This is the verse that made Luther call James an “epistle of straw,” and it is the single sentence in the New Testament that most looks like a flat contradiction of Paul (Rom 3:28: δικαιοῦσθαι πίστει ἄνθρωπον χωρὶς ἔργων νόμου). Both sentences contain δικαιοῦται/δικαιοῦσθαι, ἄνθρωπος, πίστις, and ἔργα. The verbal overlap is real and must not be minimized.

What the Greek itself gives us, before any harmonizing:

• ὁρᾶτε — the form can be indicative or imperative. NA28 punctuates it as an indicative drawing a conclusion from the Abraham exposition of vv. 21-23.
• ἐξ ἔργων is anarthrous and unqualified. James never writes ἔργα νόμου, the phrase that carries Paul’s argument. His works are the ones his own letter has named: clothing the naked, feeding the hungry (2:15-16), Abraham’s offering of Isaac, Rahab’s hiding of the spies.
• μόνον stands last and modifies πίστεως, not the verb. James denies that a person is justified out of faith-alone; he does not deny that a person is justified out of faith. This is a grammatical point, not a theological escape hatch, and it is what makes the sentence a denial of the position ventriloquized at 2:14 (“a man says he has faith”) rather than of Pauline soteriology.
• The verb is present passive with the agent unexpressed. God does the justifying; James is describing what characteristically happens, not narrating an event.

The genuinely open questions — and they are open — are (1) what δικαιόω means here (declare righteous, as in Paul’s forensic usage, or vindicate/show to be righteous), and (2) whether James is engaging Paul, a distortion of Paul, or neither. The apparatus below sets out the evidence for each without resolving them.

Proposed — pending reviewRichard Bauckham, James: Wisdom of James, Disciple of Jesus the Sage, New Testament Readings, London: Routledge, 1999.

The present active form is identical for the indicative (“you see”) and the imperative (“see!”). Greek manuscripts are unpunctuated at this depth, so the choice is a modern editorial judgment, not manuscript evidence — the same situation the corpus has already recorded at Ἴστε (Jas 1:19), where the ambiguity is exactly parallel.

Context favors the indicative here more strongly than at 1:19. The verse begins a conclusion, and James has just used βλέπεις (2:22), an unambiguous indicative, in the same argumentative slot. The parallel βλέπεις … ὁρᾶτε (singular → plural) suggests one movement: first the interlocutor is shown, then the whole assembly is told what they now see.

The shift from second singular (βλέπεις, 2:22; σὺ πιστεύεις, 2:19) to second plural here widens the address from the imagined objector back to ἀδελφοί μου. English cannot mark it; note it.

The following ὅτι is a plain complementizer introducing the content perceived, not causal. Rendering it “because” would make v. 24 a ground for v. 23 rather than a conclusion from it.

ἔργα stands anarthrous, unqualified, and — this is the load-bearing observation — without νόμου. Paul’s polemical target is ἔργα νόμου (Rom 3:20, 28; Gal 2:16 three times; 3:2, 5, 10), a phrase James never uses anywhere. The corpus glossary already fixes ἔργον as “work” across the missionary-labor (Phil 1:6), individual-conduct (Gal 6:4), and law-works senses precisely so that the reader can see when a writer does and does not qualify the noun.

What James’s ἔργα actually are is established by his own chapter, not by Paul’s: giving τὰ ἐπιτήδεια τοῦ σώματος to a brother or sister who is naked and lacking daily food (2:15-16), Abraham’s offering of Isaac (2:21), Rahab’s reception and dispatch of the messengers (2:25). These are acts of mercy and obedient trust, not badges of covenant membership or grounds of merit.

ἐκ marks source, kept distinct in this corpus from ἀπό (origin), ὑπό (personal agency), and διά (means). Note that James repeats the identical preposition in both members of the sentence: ἐξ ἔργων … ἐκ πίστεως. He is not contrasting two kinds of relation to justification but placing two nouns in the same relation and denying that the second stands alone. English must repeat “out of” at both points or the symmetry vanishes.

