Paul has just twice pronounced an ἀνάθεμα on anyone who preaches a different gospel (vv. 8-9). Verse 10 is the inference he draws from having done so, introduced by γάρ: a man who curses angels and apostles alike is plainly not canvassing for human approval. Two rhetorical questions, then a contrary-to-fact condition that seals the point.

The verse turns on three things the English must preserve.

The fronting of ἀνθρώπους. Both questions put the human object first, and the second question puts the dative first as well. Paul is not asking what he does but whom he does it for, and the whole of 1:11-12 (not κατὰ ἄνθρωπον, not παρὰ ἀνθρώπου) grows out of that word. The letter runs ἄνθρωπος five times in four verses.

The sharpness of ἄρτι. Not the colorless ‘now’ but ‘at this very moment’ — the moment of the curse just uttered. It answers the ἄρτι of v. 9 (‘and now again I say’).

The force of ἔτι. ‘If I were still pleasing men’ presupposes a time when he was, and leaves open whether that time was his pre-Christian zeal (1:13-14), a charge his opponents are making, or both. The adverb is the one word in the verse that cannot be resolved without deciding what the Galatian agitators had said about him, and it is therefore left open.

The closing Χριστοῦ δοῦλος is not a modest self-description. It is the designation Paul opened the letter with in other epistles (Rom 1:1; Phil 1:1) and it names owned, non-free legal status: a slave has one master, and the question of whose approval he serves is settled by that fact alone.

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

γάρ is inferential-explanatory and must be rendered (§10). It ties v. 10 to the anathemas: because I have just said that, the answer to the following question is obvious. Dropping it — as many English versions do, beginning the verse with a bare ‘Am I now trying…’ — severs the one connective that tells the reader the questions are not a fresh topic but a conclusion.

Both questions are introduced by ἤ rather than by μή or οὐ. This matters, because the corpus has had to distinguish οὐ-questions (expecting ‘yes’) from μή-questions (expecting ‘no’). Here neither particle is present: the first ἤ is a genuine disjunctive within the question (men or God?), and the second ἤ heads a fresh question pressing the same point. On the standing policy for ἢ δοκεῖτε ὅτι (Jas 4:5), disjunctive ἤ in this position does not offer an alternative so much as it presses a question already put, and it must always be rendered.

The expected answer is supplied not by a particle but by the alternative itself: ἢ τὸν θεόν makes the first option unthinkable, and the unreal condition in the third clause converts the rhetorical expectation into an argument.

ἄρτι is the punctiliar temporal adverb, sharper than νῦν: ‘just now, at this instant’ (Matt 26:53; John 13:37; 1 Cor 13:12). The corpus policy fixed at Gal 4:20 requires that it be kept in its Greek word-order slot so its attachment stays as open in English as in Greek, and that it not be absorbed into a colorless ‘now’.

Here the slot is clause-initial and the attachment is to the whole question. Decisively, the same adverb stands eleven words earlier, at the head of v. 9: καὶ ἄρτι πάλιν λέγω. The two occurrences should receive the same English word so the reader can see that the moment of the curse and the moment of the question are the same moment. The force is: at this very instant, with that anathema still in the air.

Some read ἄρτι as a concessive marker against a past — ‘am I now (as opposed to formerly) currying favor?’ — which would make it a near-synonym of ἔτι in the third clause. That is possible but makes the two adverbs redundant; the instantaneous reading preserves the distinction.

The present active πείθω is the ordinary transitive verb: ‘persuade, win over, prevail upon’ (Acts 18:4; 19:26; 2 Cor 5:11). It must be kept lexically distinct from the second perfect πέποιθα (‘I am confident’), from the middle/passive πείθομαι (‘be persuaded by / obey’, fixed at Gal 5:7), and from πεισμονή (‘persuading’, Gal 5:8) — Galatians uses all four forms of this root, and the English should let the reader see that one root is in play.

The difficulty is the object pairing. One does not normally persuade God, and the two accusatives (ἀνθρώπους … τὸν θεόν) are governed by one verb.

The rendering ‘trying to win over’ in the reading text is hospitable to the first and third readings and does not exclude the second. ‘Seek the approval of’ is excluded, because it imports the ἀρέσκω of the next clause into a verb Paul kept separate.

Note the article asymmetry: anarthrous ἀνθρώπους (a class, qualitatively) against articular τὸν θεόν (the one definite party). English supplies ‘God’ without an article either way, so the asymmetry can only be recorded.

