Verse 28 is the third and last participial clause hanging off στήκετε in v. 27 (standing … contending together … not being startled), so it belongs to the content of what Paul hopes to hear about the Philippians, not to a new sentence. The clause has three moving parts.

First, a striking piece of vocabulary: πτυρόμενοι is a NT hapax drawn from the language of animals shying — a horse bolting at a sudden noise. Paul is not asking for stoic calm under analysis; he is asking them not to spook.

Second, an intrusive relative clause (ἥτις …) whose antecedent is disputed, and whose grammar is untidy in exactly the way spoken argument is untidy. The one thing that is clear is the forensic word ἔνδειξις: the Philippians’ unspooked steadiness is evidence, and the same single piece of evidence points two opposite directions — destruction for the opponents, salvation for the church.

Third, a flat theological attribution: καὶ τοῦτο ἀπὸ θεοῦ — and this from God. Paul then unpacks it with the ὅτι of v. 29 (the grant of both faith and suffering). The parenthesis opened here does not close until the end of v. 29, which is why NA28 punctuates with an open bracket and no closing one in this verse.

Genre is epistolary argument, so the connectives (μή, δέ, καί) are load-bearing and are all rendered.

Built onobrien-nigtc · fee-nicnt · silva-becnt · na28

πτυρόμενοι occurs nowhere else in the Greek NT and not at all in the LXX. The verb belongs to the vocabulary of animal fright: it is used of horses shying or bolting at something startling, and by extension of humans being thrown into sudden panic. BDAG glosses ‘be frightened, terrified,’ noting the equestrian background; the related noun πτύρσις and adjective πτυρτικός sit in the same domain in Hellenistic technical writing on horsemanship.

Two features matter for rendering.

1. It is a startle-word, not a fear-word. Paul had φοβέομαι and δειλία available and did not use them. The image is a reflex — the flinch before the reasoning — which is why a rendering like ‘not being intimidated’ (an assessment of the opponents’ power) is slightly off-target, and ‘not being terrified’ overshoots the suddenness.
2. It is present passive. The durative aspect gives ‘not repeatedly being startled,’ i.e. a settled unflappability under sustained pressure, matched by the present στήκετε and συναθλοῦντες of v. 27.

Per §8, the image is live rather than dead: a first-century reader hearing πτύρομαι heard the animal bolting. The literal tier therefore keeps a startle-verb (‘being startled’) and the modern tier uses ‘flinching,’ which preserves the involuntary reflex in idiomatic English. ‘Dismayed’ is retained in the liturgical register only, where the archaism is already accepted.

The participle is anarthrous, nominative plural, and coordinate with συναθλοῦντες (v. 27), both dependent on στήκετε. Two consequences:

• μή is the expected negative with a participle in Koine and carries no special semantic freight here (per the standing glossary entry on μή + participle).
• Because the whole chain sits inside the ὅτι-clause reporting what Paul wants to hear, the participle is descriptive rather than a free-standing imperative. Rendering it as a command (‘do not be frightened!’) is not wrong in effect — Paul plainly wants the behavior — but it severs the syntax from v. 27 and should not be done in the main text.

ἐν μηδενί is neuter dative, adverbial: ‘in no respect, in nothing.’ The piling of μή with μηδενί is ordinary Greek reinforcement, not a cancelling double negative (BDF §431). Its position before the ὑπό-phrase throws the emphasis onto the comprehensiveness: nothing whatever is to shake them. English cannot stack the negatives without sounding illiterate, so the literal tier keeps ‘not … in nothing’ with a note and the smooth tier renders ‘never … in anything.’

Built onbdf · obrien-nigtc

τῶν ἀντικειμένων is a substantival present participle of ἀντίκειμαι, literally ‘those who lie over against.’ The present tense and the article together make it a standing designation — the opposition party, not a passing incident (cf. Luke 13:17; 21:15; 1 Cor 16:9; 1 Tim 5:14).

