Gal 3:15 About
3:15
contestedhapax
Draft — AI-drafted, not yet reviewed by a human editor. How this is made
Greek · SBLGNT

Ἀδελφοί, κατὰ ἄνθρωπον λέγω· ὅμως ἀνθρώπου κεκυρωμένην διαθήκην οὐδεὶς ἀθετεῖ ἢ ἐπιδιατάσσεται.

Adelphoi, kata anthrōpon legō; homōs anthrōpou kekyrōmenēn diathēkēn oudeis athetei ē epidiatassetai.

αδελφοι, κατα ανθρωπον λεγω· ομως ανθρωπου κεκυρωμενην διαθηκην ουδεις αθετει η επιδιατασσεται.
Parsing · 12 words
Ἀδελφοί ἀδελφός noun voc masc pl brothers — Masculine plural rendered as written; mixed referent noted, not expanded (§3).
κατὰ κατά prep + acc according to
ἄνθρωπον ἄνθρωπος noun acc masc sg man (a human being) — Anarthrous and generic; the same noun recurs in the genitive four words later.
λέγω λέγω pres act ind 1sg I speak
ὅμως ὅμως adv (concessive particle) even so / nevertheless — Position is irregular; see the section on its placement and accent.
ἀνθρώπου ἄνθρωπος noun gen masc sg of a man — Fronted outside the article-less noun phrase; topicalizing — it is the human origin of the instrument that is at issue.
κεκυρωμένην κυρόω perf pass ptc acc fem sg having been ratified — Perfect: completed act of ratification whose legal force stands. Agent unexpressed.
διαθήκην διαθήκη noun acc fem sg covenant / will
οὐδεὶς οὐδείς pron nom masc sg no one
ἀθετεῖ ἀθετέω pres act ind 3sg sets aside / nullifies — Gnomic present: what characteristically holds, not a report.
ἢ ἤ conj (disjunctive) or
ἐπιδιατάσσεται ἐπιδιατάσσομαι pres mid ind 3sg adds further stipulations — NT hapax; double compound ἐπί + διά + τάσσω. Middle is lexical, not reflexive.
αδελφοι αδελφος κατα κατα ανθρωπον ανθρωπος λεγω λεγω ομως ομως ανθρωπου ανθρωπος κεκυρωμενην κυροω διαθηκην διαθηκη ουδεις ουδεις αθετει αθετεω η η επιδιατασσεται επιδιατασσομαι

Brothers, I speak in human terms: even so, once a person's covenant has been ratified, no one sets it aside or adds codicils to it.

context

Overview

Paul has just closed the argument from Abraham’s faith (3:6-14) and now turns to a second, harder question: if the promise to Abraham stands, what has the Sinai legislation to do with it? He begins not with Scripture but with a commonplace from the law courts.

The verse has three moves. Ἀδελφοί re-opens direct address after a stretch of sustained argument — the first vocative since 1:11. κατὰ ἄνθρωπον λέγω flags the register: the illustration that follows is drawn from ordinary human practice, not from revelation. And then the illustration itself: a legal instrument, once properly executed, is beyond the reach of third parties — nobody voids it and nobody bolts additional clauses onto it.

Two things in that last clause carry the weight of the following verses. First διαθήκη, a word that in the renders בְּרִית (covenant) but in everyday Hellenistic Greek and the papyri overwhelmingly means a will or testament; Paul is almost certainly trading on both. Second ἐπιδιατάσσεται, a word found nowhere else in the Greek Bible, which names precisely the move Paul will deny to the law in v. 17: not abolishing the promise but supplementing it.

The argument is not that God’s covenant works like a Greek will. It is a fortiori: if even a merely human instrument is inviolable once ratified, how much more one ratified by God four hundred and thirty years before Sinai.

lexical

Ἀδελφοί

The vocative plural, standing alone at the head of the sentence with no possessive and no adjective — contrast ἀδελφοί μου ἀγαπητοί in James, or Paul’s own ἀδελφοί μου at Gal 6:1. Here the address is bare.

Its function is structural. Galatians runs long stretches without direct address (1:11 to 3:15 contains none), and the vocative marks a pivot: Paul turns from the scriptural proof of vv. 6-14 to a fresh line of argument. The same device recurs at 4:12, 4:28, 4:31, 5:11, 5:13, 6:1, 6:18.

