Verse 21 is the hinge of the paragraph. Verses 19-20 diagnosed a problem — quick speech, quick anger, and the flat verdict that human ὀργή does not work God’s righteousness. διό draws the command out of that diagnosis, and vv. 22-25 then unfold what receiving the word actually looks like.

The sentence has one main verb, δέξασθε (‘receive, welcome’), with everything else arranged around it: an aorist participle behind it (ἀποθέμενοι, ‘having stripped off’), a prepositional phrase of manner (ἐν πραΰτητι), and a doubly-articular object (τὸν ἔμφυτον λόγον τὸν δυνάμενον σῶσαι).

Four decisions carry the verse:

1. ἀποθέμενοι is garment-language — stripping off soiled clothing — and the image should stay visible rather than being abstracted to ‘get rid of.’
2. ῥυπαρία is a NT hapax whose root belongs to the dirty-clothes field, which is why the two images cohere.
3. ἔμφυτος is a second hapax, and its force is genuinely contested: a word already planted in them, or a word natively belonging to them. This is the verse’s real crux, since the participle’s tense forbids reading it as ‘engraft the word.’
4. σῶσαι τὰς ψυχάς is the standing σῴζω / ψυχή question. ‘Save your souls’ is the received English but imports a dualism the letter does not use; James’s ψυχή is the whole living self (cf. 5:20).

The rhetorical shape is worth noting: the negative and positive halves are not balanced. Two objects are stripped off in a subordinate participle; one object is received by the main imperative. The weight falls on the reception.

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

Inferential particle, stronger and more conclusive than οὖν, and characteristically deployed at the close of an argumentative movement rather than inside one. Per the corpus policy it is always rendered and never dropped (§10).

What it infers from is worth stating. The nearest ground is v. 20 (‘human anger does not work God’s righteousness’), but the participial pair ῥυπαρίαν καὶ περισσείαν κακίας is wider than anger alone, which suggests the inference reaches back over vv. 19-20 as a unit — and arguably over vv. 13-18, since the word to be received is the λόγος ἀληθείας by which God gave birth to them (v. 18).

ἀποθέμενοι — aorist middle participle, ‘having stripped off / laid aside.’

The literal sense is removing clothing: in Acts 7:58 the witnesses ἀπέθεντο τὰ ἱμάτια at Saul’s feet. From that the metaphorical use spreads through early Christian paraenesis — Rom 13:12; Eph 4:22, 25; Col 3:8; Heb 12:1; 1 Pet 2:1 — always with vices as the object and often, as in Eph 4:24 and Col 3:12, answered by a corresponding ‘put on.’ The regularity of the pattern across independent writers points to a shared catechetical idiom rather than to literary dependence.

The middle voice is doing real work: the subject removes the thing from himself. Rendering ‘let all filthiness be put away’ (a passive construal) loses the agency the voice asserts.

BDAG gives the transferred sense as ‘lay aside, rid oneself of.’ But per §8 the metaphor is live, not dead — James places it beside two more images in the same paragraph (the mirror, vv. 23-24; the bridle, v. 26) — so the garment stays visible: strip off, not ‘get rid of.’

On the participle’s relation to the imperative: an aorist participle preceding an aorist imperative is frequently coincident rather than antecedent (cf. the policy fixed at Gal 4:15). Here, though, the logic of the images favors sequence — clothing comes off before the word is welcomed in. Either way, the participle is not itself a second command; it describes what the receiving involves.

NT hapax. The family is well attested: ῥύπος (‘dirt, filth,’ 1 Pet 3:21, of bodily washing), ῥυπαρός (Jas 2:2, of the poor man’s shabby clothing; Rev 22:11, morally), ῥυπαίνω (Rev 22:11). LXX ῥύπος at Isa 4:4 renders the filth of the daughters of Zion, washed away by a spirit of judgment.

Two observations fix the rendering:

First, the ῥυπ- root sits squarely in the soiled-garment field, which is why it pairs naturally with ἀποθέμενοι. James himself uses ῥυπαρᾷ ἐσθῆτι of dirty clothes just eleven verses later (2:2). That internal usage is stronger evidence than any external parallel.

Second, the word is not specifically sexual. Some of the older English tradition, following the Vulgate’s immunditiam, drifts toward uncleanness with sexual coloring. Nothing in the context supports that: the surrounding vices are speech-vices and anger.

πᾶσαν with an anarthrous singular abstract noun is qualitative-intensive rather than numerical (BDF §275) — ‘every kind of filth,’ not a count.

Rendering: filthiness in the literal tier, filth in modern registers. Avoid ‘moral filth’ (supplies an adjective the Greek withholds) and ‘impurity’ (belongs to ἀκαθαρσία, already fixed separately and a different lexeme).

Proposed — pending reviewJ. B. Mayor, The Epistle of St. James: The Greek Text with Introduction, Notes and Comments, 3rd ed. (London: Macmillan, 1913)

περισσεία is the noun to περισσεύω, already fixed in this corpus as ‘overflow / abound’ with the surplus-image live. Four NT occurrences (Rom 5:17; 2 Cor 8:2; 10:15; Jas 1:21), and in the other three the surplus is positive — abundance of grace, of joy, of enlarged sphere. Here it governs κακίας, ‘malice, badness.’

