Therefore I command you not to unite with the sons of Joseph, but only with Levi and Judah.” |
Both columns are Charles’s English, so they are set side by side and not measured against each other. In the Greek Testament Joseph seizes the bull and rises; in the Hebrew he beats Judah for ten staves, and the wreck and the captivity are laid to him. Which is older is Charles’s question and is not tested here.
4Q215 — a Hebrew fragment from Qumran with Bilhah’s genealogy. In the Abegg/ETCBC transcription held here (11 lines; CC BY-NC 4.0, so measured and not printed) the names are not Charles’s. Bilhah’s father is Ahiyot (1 unflagged occurrence) where Charles’s Greek has Ῥουθαῖος; the handmaid Laban gives him is Hannah (3 occurrences) where the Greek has Εὐνὰν. Deborah is there, as the one who nursed Rebekah (1). What the two share: Zilpah is born first (the Hebrew has the first, 1 unflagged), Bilhah second, and Bilhah’s name is tied to haste at the breast — the Hebrew has a participle hastening (2 unflagged) beside to suck (1 unflagged), and that is the reading of Charles’s right-hand column at 1:12, ἔσπευδεν θηλάζειν, not of his α text. Names and presence only: no word run between Hebrew and Greek is valid, and none is printed.
A shepherd who kills the lion, in words David also uses. Gad 1:3 and 1 Kingdoms 17:34 share five consecutive words, and that sequence stands nowhere else in the 30,348 Greek Old Testament units. Set side by side; Charles’s English of Gad never names David.
The ox that licks up the grass. Gad 2:2 has Gad wish to lick Joseph from the land of the living ὃν τρόπον ἐκλείχει ὁ μόσχος τὰ χλωρὰ τοῦ πεδίου. Charles’s apparatus: “Num. xxii. 4 which was in the mind of the writer also supports the text.” Measured: four words with Numbers 22:4, ὁ μόσχος τὰ χλωρὰ, and the run stands in one Greek Old Testament unit.
Order, hatred, the double face, the inclination — and a verb Paul also uses.
Greek counts are over the 30,348 CATSS units screened (the Odes excluded, as on every page here) and the 7,927 SBLGNT verses, on accent-free forms. Testament counts are over Charles’s English, which is one translator’s, and the page says which each time.
Naphtali’s word. In the verses read it stands four times in four verses (3:2–5) — sun and moon keep it, the nations and Sodom and the Watchers change it. In Charles’s English, order stands 8 times in Naphtali and 14 in all twelve testaments. The noun stands 9 times in 8 New Testament verses, most in Hebrews on the order of Melchizedek, and in 18 Greek Old Testament units (six more have the same letters as the future of the verb to set, appoint, and are parsed so in the CATSS morphology; they are not counted). The phrase order of nature stands in none of either.
Gad’s word. In Charles’s English hatred stands 13 times in Gad and 3 times in the other eleven testaments together. The noun never stands in the Greek New Testament, which has the verb instead, and 11 times in 10 Greek Old Testament units. Counted as noun forms only (nominative, accusative, genitive): the dative is spelled like the verb’s third person and is left out of both counts.
Asher’s word. No occurrence in the SBLGNT or in the CATSS Greek Old Testament. In Charles’s English two stands 12 times in Asher, and twofold or double-faced 9 more.
The two inclinations of Asher 1:3–5, and the inclination repentance guides in Gad 5:7. None in the Greek New Testament; 11 Greek Old Testament units — Ezekiel 11:5, Hosea 11:6, Hosea 4:9, Hosea 5:4, Hosea 7:2, Psalm 5:11, Psalm 9:23, Sirach 15:14, Sirach 17:6, Sirach 44:4 and Wisdom 1:9 (Psalms in the Greek numbering). Charles’s English renders it inclination.
The verb of the clause Charles set beside Romans 1:32. It stands in 6 New Testament verses (Luke 11:48, Acts 8:1, Acts 22:20, Romans 1:32, 1 Corinthians 7:12 and 1 Corinthians 7:13) and 3 Greek Old Testament units (1 Maccabees 1:57, 2 Maccabees 11:24 and 2 Maccabees 11:35). Uncommon, then, but not unique to Romans and Asher.
