¹ tn It seems most likely that this refers to a real rooster crowing, although a number of scholars have suggested that “cockcrow” is a technical term referring to the trumpet call which ended the third watch of the night (from midnight to 3 a.m.). This would then be a reference to the Roman gallicinium (ἀλεκτοροφωνία, alektorophōnia; the term is used in Mark 13:35 and is found in some mss [P^37vid,45 ƒ¹] in Matt 26:34) which would have been sounded at 3 a.m.; in this case Jesus would have prophesied a precise time by which the denials would have taken place. For more details see J. H. Bernard, St. John (ICC), 2:604. However, in light of the fact that Mark mentions the rooster crowing twice (Mark 14:72) and in Luke 22:60 the words are reversed (ἐφώνησεν ἀλέκτωρ, ephōnēsen alektōr), it is more probable that a real rooster is in view. In any event natural cockcrow would have occurred at approximately 3 a.m. in Palestine at this time of year (March-April) anyway.

¹ tc A few witnesses, especially of the Western text (D 33 it sy^s Ir^lat Or Eus^syr), do not contain 21:44. However, the verse is found in א B C L W Z Δ (Θ) 0102 ƒ^{1, 13} 565 579 700 1241 1424 M lat sy^c,p,h co. The NA²⁷ and NA²⁸ put this verse in brackets, which normally indicates some doubt as to its authenticity, but Metzger’s Textual Commentary claims that the committee concluded that the verse was an “accretion,” including it not because of authenticity but because of its longstanding role in the textual tradition (TCGNT 47). Some scholars see the verse as scribally borrowed from the parallel in Luke 20:18, but both the wording and word order are different enough that such an imperfect assimilation cannot account for the great variety of witnesses that have a uniform reading here. In light of the lack of adequate explanation for the rise of this verse as it is written, the longer reading should be preferred.

¹ tc ‡ Several key witnesses along with the majority of mss (e.g., B C N W Δ 085 33 565 579 700 1241 M lat sy^h sa^mss) have μαθητάς (mathētas, “disciples”) after τοὺς δώδεκα (tous dōdeka, “the twelve”); still other authorities have μαθητὰς αὐτοῦ (mathētas autou, “his disciples”; Γ 13 892^c 1424 it sy^p sa^mss) after δώδεκα. The personal pronoun is a clarifying addition, but all these mss also throw their weight toward the μαθητάς reading. Other significant witnesses lack the word (e.g. א D L Θ ƒ^{1, 13} 892* sy^s,c bo Or). Up until this point in the Gospel, Matthew speaks of “the twelve” always with qualification (“twelve disciples,” “twelve apostles,” “these twelve”; cf. Matt 10:1, 2, 5; 11:1), but afterward just as “the twelve” (Matt 26:14, 20, 47). Thus, in spite of the strong external evidence, both longer variants look to be scribal clarifications, and hence are considered secondary. NA²⁸ puts μαθητὰς in brackets to show doubts about its authenticity.

¹ tc The majority of Byzantine minuscules, along with a few other witnesses (C³ K L Γ Θ 579 700^txt* 1424^c sa mae), lack νεκροὺς ἐγείρετε (nekrous egeirete, “raise the dead”), most likely because of oversight due to a string of similar endings (-ετε in the second person imperatives, occurring five times in v. 8). The longer version of this verse is found in several diverse and ancient witnesses such as א B C* (D) N 0281^vid ƒ^{1, 13} 33 565 579^mg lat bo; P W Δ 348 sy^h have a word-order variation, but nevertheless include νεκροὺς ἐγείρετε. Although some Byzantine-text proponents charge the Alexandrian witnesses with theologically-motivated alterations toward heterodoxy, it is interesting to find a variant such as this in which the charge could be reversed (do the Byzantine scribes have something against the miracle of resurrection?). In reality, such charges of wholesale theologically-motivated changes toward heterodoxy are immediately suspect due to lack of evidence of intentional changes (here the change is evidently due to accidental omission).

¹ sn A quotation from Hos 6:6 (see also Matt 12:7). The statement both in the Hebrew text of Hosea and the Greek text of Matthew creates an apparent antithesis between mercy and sacrifice. Even among the church fathers, some understood this to be an absolute rejection of sacrifice by Jesus, and to signal the end of the sacrificial cult with the arrival of the new covenant. This interpretation is unlikely, however, both for Hosea and for Matthew. The LXX renders the Hebrew text of Hos 6:6 as comparative: “I want mercy more than sacrifice,” and this is probably closer to Hosea’s meaning (see the note at Hos 6:6). Such an understanding is also consistent with Jesus’ teaching elsewhere in Matthew (e.g. 5:18-24; 23:23-28). Obedience to the law is important, but even more important is to show mercy to those who are in dire need, as demonstrated by Jesus himself in his ministry of healing (alluded to in Matt 9:12 with the imagery of the physician, and in Matt 9:1-8 by the healing of the paralytic).