That produces the difficulty. If περισσεία retains its neutral ‘surplus’ force, the phrase reads oddly: is there a permissible quantity of malice, of which only the excess is to be stripped off? Two lines of solution:

κακία in Hellenistic Greek covers both general moral badness (the opposite of ἀρετή) and the narrower ‘malice, ill-will’ (Rom 1:29; Eph 4:31; Col 3:8; 1 Pet 2:1 — in all four it stands in a list with speech-vices). Given vv. 19-20’s concern with speech and anger, the narrower ‘malice’ is contextually favored here. Not ‘evil’ as a bare abstraction, which loses the interpersonal edge.

The genitive is formally open between epexegetical (‘the surplus which is malice’) and possessive/subjective (‘the surplus that malice produces’). Keep it a bare ‘of’ phrase.

Proposed — pending reviewDale C. Allison Jr., A Critical and Exegetical Commentary on the Epistle of James, International Critical Commentary (London: Bloomsbury T&T Clark, 2013)

πραΰτης is already fixed in this corpus as gentleness (literal tier: meekness), with the note that modern English ‘meekness’ has narrowed to mere weakness and is excluded from the reading text.

The placement matters. ἐν πραΰτητι stands immediately before δέξασθε and modifies it: the receiving is to be done gently. It is the deliberate answer to ὀργή in v. 20 — anger does not work God’s righteousness, so the alternative disposition is named. James returns to the word at 3:13 (ἐν πραΰτητι σοφίας), where it again stands against ζῆλος and ἐριθεία.

So the disposition is not primarily one of humility before God in the abstract, but the specific non-combativeness with which a hearer takes in a word that may indict him. That is the connection to vv. 22-25: the man who glances in the mirror and walks away has not received gently; he has deflected.

ἐν + dative here is manner rather than instrument or sphere, though the three shade into each other. ‘With gentleness’ in smooth registers, ‘in gentleness’ in the literal tier to display the preposition.

δέχομαι is already fixed in the corpus as ‘receive / welcome,’ the hospitality-verb for receiving a guest, kept distinct from προσδέχομαι, παραλαμβάνω, and λαμβάνω.

That lexical choice is exegetically loaded here. James does not say ‘hear the word’ (that is ἀκούω, and the ἀκροατής of vv. 22-23 is precisely the failure mode) or ‘obey the word.’ He says welcome it, as one welcomes a guest into the house. The Thessalonian parallel is close: δεξάμενοι τὸν λόγον (1 Thess 1:6; cf. 2:13, where Paul uses παραλαβόντες … ἐδέξασθε in succession). Acts 8:14, 11:1, and 17:11 use it of receiving the word of God, the last of these of the Bereans.

Aorist imperative: a single decisive act of reception, not a durative practice. The contrast with the present imperatives of vv. 19 (ἔστω) and 22 (γίνεσθε) is real and should be marked in the study tier.

The apparent paradox — welcome what is already implanted in you — is discussed under ἔμφυτος below.

NT hapax, and the crux of the verse. A verbal adjective from ἐμφύω (‘implant, cause to grow in’), formally the passive of that verb: that which has been made to grow in.

Attested in Hellenistic Greek in two directions. In philosophical usage it commonly means innate, inborn — what belongs to a thing by nature rather than by acquisition (so of natural faculties, natural justice). In Wisdom 12:10 the phrase ἔμφυτος ἡ κακία αὐτῶν describes the Canaanites’ wickedness as ingrown or inbred. The other direction takes the -φυτ- root at face value: planted in, on the model of the sower.

Whatever is decided about origin, one thing the grammar forbids: the adjective cannot be read as an imperative or a future. The older English ‘the engrafted word’ (following the Vulgate’s insitum verbum) is defensible as a rendering, but it has often been taken to mean ‘let the word be engrafted,’ which the verbal adjective does not say. James commands the reception, not the implanting.

And there is no real paradox in welcoming what is already planted. The whole burden of vv. 22-25 is that something genuinely present can be looked at and walked away from. What is implanted still has to be taken in.

Rendering: implanted in reading tiers, which keeps the -φυτ- root and the passive shape while leaving both (a) and (c) available; ingrown or inborn would adopt (b). ‘Engrafted’ is retained in the liturgical register only.

Proposed — pending reviewPeter H. Davids, The Epistle of James: A Commentary on the Greek Text, New International Greek Testament Commentary (Grand Rapids: Eerdmans, 1982)

The second article converts the participle into an attributive, restrictive modifier: τὸν λόγον τὸν δυνάμενον — ‘the word, the one that is able.’ Per the corpus policy on article–noun–article–participle, this is rendered with an English relative clause so the restrictive force stays visible, not with a prenominal adjective.

