Paul has just appealed to ordinary legal practice: once a human διαθήκη has been validly executed, nobody cancels it or adds codicils (v. 15). Verse 17 applies that to God. The covenant with Abraham was ratified in advance — and the perfect participle insists the ratification still stands — so a legal instrument that only came into being four and a third centuries afterward cannot void it.

Four things carry the argument, and all four are easy to lose in English:

1. A ratification-word answered by its own negation. προκεκυρωμένην (κῦρος, ‘authority, validity’) is negated by ἀκυροῖ (ἀ-κῦρος). Paul says pre-validated … does not in-validate. The wordplay is the sentence.
2. Two perfects. προκεκυρωμένην and γεγονώς both assert completed acts with standing results — the covenant stands ratified; the law stands as something that came to be at a datable point. Flattening either into a simple past dissolves the contrast between a permanent instrument and a late arrival.
3. The law as grammatical subject. Greek makes ὁ … νόμος the subject of an active verb with no agent behind it. Smooth English usually passivizes (‘is not annulled by the law’), which is legitimate for readability but should be noted.
4. The number. τετρακόσια καὶ τριάκοντα is rendered as written. It agrees with the LXX of Exod 12:40, not the Hebrew, and the arithmetic problem is real and belongs in the apparatus, not in a softened rendering.

The word order is deliberately heavy: the object (διαθήκην) is fronted, then a long attributive bracket (ὁ μετὰ … νόμος) separates article from noun by six words, so the clause’s own shape enacts the delay it describes.

A discourse-management formula, not a new topic. The near-identical Λέγω δέ at Gal 4:1 and 5:16 and τοῦτο δὲ λέγω at 1 Cor 1:12 and 7:29 all introduce the application or clarification of what precedes, never fresh material. δέ is therefore transitional (‘now’), not adversative (‘but’) — rendering it ‘but’ would set v. 17 against v. 16 rather than drawing it out of vv. 15-16.

τοῦτο is cataphoric, pointing forward to the διαθήκην clause. Per standing policy the bare demonstrative is retained: ‘this is what I mean’ and ‘my point is this’ both paraphrase away a pronoun that English can carry perfectly well. A colon after λέγω is editorial; the earliest majuscules are unpunctuated.

The glossary fixes διαθήκη as ‘covenant’ — the LXX equivalent of בְּרִית some 270 times, chosen over συνθήκη precisely because the Hebrew denotes a unilateral disposition rather than a negotiated pact. That gloss stands here.

But vv. 15-17 are the one place where the Hellenistic legal sense is demonstrably in play. In the papyri διαθήκη is the ordinary word for a will, and v. 15’s appeal (ἀνθρώπου κεκυρωμένην διαθήκην οὐδεὶς ἀθετεῖ ἢ ἐπιδιατάσσεται) draws on testamentary and contractual practice: a validly executed instrument cannot be set aside or supplemented.

Anarthrous and fronted. The indefinite article is retained in English to track the qualitative force: Paul states a rule about a ratified covenant and lets the Abrahamic instance stand under it.

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

προκεκυρωμένην is a NT hapax, built on κυρόω (‘make valid, ratify’), itself from κῦρος, ‘authority, binding force’. The simplex is a documentary term: 2 Cor 2:8 (κυρῶσαι εἰς αὐτὸν ἀγάπην, ‘ratify your love toward him’) and the κεκυρωμένην of Gal 3:15 are the only NT instances, and the papyri use the verb routinely of validly executed contracts and wills.

Three features must survive in English:

• The πρό- prefix. It marks the ratification as prior to the thing that follows — not merely ‘ratified’ but ratified beforehand. The whole chronology of the verse hangs on it.
• The perfect. A completed act of ratification whose legal effect stands at the time of writing. A simple past (‘which God ratified’) severs the result and makes the covenant a historical event rather than a document still in force.
• The passive with ὑπὸ τοῦ θεοῦ. The agent is expressed and is personal; per standing policy ὑπό + gen is rendered ‘by’ and kept distinct from διά (means) and ἐκ (source). Both appear in Galatians’ argument and must stay distinguishable.

‘Confirmed’ (so the older English, following the Vulgate’s confirmatum) is retained in liturgical register but is weaker than the Greek in modern usage, where it suggests corroboration rather than legal execution. Exclude ‘established’ (institutional) and ‘made’ (deletes the validity).

Proposed — pending reviewJ. H. Moulton and George Milligan, The Vocabulary of the Greek Testament Illustrated from the Papyri and Other Non-Literary Sources, London: Hodder & Stoughton, 1930.