Proposed — pending reviewLuke Timothy Johnson, The Letter of James, Anchor Bible 37A, New York: Doubleday, 1995.

The corpus glossary fixes δικαιόω as “be justified” with the forensic-declarative sense governing in Paul, and with the present passive rendered so as to leave aspect open (Gal 5:4). That decision holds here, but the sense of the verb at Jas 2:21-25 is one of the two genuinely contested questions in the verse.

Lexically, the verb belongs to the courtroom: LXX renders צָדַק hiphil with it in the sense of giving a favorable verdict (Exod 23:7; Deut 25:1; Isa 5:23; 1 Kgdms 8:32), and the vindication sense — being shown or acknowledged to be in the right — is securely attested in the same range (Ps 50:6 LXX, cited at Rom 3:4; Matt 11:19 par. Luke 7:35, ἐδικαιώθη ἡ σοφία; Luke 7:29; 1 Tim 3:16, of Christ ἐδικαιώθη ἐν πνεύματι). Both are ordinary first-century Greek, and both are courtroom senses; the difference is whether the verdict constitutes the standing or publishes it.

The present tense is gnomic — what characteristically happens — against the aorists at 2:21 and 2:25, which report the two scriptural cases. James moves from two completed instances to a standing principle. Do not render with an English progressive, which would convert a maxim into a report on a process underway.

The passive leaves the agent unexpressed. God justifies; James does not say so, and the rendering must not supply it.

Anarthrous generic singular. Per the standing corpus policy on generic ἄνθρωπος (Gal 6:7), the masculine grammatical gender is the Greek default for a referentially inclusive statement; render “a person” in reading tiers, retain “a man” in the wooden tier, and note the generic force rather than narrowing or neutralizing silently.

The word is worth flagging because Rom 3:28 has the same anarthrous ἄνθρωπος in the same construction (λογιζόμεθα γὰρ δικαιοῦσθαι πίστει ἄνθρωπον χωρὶς ἔργων νόμου). Whatever one concludes about the relationship between the two sentences, the verbal correspondence extends to this word.

James alternates ἄνθρωπος with ἀνήρ across his letter (1:7-8, 12, 19-20, 23; 2:2, 20, 24; 3:2, 8-9). Here he uses ἄνθρωπος four verses after ὦ ἄνθρωπε κενέ (2:20). Record the alternation; do not level it.

This is the grammatically decisive clause of the verse, and it is regularly misread.

μόνον is an adverbial neuter accusative. It stands clause-final, immediately after πίστεως, and it qualifies the noun-phrase, not the verb. James is not saying “a person is not justified only by faith” in the sense of “faith is not the only thing that justifies alongside works.” He is saying that what does not justify is faith-taken-by-itself — the πίστις χωρὶς ἔργων of v. 20, the πίστις νεκρὰ καθʼ ἑαυτήν of v. 17, the credal monotheism the δαιμόνια also hold at v. 19.

This matters because it establishes what the sentence’s negated term actually is. The corpus already fixes the parallel policy at οὐ negating a prepositional phrase, not the verb (Gal 6:4) and οὐ negating an adverb (Gal 4:17): where the negative attaches to something other than the finite verb, English must keep it attached there or the claim changes.

Note also that the negation is coordinated by καί, not by ἀλλά. James does not say “not out of faith but out of works” — which would be a corrective — but “out of works, and not out of faith alone,” which is additive. The corpus has already fixed the οὐκ … δὲ μόνον … ἀλλὰ καί policy at 1 Pet 4:2 for the additive shape. Rendering with a corrective “but” inverts the logic and manufactures the very contradiction with Paul that the syntax withholds.

One further observation: this is the only place in the New Testament where πίστις and μόνος stand together. The formula sola fide is a later Latin theological slogan, not a phrase James is quoting; but the fact that its Greek elements meet here and only here is why the verse became the flashpoint it did.

