² tc ‡ Most mss (א A [C*] 0243 0285 33 1739 1881 Ï latt) include the words “and ever” here, but the shorter reading (supported by Ì⁴⁶ C³ D Ψ 6 104 365 1505 al) is preferred on internal grounds. It seemed more likely that scribes would assimilate the wording to the common NT doxological expression “for ever and ever,” found especially in the Apocalypse (cf., e.g., 1 Tim 1:17; 2 Tim 4:18; Rev 4:9; 22:5) than to the “forever” of Heb 13:8. Nevertheless, a decision is difficult here. NA²⁷ places the phrase in brackets, indicating doubts as to its authenticity.

¹ tc Most witnesses, including several important ones (א² A C D H Ψ 0243 1739 1881 Ï lat sy bo), conclude the letter with ἀμήν (amhn, “amen”). 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, there is sufficient testimony (Ì⁴⁶ א* I^vid 6 33 sa) for the lack of the particle, rendering its omission the preferred reading.

³ tc Most witnesses, including several important ones (א A C² P Ψ 33 Ï sy bo), have νεκρά (nekra, “dead”) here, while Ì⁷⁴ reads κενή (kenh, “empty”). Both variants are most likely secondary, derived from ἀργή (argh, “useless”). The reading of the majority is probably an assimilation to the statements in vv. 17 and 26, while Ì⁷⁴’s reading picks up on κενέ (kene) earlier in the verse. The external evidence (B C* 323 945 1739 sa) for ἀργή is sufficient for authenticity; coupled with the strong internal evidence for the reading (if νεκρά were original, how would ἀργή have arisen here and not in vv. 17 or 26?), it is strongly preferred.

³ tc The Byzantine text and a few other mss (P 33 Ï) have the intransitive κατῴκησεν (katwkhsen) here, which turns τὸ πνεῦμα (to pneuma) into the subject of the verb: “The spirit which lives within us.” But the more reliable and older witnesses (Ì⁷⁴ א B Ψ 049 1241 1739 al) have the causative verb, κατῴκισεν (katwkisen), which implies a different subject and τὸ πνεῦμα as the object: “The spirit that he causes to live within us.” Both because of the absence of an explicit subject and the relative scarcity of the causative κατοικίζω (katoikizw, “cause to dwell”) compared to the intransitive κατοικέω (katoikew, “live, dwell”) in biblical Greek (κατοικίζω does not occur in the NT at all, and occurs one twelfth as frequently as κατοικέω in the LXX), it is easy to see why scribes would replace κατῴκισεν with κατῴκησεν. Thus, on internal and external grounds, κατῴκισεν is the preferred reading.

² tc ‡ The oldest and best witnesses lack the verb (א* B, along with 1505 pc), but most mss (Ì⁷² א² A C P Ψ 048 33 1739 Ï) have ἐστίν here (estin, “[if] it is [necessary]”). The verb looks to be an explanatory gloss. But if no verb is present, this opens up the time frame in the author’s mind even more, since the conditional particle for both the first class condition and the fourth class condition is εἰ (ei). That may well be what was on the author’s mind, as evidenced by some of his other allusions to suffering in this little letter (3:14, 17). NA²⁷ has the verb in brackets, indicating doubts as to its authenticity.

¹ tc Although there may be only a slight difference in translation, the term translated as “trust” is the adjective πιστούς (pistous). This is neither as common nor as clear as the verb πιστεύω (pisteuw, “believe, trust”). Consequently, most mss have the present participle πιστεύοντας (pisteuonta"; Ì⁷² א C P Ψ 1739 Ï), or the aorist participle πιστεύσαντες (pisteusante"; 33 pc), while A B pc vg have the adjective. Thus, πιστούς is to be preferred. In the NT the adjective is routinely taken passively in the sense of “faithful” (BDAG 820 s.v. πιστός 1). That may be part of the force here as well: “you are now faithful to God,” although the primary force in this context seems to be that of trusting. Nevertheless, it is difficult to separate faith from faithfulness in NT descriptions of Christians’ dependence on God.