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

ζητέω is already fixed in the glossary as ‘seek / pursue’: BDAG 3, sustained and deliberate effort toward an objective, never casual wanting. The complementary infinitive construction (ζητέω + inf.) is standard (Luke 5:18; 1 Cor 14:12; Gal 2:17), and the verb’s presence is argumentatively load-bearing — Paul does not deny pleasing men as an incidental effect but denies it as an aim pursued.

ἀρέσκω with the dative is ‘be pleasing to, give satisfaction to’. The lexeme is not intrinsically culpable: Paul uses it positively of pleasing God (1 Thess 2:4; 4:1), of Christ not pleasing himself (Rom 15:3), and of accommodating a neighbor (Rom 15:2; 1 Cor 10:33). The fault here is created entirely by the object, exactly as with ζῆλος and ἐπιθυμία elsewhere in the corpus. So no pejorative is built into the gloss: ‘curry favor with’, ‘flatter’, and ‘ingratiate myself with’ are excluded, because each supplies a verdict that ἀνθρώποις alone supplies in Greek.

The background register is worth naming. In Hellenistic inscriptions and in the Greek moralists, ἀρέσκειν τοῖς ἀνθρώποις or ἀρέσκειν τοῖς πολλοῖς is the standard language for an officeholder or orator who accommodates the crowd — the stock accusation against a demagogue or flatterer (κόλαξ). Paul is repudiating a recognized social type, which is why the charge stings and why the answer is a status-claim (δοῦλος) rather than a protest of sincerity.

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

ἄνθρωπος occurs three times in this verse and five times in vv. 10-12 (ἀνθρώπους, ἀνθρώποις, ἀνθρώποις, κατὰ ἄνθρωπον, παρὰ ἀνθρώπου), after twice in v. 1 (οὐκ ἀπʼ ἀνθρώπων οὐδὲ διʼ ἀνθρώπου). It is the keyword of the chapter’s argument and one English word must serve every occurrence, or the thread disappears.

Per §3 and the standing generic-ἄνθρωπος entries: the masculine plural is the Greek default for humanity irrespective of sex and the referent here is certainly mixed. Reading and modern tiers render ‘people’; the literal tier retains ‘men’ to track the grammatical form, and the generic force is noted rather than resolved silently in either direction.

Against ἄνθρωπος stands θεός in the first question and Χριστός in the third. The verse is built on a single antithesis — human party versus divine party — and renderings that vary the English noun (‘men’ … ‘people’ … ‘human approval’) break the only structure the verse has.

εἰ ἔτι ἀνθρώποις ἤρεσκον, Χριστοῦ δοῦλος οὐκ ἂν ἤμην is a second-class (contrary-to-fact) condition: εἰ + imperfect indicative in the protasis, imperfect indicative + ἄν in the apodosis (BDF §360). The form itself asserts that the protasis does not hold. English requires ‘were … would not be’.

The imperfects are durative, not punctiliar: a sustained practice of pleasing, and a sustained standing as slave. The reference is to present time, as is normal for the imperfect in unreal conditions, though the adverb ἔτι pulls a past into view as well (see below).

ἤμην is the Koine middle form of the imperfect of εἰμί, replacing classical ἦν. The form carries no semantic weight whatever; it is recorded in the parsing table and not rendered differently. The same is true of ἄν, which some witnesses omit — the omission would leave a bare imperfect and convert the unreal statement into a report about the past, which is not the force of the construction and must not be followed.

The logic runs backward from a fact Paul takes as settled: I am Christ’s slave; a slave of Christ cannot be a pleaser of men; therefore I am not pleasing men. The conclusion is the answer to the two questions, and it rests on the incompatibility of two allegiances rather than on an appeal to his own sincerity.

ἔτι always presupposes a prior state and asks whether it persists; the corpus policy fixed at Gal 5:11 requires that it be rendered at every occurrence and not relocated. Here it is the single most consequential word in the verse, because it concedes that there was a time when Paul did please men.

The rendering ‘still’ is retained in every tier precisely because it is as open as the Greek. ‘If I were trying to please people as I once did’ and ‘if I were still trying to win human approval, as they allege’ each adjudicate and are excluded from the reading text.