In Charles’s English, 27 times in the twelve testaments (Naphtali 2, Gad 0, Asher 2). The Greek New Testament has the -r spelling once, Βελιάρ (2 Corinthians 6:15); the Greek Old Testament units never have it, and have the -l spelling once (Judges (A) 20:13). No Greek of Beliar in these three testaments was read for this page, so the count is English.
Charles’s dependence claims, measured on his own Greek; the two ways; and the formulas that run deepest.
The screen: each of the 32 Greek units read off the page against all 7,927 SBLGNT verses and all 30,348 CATSS units, on normalize.grk_seq_cmp (accents, breathings and case removed; elision and MorphGNT’s own spelling pairs resolved). Contiguous runs of whole word forms; the tie band is how many verses of the other corpus reach the same depth, and for the New Testament verses of Charles’s pairs and of the page’s main contacts the band from their side too (that verse against the Greek Old Testament). A second normalisation is reported only where it changes a number: the movable ν folded on both sides (a final ν after -σι, -ξι, -ψι, in ἐστί(ν), and on forms in -ε(ν) whose ν-less form is attested). The OCR screen (end of this stage) uses the same normaliser with single words not counted, so its 0 means no two consecutive words; the Greek it prints is in normalised forms, without accents, because that is all the OCR can vouch for.
The distribution — the units read off the page:
Against the SBLGNT: all 32 reach two, 24 reach three, 11 reach four, 1 reaches five, and nothing reaches six. The deepest is Gad 2:5 (α) — ὁ Θεὸς τῶν πατέρων ἡμῶν, the God of our fathers, tie band 3 (Acts 3:13, Acts 5:30 and Acts 22:14): a formula.
Against the CATSS Greek Old Testament: all 32 reach two, 28 reach three, 16 reach four, 9 reach five, 2 reach six, and nothing reaches seven. The deepest are Naphtali 1:11 and Naphtali 3:2 — Dinah’s naming formula and a list of sun, moon and stars. Of the units reaching five or more, four have a tie band of one: Gad 1:3 (1 Kingdoms 17:34), Gad 6:3 (α) (Deuteronomy 6:6), Gad 6:3 (β, A, S¹) (Deuteronomy 6:6) and Naphtali 1:11 (Genesis 30:21).
These units were chosen for study, not at random, so this is not the testaments’ distribution; the OCR screen below is the wider one.
⚠️ Charles’s dependence claims — measured on his own Greek:
What Charles wrote, in 1913. Of Gad 6:3–7 on forgiveness, that these verses “contain the most remarkable statement on the subject of forgiveness in all ancient literature.” and “So perfect are the parallels in thought and diction between these verses and Luke xvii. 3, Matt. xviii. 15, 35, that we must assume our Lord’s acquaintance with them.” After setting Gad 6:3 (quoted from his β column), 6:6 and 6:7 (his table prints the last reference as v. 7; the Greek beside it is Gad 6:7’s) beside Matthew 18:15, 18:35 and Luke 17:3: “That the N.T. passages are here dependent cannot be reasonably denied.” Of Paul: “St. Paul seems to have used the book as a vade mecum.” — and “Only a few of the most notable passages where St. Paul draws upon the Testament will here be appended”, a list in which he pairs Romans 1:32 with Asher 6:2, 1 Corinthians 7:5 with Naphtali 8:8, and 2 Corinthians 7:10 with Gad 5:7 (APOT 2, pp. 292–293).
