This is the verse where the long sentence begun in v. 6 finally lands. Paul has been building toward it since v. 1: he went up to Jerusalem, laid his gospel before the leadership, Titus was not compelled to be circumcised, the false brothers got nowhere, and the men of standing added nothing to me. Verse 9 states what they did instead — they gave a handshake.

Four things carry the weight.

1. The ground of the recognition is grace, not argument. What they came to know was not that Paul’s reasoning held up but τὴν χάριν τὴν δοθεῖσάν μοι — a passive with the agent unnamed. The thing they acknowledged was something God had already done.
2. The three are named, then qualified. Ἰάκωβος καὶ Κηφᾶς καὶ Ἰωάννης, and then οἱ δοκοῦντες στῦλοι εἶναι — the same δοκέω formula that has run through vv. 2 and 6. Whether the phrase is neutral, dry, or sharp is genuinely contested.
3. The act is a gesture, not a document. δεξιὰς … κοινωνίας: right hands given. A handshake, with the genitive naming what the handshake constituted.
4. The agreement is stated verblessly. ἵνα ἡμεῖς εἰς τὰ ἔθνη, αὐτοὶ δὲ εἰς τὴν περιτομήν — no verb at all. Paul reports the terms in what looks like their own compressed wording, and refuses to specify whether the division is geographical, ethnic, or missional.

The syntax is also the argument: the main verb of a sentence four verses long is ἔδωκαν, and its object is a gesture. Nothing was imposed; something was extended.

Verses 6-10 are one sentence, and a broken one. Paul begins ἀπὸ δὲ τῶν δοκούντων εἶναί τι (v. 6) — a prepositional phrase that never receives the verb it was heading toward. He interrupts himself twice (the two dashes of v. 6), restarts with ἐμοὶ γὰρ οἱ δοκοῦντες οὐδὲν προσανέθεντο, then builds a participial chain: ἰδόντες (v. 7), a γάρ parenthesis (v. 8), γνόντες (v. 9), and at last the main verb ἔδωκαν.

Three consequences for the English.

• The καί at the head of v. 9 is not a sentence-opener. It coordinates γνόντες with ἰδόντες, and the modern verse division must not be allowed to break the chain. No English full stop at the start of this verse.
• The two participles are both aorist and both antecedent to ἔδωκαν: seeing and recognizing precede the handshake. Rendering either as a finite verb (“they saw … and they recognized … so they gave”) tidies away the suspension that is doing the rhetorical work.
• The nominative subjects (Ἰάκωβος, Κηφᾶς, Ἰωάννης) are stated only here, in apposition to the participles, after two verses of unnamed οἱ δοκοῦντες. Paul withholds the names and then supplies them at the moment of the concession.

Built onna28 · bdf

γινώσκω is already fixed in this corpus as “come to know” in its relational-covenantal range, kept distinct from οἶδα (settled state) and ἐπίγνωσις (full knowledge). Here the aorist is ingressive: the three arrived at a recognition they had not previously held.

The pairing with ἰδόντες in v. 7 matters. Paul uses a seeing-verb for the fact (that he had been entrusted with the gospel of the uncircumcision) and a knowing-verb for the grace. The second is not a synonym for the first; it marks the move from observing a state of affairs to acknowledging its source. Rendering both “when they saw” levels a distinction Paul built with two verbs.

“Perceived” (so the older English, following Vulgate cognovissent) is acceptable in liturgical register but drifts toward sense-perception, which belongs to ἰδόντες. “Acknowledged” over-reaches toward a formal act.

Built onbdag

χάρις is fixed as “grace” throughout — the only English word carrying both unearned favor and, by metonymy, the concrete gift. Here the referent is contested and must stay so: the apostolic commission itself (so Rom 1:5, χάριν καὶ ἀποστολήν; Gal 1:15, διὰ τῆς χάριτος αὐτοῦ), the observable fruit of that commission, or God’s favor toward Paul generally. “The commission given me” and “my special privilege” each adjudicate and are excluded.

