¹ tc Although most witnesses, including some early versions and fathers (D² Ψ 1241 1505 M al sy Cyp Lcf Ambst), have ἀφίστασο ἀπὸ τῶν τοιούτων (aphistaso apo’ tōn toioutōn, “stay away from such things!”) after εὐσεβείαν (eusebeian, “godliness”; thus, “who suppose that godliness is a way of making a profit; stay away from such things!”), there seems to be little reason for this clause’s omission in some of the oldest and best witnesses (א A D* F G 048 6 33 81 1175 1739 1881 lat co). It is likely that it crept into the text early, perhaps as a marginal comment, but it should not be considered authentic in light of the strong external and transcriptional evidence against it.

² tc ‡ Several mss (B F G P 0278 33 81 323 1505 1739 1881 al bo) read ἀπαρχήν (aparchēn, “as a firstfruit”; i.e., as the first converts) instead of ἀπ᾿ ἀρχῆς (ap’ archēs, “from the beginning,” found in א D K L Ψ 1175 1241 M it sa), but this seems more likely to be a change by scribes who thought of the early churches in general in this way. But Paul would not be likely to call the Thessalonians “the firstfruits” among his converts. Further, ἀπαρχή (aparchē, “firstfruit”) is a well-worn term in Paul’s letters (Rom 8:23; 11:16; 16:5; 1 Cor 15:20, 23; 16:15), while ἀπ᾿ ἀρχῆς occurs nowhere else in Paul. Scribes might be expected to change the text to the more familiar term. Nevertheless, a decision is difficult (see arguments for ἀπαρχήν in TCGNT 568), and ἀπ᾿ ἀρχῆς must be preferred only slightly.

¹ tc ‡ Several significant witnesses of the Alexandrian and Western traditions, as well as many other witnesses, read ᾿Ιησοῦς (Iēsous, “Jesus”) after κύριος (kurios, “Lord”; so א A D* F G L^c P Ψ 0278 33 81 104 365 1241 2464 latt sy co). But there is sufficient evidence in the Alexandrian tradition for the shorter reading (B 1739 1881), supported by the Byzantine text, Irenaeus, and other witnesses (D^2vid 630 1175 1505). Although it is possible that scribes overlooked ᾿Ιησοῦς if the two nomina sacra occurred together (kMsiMs), since “the Lord Jesus” is a frequent enough appellation, it looks to be a motivated reading. NA²⁸ places ᾿Ιησοῦς in brackets, indicating some doubts as to its authenticity.

² tn Grk “the one who restrains.” This gives a puzzling contrast to the impersonal phrase in v. 6 (“the thing that restrains”). The restraint can be spoken of as a force or as a person. Some have taken this to mean the Roman Empire in particular or human government in general, since these are forces that can also be seen embodied in a person, the emperor or governing head. But apocalyptic texts like Revelation and Daniel portray human government of the end times as under Satanic control, not holding back his influence. Also the power to hold back Satanic forces can only come from God. So others understand this restraint to be some force from God: the preaching of the gospel or the working of the Holy Spirit through God’s people.

¹ tc ‡ δέ (de, “now”) is found in א² B D 0226 6 1505 1739 1881 2464 al, but lacking in א* A F G 33 it. γάρ (gar, “for”) is the reading of the Byzantine text and a few other witnesses (Ψ 0278 1175 1241 M al). Although normally the shorter reading is to be preferred, the external evidence is superior for δέ (being found in the somewhat better Alexandrian and Western witnesses). What, then, is to explain the γάρ? Scribes were prone to replace δέ with γάρ, especially in sentences suggesting a causal or explanatory idea, thus making the point more explicit. Internally, the omission of δέ looks unintentional, a case of homoioarcton (otandelegwsin). Although a decision is difficult, in this instance δέ has the best credentials for authenticity.

