Sitting at the right hand — how Hebrews quotes Psalm 110
A Hebrew psalm and a Greek letter cannot be compared word by word. Put the Septuagint between them and the quotation becomes visible at the level of the original.
Psalm 110:1 is quoted more often in the New Testament than almost any other single verse — Matthew 22:44, Mark 12:36, Luke 20:42, Acts 2:34, Hebrews 1:13. Every reader knows it. What is less obvious is that a computer comparing the Hebrew psalm with the Greek letter, word by word, would find nothing at all.
The problem
Here is the Hebrew:
נְאֻם יְהוָה לַאדֹנִי שֵׁב לִימִינִי עַד־אָשִׁית אֹיְבֶיךָ הֲדֹם לְרַגְלֶיךָ
And here is Hebrews 1:13:
κάθου ἐκ δεξιῶν μου, ἕως ἂν θῶ τοὺς ἐχθρούς σου ὑποπόδιον τῶν ποδῶν σου
There is not one shared token. Not one shared lemma, not one shared Strong’s number, not one shared character. Every method that works on the original languages — and those are the methods worth trusting most, because they do not run through anyone’s translation — is completely blind here.
You can get the connection from a translation, of course. Sit at my right hand and sit thou at my right hand share plenty. But then the link rests on the translators’ choices, and two translators who rendered the same Hebrew differently would break it.
The bridge
The New Testament almost always quotes the Old in the Greek of the Septuagint. Which means the comparison that cannot be made — Hebrew against Greek — can be replaced by one that can: the Greek Old Testament against the Greek New Testament. Same language, same word forms, same grammar.
Psalm 110:1 in the Septuagint is Psalm 109:1:
κάθου ἐκ δεξιῶν μου ἕως ἂν θῶ τοὺς ἐχθρούς σου ὑποπόδιον τῶν ποδῶν σου
Now the two are almost the same sentence. The evidence Moria shows for this link is the chain the two verses share:
Κάθου … δεξιῶν … ἕως … ἐχθρούς … ὑποπόδιον … ποδῶν
Six content words in the same order. That ordering is what distinguishes a quotation from a coincidence: two texts about kingship will share vocabulary, but they will not share a sequence.
What it cost to build
The bridge is not free, because the Septuagint numbers its verses differently. Psalm 110 is Psalm 109; the psalms shift by one across a long stretch, because Septuagint Psalm 9 covers what the Hebrew counts as 9 and 10. Jeremiah is reordered outright.
The alignment could have been hand-tabulated. Instead it is derived from the text, on a simple observation: proper names survive translation. Ἰσραήλ and Yisraél are recognisably the same word once both are reduced to a consonantal skeleton. Reduce every name in both texts that way, score each candidate verse pairing on name agreement and verse length, and run a monotone alignment through each book. Merges and splits fall out of the monotonicity on their own — no special case is needed for Psalm 9.
Jeremiah, the one genuinely rearranged book, got both treatments: an automatically computed chapter mapping and the classical table from textual criticism, scored against each other, better one kept.
The result was checked against 34 known correspondences — Psalm 110:1 as Septuagint 109:1, Jeremiah 31:31 as 38:31, and so on. It is right in 33 of them. That one remaining case is stated rather than buried, because a claimed accuracy without a stated failure count is not a claim at all.
Where the bridge is silent
It finds quotations that follow the Septuagint’s wording. Where the New Testament follows a different Greek text, or where the writer translated the Hebrew himself, there is nothing for it to match. Matthew’s citation formula quotations are notoriously mixed in this respect.
So silence from the bridge is not evidence that a verse is not being quoted. It means one particular route did not carry it — and the quotation method that works through the translation, which is blind in different places, may well pick it up instead.
That is the whole architecture in miniature: five methods, each blind somewhere, arranged so that their blind spots do not coincide.