¹ tc Several important witnesses omit καταστροφῇ (katastrofh, “destruction”; such as Ì^72* B C* 1241 1739 1881 pc), but this is probably best explained as an accidental omission due to homoioarcton (the word following is κατέκρινεν [katekrinen, “he condemned”]).

³ tc Most mss (B D F G Ψ 33 1739 1881 Ï lat) have διά (dia) followed by the accusative: “because of his Spirit who lives in you.” The genitive “through his Spirit” is supported by א A C⁽*⁾ 81 104 1505 1506 al, and is slightly preferred.

¹ tc Most mss, especially Western and Byzantine (D F G 048 33 1739 1881 Ï latt), read θεός (qeos, “God”) in place of κύριος (kurios, “Lord”) here. However, κύριος is found in many of the most important mss (Ì⁴⁶ א A B C P Ψ pc co), and θεός looks to be an assimilation to θεός in v. 3.

¹ tc Most later witnesses (א² Ï sy) add “fasting and” (τῇ νηστείᾳ καί, th nhsteia kai) before “prayer.” But such an addition is motivated by ascetic concerns; further, its lack in Ì^11vid,46 א* A B C D F G P Ψ 33 1739 1881 2464 al latt co argues decisively against its authenticity.

¹ tc Although usually δέ (de, “now”; found in א A C D¹ F G Ψ 0285 Ï lat) should take precedent over γάρ (gar) in textually disputed places in the corpus Paulinum, the credentials for γάρ here are not easily dismissed (Ì⁴⁶ B 0223 0243 33 1739 1881 al); here it is the preferred reading, albeit slightly.

¹ tc The phrase “in Christ Jesus” is found after “For” in some mss (א A C D F G 0278 1881 Ï lat bo), but lacking in Ì⁴⁶ B Ψ 33 1175 1505 1739* and several fathers. The longer reading probably represents a harmonization to Gal 5:6.

³ tc The bulk of the Western text (D*^,c F G K it) dropped καί (kai) here, most likely due to ascetic concerns. Strong external attestation for its inclusion from excellent witnesses as well as the majority (Ì⁴⁶ א A B C D² 0278 33 1739 1881 Ï) also marks it as original.

¹ tc Most later witnesses (D² 0278 Ï lat sy) have καί (kai, “and”) after ἀρχαῖς (arcai", “rulers”), though the earliest and best witnesses (א A C D* F G Ψ 33 104 1739 1881) lack the conjunction. Although the καί is most likely not authentic, it has been added in translation due to the requirements of English style. For more discussion, see TCGNT 586.

¹ tc ‡ Most mss (א* A 1739 1881 Ï sy) read “I give thanks to God” rather than “Now thanks be to God” (א¹ [B] Ψ 33 81 104 365 1506 pc), the reading of NA²⁷. The reading with the verb (εὐχαριστῶ τῷ θεῷ, eucaristw tw qew) possibly arose from a transcriptional error in which several letters were doubled (TCGNT 455). The conjunction δέ (de, “now”) is included in some mss as well (א¹ Ψ 33 81 104 365 1506 pc), but it should probably not be considered original. The ms support for the omission of δέ is both excellent and widespread (א* A B D 1739 1881 Ï lat sy), and its addition can be explained as an insertion to smooth out the transition between v. 24 and 25.

² tc ‡ Most witnesses (א A D² 33 Ï) have γάρ (gar) after μέν (men), though some significant Alexandrian and Western witnesses lack the conjunction (B D* G Ψ 81 365 1506 2464* pc latt). A few mss have γάρ, but not μέν (6 1739 1881). γάρ was frequently added by scribes as a clarifying conjunction, making it suspect here. NA²⁷ has the γάρ in brackets, indicating doubt as to its authenticity.

² tc ‡ Most witnesses (א A C D Ψ 33 Ï bo) have ἤδη (hdh, “already”) at this point in v. 19. But B F G 630 1739 1881 pc lat sa lack it. Since it appears to heighten the style of the narrative and since there is no easy accounting for an accidental omission, it is best to regard the shorter text as original. NA²⁷ includes the word in brackets, indicating doubt as to its authenticity.