Forgiveness, every pairing of the verses he named (he set 6:3 β against Matthew 18:15 and 6:6–7 against 18:35, with Luke 17:3 beside all; longest run of consecutive words, with the words as Charles’s page prints them):
| Gad | Matthew 18:15 | Matthew 18:35 | Luke 17:3 |
|---|---|---|---|
| 6:3 α | 2 — εἰς σέ | 1 — ἀπὸ | 2 — καὶ ἐάν |
| 6:3 β | 2 — εἰς σέ | 1 — ἀπὸ | 3 — μετανοήσῃ ἄφες αὐτῷ |
| 6:6 | 2 — Ἐὰν δὲ | 1 — Ἐὰν | 1 — Ἐὰν |
| 6:7 | 2 — Ἐὰν δέ | 1 — Ἐὰν | 2 — ἄφες αὐτῷ |
The deepest cell is three — Gad 6:3 (β, A, S¹) with Luke 17:3. And Luke 17:3 does not stand out from the rest of the New Testament: the α column’s deepest run is three, reached by 31 verses on three different three-word runs (23 on καὶ ἐν τῇ, and in the; 6 on καὶ ἐάν τις, and if anyone; 2 on τῇ ψυχῇ σου, your soul), and the β column’s three with Luke is one of 33 verses at that depth. From the Gospel side, Luke 17:3’s deepest run into the Greek Old Testament is three, band 23; Matthew 18:15’s is four, band 1 (Deuteronomy 13:7; 13:6 in English Bibles); Matthew 18:35’s is four, band 2 (3 Kingdoms 2:32 and Tobit (BA) 9:4, καὶ ὁ πατήρ μου, and my father).
Paul, the three pairs he listed:
- Asher 6:2 → Romans 1:32 — two on the shelf’s normaliser, four with the movable ν folded. On the normaliser Romans 1:32 is one of the 116 New Testament verses that reach two with Asher 6:2. Folded, the four words καὶ συνευδοκοῦσιν τοῖς πράσσουσιν stand in one New Testament verse, Romans 1:32. Dropping the ν from the SBLGNT side only gives three, because Charles’s page prints the second participle with it. His 1913 quotation spells the verb with the ν, and as quoted it reaches four on the normaliser unaided. From Romans’ side the deepest Greek Old Testament run is three, band 6 (τὸ δικαίωμα τοῦ).
- Gad 5:7 → 2 Corinthians 7:10 — four, tie band 1. Ἡ γὰρ κατὰ Θεὸν … ἡ γὰρ κατὰ θεὸν, for the … according to God: repentance in Gad, grief that works repentance in Paul. The four words stand in one New Testament verse and no Greek Old Testament unit; from Paul’s side his verse reaches three in the Greek Old Testament, band 2. The deepest of Charles’s pairs on the shelf’s normaliser, at four words.
- Naphtali 8:8 → 1 Corinthians 7:5 — one. A season for one’s wife and a season of abstinence for prayer; Paul’s ἵνα σχολάσητε τῇ προσευχῇ shares the thought and one word in sequence. Against Ecclesiastes 3:5 (a time to embrace and a time to refrain) Naphtali 8:8 measures two. From Paul’s side his verse reaches three in the Greek Old Testament, in a band of 57.
Charles argued from “thought and diction”. The screen measures diction only, and on his own Greek, read off his own pages, the diction of the pairs he named there from these testaments comes to runs of one to four consecutive words. That does not show the thought is unshared — the forgiveness of a brother who repents is plainly in both — and it does not show independence any more than a long run would show dependence. The page asserts no direction of influence; it reports Charles’s, and the number his claim rests on.
The two ways — one word, and where the Greek goes instead:
Asher 1:3 and 1:5 against the texts that set two ways before Israel: τὴν ζωὴν καὶ τὸν θάνατον (Deuteronomy 30:15) and τὴν ὁδὸν τῆς ζωῆς καὶ τὴν ὁδὸν τοῦ θανάτου (Jeremiah 21:8) — one word at every pairing, and the word is and. Against Matthew 7:13–14, the narrow and the wide way: one; from the Gospel’s side, Matthew 7:13 and 7:14 both reach three in the Greek Old Testament, bands 3 and 6.