The doubled article (τὴν χάριν τὴν δοθεῖσαν) is the fuller attributive construction, adding the participle as a separate, mildly emphatic unit rather than embedding it. The literal tier keeps both articles visible; reading tiers may resolve to a relative clause, with the compression noted.

δοθεῖσαν is a divine passive. The agent is unexpressed and must stay unexpressed: “the grace God had given me” supplies the subject Paul withheld, and withholding it is the point — what the Jerusalem leaders recognized was not Paul’s achievement and not their own conferral. They registered something already done by a third party not named in the clause.

Note the lexical play with the main verb. δοθεῖσαν (passive, Paul the recipient) is answered by ἔδωκαν (active, the three the subjects). They gave a handshake because they recognized a giving that was not theirs. One English root must serve both occurrences so the figure survives.

Built onbdag · na28

Ἰάκωβος is rendered “James” in reading tiers per the standing entry, with “Jacob” retained in the literal tier where a patriarchal echo is live; here no such echo operates and “James” stands. The referent is the Lord’s brother of Gal 1:19, not the son of Zebedee, who is dead by Acts 12:2.

Κηφᾶς is transliterated as written. Paul’s distribution across this chapter is data: Πέτρος at vv. 7-8, Κηφᾶς at vv. 9, 11, 14, and the switch is part of the evidence that vv. 7-8 cite a pre-existing formulation. Substituting “Peter” here would erase it. Witnesses reading Πέτρος at v. 9 are normalizing toward the familiar Greek name and are secondary.

The order is itself a datum. James stands first — not Peter, as in every Synoptic list of the three (Mark 5:37; 9:2; 14:33). Whether this reflects James’s standing in Jerusalem at the time of the conference, at the time of writing, or Paul’s own assessment is not recoverable from word order alone, and no conclusion should be built on it beyond the observation that the sequence is unexpected.

Built onbdag · metzger-tc

δοκέω with an infinitive is already fixed for the reflexive εἴ τις δοκεῖ formula (“think/suppose of oneself”). This is a different construction: the articular participle with an infinitive, where the subject of the supposing is unexpressed and the supposers are other people. The sense is “those reputed to be,” “those held to be.”

The formula runs through the chapter — τοῖς δοκοῦσιν (v. 2), τῶν δοκούντων εἶναί τι and οἱ δοκοῦντες (v. 6), οἱ δοκοῦντες στῦλοι εἶναι (v. 9) — and one English rendering must serve all four, or a deliberate refrain becomes invisible.

The tone is the crux. Greek δοκέω can report a reputation neutrally (“the recognized leaders”) or hold it at arm’s length (“the so-called pillars”). Both are ordinary first-century usage, and the choice determines whether Gal 2:9 is a concession or a needle.

Built onbdag · bdf

Proposed — pending reviewF. F. Bruce, The Epistle to the Galatians: A Commentary on the Greek Text, New International Greek Testament Commentary (Grand Rapids: Eerdmans, 1982)

στῦλος is the architectural column — the pillars of the tabernacle and temple in LXX usage (Exod 26:32, 37; 3 Kgdms 7:3), the pillar of cloud and fire (Exod 13:21-22), and in the NT the column-image applied to persons and to the church (1 Tim 3:15; Rev 3:12; Rev 10:1 of the angel’s legs).

The metaphor is live and must stay live (§8). “Leaders,” “authorities,” and “the men in charge” each discard the image and substitute an office-word; “the chief men” imports rank the noun does not name. What a pillar does is hold a structure up — and whether the structure in view is the eschatological temple, the Jerusalem community, or simply the church is left open by the noun and must be left open by the rendering.

Note that the word is anarthrous and predicative (στῦλοι εἶναι, not τοὺς στύλους). The claim being reported is qualitative: that these men are of the pillar sort. Supplying the definite article (“the pillars”) makes it a fixed designation and overstates the Greek.

The temple-pillar resonance is worth recording without pressing it. If the Jerusalem community thought of itself as the eschatological sanctuary, calling three men στῦλοι is a claim about the structure they uphold; but Paul reports the designation rather than asserting it, and no doctrine of the church may be built on a word he puts in other people’s mouths.

