The Epistle of Enoch, the Apocalypse of Weeks, and the Birth of Noah

• Anchor Bible Dictionary — its entries on Enoch and the books of Enoch, Noah, and the Pseudepigrapha.
• Eerdmans Dictionary of Early Judaism — its entries on 1 Enoch and on the Enoch literature.
• Dictionary of New Testament Background (IVP) — its entries on 1 Enoch and on apocalypticism. These are subjects to look up; the exact entry titles were not checked, and none of these works is held on this disk.
• On the text — any article that says which witness it is quoting. An article that cites ‘1 Enoch 91:12’ without saying whether it means the Ethiopic’s chapter 91 or the Qumran copy’s place for it is not usable for these chapters.

• R. H. Charles, The Book of Enoch, or 1 Enoch (Oxford: Clarendon Press, 1912) — the English quoted here, and the notes quoted from its pages li and 227–267.
• C. Bonner, with H. C. Youtie, The Last Chapters of Enoch in Greek (Studies and Documents 8; London: Christophers, 1937) — the Greek measured here, through its encoding.
• J. T. Milik, The Books of Enoch: Aramaic Fragments of Qumrân Cave 4 (Oxford: Clarendon Press, 1976).
• George W. E. Nickelsburg, 1 Enoch 1 (Hermeneia; Minneapolis: Fortress, 2001).
• Loren T. Stuckenbruck, 1 Enoch 91–108 (Commentaries on Early Jewish Literature; Berlin: de Gruyter, 2007).
• Michael A. Knibb, The Ethiopic Book of Enoch (2 vols; Oxford: Clarendon Press, 1978).

Genesis 5 is the family tree these chapters live in. Enoch, the seventh from Adam; his son Methuselah; Methuselah’s son Lamech; Lamech’s son Noah, whose name Genesis explains. 1 Enoch 106–107 tells that birth at length. Its Latin gives Lamech an age, 350, and the flood a date, after five hundred years: both numbers stand in Genesis, but not for those things.

• Genesis 5:28 (Lamech’s age when Noah is born) ↔ the Latin of 106:1, annorum tricentorum quinquagenta, of 350 years. The Greek Genesis gives ἑκατὸν ὀγδοήκοντα ὀκτὼ ἔτη (188); the Masoretic, the Vulgate (centum octoginta duobus annis) and Dillmann’s Ethiopic (፻፹ወ፪ዓመተ) give 182; the Samaritan Genesis on this disk, 53. Three hundred and fifty is in all four Genesis texts held for chapter 9 — as Noah’s years after the flood (the Samaritan file holds only chapters 5 and 11; Genesis 9:28: τριακόσια πεντήκοντα ἔτη; trecentis quinquaginta annis; ፫፻ወ፶ዓመተ). The page records the coincidence of the number and does not explain it. The Geʻez, Charles and the Greek papyrus give Lamech no age at 106:1.
• Genesis 5:32 (Noah five hundred years old) ↔ the Latin of 106:15, post quingentos annos, after five hundred years God will send the flood. Five hundred stands in Genesis 5:32 in the Greek (ἐτῶν πεντακοσίων), the Hebrew and the Ethiopic (፭፻ዓመተ), and in the Clementine Vulgate file inside its 5:31 (Noë vero cum quingentorum esset annorum). The Latin alone among the witnesses of 106 names the three sons — sem cham iafeth — where the Geʻez, Charles and the papyrus say only three children.
• Genesis 5:29 (Noah’s name explained) ↔ 106:18 and 107:3. Genesis gives the reason four ways: Hebrew יְנַחֲמֵ֤נוּ, he will comfort us; Greek διαναπαύσει ἡμᾶς, he will give us rest; Latin consolabitur, he will comfort; Dillmann’s Ethiopic ይናዝዘኒ, he will comfort me. In 1 Enoch the Latin says rest (106:18); at 107:3 Dillmann’s Geʻez has ያስተፌሥሓ ፡ ለምድር, he will make the earth rejoice, and Charles prints comfort, noting that the derivation of Genesis 5:29 is repeated. The glosses are this page’s; the Geʻez verbs are a causative of rejoice and a verb of comforting.
• Genesis 6:1–4 ↔ 106:13–14, 17: angels in the days of Jared who sinned with women and fathered offspring not according to the spirit but according to the flesh — the story of chapters 6–16, told again inside the birth of Noah. On the leather of 4Q204, ביומי ירד א[בי], in the days of Jared my fa[ther].
• Luke 6:24–26 and Sirach 2:12–14 ↔ the woes of 94–103. Measured in Stage 4: Luke 6:25 shares woe to you who with nine papyrus rows as encoded; Luke 6:24 reaches deeper into Sirach 2:14 than into the papyrus.
• Tertullian, De idololatria 4 ↔ 99:6–7. His text, as the OCP encodes it, carries the oath of 99:6 — Et rursus iuro vobis peccatores — which the Greek papyrus lacks (it has no row at 99:6), and the list of idols of 99:7 with stone in it, which the Geʻez lacks. The critical edition (CSEL) is not held; the Latin is the OCP’s.
• The Genesis Apocryphon from Qumran Cave 1, commonly described as telling Noah’s birth and Lamech’s fear too, is not held and is not compared.