Where the Greek goes instead. After its first two words, Asher 1:3 has five that stand in two Greek Old Testament units, both Ecclesiastes (1:13; 3:10 — God has given to the sons of). Asher 1:4, Διὰ τοῦτο πάντα δύο δύο εἰσίν, ἓν κατέναντι τοῦ ἑνός, shares four consecutive words with Sirach 33:15 and 42:24, and those four stand in no other unit; Sirach 33:15 also has δύο δύο immediately before them, where Asher 1:4 has a verb between. A doctrine of pairs, in words Sirach also uses. No dependence is asserted. Charles judged: “Our author drew also upon a Hebrew document dealing with the ‘Two Ways’ in the T. Asher” (APOT 2, p. 290). No such document is held, and the page tests nothing about it.
⚠️ Counter-examples — runs that look like contact and are not:
- Gad 6:1 → Exodus 32:27 and 1 Kingdoms 20:41 — five words, tie band 2. ἀγαπᾶτε ἕκαστος τὸν πλησίον αὐτοῦ, καὶ — love each his neighbour — and Exodus 32:27, ἕκαστος τὸν πλησίον αὐτοῦ, where the Levites are told to kill each his neighbour (in 1 Kingdoms 20:41 David and Jonathan kiss one another). Against Leviticus 19:18, love your neighbour, Gad 6:1 measures two. Depth is not meaning.
- Gad 2:5 (α) → Acts 3:13, 5:30, 22:14 — five, tie band 3. The deepest New Testament run of all the units read is the God of our fathers, which also stands in 10 Greek Old Testament units. A formula.
- Naphtali 3:2 → Genesis 37:9 and Ecclesiastes 12:2 — six, tie band 2. Sun, moon and stars, in Joseph’s dream and in the Preacher’s old age. A list.
- Gad 1:3 → Luke 22:41 — four, tie band 1. From Luke’s side the deepest Greek Old Testament run is two, in a band of 316. A stone’s throw is a measure of distance; the run is printed so that the others are not read as more than they are.
The wider screen — all 166 verses, through the OCR (a lower bound):
The OCR of the three testaments — Naphtali 1,503, Gad 1,206, Asher 1,086 words after cleaning — screened as three streams. Of the 7,927 SBLGNT verses, 45 reach four, 8 reach five, and none reaches six. The five-word runs are αβρααμ και ισαακ και ιακωβ, επι προσωπον πασησ τησ γησ, κατα το πολυ αυτου ελεοσ and ο θεοσ των πατερων ημων — Abraham and Isaac and Jacob; upon the face of all the earth; according to his great mercy; the God of our fathers.
Of the 30,348 Greek Old Testament units, 10 reach six or more and 1 reaches seven: εν τω εβδομω μηνι μια του μηνοσ (Ezekiel 45:20, seven); θυγατερα και εκαλεσεν το ονομα αυτησ (Genesis 30:21, six); και εθαψαν αυτον μετα των πατερων (4 Kingdoms 12:22, 4 Kingdoms 13:9, 4 Kingdoms 15:7, 2 Chronicles 25:28, 2 Chronicles 26:23 and Judith 8:3, six); και η σεληνη και οι αστερεσ (Ecclesiastes 12:2, six); ο ηλιοσ και η σεληνη και (Genesis 37:9, six) — a date, a burial, a naming, the lights of heaven.
How much the OCR understates. Of eleven contacts measured both on the OCR and on the lines read off the page, the OCR gives the same number in ten and a shorter one in one: 1 Kingdoms 17:34 (three on the OCR, five on the page). No OCR number on this page is a ceiling.
What this page could not check, what was wrong in its own files, and where it is weakest.
This is the section that makes the shelf trustworthy. For the Testaments it is long, because the Greek is not on the disk as text.
⚠️ The English is not a line-for-line translation of the Greek beside it:
In at least eight of the 28 verses read, the English renders something other than the main line of Charles’s printed Greek — a margin reading, an apparatus reading, another column, or his own emendation. In three of them the main-line word carries his crux dagger. The other verses were not audited word by word.
- Naphtali 1:11 — village = the apparatus reading κώμης; text πόλεως.
- Naphtali 1:12 — the right-hand column; the α column has the short line he marks corrupt.
- Naphtali 3:2 — change not = the margin ἀλλοιοῦσι; text καλύψουσι, marked corrupt.
- Naphtali 4:1 — I have read = the margin ἀνέγνων; text ἔγνων.