Proposed — pending reviewRichard N. Longenecker, Galatians, Word Biblical Commentary 41 (Dallas: Word, 1990)

δοῦλος is fixed in the glossary as ‘slave’: owned, non-free legal status in first-century usage. ‘Servant’ is excluded, because it softens an ancient social institution the argument depends on (§5, §9). The force of the apodosis is not that Paul is a devoted follower but that he is property, and property has one owner whose assessment is the only one that counts.

The genitive Χριστοῦ is fronted before the noun, which is marked against the normal order (cf. δοῦλος Χριστοῦ Ἰησοῦ, Rom 1:1; Phil 1:1). Per the standing fronted-genitive policy the literal tier keeps the sequence and the emphasis is recorded rather than compensated for with added English words. Both nouns are anarthrous and qualitative — ‘a slave of Christ’, naming a status, not identifying a particular slave.

Two registers stand behind the designation and both are live. The LXX uses δοῦλος (and παῖς) for the ‘servant of YHWH’ — Moses, David, the prophets (Josh 14:7; 2 Kgdms 7:5; Ps 88:4 LXX; Amos 3:7) — a title of honor and commission; the Hellenistic world used it for a chattel slave with no standing of his own. Paul exploits both. The honorific echo makes the phrase a credential, and the social fact makes it an argument.

Note also the glossary entry for δουλεύω and the whole δουλ- family, which Galatians runs hard (2:4; 4:1-9, 24-25; 5:1, 13). Paul’s self-designation here as Χριστοῦ δοῦλος sits at the head of a letter whose central contention is that his readers must not become slaves again — and the resolution comes at 5:13, where freedom’s content is to be slaves to one another.

Proposed — pending reviewDale B. Martin, Slavery as Salvation: The Metaphor of Slavery in Pauline Christianity (New Haven: Yale University Press, 1990)

Chrysostom takes the verse as Paul’s answer to a specific slander: the Galatians had been told that Paul preached circumcision elsewhere and withheld it from them out of a desire to please. On his reading, ἄρτι points to the anathema just pronounced as self-evident refutation — a flatterer does not curse. He takes ἔτι of the pre-Christian past: formerly Paul did court men, and the change is the measure of the gospel’s work in him.

Jerome and the later Latin tradition read πείθω in the conciliatory direction (suadeo, taken as winning over), which pulls the two questions into near-synonymy. Theodoret likewise glosses the first question by the second.

These are reception, not lexical evidence (§11). But they are worth recording on two counts: they show that the conciliatory sense of πείθω was felt as natural by readers whose Greek was native, and they show that the ‘answering a charge’ construal of ἔτι is not a modern invention.

Proposed — pending reviewJohn Chrysostom, Commentary on the Epistle to the Galatians, in Nicene and Post-Nicene Fathers, First Series, vol. 13 (ed. Philip Schaff; Buffalo: Christian Literature Company, 1889)

Vulgate: Modo enim hominibus suadeo an Deo? aut quaero hominibus placere? si adhuc hominibus placerem, Christi servus non essem.

Three points. Suadeo renders πείθω with its ordinary persuasive force and does not resolve the awkwardness of the divine object — a better guide than most English versions, which paraphrase. Adhuc keeps ἔτι in full and does not specify its referent. But servus for δοῦλος is the start of a long softening: Latin servus is in fact a slave-word, yet its English descendant ‘servant’ is not, and the English tradition has inherited the weakened sense. Per the glossary, ‘slave’ stands.

The Peshitta renders the first question with a verb of persuading men and the second with pleasing, maintaining the distinction between the two verbs rather than levelling them.

No version is permitted to settle the sense of πείθω or the referent of ἔτι; they are recorded as witnesses to how the Greek was heard, not as evidence of what it means.

The verse is textually stable in the critical editions and NA28 prints it without a variation unit of consequence; there is nothing here that bears on the sense.

Two matters are worth recording for the reader even so, since both would change the grammar if followed.

First, the apodosis depends entirely on ἄν. A bare οὐκ ἤμην without the particle would read as a statement about the past (‘I was not Christ’s slave’) rather than as an unreal consequence, which would invert the argument. The particle is secure.

Second, ἤμην is the Koine middle-form imperfect of εἰμί for classical ἦν. Where such forms appear in the manuscript tradition, Atticizing normalization is always the direction of scribal activity, so the harder Koine form is to be preferred on principle. The form is morphological only and carries no semantic weight.