² tc Most later mss (P Ï) have διὰ πνεύματος (dia pneumato", “through the spirit”) after ἀληθείας (ajlhqeia", “truth”), while the words are lacking in a broad spectrum of early and important witnesses (Ì⁷² א A B C Ψ 33 81 323 945 1241 1739 al vg sy co). On external grounds, the shorter reading cannot be easily explained if it were not original. The longer reading is clearly secondary, added to show more strongly God’s part in man’s obedience to the truth. But the addition ignores the force that the author gives to “purified” and ruins the balance between v. 22 and v. 23 (for in v. 23 the emphasis is on God’s part; here, on man’s part).

² tc The expression “consciousness/conscience of God” (συνείδησιν θεοῦ; suneidhsin qeou) is unusual, occurring only here in the NT. Because θεοῦ was liable to misinterpretation, several witnesses altered the text, either replacing it with ἀγαθήν (agaqhn; C Ψ 323 614 630 945 1241 1505 1739 al sy) or expanding the expression by adding ἀγαθήν before θεοῦ (Ì⁷² [A* 33] 81). Replacing θεοῦ with ἀγαθήν conforms to other NT phrases, notably in this same letter (Acts 23:1; 1 Tim 1:5, 19; 1 Pet 3:16, 21), suggesting that such a reading is motivated. The reading θεοῦ, however, has superior support (א A^c B P 049 Ï lat co), and best explains the rise of the other readings.

³ tc A number of mss, some of them important (A C K P 33 81 1505 1739 pm sy^h bo), read ὑμῶν (Jumwn, “your”) rather than ἡμῶν (Jhmwn, “our”), which is found in somewhat better witnesses (א B L Ψ 049 1241 pm sy^p sa). Although the majority of Byzantine minuscules are split between the two readings, the Textus Receptus reads ὑμῶν. It is possible that ὑμῶν represents a scribal assimilation to John 16:24. As far as the immediate context is concerned, either reading could possibly be original, since the recipients have already been mentioned in 1:2 (ὑμῖν, Jumin) and 1:3 (ὑμῖν), while it might seem more natural for the author to be concerned about the fulfillment of his own joy than his readers’ (cf. 2 John 4, 12; 3 John 3). Overall, the first person pronoun is preferred on both external and internal grounds.

¹ tc Most later mss (P Ï) have ἀμήν (amhn, “amen”) at the end 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 earliest and best witnesses, along with several others (א A B Ψ 33 323 630 1505 1739 al sy co), lack the inoffensive particle, rendering its omission as the authentic reading.

⁵ tn Or “by.” Datives of agency are quite rare in the NT (and other ancient Greek), almost always found with a perfect verb. Although this text qualifies, in light of the well-worn idiom of τηρέω (threw) in eschatological contexts, in which God or Christ keeps the believer safe until the parousia (cf. 1 Thess 5:23; 1 Pet 1:4; Rev 3:10; other terms meaning “to guard,” “to keep” are also found in similar eschatological contexts [cf. 2 Thess 3:3; 2 Tim 1:12; 1 Pet 1:5; Jude 24]), it is probably better to understand this verse as having such an eschatological tinge. It is at the same time possible that Jude’s language was intentionally ambiguous, implying both ideas (“kept by Jesus Christ [so that they might be] kept for Jesus Christ”). Elsewhere he displays a certain fondness for wordplays; this may be a hint of things to come.

² tc ‡ Some translations take ἅπαξ (Japax) with the following clause (thus, “[Jesus,] having saved the people once for all”). Such a translation presupposes that ἅπαξ is a part of the ὅτι (Joti) clause. The reading of NA²⁷, πάντα ὅτι [ὁ] κύριος ἅπαξ (panta {oti [Jo] kurio" {apax), suggests this interpretation (though with “Lord” instead of “Jesus”). This particle is found before λαόν (laon) in the ὅτι clause in א C* Ψ 630 1241 1243 1505 1739 1846 1881 pc co. But ἅπαξ is found before the ὅτι clause in most witnesses, including several important ones (Ì⁷² A B C² 33 81 623 2344 Ï vg). What seems best able to explain the various placements of the adverb is that scribes were uncomfortable with ἅπαξ referring to the readers’ knowledge, feeling it was more appropriate to the theological significance of “saved” (σώσας, swsas).