ἀκυροῖ is the exact α-privative counterpart of κυρόω: to deprive of κῦρος, to render legally void. NT only here and Matt 15:6 // Mark 7:13, where the tradition of the elders ἀκυρώσατε / ἀκυροῦντες the word of God — the same register of one instrument voiding another.

The wordplay with προκεκυρωμένην is the backbone of the sentence and must be reproduced if English allows it. ‘Previously ratified … does not annul’ keeps the legal register but loses the shared root; ‘pre-validated … does not invalidate’ keeps the root at some cost in naturalness and is offered in the study tier. Per the corpus discipline on figura etymologica (Gal 5:1) and on repeated roots (Gal 2:19 διὰ νόμου νόμῳ), the figure is recorded even where English cannot carry it.

Keep ἀκυρόω lexically distinct from three neighbours Paul uses in the same stretch of argument:

• ἀθετέω (v. 15) — ‘set aside, nullify’, already fixed, of voiding a standing instrument;
• καταργέω (this verse) — ‘render ineffective’, already fixed, κατά + ἀργός, putting out of operation;
• ἐπιδιατάσσομαι (v. 15) — adding a codicil.

Four distinct legal verbs in three verses. Levelling any two of them to ‘annul’ or ‘cancel’ hides that Paul is working through a sequence of distinct ways an instrument could be attacked.

The present tense is gnomic — what is characteristically the case — and renders as a simple English present, not a progressive. A modal (‘cannot annul’, so the Vulgate’s non irritum facit read as ability in the English tradition) is excluded: Greek states the fact, and the impossibility is an inference the reader is to draw.

The article ὁ at the head of the phrase does not attach to anything until νόμος six words later. Everything between — μετὰ τετρακόσια καὶ τριάκοντα ἔτη γεγονώς — is an attributive bracket defining the subject. The construction is the same device already catalogued for article + prepositional phrase (οἱ σὺν ἐμοὶ ἀδελφοί, Phil 4:21) and article + πρό + gen (τοὺς πρὸ ἐμοῦ ἀποστόλους, Gal 1:17), stretched to its limit.

The effect is that the reader waits through the whole interval before learning what arrived at the end of it — the syntax performs the delay. The literal tier preserves the sequence woodenly (‘the after four hundred and thirty years having-come-to-be law’); reading tiers resolve to a postposed relative clause, with the displacement recorded here rather than compensated for by added emphasis words.

γεγονώς is the perfect of γίνομαι, and the standing glossary entry applies: come to be / become, preserving the contrast with εἰμί and ὑπάρχω. Paul could have written ὁ … ὤν and did not. The law is presented as something that entered a state at a datable point, over against a covenant that stands ratified. ‘The law, which was given 430 years later’ substitutes a different verb (δίδωμι — which Paul does use at v. 21, ἐδόθη) and erases the choice.

Note too that ὁ νόμος is articular here, against the anarthrous qualitative νόμος of vv. 11, 18, and 21. The article is anaphoric and historical: the Sinai legislation as a datable instrument, not law as a regime. The standing anarthrous-νόμος policy (withhold the article) does not apply where Greek supplies it.

The figure is rendered as written. ‘About four hundred years’, ‘over four centuries’, and any harmonizing recalculation are excluded (§4, §6).

The source is Exod 12:40, where the texts diverge:

• MT: ‘the dwelling of the sons of Israel which they dwelt in Egypt was four hundred and thirty years.’
• LXX (and the Samaritan Pentateuch): ‘the dwelling of the sons of Israel which they dwelt in the land of Egypt and in the land of Canaan, they and their fathers, was four hundred and thirty years.’

The LXX/SP reading spreads the 430 years across the patriarchal sojourn and the Egyptian period, which makes the interval run from the Abrahamic promise to Sinai — exactly what Paul needs. On the MT reckoning the 430 years cover Egypt alone, and the Abraham-to-Sinai span is substantially longer. Gen 15:13 and Acts 7:6 give 400 years for the affliction.

μετά + accusative is temporal, to be kept distinct from μετά + gen (‘with’), which Paul uses at Gal 2:1 — the case alone marks which relation is in view. ἔτη is accusative of extent governed by the preposition.

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.

An εἰς τό + infinitive construction, which the glossary fixes as ‘so that / with a view to’, predominantly telic in Paul with result shading possible (BDF §402).

Here the clause is attached to a negated main verb, which makes the force genuinely open:

καταργῆσαι is already fixed as ‘render ineffective / nullify’: κατά + ἀργός, to put out of operation. Paul turns the verb both ways in this letter (Gal 3:17; 5:4 κατηργήθητε; 5:11 κατήργηται), and one English root is required across all of them. The aorist is complexive — the nullification viewed as a single whole act.

