Verse 18 is the proof that closes the will-and-codicil argument of vv. 15-17. Paul has said that a ratified human διαθήκη cannot be annulled or supplemented, and that a law arriving 430 years later therefore cannot void the prior undertaking to Abraham. Verse 18 states the logical ground: two sources of the inheritance are mutually exclusive, and the Abraham narrative settles which one is in force.

The whole argument runs on prepositions. The protasis puts ἐκ νόμου against ἐξ ἐπαγγελίας — two candidate sources for one inheritance — and the second clause does not simply repeat the winner but shifts to διʼ ἐπαγγελίας, promise as the channel through which God acted. English must keep ἐκ and διά distinct at every point or the structure disappears.

Two features carry the weight. The verb is a perfect (κεχάρισται): a completed grant whose result still stands, which is exactly what makes the 430-year-later law powerless against it. And the verb is χαρίζομαι, built on χάρις — Paul could have written ἔδωκεν and did not. An inheritance given as a gratuity is not a wage, which is the point the whole chapter is driving at.

Both clauses are verbless in Greek, with the verb arriving only at the end of the second. The literal tier keeps the ellipsis visible.

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

εἰ + indicative is a first-class condition: the protasis is entertained as real for the sake of argument, and the form carries no claim that the speaker holds it. The corpus policy already excludes rendering it “since,” and the exclusion is sharpest here — Paul’s whole case is that the inheritance is not ἐκ νόμου, and “since the inheritance is out of law” would make him assert what he denies.

The construction is the same reductio Paul runs at Gal 2:21 (εἰ γὰρ διὰ νόμου δικαιοσύνη), 3:21 (εἰ γὰρ ἐδόθη νόμος), and 1 Cor 15:14. In each the premise is granted hypothetically so that the apodosis can be seen to be unacceptable.

Both protasis and apodosis are nominal, with the copula elided in each. Per the standing verbless-clause policy, the supplied “is” is bracketed in the literal tier and carried by parallelism in the reading text. Note that the ellipsis is not sloppiness: the compression is what makes the two prepositional phrases stand bare against each other, which is the figure.

Three prepositional phrases in one sentence, and the variation is deliberate.

ἐκ + genitive marks source — the quarter from which a thing issues — and Galatians uses it throughout as the source-marker in the competing-origins formulae: ἐξ ἔργων νόμου against ἐκ πίστεως (2:16; 3:2, 5, 10), ἐκ νόμου here. The corpus keeps “out of” in the literal tier at every occurrence so the parallelism between protasis and apodosis is visible; “by the law … by promise” blurs source into instrumentality and hides that the two phrases are grammatically identical.

διά + genitive marks mediation — the channel through which an act passes. Paul had ἐξ ἐπαγγελίας available in the second clause; he wrote διʼ ἐπαγγελίας. The switch is not stylistic variation: it moves promise from the category of competing source (where it stood in the protasis, over against law) into the category of the means by which God actually conferred the inheritance. The same distinction is maintained at Gal 4:23 (κατὰ σάρκα … διʼ ἐπαγγελίας), where Paul again declines to make both limbs parallel.

All three nouns are anarthrous and qualitative. Per the standing anarthrous-νόμος policy, the article is withheld in every tier: supplying “the law” narrows a categorical claim about law-as-regime into a statement about the Mosaic document, which is precisely the narrowing the anarthrous form resists.

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

ἡ κληρονομία — from κλῆρος, the allotted portion, and the standing LXX equivalent of נַחֲלָה, the land-promise vocabulary (Deut 4:38; Josh 1:6; Ps 15:5 LXX). The corpus already fixes one English root across κληρονόμος / κληρονομία / κληρονομέω so the Abrahamic-promise thread stays traceable from Gal 3:18 through 3:29 (κληρονόμοι) and 4:1-7.

The article is anaphoric, not generic: the inheritance — the one under discussion since v. 16, the portion promised to Abraham and his σπέρμα. Supplying a qualifier (“the inheritance of the promise,” “God’s inheritance,” “the blessing”) narrows what Paul leaves as the single contested object.

Avoid “estate” (testamentary register only), “what was promised” (deletes the noun and pre-empts the second clause), and “God’s gift” (imports the verb into the noun). Note also that Paul does not say what the inheritance consists in here; 3:14 has named the Spirit and the Abrahamic blessing, 3:29 will name heirship itself, and the rendering must not specify.