⁴ tc ‡ Most mss (א Ï^K) do not add the words “[as] their God” (αὐτῶν θεός, autwn qeos) after “he will be with them.” The mss with these words include A 2030 2050 2329 al. The Andreas group (Ï^A) also has the words, but in a different arrangement with the preceding (ἔσται μετ᾿ αὐτῶν θεὸς αὐτῶν, estai met’ autwn qeo" autwn). Not only do the words float, but scribes may have been motivated to make a connection here more directly with Isa 7:14; 8:8; Jer 24:7; 31:33; Zech 8:8. In light of sufficient external evidence as well as the possibility that the longer reading is theologically motivated, the shorter reading is preferred. NA²⁷ places the words in brackets, indicating doubts as to their authenticity.

² tn Heb “I turned and I saw.” The phrase וָאֶרְאֶה…וְשַׁבְתִּי (vÿshavti… va’er’eh, “I turned and I saw”) is a verbal hendiadys (the two verbs represent one common idea). Normally in a verbal hendiadys, the first verb functions adverbially, modifying the second verb which retains its full verbal force. The verb וְשַׁבְתִּי (vav + perfect 1st person common singular from שׁוּב “to turn”) is used idiomatically to denote repetition: “to return and do” = “to do again” (e.g., Gen 26:18; 30:31; 43:2) or “to do repeatedly” (e.g., Lam 3:3); see HALOT 1430 s.v. שׁוב 5; BDB 998 s.v. שׁוּב 8; GKC 386 §120.e: “I observed again” or “I repeatedly observed.” On the other hand, the shift from the perfect וְשַׁבְתִּי to the preterite וָאֶרְאֶה (vav + Qal preterite 1st person common singular from רָאָה, ra’ah, “to see”) might indicate a purpose clause: “I turned [my mind] to consider.” The preterite וָאֶרְאֶה follows the perfect וְשַׁבְתִּי. When a wayyiqtol form (vav + preterite) follows a perfect in reference to a past-time situation, the preterite also represents a past-time situation. Its aspect is based on the preceding perfect. In this context, the perfect and preterite may denote definite past or indefinite past action (“I turned and considered” as hendiadys for “I observed again” or “I repeatedly observed”) or past telic action (“I turned [my mind] to consider”). See IBHS 554-55 §33.3.1a.

² tn Heb “I turned and I saw…”; or “I again considered.” The Hebrew phrase וָאֶרְאֶה…וְשַׁבְתִּי (vÿshavti…va’er’eh, “I turned and I saw”) is a verbal hendiadys (the two verbs represent one common idea). Normally in a verbal hendiadys, the first verb functions adverbially, modifying the second verb which retains its full verbal force. The verb שׁוּב (shuv, “to turn”) is used idiomatically to denote repetition: “to return and do” = “to do again” (e.g., Gen 26:18; 30:31; 43:2) or “to do repeatedly” (e.g., Lam 3:3); see HALOT 1430 s.v. שׁוב 5; BDB 998 s.v. שׁוּב 8; GKC 386 §120.e: “I observed again” or “I repeatedly observed.” On the other hand, the shift from the perfect וְשַׁבְתִּי (vav + perfect 1st person common singular from שׁוּב, “to turn”) to the preterite וָאֶרְאֶה (vav + Qal preterite 1st person common singular from רָאָה, ra’ah, “to see”) might indicate a purpose clause: “I turned [my mind] to consider….” The preterite וָאֶרְאֶה follows the perfect וְשַׁבְתִּי. When a wayyiqtol form (vav + preterite) follows a perfect in reference to a past-time situation, the preterite also represents a past-time situation. Its aspect is based on the preceding perfect. In this context, the perfect and preterite may denote definite past or indefinite past action (“I turned and considered …” as hendiadys for “I observed again” or “I repeatedly observed”) or past telic action (“I turned [my mind] to consider…”). See IBHS 554-55 §33.3.1a.

