The Textual Witnesses

The complete text-critical apparatus to "Sit at My Right Hand." Every place where the main text or a box refers to this appendix is written out here: the witnesses, the readings, what follows from them and what does not.

Witnesses used. For the Hebrew, the Leningrad Codex (WLC/OSHB, morphologically tagged; all counts mechanical and exhaustive). For the Greek of the Old Testament, the Septuagint in the edition of Swete and the text of Rahlfs where it differs. For the New Testament, eight editions: Nestle 1904, Westcott-Hort, the NA27/UBS4 variants, the RP Byzantine majority text, the text of the Greek Church, Tischendorf 8, and the two Textus Receptus editions of Scrivener and Stephanus.

Date of the last check: 8 August 2026.

B0. The rule of this book

At all the places below one thing holds, and it is the text-critical rule derived in chapter 2:

The writer of Hebrews never invents. He chooses.

Every reading he cites is attested somewhere — in a manuscript, in a translation, or in the consonants themselves. In one place something was added (B7b), and there it is said in so many words.

I. THE FOUR ROYAL PSALMS

B1. Psalm 2:6 — active or passive, and who is speaking

The matter. The consonants נסכתי carry both readings. The Masoretes vocalize actively (qal, God speaks); the Greek translators read it passively, and thereby placed the sentence in the mouth of the king. In the Greek the whole passage from here on thus stands in one voice: the king reports his appointment and then reads out the decree.

What gives it extra weight. The Greek reading presupposes precisely the form that stands in Proverbs 8:23, where the Masoretes themselves vocalize passively: נִסַּכְתִּי, "I have been appointed." These are the only two places in the Hebrew Bible where this root means "to appoint, to install" (lemma 5258 b, exhaustively counted).

What this book does. It follows the Masoretic reading in the main text and treats the Greek as a second voice, in chapter 3. Note: the counts well documented; the connection between the two places weighed.

B2. Psalm 2:7 — the ole we-yored against the construct

The matter. The Greek reads a construct chain: the decree of YHWH. The Masoretic accentuation rules that reading out. At this place stands the ole we-yored, the heaviest disjunctive of the poetic accent system of the three books (Psalms, Job, Proverbs), with the ole on אֶל and the yored on חֹק. A construct chain is never severed by that accent.

What follows. The two traditions divide the verse differently, and this is one of the purest examples in the Bible of what an accent decides. The Greek did not yet know the accent system.

Note: well documented. This seam is also the block boundary of the word grid in Appendix A.

B3. Psalm 2:9 — the root, not the form

The matter. What remains of the stem are the letters ר and ע, with the suffix "them." Two stems are written this way: רעע (to break) and רעה (to shepherd). The distinction lies solely in the vocalization, and that was written down centuries later.

The qualification that matters. Revelation follows three times the root of shepherding, but it does not take over the Greek form: the Greek psalm has ποιμανεῖς (second person), because there the address is to the King; Revelation speaks about Him. The seam therefore runs through the root, not through the form. Whoever writes that Revelation "cites the Septuagint" is not saying it quite precisely enough.

Edition status. At Revelation 2:27 all eight consulted editions agree; the only variant lies in the second verb (συντρίβεται / συντριβήσεται). Revelation 12:5 and 19:15 have been checked against Nestle 1904.

And Revelation does not choose between the two meanings. In 2:27 they stand side by side: καὶ ποιμανεῖ αὐτοὺς ἐν ῥάβδῳ σιδηρᾷ ὡς τὰ σκεύη τὰ κεραμικὰ συντρίβεται — "he will shepherd them with an iron rod, as potters' vessels are broken to pieces."

Note: well documented.

B4. Psalm 2:12 — בַּר, and what the ancient translators made of it

The Hebrew text. נַשְּׁקוּ־בַר — "kiss בַּר." The word in the meaning "son" stands four times in the whole Hebrew Bible: here, and three times in Proverbs 31:2 (exhaustively counted, lemma 1248). The Aramaic parts of Ezra and Daniel use the same word frequently; that is Aramaic and does not count here.

The Greek text reads something else. δράξασθε παιδείας — "take hold of instruction." No son, no kiss. The translator reads the close of the psalm consistently as a school: he uses the same stem as in verse 10, παιδεύθητε.