Two things are asserted by the construction. First, the ability belongs to the word, not to the hearer; the participle is not conditional on reception. Second, δύναμαι is capacity, not guarantee — the word has power to save, which is why the paragraph then turns immediately to what hearers do with it. Rendering ‘which will save you’ overstates the participle and pre-empts vv. 22-25.

σῶσαι is a complexive aorist infinitive: the whole act of rescue viewed as one. Present tense would have suggested a process.

The corpus already fixes ψυχή as life where bodily existence is at risk (Phil 2:30), with the note that ‘soul’ imports a dualism foreign to the context and that ‘soul’ is reserved for contexts where the inner self is genuinely distinguished.

James’s own usage settles the case internally. At 5:20 the one who turns a sinner back σώσει ψυχὴν αὐτοῦ ἐκ θανάτου — saves his life from death, in a letter whose eschatology is bodily and whose ethics are relentlessly concrete (2:15-16; 5:1-6). Behind it stands the Hebrew נֶפֶשׁ, the whole living self, which the LXX renders ψυχή some 750 times, including in contexts where ‘soul’ in the English sense would be nonsense (Gen 2:7; Lev 17:11).

Rendering: lives in smooth and modern tiers, souls in the literal and liturgical tiers with the caution attached. Where the study tier needs both, ‘your whole living selves.’

The verse is textually stable in its substance; NA28 prints it without a bracketed word and the UBS committee assigns it no rated variant.

The minor points worth recording:

• The spelling πραΰτητι (with diaeresis) against πραότητι is orthographic, an Atticizing alternative form with no semantic weight; both are found across the tradition.
• Some witnesses read the compound preposition or add a connective before δέξασθε, smoothing the participle-plus-imperative sequence. These are the ordinary direction of scribal easing and carry no weight.
• A few late witnesses have felt the difficulty of ἔμφυτον and the surrounding tradition shows some pressure toward interpretive glossing, but the reading itself is not in doubt.

What is genuinely undecided here is punctuation and paragraphing, which is editorial judgment rather than manuscript evidence: whether v. 21 closes the vv. 19-21 unit or opens the vv. 21-25 unit. The διό looks backward and the λόγος looks forward, which is presumably why editions differ.

Vulgate: Propter quod abicientes omnem immunditiam et abundantiam malitiae, in mansuetudine suscipite insitum verbum, quod potest salvare animas vestras.

Three observations. Immunditiam (‘uncleanness’) for ῥυπαρίαν loses the soiled-garment root and has pulled the English tradition toward a purity-vocabulary the Greek does not use — note that Jerome does not connect it to sordidus, which he uses at Jas 2:2 for ῥυπαρᾷ. Abundantiam for περισσείαν keeps the surplus sense neutral, leaving the crux open. Insitum (‘grafted in, planted in’) for ἔμφυτον is a good rendering of the passive verbal adjective and is the source of the English ‘engrafted’; it adopts the implanted reading over the innate one, but it does so without violence to the Greek.

Peshitta renders the participle with a form in the ‘put off / cast away’ domain and takes ἔμφυτον in the planted direction as well. The versional tradition is thus broadly agreed against the innate reading — which is evidence of how the verse was read early, though not decisive lexical evidence (§11).

The Greek commentators read the verse within the paragraph rather than as an isolated slogan.

On πραΰτης, the consistent patristic instinct is to tie it directly to the ὀργή of v. 20: gentleness is named because anger is the thing that prevents the word from being taken in. The disposition is treated as a condition of hearing, not a general virtue.

On ἔμφυτος, the ancient interpreters overwhelmingly took the planted rather than the innate sense, connecting it to the sowing imagery of the Gospels and to the new birth of v. 18. Where they press it, they read the reception as the cultivation of a seed already sown — which is close to construal (c) above.

On σῶσαι τὰς ψυχάς, the patristic reading is eschatological but not dualist; the language of rescue is applied to the person, in continuity with the LXX נֶפֶשׁ usage rather than with a Platonic anthropology. The narrowing of ‘soul’ to an immaterial part is a later development in the receiving languages, not in the exegesis of the verse.

These are recorded as reception (§9, §11): they show how the verse was heard, and they weigh against the innate-λόγος reading, but they are not lexical evidence.

Proposed — pending reviewBiblia Patristica / Catenae Graecae in epistulas catholicas, ed. K. Staab and J. A. Cramer (Oxford, 1840; repr.)

James 1:19-27 is paraenesis, and its texture is imperatival and image-dense. Within four verses of 1:21 there are three distinct pictures: clothing stripped off (v. 21), a mirror glanced into and forgotten (vv. 23-24), a tongue bridled (v. 26). Per §8 that density is itself an argument for keeping each metaphor live in English rather than resolving any of them into its abstract sense.

The sentence’s asymmetry is deliberate and should survive translation. Two negative objects, compressed into a subordinate participle; one positive object, carried by the main imperative and then expanded by a relative clause. The English should not balance them out by promoting the participle to a coordinate imperative (‘strip off … and receive’), which the aorist participle does not warrant and which would make the verse a two-part programme rather than a single command with a precondition.

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