The conclusion at v. 24 depends entirely on the exposition of vv. 21-23, and the structure there constrains what v. 24 can mean.

James quotes Gen 15:6 (ἐπίστευσεν δὲ Ἀβραὰμ τῷ θεῷ, καὶ ἐλογίσθη αὐτῷ εἰς δικαιοσύνην) — the same verse Paul makes the hinge of Rom 4 and Gal 3:6. But James says it was ἐπληρώθη (“filled up,” 2:23) by the Aqedah of Gen 22, which the biblical narrative places roughly thirty years later. He also says the faith συνήργει (“was working together with”) the works and ἐτελειώθη (“was brought to completion”) out of them.

Three things follow. First, faith is prior in James’s own telling: the reckoning at Gen 15:6 precedes the offering at Gen 22 and is fulfilled by it. Second, faith is not a component alongside works but the thing that works complete — συνεργέω and τελειόω both presuppose faith as the subject that the works finish. Third, τελειόω belongs to the τελ- family the corpus has already fixed (“bring to completion,” not “make flawless”), and it is the same root as the τέλειον ἔργον of 1:4. James is running one argument about things reaching their intended end.

Abraham is also called φίλος θεοῦ (2:23), a title with LXX and Second Temple pedigree (Isa 41:8; 2 Chr 20:7; Jdt 8:22; Philo). The category is covenantal relationship, not merit.

Rahab (2:25) is the harder case for a merit reading, not the easier one: she is ἡ πόρνη, and her ἔργον is an act of hospitality and risk toward God’s messengers. If James were arguing that moral achievement contributes to acquittal, Rahab is a perverse choice.

Proposed — pending reviewMartin Dibelius, James: A Commentary on the Epistle of James, rev. Heinrich Greeven, Hermeneia, Philadelphia: Fortress, 1976.

The overlap is too dense to be coincidence at the level of vocabulary; whether it is engagement is another matter.

The differences in the table are as real as the similarities. Paul excludes ἔργα νόμου; James affirms ἔργα without qualification and illustrates them with mercy to the destitute. Paul’s πίστις is the self-abandoning reliance of Rom 4:18-21; the πίστις James rejects is explicitly the assent that demons share (2:19) and that produces no clothing for the naked (2:15-16). Whether these are the same word doing different jobs, or the same job being contested, is the question.

What can be said without prejudging: James and Paul deny different propositions. Paul denies that works of law are the ground of God’s acquitting verdict. James denies that a faith which produces nothing is a faith that saves. Neither writer states the other’s proposition, and neither denies the other’s affirmation. Paul’s own summary — πίστις διʼ ἀγάπης ἐνεργουμένη (Gal 5:6) — and his insistence that the doers of the law will be justified (Rom 2:13) show that the Pauline corpus itself contains both halves.

Per §6, this project’s stated tradition holds that justification is by grace through faith and that the works James requires are the necessary fruit rather than the ground of that verdict. That stance is declared here and confined to this note; it does not shade the rendering, which keeps ἐξ ἔργων as a bare source-preposition exactly as James wrote it, and does not soften δικαιοῦται.

Proposed — pending reviewRalph P. Martin, James, Word Biblical Commentary 48, Waco: Word, 1988.

The early Church did not experience Jas 2:24 as a crisis; that is itself data about how the verse was heard before the sixteenth century.

1 Clement (late first century) sets the two together without embarrassment: Abraham was blessed “because he wrought righteousness and truth through faith” (31.2), and Rahab was saved “for her faith and hospitality” (12.1) — the same pairing and the same two exemplars as James 2, which suggests a shared catechetical tradition rather than a controversy.

Origen and the Alexandrian tradition read faith and works as inseparable in the one Christian life, and Origen speaks of a faith “without works” as no faith at all.

Augustine, who had the sharpest reason to feel a tension (he was fighting Pelagius with Paul), addressed this verse directly and repeatedly. His solution was that James is speaking of works that follow justification, and Paul of works that would precede it: Abraham’s Gen 22 obedience displayed the righteousness that Gen 15:6 had already reckoned to him. He also read Jas 2:19 as the key — the faith James rejects is the demons’ assent.

