³ tc ‡ Most mss, including several important ones (Ì⁷⁴ א A C E Ψ 33 1739 Ï h p vg sy^h mae bo Cyr), have ὑμῶν (Jumwn, “your”) here, in conformity with the LXX of Amos 5:26. But other significant and diverse witnesses lack the pronoun: The lack of ὑμῶν in B D 36 453 gig sy^p sa Ir^lat Or is difficult to explain if it is not the original wording here. NA²⁷ has the word in brackets, indicating some doubt as to its authenticity.

³ tn Or perhaps, “two whole years at his own expense.” BDAG 654 s.v. μίσθωμα states, “the customary act. mng. ‘contract price, rent’…is not found in our lit. (Ac) and the pass. what is rented, a rented house is a mng. not found outside it (even Ammonius Gramm. [100 ad] p. 93 Valck. knows nothing of it. Hence the transl. at his own expense [NRSV] merits attention) ἐν ἰδίῳ μισθώματι in his own rented lodgings Ac 28:30 (for the idea cp. Jos., Ant. 18, 235).”

³ tc ‡ αὐτό (auto) is found at this point in v. 13 in a number of significant witnesses, including A B C P 33 1739 al. But Ì⁴⁶ א D Ψ 0289 1881 Ï latt lack it. The pronoun could be a motivated reading, designed to intensify Paul’s statement. On the other hand, it could have been deleted because the article alone made the reference already clear. In this instance, the possibility of scribal addition seems more likely than scribal deletion, although a decision is difficult. NA²⁷ includes the word in brackets, indicating doubt as to its authenticity.

² tc The shorter reading, κυρίου (kuriou, “Lord”), is found in Ì⁴⁶ B 630 1739 pc; κυρίου ᾿Ιησοῦ (kuriou Ihsou, “Lord Jesus”) is read by Ì^61vid א Ψ Ï; κυρίου ᾿Ιησοῦ Χριστοῦ (kuriou Ihsou Cristou, “Lord Jesus Christ”) by D pc; and κυρίου ἡμῶν ᾿Ιησοῦ Χριστοῦ (kuriou Jhmwn Ihsou Cristou, “our Lord Jesus Christ”) by A F G P 33 al. The shorter reading is preferred as the reading that best explains the other readings, especially in view of the mention of “Jesus” twice in the previous verse.

¹ tc Grk “the brother.” Later witnesses (א² D² Ï) have ἀνδρί (andri, “husband”) here, apparently in conscious emulation of the earlier mention of ἀνήρ (ajnhr) in the verse. However, the earliest and best witnesses (Ì⁴⁶ א* A B C D* F G P Ψ 33 1739 al co) are decisively in favor of ἀδελφῷ (adelfw, “brother”), a word that because of the close association with “wife” here may have seemed inappropriate to many scribes. It is also for reasons of English style that “her husband” is used in the translation.

¹ tc The Byzantine text, as well as a few other witnesses (D² [L] Ψ 1881 Ï) lack this parenthetical material, while geographically widespread, early, and diverse witnesses have the words (so א A B C D* F G P 33 104 365 1175 1505 1739 al latt). The phrase may have dropped out accidentally through homoioteleuton (note that both the preceding phrase and the parenthesis end in ὑπὸ νόμον [Jupo nomon, “under the law”]), or intentionally by overscrupulous scribes who felt that the statement “I myself am not under the law” could have led to license.

² tc Instead of the present participle συγκλειόμενοι (sunkleiomenoi; found in Ì⁴⁶ א A B D* F G P Ψ 33 1739 al), C D¹ 0176 0278 Ï have the perfect συγκεκλεισμένοι (sunkekleismenoi). The syntactical implication of the perfect is that the cause or the means of being held in custody was confinement (“we were held in custody [by/because of] being confined”). The present participle of course allows for such options, but also allows for contemporaneous time (“while being confined”) and result (“with the result that we were confined”). Externally, the perfect participle has little to commend it, being restricted for the most part to later and Byzantine witnesses.