Per §3 the masculine plural stands as written. That the referent is mixed is not in doubt — the Galatian assemblies included women — but the masculine plural is the ordinary Greek form for a mixed group, and the grammar is data rather than a problem to be solved in the rendering. The note carries the referent; the text does not expand it.

syntax

κατὰ ἄνθρωπον λέγω

A Pauline formula, occurring at Rom 3:5; 1 Cor 9:8; 15:32; Gal 1:11; 3:15 (cf. the related ἀνθρώπινον λέγω, Rom 6:19). κατά + accusative is the preposition of norm or standard, already fixed across this corpus, and here it marks the level at which the speaker is operating: he is arguing at the human level, from human practice.

The formula is a disclaimer about method, not about truth. Paul is not apologizing for an unworthy illustration; he is flagging that the warrant for what follows is common legal custom rather than Scripture. 1 Cor 9:8 makes the same move explicit — Am I saying these things κατὰ ἄνθρωπον? Does not the law also say these things?

Does the formula look forward or back?

Most read it as introducing the illustration that follows (so the punctuation of , with a raised point after λέγω). A minority has taken it with what precedes, as a concession about the argument of vv. 6-14. The word order favors the forward reading, and the ὅμως that follows requires something to concede against. But the punctuation is editorial — the earliest majuscules carry none — and that should be said plainly rather than left to the printed colon.

Note the deliberate collision with Gal 1:11, where Paul denies that his gospel is κατὰ ἄνθρωπον. The same phrase, the opposite claim: there, the gospel’s provenance; here, the argument’s register. It is the kind of verbal echo Galatians trades in (cf. ἀφορίζω at 1:15 and 2:12).

Built onbdag · bdf

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

syntax

ὅμως

Three NT occurrences (John 12:42; 1 Cor 14:7; Gal 3:15), all of them awkward. The word is the concessive adverb nevertheless, all the same, and in classical Greek it normally stands in the clause that concedes, answering a preceding participle or καίπερ.

Here it heads the clause before anything has been conceded, which is why editors and commentators have worried at it. Two live accounts:

The force and placement of ὅμως

(a) Displaced concessive. The particle belongs logically with ἀνθρώπου — a covenant that is only a man’s, nevertheless no one sets it aside. Greek tolerates such displacement (1 Cor 14:7 is the parallel: ὅμως τὰ ἄψυχα φωνὴν διδόντα, where the particle stands before the subject it concedes). On this reading the sentence runs: though it is merely human, still… This is the reading behind the older English though it be but a man’s covenant, yet…

(b) Accentual alternative ὁμῶς. A few have proposed reading the perispomenon ὁμῶς (likewise, in the same way), which would make the clause a straightforward comparison rather than a concession. The accent is editorial in the earliest witnesses, which are unaccented, so this cannot be settled from the manuscripts — a point that must be stated rather than glossed over. The sense difference is real but not large.

Per §10 the particle is always rendered. Dropping it — as several modern versions effectively do — removes the concession on which the a fortiori argument of vv. 16-18 turns. Even so is retained because English even so can carry both the concessive and the comparative hearing.

Built onbdf · bdag

Proposed — pending reviewJ. B. Lightfoot, Saint Paul's Epistle to the Galatians, London: Macmillan, 1865 (repr. Peabody: Hendrickson, 1999).

lexical

διαθήκη

The decisive lexical question in the verse, and one on which the argument of vv. 15-18 depends.

The datum. διαθήκη is the standard equivalent of בְּרִית, used some 270 times, deliberately chosen over συνθήκη because the Hebrew denotes a unilateral disposition imposed by the stronger party rather than a negotiated pact between equals. That is the sense Paul needs in v. 17 (a covenant previously ratified by God) and v. 18 (where it is answered by ἐπαγγελία).

The Hellenistic datum. In the documentary papyri and in ordinary Greek usage of the period, διαθήκη means a will or testament, and the verbs that cluster around it in this verse — κυρόω, ἀθετέω, ἐπιδιατάσσομαι — are the vocabulary of testamentary law.

Covenant or will at Gal 3:15?

Testamentary reading. The whole verse is drawn from the law of wills: a ratified testament cannot be revoked by outsiders or supplemented by codicils. The vocabulary is decisive, and Heb 9:16-17 shows a NT writer exploiting the testamentary sense explicitly.

