James has just asserted that δέησις δικαίου — the petition of a righteous person — has great effect (5:16). Verse 17 supplies the proof text, and the whole rhetorical weight falls on the first four words. The prophet who shut heaven is introduced not as a wonder-worker but as ἄνθρωπος … ὁμοιοπαθὴς ἡμῖν, a human being subject to the same experiences as the writer and his readers. Only then comes the prayer and its effect.

Three features carry the verse:

1. The levelling predicate. ὁμοιοπαθής is a rare word (NT only here and Acts 14:15, where Paul and Barnabas use it to refuse divine honors). It denies that Elijah stood in a different category. Everything the verse concedes about his power is preceded by this denial of his exceptionality.
2. The cognate dative. προσευχῇ προσηύξατο is a Semitic construction (Hebrew infinitive absolute rendered woodenly in the LXX), and its force is genuinely contested: emphasis (‘prayed earnestly’) or mere idiom (‘prayed a prayer’). The choice affects whether James is commending fervency or simply narrating.
3. The three and a half years. The Kings narrative (1 Kgs 17-18) does not give a figure; James’s ‘three years and six months’ matches Luke 4:25 and is at home in the apocalyptic half-week (Dan 7:25; 12:7; Rev 11:2-3; 12:6, 14). This is a datum about a shared Jewish tradition of the Elijah story, not a contradiction of 1 Kgs 18:1.

The verse is one of two (17-18) forming a single argument, and no full stop should be read at the end of v. 17 — the semicolon in the printed text is a raised dot marking a pause, and the καί of v. 18 continues the sentence.

Built onna28 · bdf

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

Compound of ὅμοιος (‘like’) + the παθ- root of πάσχω / πάθος: ‘experiencing the same things.’ Two NT occurrences, and they pull in the same direction:

• Acts 14:15 — Paul and Barnabas to the crowd at Lystra: καὶ ἡμεῖς ὁμοιοπαθεῖς ἐσμεν ὑμῖν ἄνθρωποι, ‘we too are human beings of the same nature as you.’ The word there explicitly denies divine status. Note that Acts pairs it with ἄνθρωποι, exactly as James does.
• Wis 7:3 — Solomon says of his birth that he fell on the earth ὁμοιοπαθής, ‘of like nature’ with all other humans, and drew the common air.

The semantic core is therefore shared human constitution, including susceptibility to suffering and weakness, not a claim about emotional temperament in particular. This governs the rendering.

Excluded from reading tiers: ‘of like passions’ (the older English, following the Vulgate’s similis nobis passibilis) — modern English ‘passions’ names strong emotion or sexual appetite, and it produces the misreading that James is pointing at Elijah’s depressive episode under the broom tree (1 Kgs 19:4). That reading is attractive and old, but it rests on the English word, not the Greek: the παθ- root here denotes what one undergoes, not what one feels. Retain ‘of like passions’ in declared liturgical register only.

Also excluded: ‘just like us’ (register collapse, and drops ἄνθρωπος), ‘just an ordinary man’ (imports a diminishing adverb the Greek does not have — James concedes Elijah’s power in the next clause), ‘with a nature like ours’ is acceptable in smooth registers.

The dative ἡμῖν is the comparison-term required by the ὁμοιο- compound, and the first person is deliberate: James is not exempting himself.

Built onbdag · muraoka-lxx

Proposed — pending reviewHenry George Liddell, Robert Scott, and Henry Stuart Jones, A Greek-English Lexicon, 9th ed. with revised supplement (Oxford: Clarendon Press, 1996).

Word order is doing work. Greek could have written ὁμοιοπαθὴς ἡμῖν ἄνθρωπος ἦν Ἠλίας, or placed the copula first. As written:

• Ἠλίας stands first, fronted and topicalized — the great name.
• ἄνθρωπος follows immediately, anarthrous and qualitative: not ‘the man Elijah’ but ‘a human being.’
• ἦν is expressed, which in a writer who elides the copula freely (Jas 1:8; 3:16; 4:12) marks the predication as an assertion in its own right.
• ὁμοιοπαθὴς ἡμῖν lands last, the heavy compound plus the first-person dative closing the clause.

The sequence is name → species → shared condition. English should preserve the order and let the sentence land on ‘as we are.’ Reversing it (‘Elijah was just like us — a human being’) inverts the movement.

Whether ἄνθρωπος is the subject with ὁμοιοπαθής attributive (‘a man of like nature was Elijah’) or, more naturally, both are predicate to Ἠλίας (‘Elijah was a human being, of like nature with us’) makes no material difference and need not be resolved; the second is preferred and is what the rendering reflects.