³ tc ‡ Grk “so that they will not be persecuted.” The indicative after ἵνα μή (Jina mh) is unusual (though not unexampled elsewhere in the NT), making it the harder reading. The evidence is fairly evenly split between the indicative διώκονται (diwkontai; Ì⁴⁶ A C F G K L P 0278 6 81 104 326 629 1175 1505 pm) and the subjunctive διώκωνται (diwkwntai; א B D Ψ 33 365 1739 pm), with a slight preference for the subjunctive. However, since scribes would tend to change the indicative to a subjunctive due to syntactical requirements, the internal evidence is decidedly on the side of the indicative, suggesting that it is original.

⁴ tc Most witnesses, including some important ones (א A C F G I [P] 075 Ï it bo), read “and the Lord Jesus Christ” at the end of this verse, no doubt to conform the wording to the typical Pauline salutation. However, excellent and early witnesses (B D K L Ψ 33 81 1175 1505 1739 1881 al sa) lack this phrase. Since the omission is inexplicable as arising from the longer reading (otherwise, these mss would surely have deleted the phrase in the rest of the corpus Paulinum), it is surely authentic.

¹ tc Most witnesses (D Ψ Ï sy) have πιστὸς ἤ (pisto" h) before πιστή (pisth), with the resultant meaning “if a believing man or woman.” But such looks to be a motivated reading, perhaps to bring some parity to the responsibilities of men and women listed here, and as a way of harmonizing with v. 4. Further, most of the earliest and best witnesses (א A C F G P 048 33 81 1175 1739 1881 co) lack the πιστὸς ἤ, strengthening the preference for the shorter reading.

² tc Most witnesses (D¹ Ï) have “various washings, and external regulations” (βαπτισμοῖς καὶ δικαιώμασιν, baptismoi" kai dikaiwmasin), with both nouns in the dative. The translation “washings; they are… regulations” renders βαπτισμοῖς, δικαιώματα (baptismoi", dikaiwmata; found in such important mss as Ì⁴⁶ א* A I P 0278 33 1739 1881 al sa) in which case δικαιώματα is taken as the nominative subject of the participle ἐπικείμενα (epikeimena). It seems far more likely that scribes would conform δικαιώματα to the immediately preceding datives and join it to them by καί than they would to the following nominative participle. Both on external and internal evidence the text is thus secure as reading βαπτισμοῖς, δικαιώματα.

¹ tc The majority of mss, especially the later ones (א² 0278^vid 1739 Ï lat), have ὁ θεός (Jo qeo", “God”) at this point, while most of the earliest and best witnesses lack such an explicit addressee (so Ì⁴⁶ א* A C D K P Ψ 33 1175 1881 2464 al). The longer reading is a palpable corruption, apparently motivated in part by the wording of Ps 40:8 (39:9 LXX) and by the word order of this same verse as quoted in Heb 10:7.

² tc Most mss ([C] P 0246 Ï) read ὁ κύριος (Jo kurio", “the Lord”) here, while others have ὁ θεός (Jo qeo", “God”; 4 33^vid 323 945 1739 al). However, several important and early witnesses (̲³ א A B Ψ 81 co) have no explicit subject. In light of the scribal tendency toward clarification, and the fact that both κύριος and θεός are well represented, there can be no doubt that the original text had no explicit subject. The referent (God) has been specified in the translation for clarity, not because of textual basis.

¹ tc Most mss (א² A P Ï) add ὑπὲρ ἡμῶν (Juper Jhmwn, “for us”); others (א* 69 1505 pc) add ὑπὲρ ὑμῶν (Juper Jumwn, “for you”), the first hand of א also has ἀποθανόντος (apoqanonto", “since he died”) instead of παθόντος (paqonto", “since he suffered”). But the reading without ὑπὲρ ἡ/ὑμῶν best explains the rise of the other readings, for not only is there confusion as to which pronoun belongs here, but the longer readings, being clarifications, are evidently motivated readings. The shortest reading is found in important and early Alexandrian and Western witnesses (Ì⁷² B C Ψ 0285 323 1739) and is strongly preferred.