• The oaths. In Charles, six verses: I have sworn unto you at 98:4 and 98:6 (to the sinners), and I swear (unto you or to you) in four, 98:1 (to the wise and the foolish), 99:6 (to the sinners), 103:1 and 104:1 (to the righteous). The Geʻez has the same six: መሐልኩ ፡ ለክሙ at 98:4 and 98:6, እምሕል ፡ ለክሙ at the other four. Be hopeful in four (96:1, 102:4, 104:2, 104:4); fear not, said to the righteous, in four (95:3, 96:3, 102:4, 104:6).
• The papyrus’s editor. 44 of its 113 rows carry Greek words inside Bonner’s square brackets (one more only a bracketed gap) and 3 carry braces; 26 mark a gap with dots. Normalised, the rows hold 2,643 Greek words as encoded and 2,524 with every restoration removed — three brackets open in one row and close in the next (98:11–12, 100:1–2, 101:3–4) and are followed across the row end. One bracket cannot be followed: in 98:11 a bracket closes with none open (Stage 4).
• The Aramaic, column by column (letters on the leather / restored): 4Q212 1 ii 128 / 294; 1 iii 137 / 200; 1 iv 393 / 182; 1 v 172 / 294; 4Q204 5 i 44 / 402; 5 ii 271 / 456 (letters only on both sides: the SQE count of preserved signs adds seven punctuation marks, five stops and two dashes, and its count of restored signs adds two of the editors’ question marks). Column iv of 4Q212 — the one where the eighth week follows the seventh — is the only one of the six with more letters on the leather than restored.
• Chapter 105: no Greek (the papyrus has no row); 21 letters of Aramaic in three lines of 4Q204 5 i; the next line, 24, none; then a blank line before 106:1.
• Chapter 108: Geʻez and Charles only — no Greek, no Aramaic, no Latin.

The problem. In the Ethiopic the last three of the ten weeks (91:12–17) stand in chapter 91, two chapters before the first seven (93:1–10). Charles printed the chapters out of numeric order to mend it, and in his introduction he gave what he took to be the book’s first order — read here on the page image, p. li: “Thus the original order of the book is” 92, 91:1–10 and 18–19, 93:1–10, 91:12–17, 94. The two passages left out of that list, 91:11 and 93:11–14, are the two whole passages in 91–93 he prints inside square brackets.

Where the orders part. Set Charles’s printed translation against Dillmann’s Ethiopic and they give a different next verse at 8 joins in 91–94: the joins disturbed by his three moves (chapter 92 to the front; 91:12–17 to after 93:14; the last clause of 91:14 to its front). The leather of 4Q212, which he never saw, keeps letters on both sides of 7 of them — the eighth, after 91:17, falls where thirteen lines are lost — and one more join where Charles and the Ethiopic agree and the leather does not, after 93:10. Each row is read from the preserved letters of one column:

The count. At the 7 surviving joins where Charles’s printed order gives a different next verse from the Ethiopic, the leather has the Ethiopic sequence at 7 and Charles’s at 0. His 1912 ‘original order’ decides 3 of the nine joins at the level of whole verses and agrees with the leather at 0. The seven are not seven independent tests: they are the joins of three moves (chapter 92, two joins; 91:12–17, two; the clause of 91:14, three). And the one join among these nine where the leather leaves the Ethiopic is the one Charles argued for: after the seventh week comes the eighth — with one verse, the verse he called “an interpolation added by the final editor”, in between.

So the Qumran copy confirms Charles’s large move and, at the nine joins counted here, none of its details. It carries both whole passages he bracketed in this stretch — 91:11 (32 letters, line 14 of column iv) and 93:11–14 (147 letters, lines 15–23 of column v); it keeps 91:14’s last clause last; and it runs 91:19 into 92:1 and 92:5 into 93:1, where his printed translation has 93:1 and 91:1.