Built onbdf

A cognate dative reinforcing its own verb — the Greek reflex of the Hebrew infinitive absolute construction (e.g. מוֹת יוּמָת), which the LXX regularly renders with a dative or participle of the same root. Standing parallels in biblical Greek:

• ἐπιθυμίᾳ ἐπεθύμησα — Luke 22:15, ‘with desire I have desired’
• χαρᾷ χαίρει — John 3:29, ‘with joy he rejoices’
• ἀπειλῇ ἀπειλησώμεθα and similar in Acts 4:17 (v.l.), 5:28, 23:14
• θανάτῳ ἀποθανεῖσθε — Gen 2:17 LXX

The corpus already treats this as a figura etymologica to be preserved on one English root (see the standing entry fixed at Gal 5:1). The question here is not whether to preserve it but how much force to claim for it.

Policy. Render ‘prayed a prayer’ in the reading text, which reproduces the doubling without quantifying it, and keep ‘with prayer he prayed’ in the literal tier so the construction is visible. ‘Prayed earnestly’ and ‘prayed fervently’ adopt the first reading and are confined to the study tier with the alternative flagged. Never compress to a bare ‘he prayed,’ which deletes the construction altogether.

The verb is προσεύχομαι, per the standing glossary entry the general verb for prayer to God, to be kept distinct from δέησις/δέομαι (‘petition’) — which is significant here, since δέησις stands one verse earlier at 5:16 and εὔχομαι/εὐχή at 5:14-16. James rings the changes across four prayer-words in five verses, and English must keep them distinguishable.

Built onbdf · muraoka-lxx

Proposed — pending reviewPeter H. Davids, The Epistle of James: A Commentary on the Greek Text, New International Greek Testament Commentary (Grand Rapids: Eerdmans, 1982). · J. H. Moulton and Nigel Turner, A Grammar of New Testament Greek, vol. 4: Style (Edinburgh: T. & T. Clark, 1976).

Genitive articular infinitive with the obligatory negative μή. Per the standing glossary entry, Koine has broadened this construction well past classical purpose to cover result, content, and epexegesis, and the categories often overlap in a single instance.

Here at least three classifications are defensible:

• Purpose / aim — he prayed in order that it should not rain. Most natural after a verb of petition.
• Content — he prayed that it should not rain; the infinitive gives what was asked. Effectively indistinguishable in force from the first.
• Result — he prayed, so as for it not to rain. This is grammatically available and would make the clause link directly to the οὐκ ἔβρεξεν that follows.

The first two are not materially different in a prayer context and need not be separated. Per the standing policy, render with an English marker loose enough to leave more than one reading standing: ‘so as for it not to rain’ in the study tier, ‘that it would not rain’ in the reading text. Avoid ‘in order that,’ which closes it to purpose alone.

The μή is grammatically required with the infinitive and no emphasis may be read out of its presence.

Built onbdf

Ordinary verb for the falling of rain, used impersonally in the third singular with no expressed subject (Matt 5:45 ὁ πατὴρ … βρέχει is transitive-causative; the impersonal is Luke 17:29; Rev 11:6; Jas 5:17-18).

Two constraints on the rendering:

1. The same English root at both occurrences. The infinitive βρέξαι and the indicative ἔβρεξεν are the same lexeme, and the answering of prayer by outcome depends on the reader seeing one word twice. ‘He prayed for drought, and none fell’ destroys it.
2. The subject stays unexpressed. Greek gives none. English idiom supplies ‘it,’ which is a dummy subject and asserts nothing; but supplying ‘God did not send rain’ or ‘heaven gave no rain’ inserts an agent James withholds until v. 18, where ὁ οὐρανὸς ὑετὸν ἔδωκεν supplies one. The reticence is deliberate: the verse says the prophet prayed and the rain stopped, without spelling out the mechanism.

Note that at v. 18 James switches to a different noun (ὑετόν, ‘rain’ as a substance given) with a different subject (ὁ οὐρανός). The lexical shift between the two verses should be recorded, not levelled.

Built onbdag

ἐπί + genitive, locative-extensive: over the surface of. The question is the referent of γῆ.

English must choose, since ‘land’ and ‘earth’ are different words. Policy: ‘the land’ in the reading text, since the narrative referent is bounded, with ‘the earth’ recorded as the flagged alternative and retained in the literal and liturgical tiers (where the older English tradition has fixed it). Note that v. 18’s ἡ γῆ ἐβλάστησεν τὸν καρπὸν αὐτῆς is the same noun and should be rendered with the same English word if the parallel is to be visible — which is an argument for ‘the earth,’ or for ‘the land’ at both points.

Built onbdag · muraoka-lxx

Accusative of extent of time (twice), and per the standing ἐνιαυτός entry the noun denotes the year as a completed annual cycle, kept distinct from ἔτος.