¹ tc Although κανόνι (kanoni, “standard, rule”) is found in most witnesses, though in various locations in this verse (א² D² Ψ 075 1505 2464 M), it is almost surely a motivated reading, for it clarifies the cryptic τῷ αὐτῷ (tō autō, “the same”). Both the fact that the word floats, and that there are other variants which accomplish greater clarity by other means, strongly suggests the secondary nature of any of the longer readings here. Further, the shortest text has excellent and early support in P^16,46 א* A B I^vid 6 33 1739 co, rendering it decidedly the preferred reading. The translation adds “standard” because of English requirements, not because of textual basis.

¹ tn There is a significant wordplay here in the Greek text. In v. 2 a rare, strong word is used to describe those who were pro-circumcision (κατατομή, katatomē, “mutilation”; see BDAG 528 s.v.), while in v. 3 the normal word for circumcision is used (περιτομή, peritomē; see BDAG 807 s.v.). Both have τομή (the feminine form of the adjective τομός [tomos], meaning “cutting, sharp”) as their root; the direction of the action of the former is down or off (from κατά, kata), hence the implication of mutilation or emasculation, while the direction of the action of the latter is around (from περί, peri). The similarity in sound yet wide divergence of meaning between the two words highlights in no uncertain terms the differences between Paul and his opponents.

¹ tc ‡ Although most witnesses have either δέ (de; P⁴⁹ א A D² Ψ 33 1505 1739^mg 2464 M lat) or οὖν (oun; D* F G 1175) here, a few significant mss lack a conjunction (P⁴⁶ B 0278 6 1739* 1881). If either conjunction were originally in the text, it is difficult to explain how the asyndetic construction could have arisen (although the dropping of δέ could have occurred via homoioteleuton). Further, although Hellenistic Greek rarely joined sentences without a conjunction, such does occur in the corpus Paulinum on occasion, especially to underscore a somber point. “Instead” has been supplied in the translation because of stylistic requirements, not textual basis. NA²⁸ places δέ in brackets, indicating some doubt as to its authenticity.

⁴ tn Or perhaps “capstone” (NAB). The meaning of ἀκρογωνιαῖος (akrogōniaios) is greatly debated. The meaning “capstone” is proposed by J. Jeremias (TDNT 1:792), but the most important text for this meaning (T. Sol. 22:7-23:4) is late and possibly not even an appropriate parallel. The only place ἀκρογωνιαῖος is used in the LXX is Isa 28:16, and there it clearly refers to a cornerstone that is part of a foundation. Furthermore, the imagery in this context has the building growing off the cornerstone upward, whereas if Christ were the capstone, he would not assume his position until the building was finished, which vv. 21-22 argue against.

⁴ tc P⁴⁶ א* A B P 33 1739 1881 2464 Hier lack “your love” (τὴν ἀγάπην, tēn agapēn), while various other groups of mss have different arrangements of the phrase “your love toward all the saints” (τὴν ἀγάπην τὴν εἰς πάντας τοὺς ἁγίους, tēn agapēn tēn eis pantas tous hagious). Most witnesses, especially the later ones (א² D¹ Ψ 1241 1505 M latt sa), read τὴν ἀγάπην τὴν εἰς πάντας τοὺς ἁγίους. Externally, the shorter reading is superior. Internally, the omission of τὴν ἀγάπην is a significantly harder reading, for the saints become an object of faith on par with the Lord Jesus. If this reading is authentic, however, the force of πίστις (pistis) is probably closer to “faithfulness,” a meaning that could perhaps be suitable toward both the Lord and the saints. Nevertheless, if the shorter reading is authentic, later scribes would no doubt have been tempted to alter it. With the parallel in Col 1:4 at hand, τὴν ἀγάπην would have been the most obvious phrase to add. (Metzger TCGNT 533 suggests that ἣν ἔχετε would have been added instead of the second τήν if the shorter reading were original, in conformity with Col 1:4, but this is not necessarily so: Scribes often altered the text as minimally as possible, and since the second τήν was already present, replacing it with ἣν ἔχετε, when the meaning was not significantly different from the second τήν, seems unlikely.) Further, ἀγάπην comes after “saints” (thus, τὴν εἰς πάντας τοὺς ἁγίους ἀγάπην) in some witnesses (81 104 326 365 1175), and the second τήν is lacking (thus, τὴν ἀγάπην εἰς πάντας τοὺς ἁγίους) in others (D* F G). Such a floating text normally indicates inauthenticity (in this case, for ἀγάπην). On the other hand, τὴν ἀγάπην could easily have dropped out of the text by way of haplography, the Alexandrian scribes’ eyes skipping from τήν to τήν. The weak first declension feminine article-noun-article construction is common enough in the NT, occurring over 40 times, yet in four of these texts there is some ms evidence for an omission similar to Eph 1:15 (Rom 11:17; 2 Tim 3:10; Rev 11:2; 21:9). But in none of these places is the Alexandrian testimony united in the omission as it is here. Further, a wholesale Alexandrian omission of τὴν ἀγάπην presupposes a much stronger genealogical relation among the Alexandrian mss than many scholars would embrace. What seems to tip the scales in favor of the longer reading, however, is the intrinsic evidence: The question of whether πίστις could be used to mean faithfulness in the general sense toward both the Lord and the saints is quite problematic. All in all, a decision is difficult, but the longer reading is, with hesitation, preferred.