The proposed textual emendations. There is a long emendation tradition. The most frequently named proposals:

What the text itself gives, and what this book does. The rare word is not unknown, and its other occurrence is not arbitrary: Proverbs 31:2 is the only place in the Bible where a king is addressed with this word — by someone who instructs him, with the same verb (יסר) with which Psalm 2:10 addresses the kings of the earth. The book follows the Hebrew text and adds that the matter is not settled.

Note: disputed. — A complete inventory of all proposed emendations with their literature has not been made; the above four are the most frequently named. That gap stands in Appendix D as an open item.

B5. Psalm 8:5-6 (Eng. 8:4-5) — אֱלֹהִים and the angels

This is no intervention but a fixed translation habit. אֱלֹהִים means in Hebrew also the heavenly beings around God: "bow down before Him, all אֱלֹהִים" (Psalm 97:7) · "the sons of אֱלֹהִים came to present themselves before YHWH" (Job 1:6; 2:1) · "when all the sons of אֱלֹהִים shouted" (Job 38:7); compare Psalm 29:1 and 89:7 (Eng. 89:6) with בְּנֵי אֵלִים. In all those places the Greek text gives ἄγγελοι or ἄγγελοι τοῦ θεοῦ.

The second issue: βραχύ τι. The expression can be read gradually ("a little lower") or temporally ("for a short time"). Hebrews 2:9 allows both, and the choice determines whether the verse is about the rank or about the duration of the humiliation.

What this book does. It keeps both open with "a little lower," and says in the note that the text does not decide. Note: the translation habit well documented; βραχύ τι disputed.

B6. Psalm 8:7 (Eng. 8:6) and Psalm 110:1 — two prepositions, two psalms

The matter. Two citations that stand close together and both concern feet, with two different prepositions. The writer follows his source per psalm and does not make them alike.

Edition status. For Hebrews 1:13, eight editions have been checked and agree. For Hebrews 2:8, the reading ὑποκάτω is confirmed in Nestle 1904; the full edition comparison at this place is still open.

By comparison: where Paul cites the same verse of Psalm 8 (Ephesians 1:22), he paraphrases and chooses a different preposition. Hebrews cites; Paul summarizes.

Note: well documented for 1:13; for 2:8 partial — see Appendix D.

B7. Psalm 102 — the answer and the invocation

#### B7a. Verse 24 (Eng. 23) — one row of consonants, two verbs

The three consonants ע־נ־ה carry both verbs; the distinction lies in the vowels. In the Greek a second speaker thereby arises, and what follows is no longer the lament of the dying man but the answer he receives. Hebrews 1:10-12 cites precisely that answer.

This same verse carries moreover a ketiv/qere: written is כֹּחוֹ ("his strength"), read is כֹּחִי ("my strength"). The tradition was already wavering here before it was translated. It is one of the 1,268 such notes in the Leningrad Codex (counted mechanically; the exact number differs slightly per edition and method of counting).

#### B7b. Verse 26 (Eng. 25) — the only addition in the whole series

The invocation Κύριε has no basis whatsoever in the Hebrew consonants. This is not a choice between two readings but an addition, and the book names it as such. It is moreover load-bearing: without that vocative the verse is a statement about God; with it, it is an address to Someone — and Hebrews 1:10 sets it directly after the ὁ θεός address from Psalm 45.

Note: well documented. This is the weakest place in the whole of chapter 2 and is presented there as such.

B8. Psalm 104:4 — the direction of reading

עֹשֶׂה מַלְאָכָיו רוּחוֹת מְשָׁרְתָיו אֵשׁ לֹהֵט

The verb עשה with a double object allows both orders; the Hebrew does not mark which of the two nouns is the result.

- "who makes his messengers winds" — the direction of the Greek text and of Hebrews 1:7 - "who makes his messengers of the winds" — grammatically just as correct

What speaks for the first: רוּחַ means in the preceding verse unmistakably "wind." What it does not decide: the sentence construction itself. Only the first direction carries the μέν-δέ contrast of Hebrews 1:7-8 — angels as something that God makes from moment to moment, over against a Son with a throne forever.

Note: no manuscript issue but a translation choice; weighed.

B9. Psalm 45:5 (Eng. 45:4) — עַנְוָה

עַל־דְּבַר־אֱמֶת וְעַנְוָה־צֶדֶק. The form is unusual and the colon runs awkwardly; "meekness" beside sword and arrows is a striking attribute. The word stands twice in the canon (counted mechanically).

Note: disputed. What the ancient translations and the revisers do here has not yet been traced — see Appendix D.