Covenantal reading. The term is already -loaded for Paul, and in v. 17 the διαθήκη is one God made with Abraham, which is a בְּרִית and not a will — God does not die. On this reading Paul is using a covenant-word and merely illustrating from an analogous human instrument.

Double-sense reading. Paul is exploiting the overlap: the human instrument of v. 15 is a will, the divine instrument of v. 17 is a covenant, and the one word carries him across. The two are held together by the shared features of unilateral disposition and irrevocability once executed.

There is a further complication rarely admitted: under Greek and Roman law a testator could in fact normally revoke his own will. Paul’s point survives because οὐδείς concerns third parties, not the testator — but the difficulty has driven some to posit specifically Jewish or Egyptian testamentary practice (the irrevocable mattenat bari or the Egyptian deposited will) as the background. That hypothesis is attractive and unproven.

Policy. Render covenant in the reading text, per the standing corpus entry, and record the testamentary sense at every occurrence in this paragraph rather than switching words mid-argument. Switching to will at v. 15 and back to covenant at v. 17 — which several modern versions do — makes Paul’s single Greek word into two English ones and conceals that the argument runs on one term.

Built onbdag · muraoka-lxx

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.

lexical

κυρόω

From κῦρος, authority, validity; hence to make a thing valid, to ratify. Only two NT occurrences: here and 2 Cor 2:8, where Paul asks the Corinthians to ratify their love toward the offender — a formal, public confirmation, not a feeling. The uses it of confirming a vow (Gen 23:20; 4 Macc 7:9), and the compound προκυρόω stands at Gal 3:17 three verses later.

Three features of the form are load-bearing:

  • Perfect. A completed act of ratification whose legal effect stands in force at the moment described. English present perfect is required; a ratified covenant flattens it into an adjective and a covenant that was ratified severs the standing result.
  • Passive. The agent is unexpressed and must stay unexpressed. At v. 17 Paul will supply ὑπὸ τοῦ θεοῦ; here he deliberately does not, because the instrument in view is a human one and naming who ratified it is beside the point.
  • Attributive position. The participle stands between the fronted genitive ἀνθρώπου and the noun διαθήκην, which is a tight and marked arrangement: of-a-man having-been-ratified covenant. Keep the Greek order in the literal tier.

Exclude confirmed as base gloss in modern registers — English confirm has drifted toward corroborating a report. Validate and execute are too technical in the wrong direction. Ratify preserves the legal act of giving binding force.

Built onbdag · muraoka-lxx

lexical

ἀθετέω

Already fixed in this corpus at Gal 2:21 (set aside / nullify; register documentary and legal, naming the voiding of something standing in force). The entry is confirmed here, where the object is explicitly a ratified instrument and the legal register is beyond dispute.

The verb is the exact counterpart of κυρόω: one gives an instrument force, the other takes it away. English should ideally keep the pair audible. Nullify against ratify does this reasonably; set aside is the alternative where a less Latinate register is wanted.

The present is gnomic — what characteristically holds, not a report of current practice (cf. the gnomic-present policy fixed at Gal 5:9). Render with a simple English present, never a progressive.

Excluded as before: reject (attitudinal, and belongs to ἀρνέομαι/ἀποδοκιμάζω), ignore and disregard (far too light for a legal act), cancel (modern commercial register). Disannul is retained only in declared liturgical register, where the older English has fixed it.

Built onbdag

lexical

ἐπιδιατάσσομαι

A hapax in the Greek Bible — absent from the and occurring nowhere else in the NT — and rare in surviving Greek generally. The form is transparent: ἐπί (in addition, on top of) + διατάσσομαι (make arrangements, give orders, stipulate), the latter already a legal and administrative verb (Matt 11:1; Acts 7:44; 1 Cor 7:17; 11:34; Titus 1:5).

The compound therefore names the adding of further dispositions to an instrument already executed — in Roman testamentary terms, a codicil. This is not revocation (that is ἀθετέω, the first member of the pair) but supplementation.

The distinction is the whole point. Paul’s opponents are not proposing to cancel the Abrahamic promise; they are proposing to add a requirement to it. And v. 17 answers in exactly these terms: the law, coming 430 years later, οὐκ ἀκυροῖ the covenant εἰς τὸ καταργῆσαι τὴν ἐπαγγελίαν — it does not invalidate it so as to render the promise inoperative. The two verbs of v. 15 anticipate the two claims of v. 17.