The identity of the opponents is deliberately unspecified. The immediate context (πολιτεύεσθε, v. 27; πραιτώριον, v. 13; Roman colonial Philippi) makes local civic and pagan hostility the most natural reference, and the parallel with Paul’s own ἀγών ‘in me’ (v. 30) supports it: the Philippians are facing the same kind of pressure Paul is facing from outside, not internal doctrinal rivals. Some read the opponents as the Judaizing teachers of ch. 3 or as the false preachers of 1:15–17.

Per §5, the rendering must not adjudicate the identification. ‘Those who oppose you’ keeps the referent open exactly as the Greek does; ‘adversaries’ and ‘enemies’ both sharpen it beyond the participle’s warrant.

Built onbdag · obrien-nigtc · fee-nicnt

The relative pronoun is feminine singular, but there is no feminine singular noun in the preceding clause. Three explanations are on the table, and they are not equally strong but none is decisive.

The difference does not change the outcome very much: on any reading, the observable situation — a church that will not flinch, and the opposition it is not flinching at — functions as evidence. What the reading does change is the direction of emphasis, so the translation should keep the vague ‘which’ / ‘a thing which’ rather than substituting a resolved noun phrase (‘your steadfastness is …’), which closes the question the Greek leaves open (§4).

ὅστις rather than ὅς is deliberate: the indefinite/qualitative relative marks the antecedent as belonging to a class — ‘the sort of thing that constitutes proof.’

Built onbdf · obrien-nigtc · silva-becnt · fee-nicnt

ἔνδειξις is rare in the NT (Rom 3:25, 26; 2 Cor 8:24; here). BDAG: ‘demonstration, proof, sign.’ The word carries forensic and public-display overtones — evidence adduced, a case demonstrated — rather than a private intimation. In documentary Greek it is used of formal information laid before a magistrate and of the demonstrable proof of a claim.

Two related nouns are worth distinguishing, since English blurs them:

• σημεῖον — a sign pointing to something else
• ἔνδειξις — the demonstration or evidentiary proof of it
• ἔνδειγμα (2 Thess 1:5, in a closely parallel argument about suffering and just judgment) — the concrete token that results

The forensic sense is what makes the double genitive work. One single piece of evidence, read in a courtroom, establishes two verdicts at once: ἀπωλείας against the opponents and σωτηρίας for the church. That is why ‘proof’ is preferable to ‘sign’ (too weak, too indirect) or ‘token’ (archaic and object-like). Where 2 Thess 1:5 uses ἔνδειγμα for the same theological move, the vocabulary difference should be visible in English (‘evidence’ there, ‘proof’ here) rather than harmonized.

Built onbdag · obrien-nigtc · silva-becnt

The pairing is a fixed Pauline eschatological antithesis (2 Cor 2:15–16; Phil 3:19 vs. 3:20; 2 Thess 2:10).

ἀπώλεια is the noun of ἀπόλλυμι: not annihilation as a metaphysical claim, but ruin, loss, the coming-to-nothing of a thing. The diachronic check (§9) is important here: the word is used in Hellenistic Greek of the wasting of property (Matt 26:8, the ‘waste’ of ointment) as much as of persons. Rendering it ‘damnation’ imports later doctrinal specificity the noun itself does not carry; ‘perdition’ is accurate but now opaque. Destruction or ruin is the base gloss, with the eschatological reference supplied by context rather than by the word.

σωτηρία follows the standing corpus gloss (‘salvation’). Note that in Phil 1:19 the same noun carried the narrower sense of deliverance/vindication; here the antithesis with ἀπώλεια fixes the full eschatological sense, so no deviation note is required.

Word order matters: ὑμῶν δὲ σωτηρίας fronts the pronoun before its noun. Greek can do this for emphasis, and the emphasis is the point — yours, by contrast. English carries it with ‘but of your salvation.’

Built onbdag · muraoka-lxx · obrien-nigtc

Copula elided; supply ‘is’ or ‘comes.’ τοῦτο is neuter, so it does not agree with ἥτις or σωτηρίας but resumes the entire preceding situation — the twofold verdict and the standing that constitutes its proof.

The scope of τοῦτο is where the theological weight sits, and it is genuinely contested.