B10. Psalm 45:7-8 (Eng. 45:6-7) — "Your throne, O God"

This is no MT/LXX seam. The Greek renders the Hebrew and leaves the ambiguity untouched; what is disputed here is grammar, not manuscript.

The Greek ὁ θεός is in form a nominative and can function as a vocative; the Septuagint thus reproduces the tension instead of resolving it.

There is a genuine variant in the citation itself, but it lies in the New Testament:

In addition, a word-order matter with ῥάβδος εὐθύτητος (LXX and TR) over against ἡ ῥάβδος τῆς εὐθύτητος (critical editions).

Note: the grammatical issue disputed; the New Testament variant well documented. The book follows the vocative, because Hebrews 1:8 does, and adds the other readings.

B11. Psalm 110:1 — the vocalization לַאדֹנִי / אֲדֹנָי

On this vocalization distinction rests the question that Jesus himself poses in Matthew 22:41-46. The Greek text cannot render it: there stands κύριος twice.

Note: the distinction in the Masoretic tradition well documented; how old it is and what the other ancient translations do with it — still to be investigated, see Appendix D.

B12. Psalm 110:3 — where the Greek stands alone

The Hebrew and the Greek text diverge here so far that they are scarcely still the same verse.

And this is the place where the balance of this book tips. Theodotion, Aquila, Symmachus and the so-called Quinta all four follow the Hebrew reading; the Septuagint is here the exception. Moreover no citation in the New Testament quotes this verse.

What follows. What is said in chapter 2 about Deuteronomy 32 and Psalm 102 — that the Greek can preserve an older Hebrew reading — holds not as a rule but per place. Here the Greek stands alone, and that is said.

Note: well documented.

II. THE OTHER CITATIONS OF HEBREWS 1

B13. Deuteronomy 32:43 — the textbook case

The full Greek text: εὐφράνθητε, οὐρανοί, ἅμα αὐτῷ, καὶ προσκυνησάτωσαν αὐτῷ πάντες ἄγγελοι θεοῦ· εὐφράνθητε, ἔθνη μετὰ τοῦ λαοῦ αὐτοῦ, καὶ ἐνισχυσάτωσαν αὐτῷ πάντες υἱοὶ θεοῦ· ὅτι τὸ αἷμα τῶν υἱῶν αὐτοῦ ἐκδικᾶται, καὶ ἐκδικήσει καὶ ἀνταποδώσει δίκην τοῖς ἐχθροῖς καὶ τοῖς μισοῦσιν ἀνταποδώσει, καὶ ἐκκαθαριεῖ κύριος τὴν γῆν τοῦ λαοῦ αὐτοῦ.

The four cola of the Hebrew text are the third, the fifth, the sixth and the eighth. The four that the Hebrew lacks are: the call to the heavens, the call to the angels, the call to the sons of God, and "and to those who hate Him He will repay." Three of those four are about heavenly beings.

The scroll from cave 4. 4QDeutq opens the verse with הרנינו שמים עמו ("shout, heavens") where the Masoretic text has גוים ("nations"); it reads בניו ("his sons") where the MT has עבדיו ("his servants"), and it also has the colon about those who hate. On both points it agrees with the Greek. The reading rests on the published edition as it is represented in the literature; the plate itself has not been consulted — see Appendix D.

The branching within the Greek. In the one branch "all angels of God" stands in the second line and "all sons of God" in the fourth; in the other, reversed. Hebrews 1:6 corresponds literally to the first. In the current critical edition "sons" stands at that place, on account of which one reads in the literature that the wording of Hebrews 1:6 occurs in no manuscript; that is put too strongly.

On the New Testament side there is no variant. Hebrews 1:6 reads πάντες ἄγγελοι θεοῦ in all eight consulted editions. The branching thus lies entirely on the Septuagint side.

Three explanations for the difference in the Hebrew — verdict open:

1. Two Hebrew forms side by side. The one opened with "heavens," the other with "nations"; the Greek took in both. Requires no loss and explains 8 versus 6 versus 4. 2. A parablepsis. Both opening lines begin with הַרְנִינוּ and end with עַמּוֹ; whoever jumps from the one עַמּוֹ to the other loses exactly what stands between. 3. Deliberate omission, because "all the divine ones" sounded awkward. The weakest: it presupposes a motive that is written down nowhere.

(1) and (2) do not exclude each other. *Note: that a longer Hebrew text existed — well documented; which explanation is correct — weighed, verdict open.*

Aside, and not without importance: this song also circulated separately from Deuteronomy, as an ode at the back of the Greek psalters, to be sung. The writer thus possibly knew this line as a song and not as a verse.

