The verse is the hinge of Galatians 5. Paul has just closed the freedom-and-love paragraph (vv. 13-15) with a warning that the Galatians are biting and devouring one another. Λέγω δέ now draws the practical consequence, and it does so with a command plus a promise: one imperative (περιπατεῖτε) and one emphatic future denial (οὐ μὴ τελέσητε).

Three things carry the weight, and each is genuinely open:

1. πνεύματι — a bare dative. Instrument (‘by the Spirit’), sphere (‘in the Spirit’), or rule (‘according to the Spirit’) are all grammatically live, and Galatians uses this bare dative four times in twenty verses (5:5, 16, 18, 25) without settling it.
2. οὐ μὴ τελέσητε — not a second command but the strongest negation Greek can build. Paul is not saying ‘do not gratify’; he is saying ‘you most certainly will not bring it to completion.’ The Latin tradition’s non perficietis got this right; a good deal of English tradition has not.
3. τελέσητε — a verb of reaching a terminus, not of feeling or entertaining a desire. What is denied is the craving’s consummation, not its presence — which is exactly what v. 17 goes on to concede.

The grammar therefore yields a promise, not a threat: the Spirit-walk is not a strategy for defeating the flesh but the condition under which the flesh’s project fails.

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

Paul’s standing discourse-management formula, marking not a new subject but the application of what precedes (Gal 3:17; 4:1; 5:16; 1 Cor 7:8; 10:15). Per the corpus policy, δέ here is transitional and rendered ‘now,’ not adversative ‘but’ — v. 16 does not correct v. 15 but draws the remedy out of it.

The object of λέγω is unexpressed. Greek permits this; English needs a bracketed ‘[this]’ in the literal tier. The following clause supplies the content, so a colon is the right punctuation.

The first person singular is worth noting: from v. 13 Paul has been addressing ὑμεῖς, and the switch to ἐγώ-level speech (cf. the emphatic ἐγὼ Παῦλος of v. 2 and ἐγὼ πέποιθα of v. 10) marks vv. 16-18 as Paul’s own ruling rather than shared premise.

Built onbdf · bdag

The single most consequential word in the verse, and its case-function is unresolved.

Anarthrous but definite. The absence of the article does not make the referent indefinite or the noun abstract. Galatians has established the Spirit’s identity at 3:2-5 and 4:6 (τὸ πνεῦμα τοῦ υἱοῦ αὐτοῦ), and this bare dative recurs at 5:5, 16, 18, and 25 as a refrain. English must supply the article; the literal tier brackets it.

The three readings:

The rendering ‘by’ is adopted because English ‘by’ leans instrumental while remaining hospitable to means and rule — the same solution already fixed for πνεύματι θεοῦ at Phil 3:3 and for πνεύματι ἄγεσθε at Gal 5:18. One English preposition must serve all four occurrences in this paragraph, or a deliberate refrain becomes invisible.

Built onbdf · bdag

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

περιπατεῖτε is Paul’s default ethical idiom, taken over from the LXX rendering of הָלַךְ for the moral course of a life (Deut 8:6; Prov 8:20; and in the NT, Rom 6:4; 8:4; 13:13; Eph 4:1; Col 1:10).

Two features matter here:

The image is live, not dead. Galatians has a running motion-field: τρέχω of the footrace (2:2; 5:7), στοιχέω of keeping rank (5:25; 6:16), ἄγομαι of being led (5:18). Paul is not reaching for a faded synonym of ‘live’; he is placing conduct inside a field of movement, direction, and pace. Per §8 the image is retained: ‘walk,’ never ‘live’ or ‘conduct yourselves.’

The present imperative is durative. Not ‘take a walk’ but ‘keep walking’ — a sustained practice, aspectually matched to the standing condition described in v. 17. The literal tier marks the aspect; the smooth tier carries it by context rather than importing ‘keep on,’ per the standing corpus rule.

Note that Paul had πολιτεύομαι available (he uses it at Phil 1:27) and στοιχέω, which he deploys nine verses later at 5:25. The choice of περιπατέω here rather than στοιχέω is a choice, and the two must stay lexically distinct in English.