¹ tc Most mss (D F G I 0176 0278 1505 M it sy) read “ratified by God in Christ” whereas the omission of “in Christ” is the reading in P⁴⁶ א A B C P Ψ 6 33 81 1175 1241 1739 1881 2464 co. The shorter reading is strongly supported by the ms evidence, and it is probable that a copyist inserted the words as an interpretive gloss. However, this form of the “in Christ” expression is somewhat atypical in the corpus Paulinum (εἰς Χριστόν [eis Christon] rather than ἐν Χριστῷ [en Christō]), a fact which tempers one’s certainty about the shorter reading. Nevertheless, the expression is used more in Galatians than in any other of Paul’s letters (Gal 2:16; 3:24, 27), and may have been suggested by such texts to early copyists.

¹ tc ‡ Several significant witnesses have ὁ θεός (ho theos) after εὐδόκησεν (eudokēsen; so א A D Ψ 0278 33 1175 1241 1739 1881 2464 M co) while the shorter reading is supported by P⁴⁶ B F G 629 1505 lat. There is hardly any reason why scribes would omit the words (although the Beatty papyrus and the Western text do at times omit words and phrases), but several reasons why scribes would add the words (especially the need to clarify). The confluence of witnesses for the shorter reading (including a few fathers and versions) adds strong support for its authenticity. It is also in keeping with Paul’s style to refrain from mentioning God by name as a rhetorical device (cf. ExSyn 437 [although this section deals with passive constructions, the principle is the same]). NA²⁸ includes the words in brackets, indicating some doubts as to their authenticity.

² tc Most witnesses, especially later ones (א² D Ψ M lat sy bo), conclude this letter with ἀμήν (amēn, “amen”), while several early and significant mss (P⁴⁶ א* A B F G 0243 6 33 630 1175 1739 1881 sa) lack the particle. 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. That so many diverse witnesses lacked the word here is strong testimony to its absence for the original text of 2 Corinthians.

¹ tc The sense of this translation is attested by the fact that most of the later mss, along with several early and significant ones (א C D² Ψ 0225 0243 1739 1881 M lat), have the imperative verb ἐνδείξασθε (endeixasthe) in place of the participle ἐνδεικνύμενοι (endeiknumenoi), which is found in B D* F G 33. Since an imperatival participle is more Hebraic in style, many scribes would not have understood the idiom as easily and would have been likely to change the participle to an imperative (so TCGNT 513-14). But there is no good reason why scribes would change the imperative into a participle. Thus, ἐνδεικνύμενοι is almost surely the wording of the original text.

² tc Most mss have the words τὰ πάντα (ta panta, “all things”; cf. KJV “behold, all things are become new”), some after καίνα (kaina, “new”; D² K L P Ψ 104 326 945 2464 pm) and others before it (6 33 81 614 630 1241 1505 1881 pm). The reading without τὰ πάντα, however, has excellent support from both the Western and Alexandrian text-forms (P⁴⁶ א B C D* F G 048 0243 365 629 1175 1739 co), and the different word order of the phrase which includes it (“all things new” or “new all things”) in the ms tradition indicates its secondary character. This secondary addition may have taken place because of assimilation to τὰ δὲ πάντα (ta de panta, “and all [these] things”) that begins the following verse.