Policy. Both morphemes must survive in English. Adds further stipulations to in the literal and study tiers; adds codicils to in reading tiers where the testamentary frame is wanted; adds to alone is permitted only where the compression is recorded. Excluded: modify and amend (modern legislative register, and both admit alteration rather than addition), change (deletes the τάσσω root and the addition-sense), supersede (names displacement, which is ἀθετέω’s job).

The middle voice is lexical, not reflexive, and must not be over-read: the simplex διατάσσομαι is likewise middle-only in most of its NT occurrences.

Built onbdag · bdf

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

syntax

The pair: nullify or add

The two verbs are joined by bare ἤ and governed by a single οὐδείς. The disjunction is exhaustive rather than alternative: between them they cover every way a third party could interfere with an executed instrument — taking something away from it and putting something onto it.

That exhaustiveness is why the illustration works. Paul does not need to argue that the law cannot cancel the promise (nobody claimed it did); he needs to argue that it cannot supplement it, and the pairing makes supplementation an interference of the same order as cancellation.

ἤ must be rendered, not reduced to and or nor. The disjunctive form presents two distinct acts, either of which would be a violation; and would suggest that both together are required to constitute one.

Note also that οὐδείς is masculine singular and distributive — no one, person by person — not a collective nobody ever. The claim is about what any third party may do.

Built onbdf

textual

Text

The verse is textually stable in and carries no significant variation in the editions. Three minor points are worth recording for transparency:

  • ὅμως / ὁμῶς. As noted above, the accent is an editorial decision. The earliest majuscules (𝔓46, Sinaiticus, Vaticanus, Alexandrinus) are unaccented, so the choice between the concessive particle and the adverb likewise rests on grammatical and contextual judgment, not on manuscript evidence. This must be reported as such and never cited as a textual datum.
  • Punctuation. The raised point after λέγω, which assigns κατὰ ἄνθρωπον λέγω to what follows rather than to what precedes, is likewise editorial. So is the question of whether a comma should stand after ὅμως, which bears directly on its force.
  • Word division and order. Some later minuscules normalize the fronted ἀνθρώπου to follow διαθήκην, which is the easier order; the harder, marked order of the critical text is to be retained.

None of these affects the sense materially, but all three are judgments made for the reader by the printed page, and §4 requires that they be visible.

Built onna28 · metzger-tc

note

How the early Church read it

Chrysostom (Homilies on Galatians, ad loc.) takes the point to be an a fortiori from human to divine: if no one dares tamper with a man’s covenant, far less may anyone tamper with God’s. He reads κατὰ ἄνθρωπον straightforwardly as from human affairs and sees no embarrassment in Paul’s arguing from everyday practice — it is, he says, the method of one who condescends to his hearers’ level.

Jerome (Commentary on Galatians) is alert to the testamentary sense and discusses whether the illustration concerns a testamentum or a pactum, noting that the Greek word serves both. His renders testamentum hominis confirmatum nemo spernit aut superordinat — and superordinat is a Latin coinage built on the same principle as the Greek, preserving both the super- and the ordin-. It is a better guide to ἐπιδιατάσσομαι than most English versions.

Theodoret and the Greek catenae generally read διαθήκη here covenantally rather than testamentarily, which is unsurprising given their -saturated usage; the testamentary reading has been pressed hardest in the modern period on the strength of the papyri.

The patristic tradition is uniform on one point worth recording: nobody in antiquity took κατὰ ἄνθρωπον λέγω as an apology for a weak argument. It was heard as a signal of register, not of diffidence.

Proposed — pending reviewJohn Chrysostom, Commentary on the Epistle to the Galatians, in Nicene and Post-Nicene Fathers, Series 1, vol. 13, ed. Philip Schaff, 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.

theology

Where this sits in the argument

The illustration is doing work that the next four verses will cash out:

  • κεκυρωμένην → προκεκυρωμένην ὑπὸ τοῦ θεοῦ (v. 17). The human instrument is ratified by someone unnamed; the divine one, by God, and beforehand.
  • ἀθετεῖ → οὐκ ἀκυροῖ (v. 17). Paul switches from ἀθετέω to the ἀκυρόω/κυρόω pair in v. 17, which binds the verb more tightly to κεκυρωμένην.
  • ἐπιδιατάσσεται → εἰς τὸ καταργῆσαι τὴν ἐπαγγελίαν (v. 17). The addition would operate by rendering the promise inoperative.
  • ἀνθρώπου → the whole weight of the a fortiori. A man’s instrument is inviolable; God’s the more so.