Built onbdag · muraoka-lxx

ἐπιθυμία is a morally neutral stem. Its charge comes from the object or qualifier, never from the lexeme: Christ uses the cognate verb of his own desire at Luke 22:15 (ἐπιθυμίᾳ ἐπεθύμησα), and the LXX Decalogue uses it of the forbidden coveting (Deut 5:21). Here the qualifier σαρκός marks it as disordered, so the corpus rendering ‘craving’ applies.

‘Lust’ is excluded as a base gloss. Modern English has narrowed it irrecoverably to sexual appetite, whereas the vice-catalogue three verses later (5:19-21) shows that only the first three of fifteen items are sexual — the rest are idolatry, sorcery, hostility, rage, factionalism, envy, drunkenness. Rendering ἐπιθυμία ‘lust’ here mis-scopes the whole paragraph.

Both nouns are anarthrous. This is qualitative, not indefinite: craving of the kind that belongs to flesh. Do not render ‘the desires of the flesh’ (plural, definite) — Paul writes a singular, and the singular names the appetite as such rather than itemizing instances. The plural ἐπιθυμίαι arrives at 5:24, where instantial force is appropriate.

The genitive σαρκός is subjective or possessive — the craving that flesh has, or that belongs to it. Per the standing σάρξ policy, one English gloss ‘flesh’ is retained across the whole Galatians range: Christ’s humanity (4:4), Paul’s illness (4:13-14), Ishmael’s generation (4:23, 29), the credential-sphere (3:3; 6:12-13), and the hostile appetite here. Substituting ‘sinful nature’ would hide from the reader that one word does all of that work, and would import a modern anthropology (§9).

Built onbdag · muraoka-lxx

This is where most English renderings go wrong, and it is a grammatical question with a clear answer.

οὐ μή + aorist subjunctive is the emphatic future denial — the strongest negation Greek possesses (BDF §365). It is not a prohibition. Prohibition uses μή + aorist subjunctive (without οὐ) or μή + present imperative; the piled negatives here reinforce rather than cancel. The construction is a Septuagintism in origin, at home in solemn divine pronouncement (Ps 120:3-4 LXX; Isa 28:16), and the corpus has already fixed it at Gal 4:30 (οὐ μὴ κληρονομήσει) as ‘shall certainly not.’

The consequences are substantial:

• Paul is not issuing two commands (‘walk by the Spirit and do not gratify the flesh’). He issues one command and one promise.
• The καί is coordinating with consecutive force. Koine regularly expresses result by simple καί after an imperative, and the corpus policy is to render plain ‘and,’ letting the consequence be heard rather than over-specifying with ‘then’ or ‘so that.’
• The emphasis must be marked in English. A bare ‘will not’ under-translates the strongest denial Greek can build; ‘certainly not’ or ‘most certainly not’ is required.

The aorist is complexive, viewing the non-completion as a whole. It is not iterative and does not mean ‘you will not gratify it on any given occasion.’

Built onbdf · bdag · metzger-tc

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

τελέσητε is from τέλος — to bring a thing to its intended end. Not ‘gratify,’ not ‘satisfy,’ not ‘indulge.’

The distinction is theologically load-bearing, because v. 17 immediately concedes that the flesh does crave (ἐπιθυμεῖ) against the Spirit. If v. 16 denied the presence of craving, v. 17 would contradict it. What v. 16 denies is the craving’s consummation — its arrival at the outcome it is reaching for.

Related corpus decisions bear on the gloss:

• τελειόω is already fixed as ‘bring to completion’ (Phil 3:12), with ‘perfect’ excluded from modern registers.
• ἐπιτελέω is fixed as ‘bring to completion / carry through’ (Phil 1:6; 1 Pet 5:9).
• πληρόω is fixed as ‘fill up’ (the vessel image), and stands two verses earlier at Gal 5:14 (πεπλήρωται).

τελέω here should be rendered with completion-vocabulary that stays distinguishable from all three. ‘Carry out’ is adopted for the smooth tier and ‘bring to completion’ for the literal, since ‘carry out’ names reaching the end of an undertaking without the vessel-image of πληρόω or the perfective prefix of ἐπιτελέω.

Excluded: ‘gratify’ and ‘satisfy’ (both relocate the τέλος from the action’s completion into the appetite’s pleasure), and ‘fulfil’ (which in modern English has drifted toward the fulfilment of predictions and duplicates πληρόω).