ἐπαγγελία is not a prediction but a formally binding undertaking — which is exactly why Paul can argue it on the analogy of a ratified διαθήκη (vv. 15-17). The whole force of vv. 15-18 depends on the noun belonging to the legal-documentary domain, not the predictive one.

The noun is anarthrous at both occurrences in this verse, hence qualitative: promise as a mode, over against law as a mode. The literal tier leaves the article out; reading tiers may supply “promise” without an article, which is natural English and tracks the Greek.

Avoid “God’s promise” (supplies a possessive Paul withholds here, though he supplies it himself at 3:21 τῶν ἐπαγγελιῶν τοῦ θεοῦ) and avoid pluralizing: Paul uses the plural at 3:16 and 3:21 and the singular here, and the alternation should be recorded rather than levelled.

χαρίζομαι is the verb of giving built on χάρις. Paul had δίδωμι available — he uses it of this very transaction at 3:21 (ἐδόθη νόμος) and 3:22 (ἡ ἐπαγγελία … δοθῇ) — and here chose the grace-verb instead. The corpus gloss is “graciously give / grant as a favor,” and the χάρις root must stay audible: a bare “gave” drops the one morpheme that makes the clause an answer to ἐκ νόμου.

Three further features are load-bearing:

The perfect. κεχάρισται names a completed act whose result stands at the time of writing. This is not a stylistic preference — it is the grammatical backbone of vv. 15-18. A grant that still stands cannot be annulled by a later instrument. An English simple past (“God gave”) severs the standing result and quietly removes the argument; the present perfect is required in the reading text, and “God gave” is confined to declared liturgical register.

The deponent middle. χαρίζομαι is middle-only; no reflexive or self-interested force may be read into the voice.

The unexpressed object. Greek supplies no accusative. English requires one and it is bracketed in the literal tier. Supplying “the inheritance” is the natural completion from the first clause, but note that “granted it to Abraham by promise” and “made a gift of it to Abraham” are both resolutions of an ellipsis, not renderings of a stated object.

οὐκέτι marks the termination of a prior state and in Paul is frequently logical as well as temporal (Rom 6:9; 7:17, 20; 11:6; Gal 2:20; 3:25; 4:7). Here the force is almost entirely logical: not “at a later time it ceased to be from promise,” but “on that supposition it is not, after all, from promise.”

The closest parallel is Rom 11:6, where the identical move is made with χάρις and ἔργα: εἰ δὲ χάριτι, οὐκέτι ἐξ ἔργων, ἐπεὶ ἡ χάρις οὐκέτι γίνεται χάρις. Two sources are declared mutually exclusive, and οὐκέτι carries the exclusion.

Render uniformly “no longer.” Avoid “then it is not” (deletes the adverb) and “not any more” (colloquial). The logical force is recorded in the note rather than built into the English, since Paul chose a temporal adverb to make a logical point and the reader should see that.

The second clause is τῷ δὲ Ἀβραὰμ διʼ ἐπαγγελίας κεχάρισται ὁ θεός — recipient, then channel, then verb, then subject. Both ends are marked.

The dative τῷ Ἀβραάμ is fronted before everything, topicalizing the recipient: to Abraham — the man whose case has been under discussion since 3:6 — is where the giving landed. Per the corpus topicalization policy, English keeps the recipient early where syntax permits and records the fronting in apparatus rather than compensating with added emphasis words.

ὁ θεός is held to the very end. The sentence is built so that its last word is the giver. The Byzantine-era tendency in such clauses is to move the subject forward for readability; the harder word order stands in the critical text and should be preserved where English tolerates it.

Between them, διʼ ἐπαγγελίας sits immediately before the verb, in the emphatic pre-verbal slot. The modern rendering marks this with dashes in the study tier rather than relocating the phrase.

γάρ grounds v. 18 in v. 17: the law cannot void the promise because inheritance by law and inheritance by promise are mutually exclusive. Epistolary connectives are argument structure and are always rendered (§10); dropping γάρ here would leave v. 18 looking like a fresh assertion rather than the logical floor under v. 17.

Render “for.” Reserve “because” only where English syntax makes “for” ungrammatical.

The δέ in the second clause is genuinely adversative, not merely transitional. The corpus policy forbids manufacturing antitheses with δέ where the particle only advances (Jas 1:15) or heightens (Phil 2:8) — but here the clause flatly contradicts the hypothetical just entertained, and “but” is correct. The classification is recorded rather than assumed.