The figure is not in 1 Kings. 1 Kgs 18:1 dates the word of the LORD to Elijah ἐν τῷ ἐνιαυτῷ τῷ τρίτῳ, ‘in the third year,’ which is not obviously three and a half. James’s ‘three years and six months’ agrees exactly with Luke 4:25, where Jesus cites the same drought.

Three observations, in order of weight:

1. A shared tradition. The agreement of James and Luke against the plain text of Kings is best explained by a common Jewish tradition about the length of Elijah’s drought, current in the first century AD and independent of either writer. The apparatus should say so rather than manufacture a harmonization.
2. The apocalyptic half-week. Three and a half years, forty-two months, or 1,260 days is the standard biblical measure of a bounded period of distress: Dan 7:25; 12:7 (a time, times, and half a time), Rev 11:2-3; 12:6, 14; 13:5. Notably Rev 11:6 describes the two witnesses in terms drawn straight from Elijah — ἐξουσίαν ἔχουσιν κλεῖσαι τὸν οὐρανόν, ἵνα μὴ ὑετὸς βρέχῃ — using βρέχω, the same verb as here. The tradition and the symbolic number are plainly connected; which direction the influence ran is not recoverable.
3. Harmonization is possible but not required. ‘In the third year’ (1 Kgs 18:1) could be counted from Elijah’s arrival at Zarephath rather than from the drought’s onset, leaving room for a longer total; or the three years may be counted as three rainless winters spanning three and a half calendar years. These are legitimate reckonings, but the apparatus should not present either as the solution: James is citing the story as his readers knew it.

No hedging in the rendering. The numbers are rendered as written.

Built onmetzger-tc

Indeclinable-stem Greek form of אֵלִיָּהוּ, declined in the NT as a first-declension masculine (Ἠλίας, Ἠλίου, Ἠλίᾳ, Ἠλίαν). Retain the English proper noun ‘Elijah’; ‘Elias’ (the Vulgate’s Helias and the older English tradition) is confined to declared liturgical register.

The choice of exemplar is not incidental. In first-century Judaism Elijah was the paradigm of the prophet who commands the elements and who did not die (2 Kgs 2:11), the figure expected to return (Mal 3:23-24 / 4:5-6; Sir 48:1-12). Sirach’s encomium in particular piles up the marvels: ὡς ἐδοξάσθης, Ηλια, ἐν θαυμασίοις σου. James takes exactly this figure and levels him with ἄνθρωπος ὁμοιοπαθὴς ἡμῖν.

That is the argumentative move: the strongest available counter-example to ‘prayer like mine could never do that’ is disarmed by classifying the prophet with the reader. It is the same move Paul and Barnabas make with the same word at Acts 14:15, in the opposite direction — refusing an elevation being offered them.

Built onbdag

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

The verse is textually quiet in the critical editions; NA28 prints it without a bracketed reading and Metzger’s commentary does not treat it.

Two points worth recording:

• προσευχῇ προσηύξατο stands in the whole tradition. Since a cognate dative is exactly the kind of Semitism scribes elsewhere smooth away, its unanimous retention is evidence that it was felt as a fixed biblical idiom rather than as an awkwardness.
• Some minuscules read ἔτη for ἐνιαυτούς, a substitution of the commoner word for the year. This is a levelling toward ordinary usage and is secondary; the harder ἐνιαυτούς is to be retained, and English cannot mark the difference, so it is recorded here.

The punctuation at the end of the verse is a raised dot (·) in NA28 — a pause, not a full stop. Since the earliest majuscules are unpunctuated, this is editorial judgment. It matters: vv. 17-18 are one argument, and the καί that opens v. 18 continues the narration rather than beginning a new sentence. English should not close the sentence hard at the verse boundary.

Built onna28 · metzger-tc

The Fathers read the verse the way James wrote it — as a levelling, not as an exaltation of Elijah’s prayer technique. Two lines of reception are worth distinguishing:

• The consolatory line. Elijah is adduced precisely to remove the excuse of the ordinary believer. The argument runs: if a man of our own constitution could do this, the objection ‘I am not a prophet’ has no purchase. This is the plain sense of ὁμοιοπαθὴς ἡμῖν and it governs the paragraph from 5:16 onward.
• The moralizing line. Later readers, working from the Latin oratione oravit and from the ‘like passions’ rendering, drew from the verse a lesson about the intensity required in prayer, sometimes coupled with reference to Elijah’s despondency at 1 Kgs 19. This is a genuine strand of reception, and it may be named as such (§11), but it must not be used as evidence for the rendering: the ‘passions’ reading depends on the Latin and on English semantic drift, and the ‘fervency’ reading depends on one of two live classifications of the cognate dative.

Built onsilva-becnt