What the verse does not do is settle the status of the law. That comes later, and when it does (vv. 19-25) the answer is neither that the law is a codicil nor that it is void, but that it was προσετέθη — added — for a limited purpose and a limited term. The careful reader will notice that προσετέθη in v. 19 is itself an addition-word, which makes v. 15’s denial of ἐπιδιατάσσεται more delicate than it first appears. Paul’s resolution is temporal and functional: the law was added alongside, not written into, the covenant, and it had a terminus.

The illustration also quietly establishes the direction of the whole chapter: the promise is prior, and priority in a legal instrument is decisive. Everything in 3:16-29 turns on that.

Built onbdag

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

The formula κατὰ ἄνθρωπον
Rom 3:5 Same formula, flagging an argument from human practice.
1 Cor 9:8 Paul explicitly contrasts κατὰ ἄνθρωπον with the law's own testimony.
1 Cor 15:32 κατὰ ἄνθρωπον of a hypothetical argument.
Rom 6:19 ἀνθρώπινον λέγω — the cognate disclaimer.
Gal 1:11 The same phrase denied of the gospel; deliberate echo within the letter.
Within the argument of Galatians 3
Gal 3:17 Cashes out the illustration: a covenant προκεκυρωμένην by God, not invalidated by law 430 years later.
Gal 3:18 Inheritance by promise, not out of law — the conclusion the illustration serves.
Gal 3:19 προσετέθη — the law was added; the counterpart to ἐπιδιατάσσεται here.
Gal 2:21 ἀθετέω of God's grace; the verb's prior occurrence in the letter.
κυρόω and the ratification vocabulary
2 Cor 2:8 The only other NT occurrence of κυρόω: a formal, public act of confirmation.
Gen 23:20 LXX ἐκυρώθη of a legally confirmed transfer of property.
Heb 9:16-17 A NT writer exploiting the testamentary sense of διαθήκη explicitly.
διαθήκη as covenant
Gen 15:18 The Abrahamic covenant Paul has in view at v. 17.
Gen 17:7 The covenant as unilateral divine disposition.
Gal 4:24 δύο διαθῆκαι — Paul's later use of the same noun in the allegory.
ὅμως elsewhere
John 12:42 Concessive, clause-initial.
1 Cor 14:7 The closest parallel for the displaced position at Gal 3:15.
Where this analysis was built
Text

Nestle-Aland, Novum Testamentum Graece. 28th rev. ed. Stuttgart, 2012.

Textual apparatus

Metzger, B. M. A Textual Commentary on the Greek New Testament. 2nd ed. Stuttgart, 1994.

Lexica

Bauer, W., F. W. Danker, W. F. Arndt, F. W. Gingrich. A Greek–English Lexicon of the New Testament and Other Early Christian Literature. 3rd ed. Chicago, 2000.

Muraoka, T. A Greek–English Lexicon of the Septuagint. Leuven, 2009.

Grammars

Blass, F., A. Debrunner, R. W. Funk. A Greek Grammar of the New Testament. Chicago, 1961.

Proposed — pending editor review

F. F. Bruce, The Epistle to the Galatians: A Commentary on the Greek Text, New International Greek Testament Commentary, Grand Rapids: Eerdmans, 1982.

J. Louis Martyn, Galatians: A New Translation with Introduction and Commentary, Anchor Bible 33A, New York: Doubleday, 1997.

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

J. B. Lightfoot, Saint Paul's Epistle to the Galatians, London: Macmillan, 1865 (repr. Peabody: Hendrickson, 1999).

John Chrysostom, Commentary on the Epistle to the Galatians, in Nicene and Post-Nicene Fathers, Series 1, vol. 13, ed. Philip Schaff, 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.

James Hope Moulton and George Milligan, The Vocabulary of the Greek Testament Illustrated from the Papyri and Other Non-Literary Sources, London: Hodder & Stoughton, 1930.

Send feedback

A correction, a question, or a request. It goes to the people who make this site.