² tn Grk “what to us and to you?” (an idiom). The phrase τί ἡμῖν καὶ σοί (ti hēmin kai soi) is Semitic in origin, though it made its way into colloquial Greek (BDAG 275 s.v. ἐγώ). The equivalent Hebrew expression in the OT had two basic meanings: (1) When one person was unjustly bothering another, the injured party could say “What to me and to you?” meaning, “What have I done to you that you should do this to me?” (Judg 11:12, 2 Chr 35:21, 1 Kgs 17:18). (2) When someone was asked to get involved in a matter he felt was no business of his own, he could say to the one asking him, “What to me and to you?” meaning, “That is your business, how am I involved?” (2 Kgs 3:13, Hos 14:8). These nuances were apparently expanded in Greek, but the basic notions of defensive hostility (option 1) and indifference or disengagement (option 2) are still present. BDAG suggests the following as glosses for this expression: What have I to do with you? What have we in common? Leave me alone! Never mind! Hostility between Jesus and the demons is certainly to be understood in this context, hence the translation: “Leave us alone….”

¹ tc The textual tradition here is quite complicated. A number of mss (B C (Δ) Θ sy^s,p,h) read “Gadarenes,” which is the better reading here. Many other mss (א² L W ƒ^{1, 13} 565 579 700 1424 M al bo) have “Gergesenes.” Others (892^c latt sy^hmg sa mae) have “Gerasenes,” which is the reading followed in Luke 8:26. The difference between Matthew and Luke may be due to uses of variant regional terms. Of the three readings, Gergesa is most likely the right location for this exorcism (the only region close to the Sea of Galilee and with a steep bank [κρημνός in Mark 5:13]) but almost surely a secondary reading in all the Synoptics. As Baarda articulated, this variant is quite possibly due to a conjecture made by Origen, a reading which then made its way into sevral mss (Tjitze Baarda, “Gadarenes, Gerasenes, Gergesenes and the ‘Diatassaron’ Traditions,” in Neotestamentica et Semitica: Studies in Honour of Matthew Black, ed. E. Earle Ellis and Max Wilcox [Edinburgh: T&T Clark, 1969], 181-97).

¹ tc The majority of mss read the word εἰκῇ (eikē, “without cause”) here after “brother.” This insertion has support from א² D L W Γ Δ Θ 0233 ƒ^{1, 13} 33 565 579 700 1241 1424 M it sy co Ir^lat Or^mss Cyp Cyr. Thus the Western and Byzantine groups, as well as several other witnesses, all include the word, while the best Alexandrian and some other witnesses (P⁶⁴ א* B aur vg Or Hier^mss) lack it. The ms evidence favors its exclusion, though there is a remote possibility that εἰκῇ could have been accidentally omitted from these witnesses by way of homoioarcton (the next word, ἔνοχος [enochos, “guilty”], begins with the same letter). An intentional change would likely arise from the desire to qualify “angry,” especially in light of the absolute tone of Jesus’ words. While “without cause” makes good practical sense in this context, and must surely be a true interpretation of Jesus’ meaning (cf. Mark 3:5), it does not commend itself as the original wording.

² sn The origin of the concept of “the day of the Lord” is uncertain. It may have originated in the ancient Near Eastern idea of the sovereign’s day of conquest, where a king would boast that he had concluded an entire military campaign in a single day (see D. Stuart, “The Sovereign’s Day of Conquest,” BASOR 221 [1976]: 159-64). In the OT the expression is applied to several acts of divine judgment, some historical and others still future (see A. J. Everson, “The Days of Yahweh,” JBL 93 [1974]: 329-37). In the OT the phrase first appears in Amos (assuming that Amos predates Joel and Obadiah), where it seems to refer to a belief on the part of the northern kingdom that God would intervene on Israel’s behalf and judge the nation’s enemies. Amos affirms that the Lord’s day of judgment is indeed approaching, but he declares that it will be a day of disaster, not deliverance, for Israel. Here in Zephaniah, the “day of the Lord” includes God’s coming judgment of Judah, as well as a more universal outpouring of divine anger.