⁵ tn Alternately, “captivated.” The verb אָסַר (’asar, “to bind, capture, hold captive, put in prison”) is commonly used of binding a prisoner with cords and fetters (Gen 42:34; Judg 15:10-13; 16:5-12; 2 Kgs 17:4; 23:33; 25:7; 2 Chr 33:11) (HALOT 75 s.v. אסר). It is frequently used as a figure to depict absolute authority over a person (Ps 105:22). The passive participle סוּר means “to be bound, held captive, imprisoned” (2 Sam 3:34; Jer 40:1; Job 36:8). Like a prisoner bound in cords and fetters and held under the complete control and authority of his captor, Solomon was captivated by the spellbinding power of her hair. In a word, he was the prisoner of love and she was his captor. Similar imagery appears in an ancient Egyptian love song: “With her hair she throws lassoes at me, with her eyes she catches me, with her necklace she entangles me, and with her seal ring she brands me” (Song 43 in the Chester Beatty Cycle, translated by W. K. Simpson, ed., The Literature of Ancient Egypt, 324). J. S. Deere suggests, “The concluding part of the metaphor, ‘The king is held captive by your tresses,’ is a beautiful expression of the powerful effect of love. A strong monarch was held prisoner by the beauty of his Beloved” (“Song of Solomon,” BKCOT, 206-207). This is a startling statement because Solomon emphasizes that the one who was being held captive like a prisoner in bonds was the “king”! At this point in world history, Solomon was the ruler of the most powerful and wealthy nation in the world (1 Kgs 3:13; 10:23-29). And yet he was held totally captive and subject to the beauty of this country maiden!

³ tn Heb “the courtyard of the guarding” or “place of guarding.” This expression occurs only in the book of Jeremiah (32:2, 8, 12; 33:1; 37:21; 38:6, 12, 28; 39:14, 15) and in Neh 3:25. It is not the same as an enclosed prison which is where Jeremiah was initially confined (37:15-16; literally a “house of imprisoning” [בֵּית הָאֵסוּר, bet ha’esur] or “house of confining” [בֵּית הַכֶּלֶא, bet hakkele’]). It is said to have been in the palace compound (32:2) near the citadel or upper palace (Neh 3:25). Though it was a place of confinement (32:2; 33:1; 39:15) Jeremiah was able to receive visitors, e.g., his cousin Hanamel (32:8) and the scribe Baruch (32:12), and conduct business there (32:12). According to 32:12 other Judeans were also housed there. A cistern of one of the royal princes, Malkijah, was located in this courtyard, so this is probably not a “prison compound” as NJPS interpret but a courtyard adjacent to a guardhouse or guard post (so G. L. Keown, P. J. Scalise, T. G. Smothers, Jeremiah 26-52 [WBC], 151, and compare Neh 12:39 where reference is made to a Gate of the Guard/Guardhouse) used here for housing political prisoners who did not deserve death or solitary confinement as some of the officials though Jeremiah did.

¹ tn Grk “the Passover of the Jews.” This is first of at least three (and possibly four) Passovers mentioned in John’s Gospel. If it is assumed that the Passovers appear in the Gospel in their chronological order (and following a date of a.d. 33 for the crucifixion), this would be the Passover of the spring of a.d. 30, the first of Jesus’ public ministry. There is a clear reference to another Passover in 6:4, and another still in 11:55, 12:1, 13:1, 18:28, 39, and 19:14. The latter would be the Passover of a.d. 33. There is a possibility that 5:1 also refers to a Passover, in which case it would be the second of Jesus’ public ministry (a.d. 31), while 6:4 would refer to the third (a.d. 32) and the remaining references would refer to the final Passover at the time of the crucifixion. It is entirely possible, however, that the Passovers occurring in the Fourth Gospel are not intended to be understood as listed in chronological sequence. If the material of the Fourth Gospel originally existed in the form of homilies or sermons by the Apostle John on the life and ministry of Jesus, the present arrangement would not have to be in strict chronological order (it does not explicitly claim to be). In this case the Passover mentioned in 2:13, for example, might actually be later in Jesus’ public ministry than it might at first glance appear. This leads, however, to a discussion of an even greater problem in the passage, the relationship of the temple cleansing in John’s Gospel to the similar account in the synoptic gospels.