² tc Most mss, especially the later ones (א² D Ψ 33 Ï sy), read νόμου (nomou, “of the law”) here, echoing Paul’s usage in Rom 3:20, 28 and elsewhere. The qualifying phrase is lacking in א* A B F G 6 629 630 1739 1881 pc lat co. The longer reading thus is weaker externally and internally, being motivated apparently by a need to clarify.

² tc Most mss (א¹ A D¹ Ψ 33 1881 Ï sy) have θεοῦ (qeou) here rather than Χριστοῦ (Cristou; found in Ì^46vid א* B C D* 6 81 629 1506 1739 pc lat co). External evidence strongly favors the reading “Christ” here. Internal evidence is also on its side, for the expression ῥῆμα Χριστοῦ (rJhma Cristou) occurs nowhere else in the NT; thus scribes would be prone to change it to a known expression.

² tc ‡ Although the witnesses for the shorter reading (Ì⁴⁶ D F G Ψ 075 0243 1739 1881 Ï) are not as strong as for the addition of ἀδελφοί (adelfoi, “brothers”) at this juncture (א A B K P 33 81 104 365 1175 2464 lat sy co), it is difficult to find a reason why scribes would either intentionally or unintentionally drop the address here. Thus, the shorter reading is slightly preferred.

² tc ‡ Several important witnesses (א C D F G Ψ 1881), as well as the Byzantine text, add κύριον (kurion) here, changing the reading to “the Lord Jesus.” Although the external evidence in favor of the shorter reading is slim, the witnesses are important, early, and diverse (Ì⁴⁶ B [0243 33] 629 [630] 1175* [1739] pc r sa). Very likely scribes with pietistic motives added the word κύριον, as they were prone to do, thus compounding this title for the Lord.

¹ tc Most mss (Ì⁴⁶ D Ψ 1881 Ï) lack διό (dio, “Therefore”), but the widespread distribution and quality of mss which include it (א A B F G 0243 33 81 1175 1739 pc) argues for its authenticity. Internally, its case is equally strong in that its inclusion is grammatically rough (διό is hardly necessary to convey purpose, especially since Paul uses ἵνα [{ina, “so that”] next).

¹ tc ‡ The conjunction δέ (de) is found in Ì⁴⁶ א*^,2 A D¹ Ψ 1739 1881 Ï sy bo, while γάρ (gar) is the conjunction of choice in א¹ B D*^,c F G 33 pc lat sa. There are thus good representatives on each side. Scribes generally tended to prefer γάρ in such instances, most likely because it was more forceful and explicit. γάρ is thus seen as a motivated reading. For this reason, δέ is preferred.

² tc ‡ φόνοι (fonoi, “murders”) is absent in such important mss as Ì⁴⁶ א B 33 81 323 945 pc sa, while the majority of mss (A C D F G Ψ 0122 0278 1739 1881 Ï lat) have the word. Although the pedigree of the mss which lack the term is of the highest degree, homoioteleuton may well explain the shorter reading. The preceding word has merely one letter difference, making it quite possible to overlook this term (φθόνοι φόνοι, fqonoi fonoi).

² tc Although several important witnesses (א¹ A C P 6 81 326 1739^c 1881) have πᾶσα ἡ οἰκοδομή (pasa Jh oikodomh), instead of πᾶσα οἰκοδομή (the reading of א* B D F G Ψ 33 1739* Ï), the article is almost surely a scribal addition intended to clarify the meaning of the text, for with the article the meaning is unambiguously “the whole building.”

³ tc Most Western and Byzantine witnesses, along with a few others (א² D F G Ψ 0278 1881 Ï lat sy), have “of our Lord Jesus Christ” after “Father,” but such an edifying phrase cannot explain the rise of the reading that lacks it, especially when the shorter reading is attested by early and important witnesses such as Ì⁴⁶ א* A B C P 6 33 81 365 1175 1739 co Or Hier.