¹ tc The consonantal text repeats אסף אסף with the vowels for the Qal infinitive absolute of אָסַף (ʾasaf, “gather up, retract”) followed by a Hiphil first person jussive form of סוּף (suf, “come to an end”). A similar combination appears in Jer 8:13, but it is textually disputed based on the LXX. Here a literal translation might be, “Withdrawing, I will bring to an end.” While this English rendering may sound reasonable, this is very unusual Hebrew grammar and the small number of similar cases are textually disputed. Some prefer to emend the text so that the infinitive and finite form of the verb are from the same root and same stem. Some render as “I will certainly sweep away” (NIV, ESV, Holman), probably assuming a Hiphil of אָסַף, though this root does not otherwise occur in the Hiphil, and if it did, it should mean “I will remove” (NASB). HALOT includes a Qal stem gloss “to destroy” (HALOT 74 qal 4), but this meaning is applied only to this example and one other textually disputed reference, that is, the dictionary’s gloss is merely accommodating this problem and is not evidence. Read as Hiphil forms of סוּף, the text would mean “I will certainly bring to an end,” which is conceptually similar to destroying. For a discussion of proposals see J. J. M. Roberts, Nahum, Habakkuk, and Zephaniah (OTL), 167, 169.

⁵ tc The MT reads the Qal participle הַמֹּכֶרֶת (hammokheret) which is derived from מָכַר (makhar, “to sell, to betray”): “the one who sells/betrays [nations].” The MT is supported by the LXX. The Dead Sea Scrolls read הממכרת (4QpNah 2:7): “the one who sells/betrays [nations]” (see DJD 5:38). Dahood repoints the MT as a Hophal participle, הַמֻּכֶּרֶת (hammukkeret) from נָכַר (nakhar, “to know, to recognize”): “the one who is known [by the nations for her harlotries]” (M. Dahood, “Causal Beth and the Root NKR in Nahum 3.4,” Bib 52 [1971]: 395-96). The BHS editors suggest emending the MT, due to metathesis, to הַכֹּמֶרֶת (hakkomeret) from II כמר (“to ensnare”; HALOT 482 s.v. II כמר) which is related to Assyrian kamaru [A] (“to ensnare”): “The one who ensnares [nations].” The related nouns “snare; net” (מִכְמָר, mikhmar) and “net” (מִכְמֶרֶת, mikhmeret) are used as metaphors of the wicked destroying their victims (Ps 141:10; Isa 51:20; Hab 1:15, 16). This approach is adopted by NJPS: “who ensnared nations.” Others suggest emending to the Qal participle הַכֹּמֶרֶת from III כמר (“to destroy, to overthrow”; BDB 485 s.v. III כמר) related to Assyrian kamaru [B] (“to destroy; to annihilate”): “the one who destroys nations.” The MT may be retained due to strong external support (LXX and 4QpNah) and adequate internal support; the conjectural emendations are unnecessary.

⁶ tc The MT reads the Hophal perfect third person masculine singular וְהֻכַן (v^ehukhan, “and [it] is prepared”). On the other hand, the LXX reading reflects the Hiphil perfect third person common plural וְהֵכִינּוּ (v^ehekhinnu, “and they will prepare”). Arguing that the active sense is necessary because the three preceding verbs are all active, K. J. Cathcart (Nahum in the Light of Northwest Semitic [BibOr], 95) suggests emending to the Hiphil infinitive absolute וְהָכִין (v^ehakhin, “and [they] prepare”). However, the Masoretic form should be retained because it is the more difficult reading that best explains the origin of the LXX reading. The shift from active to passive verbs is common in Hebrew, marking a cause-result sequence (e.g., Pss 24:7; 69:14 [15]; Jer 31:4; Hos 5:5). See M. Weinfeld, “The Active-Passive (Factitive-Resultive) Sequence of Identical Verbs in Biblical Hebrew and Ugaritic,” JBL 84 (1965): 272-82.