- Naphtali 8:4 — my children = the margin καλόν, τέκνα μου.
- Gad 2:4 — the β column; the α column has no verse 4.
- Gad 6:1 — his brother = the margin τὸν ἀδελφόν; text τὸν πλησίον αὐτοῦ.
- Asher 6:2 — guilty of a twofold sin = Charles’s emendation of κολάζονται, marked corrupt; his note: “There is no mention here of punishment.”
So the English is Charles’s English of his preferred text, not of the α text he printed. A reader who sets this English beside his Greek, or beside de Jonge’s, must expect it to part company in places, and this page says where it found that.
⚠️ Faults in the files, found on the page images — reported, not buried:
- The English file, Naphtali 4:1: for I. have read; the 1917 page (p. 79) has no stop. Printed corrected.
- The English file, Naphtali 8:4: a full stop after through you; the 1917 page (p. 81) has a comma. Printed corrected.
- The English file, Gad 1:6: zilpah in lower case; the 1917 page (p. 83) has a capital inside the half-brackets. Not printed here.
- The scan: PDF pages 226 and 227 are second scans of printed pages 154–155. The OCR screen uses one copy.
- 4Q215 is on this disk. The acquisition notes for the Testaments say it is not held; it is, inside the Abegg/ETCBC Dead Sea Scrolls features filed with the Ben Sira witnesses, and no path in the shelf’s tooling points to it.
⚠️ What this page could NOT check:
- Any Greek not read off the page. 138 of the 166 verses are known here only through OCR. A contact deeper than anything on this page could hide in them.
- De Jonge’s text. Every Greek word here is Charles’s printed text, or a column, margin or apparatus reading he printed. His text is built on the α family but is not always α’s: at Asher 7:3 it prints the preposition of MS g and the Armenian, and his apparatus says so. Other departures of that kind in the verses read were not audited word by word. A run measured on another edition may differ.
- Charles’s accents, at the edge of legibility. The lines were read at 300–900 dpi; where a mark was uncertain the sidecar says so (the iota subscript in the second-last word of Gad 1:3 is one). Accents do not affect any run, which is measured accent-free.
- The Armenian and Slavonic. Not held; known only through Charles’s sigla A and S.
- The late Hebrew Testament of Naphtali in Hebrew. Only Charles’s English is held, so the Stage 2 table compares contents, not words.
- Charles’s 1908 list. His English edition of 1908 (held here as OCR only; introduction §26, pp. lxxviii–xcii, to which APOT 2, p. 291 refers the reader) names further New Testament parallels from these testaments — among them James 4:7 with Naphtali 8:4, Romans 9:21 with Naphtali 2:2, Jude 7 with Naphtali 3:4, Matthew 12:35 with Asher 1:9, Luke 24:32 with Naphtali 7:4 and Acts 8:23 with Naphtali 2:8. The first three fall on verses read off the page and measure three, two and two consecutive words, in that order; the others were not read off the image. Only the 1913 pairs are counted here as his.
- Charles’s dating and his Hebrew original. Reported from his introduction; not tested.
- The Didache and other two-ways texts outside the Bible. Not held; not measured.
⚠️ Where this page is weakest:
- The units were chosen. The 28 verses read were picked for what Charles and earlier pages pointed at, so their distribution says nothing about the testaments as a whole; the OCR screen is the whole, and it understates.
- The OCR rules are the page’s own. Where the apparatus starts was set by eye on seven pages, and the token repairs accept a stripped form only if it is a word of the SBLGNT or CATSS. A different set of rules would give different lower bounds; none could give a run that is not in the scan.
- The movable ν. The headline numbers are on the shelf’s normaliser; the folded figure is reported beside them where it differs, and Asher 6:2 is the one place it changes what the page says.
- The structure is a division the page made. Its blocks are by content, and Gad’s teaching holds his own story of repentance.
Commentaries & studies:
- R. H. Charles, The Greek Versions of the Testaments of the Twelve Patriarchs (1908) — held as images; the Greek, the apparatus notes and the two notes quoted (pp. 161, 178–179).