Note the three negation-verbs now stacked: ἀθετεῖ (v. 15), ἀκυροῖ, καταργῆσαι. Each names a different mode of attack, and the English must keep them apart.

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

τὴν ἐπαγγελίαν is articular and anaphoric — the promise of vv. 14 and 16, which v. 16 has just identified as spoken to Abraham καὶ τῷ σπέρματι αὐτοῦ. Per the standing entry, ἐπαγγελία is not a prediction but a formally binding undertaking, which is what allows Paul to argue it on the analogy of a ratified διαθήκη.

The argumentative force of the verse is chronological-legal rather than moral. Paul is not saying the law is bad; he is saying it is late, and that a later instrument cannot void an earlier one that was validly executed. That restraint matters for how the verse is read: the question of what the law positively is for is deferred to v. 19, and the rendering must not anticipate the answer.

Note also that God is named only as the agent of the ratification (ὑπὸ τοῦ θεοῦ) and not as the giver of the law — Paul reserves that for the carefully distanced διαταγεὶς διʼ ἀγγέλων ἐν χειρὶ μεσίτου of v. 19. The reticence is deliberate and should not be repaired by supplying ‘the law God gave’.

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

The critical text (NA28/UBS5) reads ὑπὸ τοῦ θεοῦ with no further qualification, supported by 𝔓46 ℵ A B C P Ψ 33 and the bulk of the early versions.

A substantial body of later witnesses — including D F G and the Byzantine tradition, reflected in the Textus Receptus and hence in the older English versions — reads ὑπὸ τοῦ θεοῦ εἰς Χριστόν, ‘by God unto Christ’. The addition is almost certainly a scribal expansion drawn from v. 16 (ὅς ἐστιν Χριστός) and v. 24 (εἰς Χριστόν), specifying what the ratified covenant was directed toward. It is the easier and fuller reading, and its absence from the earliest and most diverse witnesses is decisive.

Minor variation also affects the word order of τετρακόσια καὶ τριάκοντα and the spelling of ἀκυροῖ (ἀκυροῦται in a few minuscules, converting the active to a passive). Neither affects the sense, and the active ἀκυροῖ — with ὁ νόμος as grammatical subject — is secure.

Exod 12:40 LXX supplies the figure (see above) and is the decisive datum: Paul’s 430 years presuppose the Greek text’s inclusion of Canaan in the reckoning, not the Hebrew’s restriction to Egypt. This is a clear case where the NT author’s own textual base is data to record rather than to correct (§1, and the OT-citation policy).

Vulgate: hoc autem dico testamentum confirmatum a Deo quae post quadringentos et triginta annos facta est lex non irritum facit ad evacuandam promissionem. Three observations. Jerome’s testamentum adopts the testamentary sense of διαθήκη outright, which the Greek leaves open — reception, not evidence (§11). Non irritum facit renders ἀκυροῖ well but cannot reproduce the κυρόω/ἀκυρόω pairing, since confirmatum and irritum share no root; the Latin thus loses the verse’s central figure, which is one reason the English tradition descending from it (‘confirmed … cannot disannul’) has had to coin an awkward compound to recover it. Ad evacuandam takes the εἰς τό clause as purpose.

Peshitta likewise renders with distinct roots for the ratifying and the annulling, and shows the same loss.

Chrysostom (Hom. in Gal. 3) reads the verse as an argument from precedence and from God’s unchangeableness: if human covenants stand once executed, far more does a divine one, and the lateness of the law shows it was never meant to overturn what was already settled. He is careful to insist that Paul is not disparaging the law — the point is sequence, not quality.

Jerome (Comm. in Gal.) wrestles with the chronology at some length, noting the divergence from the Hebrew and the 400 years of Gen 15:13, and concludes that Paul follows the Septuagint reckoning. His discussion is early evidence that the arithmetical difficulty was felt from the start and was not invented by modern criticism.

Augustine emphasizes the irrevocability of the divine disposition and reads the εἰς τό clause as purpose: the law was not introduced with a view to voiding the promise.

Across the Fathers the verse functions apologetically — against readings that set law and promise as rival systems — and the consistent move is to make lateness, not inferiority, the operative category.

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 Publishing, 1889. · Jerome, Commentarii in Epistulam Pauli ad Galatas, ed. Giacomo Raspanti, Corpus Christianorum Series Latina 77A, Turnhout: Brepols, 2006.