What the joins cannot show — and one join they do not count. Column iv ends at line 26 with 91:17 and the first thirteen lines of column v are lost, so what followed the tenth week on the leather is not known. The joins above are counted by the verse that follows. Counted by the verse that precedes, one more survives: 91:18 opens on line 18 of column ii, ארחת קשט[א], paths of truth, and line 17 above it keeps 9 letters, כל דרי עלמי[ן], all generations of eterni[ty]. Whatever they are, they are not 91:11 or 91:17 — both stand in column iv — so at the entry to 91:18 the leather has neither the Ethiopic’s verse before it (91:17) nor Charles’s printed one (91:11). Unless the copy carried a verse twice, it follows too that the lost lines after 91:17 did not hold 91:18. The four lines above (13–16) keep 19 letters, which the acquisition placed at 91:10 (the editors restore line 13 as the righteous shall arise from their sleep, 91:10’s matter); from the preserved letters alone this page could not match them to the Ethiopic 91:10. If that placement is right, the leather has Charles’s 1912 ‘original order’ at this join — 91:10, then 91:18 — and that order agrees with the leather at one join, not none. The column sequence (ii, iii, iv, v) is the editors’; every join on this page is inside one column.

Charles printed 106:17 after 106:14 — “I have restored this verse to its original place” — so that the giants not according to the spirit, but according to the flesh follow the angels who married women. The papyrus, published six years after his death, has the first half of 106:17 exactly there: its row for 106:14 runs on, without a break, into the clause the encoding labels 17a, and only then comes 106:15. The second half it has after 106:16, where the Ethiopic has the whole verse, in a row labelled 17b. On the leather of 4Q204, the line where the editors restore 17a (5 ii 19) has no letter preserved (0 / 55); the lines after 106:16 keep [ותתד]כא ארעא [מן] חבלא [ר]בא — [cleans]ed the earth from the [gr]eat corruption — the matter of 17b, in the Ethiopic’s place.

113 papyrus rows against the 7,927 verses of the Greek New Testament: all 113 share a word, 112 reach two, 95 three, 26 four, 7 five, and one — 106:1 — reaches seven; no other row passes five, and nothing reaches eight.

Against the 30,348 units of the Greek Old Testament: 112 reach two, 103 three, 49 four, 14 five, 3 six, and one reaches eight.

The deepest New Testament run in these chapters is a formula, and it is the page’s first counter-example. 106:1 against Matthew 1:25 — ἔτεκεν υἱόν· καὶ ἐκάλεσεν τὸ ὄνομα αὐτοῦ, she bore a son, and he called his name — seven words, band 1. The same seven stand in six Greek Old Testament units (1 Chronicles 7:23, 1 Samuel 1:20, 2 Samuel 12:24, Genesis 38:3, 4, 5), so from Matthew’s side the Greek Old Testament reaches seven with band 6; and with the and before them, καὶ ἔτεκεν υἱόν καὶ ἐκάλεσεν τὸ ὄνομα αὐτοῦ, eight words, the papyrus row and three Greek Old Testament units agree and Matthew drops out. And in the papyrus the child so named is Lamech (codex, 106:1). A birth-and-naming formula; nothing more.

The whole Greek New Testament, all 7,927 verses, screened against the 113 rows (single words not counted): 2,487 share no two consecutive words with them, 4,714 share two, 677 three, 38 four, 10 five and 1 seven. Of the 726 that reach three or more, 18 reach deeper here than anywhere in the 30,348 Greek Old Testament units; six of them do it at five words:

Read, not counted, four of the six are common words — the sea and the land, great upon the earth — and the run engine ranks them beside the other two. Two carry a subject: 1 John’s and the papyrus’s day of judgement, with the article, in no Greek Old Testament unit (Matthew says in a day of judgement, without it); and the naming formula of Luke 2:21, which the Greek Old Testament has only four words long. No citation formula stands in either New Testament verse, and the page assigns neither to Enoch.

The other five New Testament verses that reach five do not run deeper here: Matthew 1:25 (above); Matthew 13:42 and 13:50 (next); and Acts 4:24 and 14:15, whose five words with the papyrus, καὶ πάντα τὰ ἐν αὐτοῖς, and all that is in them, are the tail of a thirteen-word run each has with Exodus 20:11.