Built onbdag · muraoka-lxx

δεξιάς is the substantival feminine plural of δεξιός with χεῖρας understood — the regular Greek ellipsis (1 Macc 6:58; 11:50, 62; 13:50, all of right hands given to ratify terms). The literal tier brackets the supplied noun.

The plural is not decorative. Three men gave right hands to two men; Greek keeps the count and English idiom normally does not. Reading tiers may use the English idiom “the right hand of fellowship,” but the plural stands in the literal tier and the compression is recorded — the Vulgate’s dextras preserves it, and the English singular is the levelling.

κοινωνία is fixed in this corpus as “fellowship / participation,” with “partnership” where a joint enterprise with material stake is in view, and with “communion” excluded as importing post-biblical sacramental freight. The genitive here is delayed — separated from δεξιάς by the whole dative phrase ἐμοὶ καὶ Βαρναβᾷ — which is marked word order and throws the clause’s weight onto the final noun. The English should keep κοινωνίας late where syntax permits.

What kind of genitive it is matters and is not settled:

The gesture itself is realia (§9). In the Hellenistic world the extended right hand ratified a pledge, sealed an alliance, and guaranteed safe dealing; it was not a greeting. Record the institution in the note; do not import “agreement” or “treaty” into the rendering.

Built onbdag · muraoka-lxx

Proposed — pending reviewHans Dieter Betz, Galatians: A Commentary on Paul's Letter to the Churches in Galatia, Hermeneia (Philadelphia: Fortress, 1979)

The ἵνα clause has no verb at all. ἵνα ἡμεῖς εἰς τὰ ἔθνη, αὐτοὶ δὲ εἰς τὴν περιτομήν — two subjects, two prepositional phrases, and nothing else.

Three features must survive into English.

The ellipsis. Something must be supplied for English to function, and whatever is supplied is interpretive. Candidates: a verb of going (εὐαγγελιζώμεθα / πορευώμεθα — “that we should go to the nations”), a verb of being (“that we should be for the nations”), or a verb of proclaiming. Bracket the supplement in the literal tier. Do not supply “preach the gospel” unbracketed: Paul had εὐαγγελίζομαι available in v. 2 and withheld it here.

The expressed pronouns. ἡμεῖς and αὐτοί are both redundant to a verb that is not there, and both are therefore doing the whole work of the clause. The intensive αὐτοί is in predicate position — “they themselves,” not the identity-adjective ὁ αὐτός. Both must be rendered visibly; reducing either to a bare subject deletes the contrast that the clause consists of.

The δέ. Mildly adversative, marking the pivot between the two spheres. Always rendered (§10); dropping it leaves two coordinate assignments where Greek sets one over against the other.

The compression itself is evidence. A verbless clause of this shape reads like a formula being quoted rather than a sentence being composed — which fits the σ switch to Πέτρος in vv. 7-8 and suggests Paul is reporting the terms in something close to the words in which they were struck.

Built onbdf · na28

ἔθνη is fixed as “the nations” in the literal tier (LXX equivalent of גּוֹיִם, outsiders to the covenant people rather than an ethnic category), with “Gentiles” permitted in smooth registers and “pagans” and “heathen” excluded (§5, §9).

περιτομή is fixed as “circumcision,” retained even where the reference is to the circumcised by the abstract-for-concrete idiom — the same construction Paul parodies with κατατομή at Phil 3:2. “The Jews” is excluded from the reading text: Paul writes the abstract noun, and the abstract-for-concrete idiom is part of how he handles the category throughout Galatians (2:7, 8, 9, 12; 5:6; 6:15).

The terms of the division are not symmetrical in form — a neuter plural of peoples against a feminine singular abstract — and the asymmetry should not be levelled (“to the Gentiles … to the Jews”).

Built onbdag · muraoka-lxx

Proposed — pending reviewJames D. G. Dunn, The Epistle to the Galatians, Black's New Testament Commentary (London: A. & C. Black, 1993)

The verse is stable. Three points worth recording.