³ tc The MT reads the Niphal imperfect third person masculine plural יִכָּשְׁלוּ (yikkash^elu, “they stumble”) from the root כָּשַׁל (kashal, “stumble”). G. R. Driver argues that the MT makes little sense in the portrayal of a successful assault; the motif of stumbling warriors usually connotes defeat (Isa 5:27; Jer 46:6). Driver argues that MT’s יִכָּשְׁלוּ (“they stumble”) arose from metathesis (reversal of consonants) from an original יִשָּׁלְכוּ (yishal^ekhu, Niphal from שָׁלַךְ [shalakh, “to cast forth”]) which also appears in 1 Kgs 13:24, 25, 28 (“hurled himself,” i.e., rushed headlong). Driver suggests that this is related to Arabic salaka VII (“to rush in”). He notes that the emendation would produce a tighter parallelism with the following noun: יְמַהֲרוּ (y^emaharu, “they hasten”). See G. R. Driver, “Linguistic and Textual Problems: Minor Prophets II,” JTS 39 (1938): 270. On the other hand, Armerding argues that the anomalous MT reading יִכָּשְׁלוּ (“they stumble”) can be explained without recourse to textual emendation. The stumbling of the attacking army is caused, not by their weakness, but by the corpses of the Assyrians strewn in their path which obstructs their advance. Armerding suggests that this motif appears in Nah 3:3 (C. E. Armerding, “Nahum,” EBC 7:475).

⁵ tn Heb “the steel.” The Hebrew term פְּלָדוֹת is a hapax legomenon. The corresponding noun פְּלָדָה (p^eladah) probably means “metal, steel” (BDB 811 s.v. פְּלָדָה; HALOT 761 s.v. פְּלָדָה), and it is probably related to Arabic puladu, Syriac pldʾ, and early Persian fulad (all of which mean “steel”). This rendering is followed by NASB, NIV, NRSV. The term פְּלָדוֹת (“steel”) probably refers to the metallic pole attachments for the chariot spears, the side armor of the chariots, or the steel scythes fastened to the axle of a chariot. Xenophon described the army of Cyrus in a similar manner; the side armor of the chariots and the breastplates and thigh-pieces of the chariot-horses were “flashing with bronze” (Xenophon, Cyropaedia 6.4.1). On the other hand, Cathcart connects Hebrew פְּלָדָה to Ugaritic paladu, which means “a garment made of linen hair,” and suggests that פְּלָדוֹת הָרֶכֶב (peladot harekhev) refers to the coverings, blankets, or caparisons of chariot horses (K. J. Cathcart, Nahum in the Light of Northwest Semitic [BibOr], 88). This demands that הָרֶכֶב be nuanced “chariot horses”—a problem when it means “chariots” in Nah 2:4; 3:2.

⁶ tc The BHS editors suggest revocalizing the Masoretic noun מְצֻרָה (m^etsurah, “rampart”) to the noun מַצָּרָה (matsarah, “the watchtower”) from the root נָצַר (natsar, “to watch, guard”). This would create a repetition of the root נָצַר which immediately precedes it: מַצָּרָה נָצוֹר (natsor matsarah, “Watch the watchtower!”). However, the proposed noun מַצָּרָה (“the watchtower”) appears nowhere in the Hebrew Bible. On the other hand, the Masoretic reading מְצֻרָה (“rampart”) and the related noun מָצוֹר (matsor, “rampart”) appear often (Pss 31:22; 60:11; Hab 2:1; Zech 9:3; 2 Chr 8:5; 11:5, 10, 11, 23; 12:4; 14:5; 21:3; 32:10). Thus, the Masoretic vocalization should be preserved. The LXX completely misunderstood this line. The LXX reading (“one who delivers out of tribulation”) has probably arisen from a confusion of the MT noun נָצוֹר (“guard”) with the common verb נָצַר (“deliver”). It also reflects a confusion of MT מְצֻרָה (“road, rampart”) with מִצְּרָה (mits^erah, “from distress”).

³ tn Or “are strong” (cf. NCV); or “are at full strength” (NAB, NRSV); or “are intact.” Alternately, “Even though they have allies” (cf. NIV, NLT). The Hebrew noun שְׁלֵמִים (sh^elemim, from שָׁלֵם [shalem]) means “complete, healthy, sound, safe, intact, peaceful” (BDB 1023-24 s.v. שָׁלֵם; HALOT 1538-1539 s.v. שָׁלֵם). It can connote “full strength” or “full number” of an object (Gen 15:16; Deut 25:15; Prov 11:1; Amos 1:6, 9). Most commentators view this as a reference to the strength or numbers of the Assyrian army: “strong” (R. L. Smith, Micah-Malachi [WBC], 77-78), “full strength” (NASB, NRSV) or “intact” (T. Longman, “Nahum,” The Minor Prophets, 2:798). On the other hand, NIV and NLT follow the lead of Wiseman who points out that שְׁלֵמִים can refer to military allies: “Even though they will have allies and so be all the more numerous” (D. J. Wiseman, “Is It Peace? Covenant and Diplomacy,” VT 32 [1982]: 311-26). Nahum refers to the allies of the Assyrians elsewhere (Nah 3:15-17).