³ tc ‡ The vast bulk of mss add ἀμήν (amhn, “amen”) at the end of this letter, as they do almost all the rest of the NT books (only Acts, James, and 3 John lack a majority of witnesses supporting a concluding ἀμήν). The omission in B 1241 1243 1739* 1881 2298 appears to be original, although the fact that some of the best and earliest Alexandrian witnesses (Ì⁷² א A C P Ψ 33 co), along with the Byzantine text and early versions (vg sy), add the particle renders such a judgment less than iron-clad. NA²⁷ places the word in brackets, indicating doubts as to its authenticity.

¹ tc The Byzantine text has ἀμήν (amhn, “amen”) at the conclusion of this letter. Such a conclusion is routinely added by scribes to NT books because a few of these books originally had such an ending (cf. Rom 16:27; Gal 6:18; Jude 25). A majority of Greek witnesses have the concluding ἀμήν in every NT book except Acts, James, and 3 John (and even in these books, ἀμήν is found in some witnesses). It is thus a predictable variant. Further, the particle is lacking in excellent, early, and diffuse witnesses (א A B P Ψ 33 81 323 1739 1881 al co), rendering its omission the strongly preferred reading.

⁷ tc Most later witnesses (P Ψ Ï sy) have θεόν (qeon, “God”) after δεσπότην (despothn, “master”), which appears to be a motivated reading in that it explicitly links “Master” to “God” in keeping with the normal NT pattern (see Luke 2:29; Acts 4:24; 2 Tim 2:21; Rev 6:10). In patristic Greek, δεσπότης (despoth") was used especially of God (cf. BDAG 220 s.v. 1.b.). The earlier and better witnesses (Ì^72,78 א A B C 0251 33 81 323 1241 1739 al co) lack θεόν; the shorter reading is thus preferred on both internal and external grounds.

² tc Several mss (Ì⁴⁷ א 1 1006 1611 2351 Ï^K pc) have the conjunction “for” (γάρ, gar) here so that the phrase reads: “for they are blameless.” Other important mss (A C P 1854 2053 al lat) lack the word. The shorter reading is to be preferred since the scribes were more likely to make the connection explicit through the addition of “for” than they would have been to omit the conjunction. As it is, the passage without the conjunction makes good sense and evokes a very somber tone.

² tc Most mss (Ì47 א* Ï sa) lack ἄλλον (allon, “another”) here, but the support for it is stronger (Ì^115vid א² A C P 051 1006 1611 1841 2053 2329 al latt sy bo). The problem that its inclusion represents is that there is no reference to any other angel in the immediate context (the last mention was in 11:15). In this instance, the longer reading is harder. The word was probably intentionally omitted in order to resolve the tension; less likely, it might have been accidentally omitted since its spelling is similar to “angel” (ἄγγελος, angelos).

¹ tc It appears that “dipped” (βεβαμμένον, bebammenon), supported by several uncials and other witnesses (A 051 Ï), is the original reading. Due to the lack of the preposition “in” (ἐν, en) after the verb (βεβαμμένον αἵματι, bebammenon {aimati), and also probably because of literary allusions to Isa 63:3, several mss and versions seem to have changed the text to “sprinkled” (either ῥεραντισμένον [rJerantismenon] in P 2329 al; ἐρραντισμένον [errantismenon] in 1006 1841; ἐρραμμένον [errammenon] in 2053 2062; or ῥεραμμένον [rJerammenon] in 1611; or in one case περιρεραμμένον [perirerammenon] in א^[2]). The reading most likely to give rise to the others is “dipped.”