98:3 against Matthew 13:42 and 13:50 — five words, band 2: εἰς τὴν κάμινον τοῦ πυρός, into the furnace of fire. It looks like contact. The same five words stand in eight Greek Old Testament units, all Daniel 3 (εἰς τὴν κάμινον τοῦ πυρὸς, 3:6, 11, 15, 20, in both Greek Daniels), so it is a formula.

And it is not on the papyrus. In the encoding of 98:3 the whole clause about spirits cast into the furnace of fire stands inside Bonner’s square brackets, except the first letter of its first word. Remove every restoration from every row (following the three brackets that run on into the next row) and re-run: 98:3 falls from five to two against the New Testament and from five to three against the Greek Old Testament; altogether ten rows change depth against the New Testament and twelve against the Greek Old Testament. That re-run also drops the words inside the braces of three rows (102:2, 106:10, 106:18); the files held do not say whether braces mark letters supplied or letters struck, so the re-run was made both ways — keep the braced words and the counts are nine and ten, and 98:3 falls exactly as before. Bonner’s own account of his text, as the archive.org record of his book quotes it: corrected chiefly by the use of the Ethiopic version — and Charles’s Ethiopic-based English of 98:3 ends “Their spirits shall be cast into the furnace of fire.” The run measures the editor’s Greek, not the scribe’s. None of the six runs in the table above, and not the seven of Matthew 1:25, touches a restored word. The woes below carry a caveat. In the encoding of 98:11 a square bracket closes inside the row’s thirteenth word with no bracket opened in that row or left open by the row before; read as closing a restoration that began at the row’s start, it puts woe to you who in that row inside Bonner’s brackets. Read that way the re-run’s counts are eleven and thirteen, and Luke 6:25’s three words stand in eight papyrus rows, not nine. The encoding does not say where that bracket opened, so the page reports both.

Luke 6:25 — οὐαὶ ὑμῖν, οἱ, woe to you who — shares those three words with nine papyrus rows as encoded (97:7, 97:8, 98:11, 98:13, 98:15, 99:1, 99:2, 99:15, 100:7; eight if the unopened bracket of 98:11, above, is read as a restoration). From Luke’s side the band is 9; its best in the Greek Old Testament is two words, band 18. The three-word sequence stands in no Greek Old Testament unit and in no other New Testament verse. Shallow, and rare.

Luke 6:24 — woe to you the rich — reaches three words with one Greek Old Testament unit, οὐαὶ ὑμῖν τοῖς (Sirach 2:14), and only two with the papyrus. The verse it most resembles is 94:8, Woe to you, ye rich, and 94:8 has no Greek: the comparison the two texts invite is the one this shelf cannot measure.

Three thematic pairs that measure next to nothing. The rich man of Luke 12:16–21 against the rich of 97:8–10: two words at most (and the; that not). Revelation 14:20’s blood ἄχρι τῶν χαλινῶν τῶν ἵππων, up to the horses’ bridles, against 100:3’s horse walking in the blood of sinners: one word; against all 400 Greek Enoch rows held, two. Philippians 2:15’s lights in the world against 104:2’s lights of heaven: one. A theme is not a text.

Normalisation, declared exactly as on the Jubilees pages: words split on the Geʻez word divider and on spaces; the full stop and other Ethiopic punctuation removed; every word form kept as written, prefixes included. 0 means no two consecutive word forms in common. An optional fold merges the letters manuscripts interchange (ሀ/ሐ/ኀ, ሰ/ሠ, አ/ዐ, ጸ/ፀ) and is reported with and without.

The 206 verses (3,400 word forms) against Dillmann’s Ethiopic Genesis, 1,526 rows: 122 reach nothing, 78 reach two, 6 reach three, nothing reaches four. Against VanderKam’s Geʻez Jubilees, 1,308 rows: 89 reach nothing, 101 two, 14 three, 2 four. Folded, not one depth or band moves in either screen.

Two of Genesis’s six three-word partners are births. 106:18 with Genesis 5:29, band 1 — ስሞ ፡ ኖኅ ፡ እስመ, his name Noah, for — the one verse of Genesis in which Lamech names him; and 106:2 with Genesis 25:25, band 1, like a rose flower (codex). The other four are phrases: on the day in which (102:5, Genesis 5:1–2), upon all the earth (106:15), those who are in (103:9), and the naming formula and he called the name of that (107:3, Genesis 32:3 and 35:7). The Jubilees fours are 92:5 with Jubilees 1:23, እምይእቲ ፡ ዕለት ፡ እስከ ፡ ለዓለም, from that day for ever, and 106:17 with Jubilees 23:22, ወይከውን ፡ መቅሠፍት ፡ ዐቢይ ፡ ዲበ, and there will be a great punishment upon (Jubilees measured, not printed).