² tc Two viable variants exist at this place in the text: ἁγιότητι (hagiotēti, “holiness”) vs. ἁπλότητι (haplotēti, “pure motives”). A confusion of letters could well have produced the variant (TCGNT 507): In majuscule script the words would have been written agiothti and aplothti. This, however, does not explain which reading created the other. Overall ἁπλότητι, though largely a Western-Byzantine reading (א² D F G M lat sy), is better suited to the context; it is also a Pauline word while ἁγιότης (hagiotēs) is not. It also best explains the rise of the other variants, πραότητι (praotēti, “gentleness”) and σπλάγχνοις (splanchnois, “compassion”). On the other hand, the external evidence in favor of ἁγιότητι is extremely strong (P⁴⁶ א* A B C K P Ψ 0121 0243 33 81 1739 1881 al co). This diversity of mss provides excellent evidence for authenticity, but because of the internal evidence listed above, ἁπλότητι is to be preferred, albeit only slightly.

¹ tc The Western and Byzantine texts, as well as one or two Alexandrian mss (D F G Ψ 33 M latt sy), combine in reading ἀδελφοί (adelphoi, “brothers”) here, while the Alexandrian witnesses (P⁴⁶ א A B C P 81 630 1175 1739 1881 2464 co) largely lack the address. The addition of ἀδελφοί is apparently a motivated reading, however, for scribes would have naturally wanted to add it to ἐπαινῶ δὲ ὑμᾶς (epainō de humas, “now I praise you”), especially as this begins a new section. On the other hand, it is difficult to explain how the shorter reading could have arisen from the longer one. Thus, on both internal and external grounds, the shorter reading is strongly preferred.

² tc The reading ἤ—τὸ καύχημά μου οὐδεὶς κενώσει (ē—to kauchēma mou oudeis kenōsei, “than—no one will deprive me of my reason for boasting!”) is syntactically abrupt, but fully in keeping with Pauline style. It is supported by P⁴⁶ א* B D*^,c 33 1739 1881 as well as early patristic authors. Most witnesses, especially the later ones (א² C D² Ψ M lat), have a significantly smoother reading than this: ἢ τὸ καύχημά μου ἵνα τις κενώσῃ (or κενώσει); ē to kauchēma mou hina tis kenōsē (or kenōsei), “than that anyone should deprive me of my boasting.” The simple replacement of οὐδείς with ἵνα essentially accomplishes the smoothing out of the text, and as such the ἵνα reading is suspect. Not only is the harder reading in keeping with Pauline style, but it is also found in the earlier and better witnesses.

² tn Verses 4b-5a are capable of various punctuations: (1) “and I am with you in spirit, through the power of our Lord Jesus turn this man over to Satan”; (2) “and I am with you in spirit with the power of our Lord Jesus, turn this man over to Satan”; (3) “and I am with you in spirit, along with the power of our Lord Jesus, turn this man over to Satan” (as adopted in the text). The first option suggests the Lord’s power is needed when the church is to hand the man over to Satan; the second option suggests that the Lord’s power is present when Paul is gathered with the Corinthians in spirit; the third option leaves the relation of the Lord’s power to the surrounding phrases vague, perhaps implying that both are in view.

¹ tc ‡ After οὖν (oun), several significant Alexandrian and Western mss (B C D F G 81 365) have τήν (tēn). The article is lacking in א A Ψ 33 1175 1241 1505 1739 1881 M however. P⁴⁶ supplies a relative pronoun and has a different reading entirely (“which I have [as a] boast”). Articles were frequently introduced to clarify the meaning of the text. In this instance, since the word modified (καύχησιν, kauchēsin) is third declension, a visual oversight (resulting in omission) is less likely. Hence, the shorter reading is probably earlier. The difference in translation between these first two options is negligible (“I have the boast” or “I have a boast”). NA²⁸ puts the article in brackets, indicating some doubt as to its authenticity.