Built onbdag · bdf

The verse is textually stable in all major witnesses (𝔓⁴⁶ 𝔓⁵¹ ℵ A B C D F G Ψ 33 1739 and the Byzantine tradition). NA28 prints without variation in the apparatus for the clause as a whole.

Two minor points worth recording:

• Some witnesses read τελέσετε (future indicative) for τελέσητε. This is not a semantic change — οὐ μή + future indicative is an attested alternative to οὐ μή + aorist subjunctive with identical force (cf. the corpus entry at Gal 4:30) — and it looks like a phonetic/orthographic levelling in a period when ε and η were converging in pronunciation. The subjunctive is the better-attested reading and is the harder form.
• A handful of witnesses insert the article before πνεύματι, assimilating to the fuller expressions of 4:6 and 5:17 (τὸ πνεῦμα). The anarthrous form is original and is the reading that the refrain across 5:5, 16, 18, 25 supports.

Neither variant would change the rendering.

Built onna28 · metzger-tc

Vulgate: Dico autem: spiritu ambulate, et desideria carnis non perficietis. Three observations. (1) spiritu is a bare ablative, reproducing the Greek dative’s ambiguity exactly — Jerome did not resolve it, and the Latin tradition therefore preserved the openness that English versions have tended to close. (2) non perficietis is a future indicative and correctly reads οὐ μή as denial rather than prohibition; but the double negative’s emphasis is lost, since Latin has no comparable construction. (3) desideria is plural where the Greek is singular, and later Latin exegesis reads the verse as concerning multiple appetites rather than the flesh’s appetite as such. This pluralization is reception, not evidence (§11), and must not be followed.

Peshitta renders the second clause with a straightforward negated future, again agreeing that this is denial rather than command, and uses the standard Syriac equivalent for ‘walk’ — preserving the motion-idiom.

The versional consensus that οὐ μή is denial and not prohibition is worth weighing against the modern English tradition, which has largely gone the other way.

Built onmetzger-tc

Chrysostom (Homilies on Galatians, ad loc.) reads the verse as consolation rather than demand. His point is that Paul does not say ‘struggle against the flesh’ but shows that the Spirit’s presence makes the flesh’s project fail — the walking is the whole of what is asked, and the outcome follows. He also presses the parallel between the Spirit’s leading here and the pedagogue of 3:24-25: the Galatians are being offered a governing presence, not a heavier rule.

Jerome (Commentary on Galatians) sets v. 16 against v. 17 explicitly and takes the pairing to mean that the desire remains while its accomplishment is prevented — which is precisely the τελέω point above, reached from the Latin perficietis rather than the Greek. His reading is evidence that the completion-sense was felt in antiquity, even in Latin.

Augustine, engaging the verse in the Pelagian controversy, reads πνεύματι as the divine Spirit rather than the human spirit and takes the promise as a statement about grace’s efficacy — a reading that later becomes contested precisely because it makes v. 16 an assurance rather than an exhortation. Both sides of that debate turn on the οὐ μή construction, which shows how much the grammar carries.

What none of them do is read the second clause as a second command. That reading has no ancient support of which I am aware.

Proposed — pending reviewJohn Chrysostom, Commentary on the Epistle of St. Paul the Apostle to the Galatians, in Nicene and Post-Nicene Fathers, First Series, vol. 13, ed. Philip Schaff, Buffalo: Christian Literature Publishing, 1889 · Jerome, Commentary on Galatians, trans. Andrew Cain, Fathers of the Church 121, Washington, DC: Catholic University of America Press, 2010

Epistolary argument (§10), and the connectives are load-bearing. Paul’s structure across vv. 16-18 is:

• v. 16: command + promise (Λέγω δέ … καί …)
• v. 17: γάρ — the ground of the promise (the two are drawn up against each other)
• v. 18: εἰ δέ — the consequence (not under law)

Every particle in that chain is rendered. Dropping δέ at v. 16 or γάρ at v. 17 would leave three free-standing assertions where Paul built a single argument.

The rhythm of the verse itself is worth preserving: a short formula, a two-word command, and then a long emphatic clause with the negation stacked at the end (οὐ μὴ τελέσητε). English word order can hold that shape, and the smooth rendering keeps the negation late for the same reason.