Bede, whose commentary on the Catholic Epistles carried the Western reading through the Middle Ages, follows Augustine closely and stresses the μόνον: James condemns faith alone, not faith.

The patristic consensus, in short, took μόνον seriously as qualifying πίστεως and located the difference from Paul in the kind of faith and the timing of the works. Whether that is exegesis or harmonization is a fair question; that it is the reading of the ancient Church, East and West, is not in doubt.

Luther’s judgment (the 1522 Preface to the New Testament, softened in later editions) that James “mangles the Scriptures and thereby opposes Paul” is a reception-historical fact of enormous consequence and no lexical weight (§11). It is named here, not used.

Proposed — pending reviewBede the Venerable, Commentary on the Seven Catholic Epistles, trans. David Hurst, Cistercian Studies 82, Kalamazoo: Cistercian Publications, 1985. · Augustine, De diversis quaestionibus LXXXIII and De fide et operibus, in Corpus Christianorum Series Latina 44A and 41.

The text is stable. NA28 prints ὁρᾶτε ὅτι ἐξ ἔργων δικαιοῦται ἄνθρωπος καὶ οὐκ ἐκ πίστεως μόνον with no variant flagged in the apparatus of UBS5, and the verse is not among the passages Metzger treats for James 2. 𝔓74, ℵ, A, B, C, and the Byzantine tradition agree.

Minor witnesses show the expected pressure points: some cursives add τοίνυν (“therefore”) after ὁρᾶτε, an inferential connective supplying explicitly what the asyndeton leaves implicit — smoothing in the direction the corpus policy on asyndeton warns against. A few read the article before ἄνθρωπος, converting the generic to a definite. Neither has any claim.

Vulgate: Videtis quoniam ex operibus iustificatur homo et non ex fide tantum. Word-for-word, including the repeated ex and the final tantum answering μόνον. Jerome resolves ὁρᾶτε as an indicative and preserves the adverb in its qualifying position; the Latin does not create the ambiguity that later English versions introduced by moving “only” toward the verb.

Peshitta: renders with the same structure, and the Syriac tradition — which for a long period did not include James in its canon at all — shows no sign of the verse having been felt as problematic when it was received.

One text-critical observation worth recording: the absence of variation is itself informative. A sentence that scribes found doctrinally intolerable would be expected to attract smoothing, and this one attracted almost none.

Proposed — pending reviewBiblia Sacra Iuxta Vulgatam Versionem, ed. R. Weber and R. Gryson, 5th ed., Stuttgart: Deutsche Bibelgesellschaft, 2007.

1. ὁρᾶτε rendered as an indicative (“you see”), following NA28’s punctuation and the parallel with βλέπεις at 2:22, with the imperative recorded as an equally available reading of the form. The punctuation is editorial, not manuscript, judgment.

2. ἐξ … ἐκ rendered with the same English preposition (“out of”) at both occurrences. Varying to “by works … by faith” would obscure that James placed both nouns in the identical relation to the verb, which is the syntactic ground of the μόνον argument.

3. ἔργων left bare. No qualifier is supplied — not “good works,” not “works of the law,” not “deeds of obedience.” Each would settle a contested question in the reading text.

4. δικαιοῦται rendered “is justified,” passive with the agent unexpressed and the declarative/demonstrative question left open. “Is shown to be righteous” adopts one reading; “is declared righteous” adopts another and imports Pauline forensic specificity.

5. μόνον kept in final position attached to “faith.” “Not by faith only” and “not out of faith alone” are both acceptable; “not only by faith” is excluded, since English word order there floats the adverb toward the verb and changes what is denied.

6. καί rendered “and,” not “but.” The negation is additive, not corrective.

7. The em dash in the study tier marks the coordinate negation without asserting a contrast the Greek does not supply.