Either way, ἀπό marks source, not agency or instrument (which would be ὑπό or διά), so the rendering is ‘from God.’ θεοῦ is anarthrous but definite — normal for θεός after a preposition in Paul.

Per §6, the main text keeps the bare, unresolved ‘and this from God,’ which is what the Greek says; the choice between the readings is confined to this note.

Built onobrien-nigtc · fee-nicnt · silva-becnt · bdag

The text of this verse is stable in substance. Two things are worth recording.

1. πτυρόμενοι. Because the verb is rare, it drew scribal substitution. A scattering of witnesses read πτοούμενοι (from πτοέω, ‘be terrified’) — a normalization toward a commoner verb, and precisely the kind of change that betrays itself: a scribe replaces a rare word with a familiar one, not the reverse. NA28 prints πτυρόμενοι without hesitation, and the internal evidence (lectio difficilior) is decisive even apart from the external support.

2. ἥτις / ἥ. A few witnesses smooth the awkward relative. The harder, better-attested ἥτις stands.

3. The parenthesis. NA28 opens a bracket at ἥτις and does not close it until the end of v. 29; the sentence resumes at v. 30 with τὸν αὐτὸν ἀγῶνα ἔχοντες, which is again dependent on στήκετε. Ancient manuscripts carry no such punctuation, so the bracketing is an editorial judgment about syntax, not textual data. It is a good judgment — the ὅτι of v. 29 explains καὶ τοῦτο ἀπὸ θεοῦ, and v. 30’s participle plainly reaches back over vv. 28b–29 — but a reader should know it is interpretive. The rendering here reproduces the effect with dashes rather than importing brackets into the reading text.

Built onna28 · metzger-tc · bdag

There is no formal OT citation here, but the ‘do not be frightened’ formula belongs to a well-worn Hebrew pattern. The LXX rendering of אַל־תִּירָא / אַל־תֵּחָת (‘do not fear, do not be dismayed’) in the holy-war and prophetic-commission texts — Deut 1:21; 31:6; Josh 1:9; Isa 41:10; Jer 1:8, 17 — establishes the register: a summons to steadiness before enemies grounded in God’s presence, not in an assessment of the odds.

Paul does not use the LXX’s verbs (φοβέομαι, δειλιάω, πτοέομαι), which is itself notable: πτύρομαι is his own choice, sharper and more physical. But the structure of the exhortation — stand, fight together, do not be startled by the opposition, and this is from God — reproduces the pattern, and the closing attribution to God does exactly what the divine-presence clause does in the Hebrew formula.

The closest conceptual parallel in Paul is 2 Thess 1:4–6, where the church’s endurance under persecution is likewise ἔνδειγμα of God’s righteous judgment, cutting both ways. That parallel is Paul reading his own earlier formulation, and per §9 it is recorded as biblical theology rather than blended into this rendering.

Built onmuraoka-lxx · obrien-nigtc

The Greek fathers read the double genitive as the hinge of the verse.

Chrysostom (Homilies on Philippians, hom. 4) takes the opponents’ hostility as self-condemning: the very fact that they persecute demonstrates their ruin, while the endurance of the persecuted demonstrates their salvation. He presses the courtroom sense of ἔνδειξις hard — the situation is a proof already entered, awaiting only the sentence — and reads καὶ τοῦτο ἀπὸ θεοῦ broadly, of the whole arrangement, which he ties directly to the ἐχαρίσθη of v. 29 and to the notion that suffering for Christ is a gift rather than a misfortune.

Theodoret of Cyrus likewise connects v. 28 to v. 29 as ground and consequence, emphasizing the grant.

Ambrosiaster and the Latin tradition render πτυρόμενοι with terreamini (Vulgate: in nullo terreamini ab adversariis), which flattens the startle-image into general terror — an early instance of exactly the loss the modern rendering here tries to avoid. The Latin indicium for ἔνδειξις, by contrast, preserves the forensic force better than most English options.

The patristic consensus is worth registering for one reason especially: no early reader treated the twofold verdict as harsh or in need of softening. It read to them as the ordinary logic of divine judgment.

Built onobrien-nigtc · silva-becnt