² tn Less likely, “[What are you] thinking about.” When used with אֶל (ʾel) the verb חָשַׁב (khashav) may be taken (1) in a hostile sense: “What are you plotting against the Lord?” or (2) in a nonhostile sense: “What are you thinking about the Lord?” The hostile sense is clearly intended when it is used in collocation with the direct object רָעָה (raʿah, “evil”; Zech 7:10; 8:17; Pss 35:4; 140:3; Prov 16:9) or when it is followed by the preposition עַל (ʿal; Gen 50:20; 2 Sam 14:13; Jer 11:19; 18:11, 18; 29:11; 48:2; 49:30; Mic 2:3; Nah 1:11; Ps 36:5; Esth 8:3; 9:24, 25; Dan 11:25). It is also used in a hostile sense when followed by the preposition אֶל, as it is here (Jer 49:20; 50:45; Hos 7:15; Nah 1:9). The major lexicons classify this usage in a hostile sense (BDB 363 s.v. חָשַׁב; HALOT 360 s.v. חשׁב). The verb is repeated in Nah 1:11 where it is clearly used in a hostile sense.

³ tn Or “is upheaved”; or “heaves.” There is debate whether the originally unpointed Hebrew verb וַתִּשָּׂא (vattissaʾ) should be vocalized as וְתִּשָּׂא (v^ettissaʾ; NASB “is upheaved”; NRSV, NJPS “heaves”) from the root נָשָׂא (nasaʾ, “to lift up”) or as וַתִּשָּׁא (vattishaʾ, “is devastated, laid waste”) from the root שָׁאָה (shaʾah, “to devastate, lay waste,” see HALOT 1367 s.v. 1 שׁאה). The vocalization וְתִּשָּׂא is attested in the Masoretic tradition and the Greek versions: Origen (“was raised up”), Symmachus (“was moved”), and Aquila (“shivered”). However, וְתִּשָּׂא demands an intransitive (“heaves”) or passive (“is upheaved”) sense which is not attested for the Qal stem. The vocalization וַתִּשָּׁא (“is devastated, laid waste”) is supported by the Syriac and Vulgate. The revocalization of the MT וְתִּשָּׂא (“is lifted up”) to וַתִּשָּׁא (“is devastated”) is suggested by the BHS editors and several Hebrew lexicons (HALOT 726 s.v. נשׁא; BDB 670-71 s.v. נָשָׂא). The revocalization involves only the difference between the form שׂ (sin) and שׁ (shin) and is followed in the present translation.

⁵ tn The verb נָטַר (natar) is a synonym of נָצַר (natsar) andשָׁמַר (shamar), each including a meaning “to keep, guard.” נָטַר (natar) is used elsewhere of keeping a vineyard (Song 1:6; 8:11-12) and guarding a secret (Dan 7:28). It describes a person bearing a grudge, seeking revenge, and refusing to forgive (Lev 19:18). It is also used in collocation with לְעוֹלָם (l^eʿolam, “forever, always”) and לָעַד (laʿad, “continually”) to picture God harboring rage against his enemies forever (Jer 3:5, 12; Amos 1:11; Ps 103:9). Contra HALOT (695 s.v.), it does not mean “to control anger” or “to be slow to anger” and is not collocated with אָף (ʾaf, “anger”) as the entry implies (see TWOT 2:576 and NIDOTTE 1581, s.v.). The long-term rage depicted by נָטַר (“maintain rage”) serves as an appropriate bridge to the following statement in Nahum that the Lord is slow to anger but furious in judgment. God seeks vengeance against his enemies; he continually rages and maintains his anger; he is slow to anger, but will eventually burst out with the full fury of his wrath.

² sn Required three days to walk through it. Although this phrase is one of the several indications in the book of Jonah of Nineveh’s impressive size, interpreters are not precisely sure what “a three-day walk” means. In light of the existing archaeological remains, the phrase does not describe the length of time it would have taken a person to walk around the walls of the city or to walk from one end of the walled city to the other. Other suggestions are that it may indicate the time required to walk from one edge of Nineveh’s environs to the other (in other words, including outlying regions), or that it indicates the time required to arrive, do business, and leave. More information might also show that the phrase involved an idiomatic description (consider Gen 30:36 and Exod 3:18; a three-day-journey would be different for families than for soldiers, for example), rather than a precise measurement of distance, for which terms were available (Ezek 45:1-6; 48:8-35). With twenty miles as quite a full day’s walk, it seems possible and simplest, however, to take the phrase as including an outlying region associated with Nineveh, about sixty miles in length.