B14. 2 Samuel 7:14 — the truncated clause

Hebrews 1:5 cites: "I will be a Father to Him, and He will be a Son to Me."

In 2 Samuel the sentence does not stop there. There follows: אֲשֶׁר בְּהַעֲוֺתוֹ וְהֹכַחְתִּיו בְּשֵׁבֶט אֲנָשִׁים וּבְנִגְעֵי בְּנֵי אָדָם — "whom, when he commits iniquity, I will chasten with the rod of men and with the stripes of the sons of men."

That the letter breaks off there stands firm. The motive does not stand firm. The reading that the omission is deliberate and serves the argument — this Son is not like the kings who followed David — is materially strong, but it has not yet been traced in the commentary tradition (Attridge, Koester, Ellingworth).

Note: the textual datum well documented; the motive weighed — see Appendix D.

B15. Acts 13:33 — "in the first psalm"

A trace of the count in which Psalm 1 and Psalm 2 counted as one psalm. For a book that works with Psalm 2 and uses the Psalm 1 hook, that closes the circle.

Note: the variant well documented; whether Psalm 1 and 2 ever really formed one psalm — disputed.

III. THE NEW TESTAMENT

B16. Genesis 37:20 over against Mark 12:7

Three words, identical. In Mark it stands moreover the same in all eight editions; the only variation lies in the word order of "they said to one another" around it. The connection between Joseph and the tenants is thus lexical, not thematic.

And it runs over the inheritance: the tenants add why — then the inheritance is ours (ἡ κληρονομία) — and that is the word with which Hebrews 1:2 calls the Son heir of all things and with which the Greek Psalm 2:8 promises the nations to Him.

Note: well documented.

B17. ἀμνός and ἀρνίον — the sacrifice word and the diminutive

He who sits on the throne and takes the scroll carries, in the last book of the Bible, no sacrifice word but a diminutive. In Revelation 13:11 it stands moreover in mirror image: the beast from the earth has "two horns, like a little lamb."

Note: well documented (concordance data checked).

B18. The Revelation passages of Psalm 2 — the Greek forms

Two observations from the same check.

Revelation 19 knows two assemblies with the same verb. In verse 17 an angel summons the birds: δεῦτε συνάχθητε εἰς τὸ δεῖπνον τὸ μέγα τοῦ Θεοῦ. In verse 19 the kings of the earth stand συνηγμένα. Two gatherings in two verses, one verb.

And the two fallings of John are lexically distinguishable. Revelation 1:17: ἔπεσα πρὸς τοὺς πόδας αὐτοῦ ὡς νεκρός — and a right hand is laid upon him. Revelation 19:10: ἔπεσα ἔμπροσθεν τῶν ποδῶν αὐτοῦ προσκυνῆσαι αὐτῷ — and he is held back: ὅρα μή. The same movement, toward the same feet; the one time raised up, the other corrected.

Note: the forms well documented (Nestle 1904; 2:27 against eight editions); the attribution of 11:18 to Psalm 2 weighed; the coherence of the two assemblies weighed — the forms stand firm, that the one is the meal of the other neither of the two verses says in so many words.

B19. 1 Corinthians 10:4 — the following rock

*ἔπινον γὰρ ἐκ πνευματικῆς ἀκολουθούσης πέτρας, ἡ πέτρα δὲ ἦν ὁ Χριστός.*

The participle stands there; what it means — a rock travelling along, a spring travelling along, or the Christ who moved along — remains undecided.

And there is a Hebrew datum that belongs with it and corrects the common picture: Exodus 17 has צוּר, Numbers 20 has סֶלַע — in all three key verses (20:8 twice, 20:10 twice, 20:11). Whoever, on the ground of the Hebrew text, makes one rock of both chapters does something the text does not do. Paul connects them in the Greek; the Hebrew does not.

Note: disputed.

B20. 2 Corinthians 3:7, 13 — the fading radiance

καταργουμένην. The rendering "fading" — the radiance on Moses' face that was disappearing, over against the Son as radiation (ἀπαύγασμα, Hebrews 1:3) — is not uncontested. This is a translation and interpretation matter, not a text matter.

Note: disputed; belongs with the commentary debt of Appendix D.

IV. SUMMARY TABLE

What stands in this appendix as "to be verified" or "not yet traced" is gathered in Appendix D.