Its position is third in the clause, displaced by the fronted dative phrase τῷ Ἀβραάμ, which the postpositive treats as a single unit.

The text of this verse is stable. NA28 prints it without variation marked in the critical apparatus of any consequence for sense.

Two minor matters worth recording. First, a scattering of later witnesses supply an object for κεχάρισται (αὐτήν or τὴν κληρονομίαν), which is scribal completion of a felt ellipsis — the kind of filling-in the tradition does repeatedly in this corpus — and must not be followed. Second, the elision διʼ before the vowel is orthographic and carries no weight.

There is also no witness substituting ἔδωκεν for κεχάρισται, which is worth noting: the grace-verb and the perfect tense were both felt as original and neither was levelled toward the commoner δίδωμι that Paul himself uses three verses later.

Paul’s second clause is not a quotation but a summary of the Genesis Abraham narrative, where the land and seed are conferred by divine undertaking (Gen 12:1-3, 7; 13:14-17; 15:4-7, 18-21; 17:4-8) with no condition of performance stated. Gen 15:18 is the controlling text for the διαθήκη language of vv. 15-17 (LXX: διέθετο κύριος τῷ Ἀβραμ διαθήκην), and Gen 15:7 and 17:8 supply the inheritance vocabulary (κληρονομῆσαι, εἰς κατάσχεσιν αἰώνιον).

The LXX κληρονομ- family in Deuteronomy and Joshua is land-allotment vocabulary throughout, and Paul’s argument trades on the fact that an allotment is received, not earned. That is why χαρίζομαι rather than δίδωμι: the verb specifies the mode under which the portion came.

Note what Paul does not say. He does not argue that the Abraham narrative contains no commands — Gen 17:9-14 commands circumcision, and Gen 22:16-18 grounds a renewed oath in obedience. His claim is narrower: the inheritance was conferred διʼ ἐπαγγελίας, and the conferral stands (perfect).

Proposed — pending reviewJ. B. Lightfoot, Saint Paul's Epistle to the Galatians, 10th ed., London: Macmillan, 1890

The verse asserts that law and promise cannot both be the source of one inheritance. The logic is not that law is bad but that the two modes are categorically different: a promise confers; a law stipulates. If the portion comes out of law it is a return on compliance, and a return on compliance is not a gift — which is why Paul reaches for χαρίζομαι rather than δίδωμι.

The structural parallel is Rom 11:6 (εἰ δὲ χάριτι, οὐκέτι ἐξ ἔργων), where the same οὐκέτι does the same exclusive work, and Rom 4:14 (εἰ γὰρ οἱ ἐκ νόμου κληρονόμοι, κεκένωται ἡ πίστις καὶ κατήργηται ἡ ἐπαγγελία), which is Gal 3:18 in a Roman key with the same conditional form and the same inheritance-vocabulary.

The perfect κεχάρισται carries the theological weight as much as the vocabulary does. The point is not that God once made a generous gesture but that a grant made then is in force now — which is precisely what a later instrument cannot touch.

Proposed — pending reviewHans Dieter Betz, Galatians: A Commentary on Paul's Letter to the Churches in Galatia, Hermeneia, Philadelphia: Fortress, 1979 · J. Louis Martyn, Galatians: A New Translation with Introduction and Commentary, Anchor Bible 33A, New York: Doubleday, 1997 · John Chrysostom, Commentary on the Epistle to the Galatians, in Nicene and Post-Nicene Fathers, First Series, vol. 13

The Vulgate reads nam si ex lege hereditas iam non ex repromissione Abrahae autem per repromissionem donavit Deus. Two observations. It renders both ἐκ phrases with ex and the διά phrase with per, preserving the source/mediation distinction that most English versions blur — a better guide here than the English tradition. But donavit is a simple perfect in Latin, which is ambiguous between the Greek aorist and perfect forces; Latin cannot show that κεχάρισται asserts a standing result, and the English “gave” descending from it loses what the Greek tense supplies.

The Peshitta likewise supplies an object pronoun for the verb, as Syriac idiom requires; this is translational necessity, not evidence for the Greek, and must not be cited in support of the witnesses that add αὐτήν.

Neither version can reproduce the anarthrous νόμου, since Latin has no article and Syriac handles definiteness differently — so neither is evidence on the contested question of whether “law” or “the law” is meant.