³ tc The witnesses display a bewildering array of variants here. Instead of “the Holy One of God” (ὁ ἅγιος τοῦ θεοῦ, Jo {agio" tou qeou), Tertullian has ὁ Χριστός (Jo Cristo", “the Christ”); C³ Θ* ˹ 33 565 lat read ὁ Χριστὸς ὁ υἱὸς τοῦ θεοῦ (Jo Cristo" Jo Juio" tou qeou, “the Christ, the Son of God”); two versional witnesses (b sy^c) have ὁ υἱὸς τοῦ θεοῦ (“the Son of God”); the Byzantine text as well as many others (Ψ 0250 ˹³ 33 Ï) read ὁ Χριστὸς ὁ υἱὸς τοῦ θεοῦ τοῦ ζῶντος (Jo Cristo" Jo Juio" tou qeou tou zwnto", “the Christ, the Son of the living God”); and Ì⁶⁶ as well as a few versions have ὁ Χριστὸς ὁ ἅγιος τοῦ θεοῦ (“the Christ, the Holy One of God”). The reading ὁ ἅγιος τοῦ θεοῦ is, however, well supported by Ì⁷⁵ א B C* D L W as well as versional witnesses. It appears that Peter’s confession in the Synoptic Gospels (especially Matt 16:16) supplied the motivation for the variations. Although the witnesses in Matt 16:16; Mark 8:29; and Luke 9:20 vary considerably, the readings are all intra-synoptic, that is, they do not pull in “the Holy One of God” but reflect various permutations of “Christ”/“Christ of God”/“Christ, the Son of God”/“Christ, the Son of the living God.” The wording “the Holy One of God” (without “Christ”) in important witnesses here is thus unique among Peter’s confessions, and best explains the rise of the other readings.

¹ tc There is a considerable degree of difference among the mss regarding the presence and position of the doxology of 16:25-27. Five situations present themselves from the ms tradition. The doxology is found in the ancient witnesses in three separate locations: (1) here after 16:23 (Ì⁶¹ א B C D 81 365 630 1739 2464 al co), (2) after 14:23 (Ψ 0209^vid Ï), or (3) after 15:33 (Ì⁴⁶). The situation is further complicated in that some of the mss have these verses in two places: (4) after 14:23 and after 16:23 (A P 33 104 2805 pc); or (5) after 14:23 and after 15:33 (1506). The uncertain position of the doxology might suggest that it was added by later scribes. But since the mss containing the doxology are so early and widespread, it almost certainly belongs in Romans; it is only a question of where. Further, the witnesses that omit the doxology are few: F G 629 Hier^mss. (And of these, G has a blank space of several lines large enough for the doxology to belong there.) Only two positions (after chapter 14 only and at the end of the letter only) deserve particular notice because the situation of the mss showing the doxology in two places dates back to the 5th century. Later copyists, faced with the doxology in two different places in the mss they knew, may have decided to copy the doxology in both places, since they were unwilling to consciously omit any text. Because the textual disruption of the doxology is so early, TCGNT 472 suggests two possibilities: either (1) that Paul may have sent two different copies of Romans – a copy lacking chapter 16 and a copy with the full text of the epistle as we now have it, or (2) Marcion or some of his followers circulated a shortened form of the epistle that lacked chapters 15 and 16. Those mss that lacked chapters 15-16 would naturally conclude with some kind of doxology after chapter 14. On the other hand, H. Gamble (The Textual History of the Letter to the Romans [SD], 123-32) argues for the position of the doxology at 14:23, since to put the doxology at 16:25 would violate Paul’s normal pattern of a grace-benediction at the close of the letter. Gamble further argues for the inclusion of 16:24, since the mss that put the doxology after chapter 14 almost always present 16:24 as the letter’s closing, whereas most of the mss that put the doxology at its traditional position drop 16:24, perhaps because it would be redundant before 16:25-27. A decision is difficult, but the weight of external evidence, since it is both early and geographically widespread, suggests that the doxology belongs here after 16:23. For a full discussion, see TCGNT 470-73.