- R. H. Charles, APOT 2 (1913) — held; the introduction (pp. 282, 290, 292–293) and Appendix I (pp. 361–363).
- M. de Jonge et al. (1978) — not held.
- H. C. Kee in OTP 1 (1983) — not held.
A scholar’s certainty, a measured four words, and a teaching on hatred worth hearing anyway.
Name what you learned, why it matters, and one concrete response.
What should we carry away — and do?
- “Cannot be reasonably denied” is a claim, not a measurement. Charles’s strongest words about these testaments, in 1913, rest on pairs that share one to four consecutive words on his own Greek. Ask what a parallel is made of before you repeat it.
- Which Greek? One letter separates two of Charles’s printings of Asher 6:2, and one column separates two of his readings of Gad 6:3; each changes a number on this page.
- Gad’s confession is still worth hearing. “These things I learnt at last, after I had repented concerning Joseph.” A man who wanted his brother dead, telling his sons what hatred did to him.
- The church’s view, labelled as such. Pleasant Springs reads the Testaments as a witness, not Scripture. The command to forgive a brother from the heart is Scripture’s own — ἐὰν μὴ ἀφῆτε ἕκαστος τῷ ἀδελφῷ αὐτοῦ ἀπὸ τῶν καρδιῶν ὑμῶν (Matthew 18:35) — and vengeance left to God is Paul’s, quoting Deuteronomy 32:35 (Romans 12:19). Gad 6:7 says something like both; the church teaches them from the canon. Likewise the bracketed words of Asher 7:3 are reported as Charles’s brackets: the page draws no teaching about Christ from them.
- One concrete response. Name one person you are holding a grudge against, and this week do what Matthew 18:15 commands: go to him or her, alone. Gad 6:3 counsels speaking peaceably; the command is the Gospel’s.
Asher 6:2 · to share
Charles’s Greek (1908, read off the page) · his English (1917)Every Deep Research page traces its passage across the whole sweep — the Old Testament, the Second-Temple world between the testaments (historical witnesses, not Scripture), and the New Testament. Follow the thread:
Pleasant Springs Church — Deep Research
Research method after Michael S. Heiser’s Bible Research Method (examine dictionaries → ancient literature → important words → textual connections → academic publications → share), with the Pleasant Springs “Author & Audience” foundation. Scripture set in the Hebrew, the Greek editions named on each page and the ESV (Old Testament), and the Greek New Testament and the ESV (New Testament), per the practice of Pleasant Springs Church. ⚠️ Status: the Testaments of the Twelve Patriarchs are not Scripture for Pleasant Springs and are not deuterocanonical; filed under Pseudepigrapha & Second Temple and read as a witness. Greek: R. H. Charles, The Greek Versions of the Testaments of the Twelve Patriarchs (Oxford, 1908), public domain, read off the archive.org page images for every Greek line of the Testaments printed here, each recorded with its page in the sidecar file; the OCR of the same scan is used only to screen, as a lower bound. Not de Jonge’s edition. English: Charles, The Testaments of the Twelve Patriarchs (1908; SPCK 1917), public domain; two lines printed as the 1917 page has them. Charles’s introduction and Appendix I are quoted from APOT vol. 2 (Oxford, 1913), public domain, read off the page images. Canonical Greek: the CATSS morphological Septuagint and the SBLGNT. Measured, not printed: the Abegg/ETCBC transcription of 4Q215 (CC BY-NC 4.0). No ESV verse is printed on this page; italic English beside canonical Greek is the page’s own gloss. Direction of influence is nowhere asserted.
Scripture quotations marked ESV are from the ESV® Bible (The Holy Bible, English Standard Version®), © 2001 by Crossway, a publishing ministry of Good News Publishers. ESV Text Edition: 2016. Used by permission. All rights reserved. The ESV text may not be quoted in any publication made available to the public by a Creative Commons license, and the ESV may not be translated in whole or in part into any other language. www.esv.org
|
Did this study bless you today? 💚 Give to Support PS Church100% of gifts go to the General Fund — thank you. |