Κηφᾶς / Πέτρος. A minority of witnesses read Πέτρος here, assimilating to vv. 7-8 and to the familiar Greek name. The Aramaic form is the harder reading, it is what the surrounding verses (11, 14) require, and the alternation across vv. 7-9 is itself the datum. NA28 reads Κηφᾶς.

Order of names. Part of the tradition transposes to Πέτρος καὶ Ἰάκωβος καὶ Ἰωάννης, bringing the sequence into line with the Synoptic lists. This is plainly secondary, and its existence shows that ancient readers also found Ἰάκωβος-first unexpected.

καταδουλώσουσιν / ἀναγκάσουσιν family (v. 4) and the verse-9 ἵνα clause. No witness supplies a verb in the ἵνα clause of v. 9, which is strong evidence that the ellipsis is original and was felt as formulaic rather than defective — the same scribal tradition fills gaps readily elsewhere in this letter.

Built onna28 · metzger-tc

Paul’s thesis in Gal 1-2 is that his gospel came neither from men nor through a man (1:1, 11-12). The obvious objection is that he nevertheless went to Jerusalem and dealt with the men who had been apostles before him (1:17). Verses 6-10 answer it by distinguishing two things that English easily confuses: addition and recognition.

The negative side is v. 6: ἐμοὶ … οὐδὲν προσανέθεντο — they added nothing. Nothing was supplied, imposed, corrected, or completed. The positive side is v. 9: they gave right hands. What passed between them was acknowledgment of something already in place, not the conferral of something new.

The grammar carries this. δοθεῖσαν is passive with an unnamed agent; ἔδωκαν is active with the three as subjects. They gave because a giving had already happened that was not theirs to give. The handshake registers a fact; it does not create one.

This is also why the ἵνα clause can be verbless. Had Paul been reporting a commissioning he would have needed a verb of sending. What he reports is a division of spheres already operative — he was already preaching among the nations (v. 2, ὃ κηρύσσω, present tense) before the conference convened.

The verse therefore does two things at once that commentators sometimes try to make it do singly: it grants that the Jerusalem leadership had standing worth having, and it denies that Paul’s gospel derived from it. Both are required by the argument. A rendering that makes οἱ δοκοῦντες wholly dismissive loses the first; a rendering that makes the handshake a conferral loses the second.

Built onbauckham-god-crucified · bdag

Proposed — pending reviewJ. Louis Martyn, Galatians: A New Translation with Introduction and Commentary, Anchor Bible 33A (New York: Doubleday, 1997)

Chrysostom reads οἱ δοκοῦντες as reportive rather than derogatory, taking Paul to be citing the general estimate because his argument needs men of real weight to have endorsed him; the reserve, he holds, lies in v. 6’s declaration of indifference to rank, not in the participle. On the handshake he stresses that the three did not teach Paul anything but recognized him as a colleague — which is precisely the distinction between addition and recognition that the sentence’s structure enforces.

On στῦλοι, the Greek commentators regularly connect the image to the church as building and temple (1 Tim 3:15), and note that pillars bear weight rather than issue commands — a reading that resists turning the metaphor into an office-word, and that supports keeping the architectural image in English.

Jerome, in his commentary on Galatians, is exercised by the implications of the division of spheres for Peter’s and Paul’s later conduct at Antioch, and treats the εἰς-phrases as indicating principal fields rather than exclusive jurisdictions — the third of the three construals set out above. His Latin dextras … societatis preserves both the plural and the bare genitive that most English versions level.

These readings are reception and are reported as such (§11). They are useful as a check on modern drift — particularly on the pull toward reading στῦλοι as an administrative title — but they are not lexical evidence.

Proposed — pending reviewJohn Chrysostom, Commentary on the Epistle to the Galatians, in Nicene and Post-Nicene Fathers, First Series, vol. 13 (Edinburgh: T&T Clark, 1889) · Jerome, Commentary on Galatians, trans. Andrew Cain, Fathers of the Church 121 (Washington, DC: Catholic University of America Press, 2010)