• The Qumran editions. Milik (1976) and the official edition are not held, and no local file records when or where 4Q204 and 4Q212 were found; after Charles’s death rests on the commonly reported discovery of Qumran Cave 4 in 1952 (not checked here). The placement of every Aramaic line against Charles’s verses is a reading of the SQE text — by the acquisition and, line by line for the nine joins, by this page. The observation that 4Q212 has 91:11–17 after 93:10 is the acquisition’s note and is commonly reported in the literature (not checked here); what this page adds is the join-by-join comparison with Charles’s own two orders.
• The column sequence of 4Q212. The joins are each inside one column; the order of the columns themselves (ii, iii, iv, v) is the editors’ and is not checked.
• The Greek papyrus itself. Only Bonner’s text as the OCP encodes it, his restorations in it; not the papyrus, not Bonner’s apparatus, not Kenyon’s edition of the Chester Beatty leaves. The restorations were measured with and without; the papyrus’s date is the acquisition record’s.
• The Ethiopic manuscripts. Dillmann’s apparatus and Knibb’s edition are not held; every Geʻez reading is Dillmann’s text as digitised, cut to Charles’s verses by inference. Charles translated his own Ethiopic text (1906), which is not held either.
• The Latin editions. James (1893) for the British Library fragment and CSEL for Tertullian are not held; the Latin is the OCP’s.
• The commentaries named in Stage 1 — none held, none consulted, except Charles’s 1912 pages li, 227, 231–233, 245–246 and 267, read on the page images.

• The order finding rests on eight surviving joins in one manuscript (the seven where Charles’s printed order gives a different next verse, and 93:10) and the entry to 91:18, each read by eye from a transcription, two of them (92:5 at 10 letters in lines 16–18, 92:1 at 8 letters in its first line) on thin evidence. The page prints every letter it relies on so a reader can check each one.
• The Greek is measured on an editor’s text. Its sharpest-looking run (98:3, Matthew 13) is the editor’s; the re-run without restorations is printed, but the reverse risk — an editor’s reading that is wrong but unbracketed — cannot be tested without the papyrus.
• The glosses of Geʻez and Aramaic are this page’s, made with the help of Charles’s English and the editors’ restorations; they are offered to make the printed letters readable, not as translations.
• The structure’s frame is one name, counted in two witnesses. It is a frame, not a design argument; Charles took the chapters at both ends for additions.

• R. H. Charles, The Book of Enoch, or 1 Enoch (Oxford: Clarendon Press, 1912); English reprinted by SPCK, 1917.
• A. Dillmann, Liber Henoch aethiopice (Leipzig, 1851).
• C. Bonner, with H. C. Youtie, The Last Chapters of Enoch in Greek (London, 1937).
• J. T. Milik, The Books of Enoch (1976).
• George W. E. Nickelsburg, 1 Enoch 1 (Hermeneia, 2001).
• Loren T. Stuckenbruck, 1 Enoch 91–108 (CEJL, 2007).
• Online Critical Pseudepigrapha (ed. I. W. Scott et al.), the Greek and Latin encodings used here.
• Scripta Qumranica Electronica, transcriptions of 4Q204 and 4Q212.

• The church’s view, labelled as such. Pleasant Springs reads 1 Enoch as a Second-Temple witness, not as Scripture. No New Testament verse introduces words of these chapters as a quotation; the runs above are measurements of shared wording, and the page does not argue from them that any New Testament writer read this book.
• A reconstruction and a manuscript can agree on the large move and differ at the joins. Charles argued that the weeks were out of order; a manuscript found after his death agrees — and at the seven surviving joins where his printed order gives a different next verse from the Ethiopic, it does not follow him. Hold reconstructions — including our own — with an open hand.
• Thirty-two woes, all aimed at people who are winning. The rich, the violent, the makers of idols. Luke 6:24 has Jesus say woe to you the rich — and that word, not 1 Enoch’s, is the one this church stands on (the church’s view).
• Measure, then read. A five-word run with Matthew 13 turned out to be an editor’s restoration; a three-word run with Luke 6 turned out to be rarer than it looked.
• One concrete response. The woes are addressed to those who trusted in your riches and have not remembered the Most High in the days of your riches (94:8) — the witness’s words; the word we stand on is Luke 6:24’s. This week, name one thing you own that you trust more than you should, and give part of it away.