232 . See also Safrai 1974/76a: 965; Sifre Deut. 43.3.7.

232 . See also Safrai 1974/76a: 965; Sifre Deut. 43.3.7.

. Garland 1993: 14-15 rightly cites Quint. Inst. Or. 3.7.10 here. Considering the broader context, Overman 1996: 30 points out the importance of origin stories in Greco-Roman antiquity (e.g., the Aeneid).

. One could extend this principle further, e.g., in the Lukan perspective of the Gentile mission rooted in Israel’s faith; at least some historians explicitly noted that history set individual lives in a broader context (Diod. Sic. 1.2.3).

. The chiastic appearance of Abraham and David in the genealogy (cf. Bauer 1996: 141) may reinforce this emphasis, although a nonchiastic parallelism could have accomplished the same, and Matthew may mention David first to recognize Jesus’ unique messianic status not shared with Israel.

. Cf. Kingsbury 1975: 12; Longenecker 1975: 141-43; F. F. Bruce 1972a: 64-65.

. Later Jewish teachers (especially in those circles that later produced rabbinic Judaism) not only regarded Abraham as the ideal Israelite (even the model Pharisee in p. Soṭa 5:5, §2) but also as the first Gentile convert to Judaism (Pesiq. R. 11:4; Montefiore 1979: 43; Urbach 1979: 1:32; Sandmel 1978b: 181; cf. also on his proselytization of others, e.g., ARN 12A; 26, §54). Given Matthew’s emphases in this genealogy, he could appeal to either or both nuances, though the Israelite nuance is stronger. Second-century rabbis may have toned down messianic speculation, hence the Son of David image (widespread in some parts of Judaism) after Bar Kochba, but even here it reemerged in later sources; cf. b. Sanh. 97b-98a; p. Sukk. 5:1, §7; Gen. Rab. 88:7; Song Rab. 2:13, §4; Pesiq. R. 15:14/15.

. Many scholars favor this view: see, e.g., Martin 1977: 246; Davies 1966b: 14-15; Argyle 1963: 24; Meier 1980: 4; Gundry 1982: 13, 19; Davies and Allison 1988: 161-65; Lachs 1987: 3. Numbers could be used symbolically to spell names (e.g., Lucian Alex. False Prophet in LCL 4:190-91); on gematria see, e.g., Russell 1993; N. Cohen 1992; Lieberman 1987.

. Some find six weeks of generations preparing for an eschatological sabbath (e.g., Foresti 1984; cf. Meier 1980: 3); other interpreters (e.g., Lake and Cadbury 1979: 151; see Johnson 1988: 200) detect three 490-year periods. If the traditions behind 2 Bar. 53–54 existed in Matthew’s day, 14 may have appealed to Matthew from its use there.

. However one counts, not all Matthew’s sets come to precisely 14; but Jewish interpreters did not expect numerical exactitude (cf. the more than three sayings of R. Eliezer b. Hyrcanus in m. ʾAbot. 2:10), and ancients could vary between inclusive and exclusive counting (Blomberg 1992a: 53 cites m. ʾAbot 5:1-6). Fourteen was a reasonable round number of generations (cf. 16 in Dion. Hal. 1.9.4).

. This is generally accepted, even if cautiously (e.g., Meier 1991a: 216-19). B. Sanh. 43a, bar., may preserve a tradition that Jesus was of royal lineage (unless it suggests connections with the Herodian or Roman rulers, or that he was about to take control of the people; both views are unlikely).

. Albright and Mann 1971: 4 are surely correct that at least one and possibly more of the omissions stem from Matthew’s use of the Greek version of the Old Testament here (1 Chron 3:11 LXX; see also Johnson 1988: 186). But because Matthew must have counted as well as transcribed the names from the Chronicler, it is unlikely that he skipped three generations by having his eye slip a line (ibid.). Either he adjusted the numerical scheme (as I think more likely) or his manuscript of Chronicles had already made the error Albright and Mann suggest.

. Cf. also Feuillet 1988. Although we lack definite proof, preference for clan-based endogamous marriages (cf. Malina and Rohrbaugh 1992: 100) plus some archaeological evidence for group migrations from Judea to the Nazareth area could allow the possibility that both are from the Davidic line.

. Johnson 1988: 142 summarizes their position. Nettelhorst 1988 suggests that Luke traces the lineage via Joseph’s father, Matthew via Joseph’s maternal grandfather.

. See Schweizer 1975: 24; Gundry 1982: 15-16; Davies and Allison 1988: 175. Taken collectively, these changes cannot be viewed as merely orthographic, pace Johnson 1988: 182. By contrast, Josephus presents Asa largely positively and even tones down Manasseh’s idolatry; cf. Feldman 1994b; 1991.

. Even in Israel’s exile he will find messianic significance (2:17-18).

. One might wonder, though one cannot test, whether Joseph is the son of “Jacob” to allude to the climactic hero of the Genesis story (Gen 37–50; cf. Lk 3:23).

. Jer. 22:24-30 could suggest that Jeconiah’s line was unfit for royal heirs, but already in the OT his line was restored (1 Chron 3:17-19; cf. 2 Kings 25:27-30; Jer 52:33; Hag 2:23; cf. the ration tablets in ANET 308). Possibly Jeconiah repented (cf. 2 Chron 33:12-13; Jer. 18:5-10); this was also the rabbis’ explanation of his restoration (Johnson 1988: 184, following Strack-Billerbeck). That Zerubbabel (exalted as a deliverer in 1 Esdr. 3–4) is listed as Shealtiel’s son indicates that he inherited through the royal line of the firstborn; perhaps Shealtiel died without offspring (see 1 Chron 3:17-19; cf. Deut 25:5-10). Some later traditions are confused; despite Ezra’s priestly lineage in Ezra 7:1-5, 4 Ezra 3:1 identifies him with Shealtiel. For comments on the textual problem, see Hagner 1993: 6.

. Greek readers knew the first book of the Bible as “Genesis” (its title in the LXX). The phrase “the book of” could introduce an entire book (e.g., Tob 1:1; Rev 1:1; cf. PGM 13.343). But in light of the inclusio in 1:1, 18 and Genesis’s application of the term to genealogies (in 2:4; 5:1; cf. 10:1), most scholars concur that this title applies to the genealogy, not to the whole Gospel (e.g., Kingsbury 1975: 10; Meier 1980: 4; Gundry 1982: 13; Stanton 1993: 12-13; Nolland 1996: 469-70), although in view of strong pointers in both directions Davies and Allison 1988: 150-53 ultimately contend for a double entendre (cf. Bauer 1988: 73-77 for the proposal that it applies to the section 1:1–4:16). Malina and Rohrbaugh 1992: 24 and Bauer 1996: 159 connect the “genesis” allusion here with the “end of the age” passage in 28:18-20, the former also noting parallels there with the edict that probably closes the Hebrew canon (2 Chron 36:23; but stronger parallels exist).

. Later teachers often praised these matriarchs (Jub. 36:23-24; Pesiq. Rab Kah. 11:6; see also Kaunfer 1995), e.g., Sarah (1QapGen 20.2-10; Test. Abr. 3, 5, 6A; ARN 26, §54B; Gen. Rab. 20:6; 41:2; Num. Rab. 14:11; Pesiq. Rab Kah. 22:1; Pesiq. R. 43:4); for the popularity of her name, cf. Tob 3:7; CIJ 1:402, §543; 1:431, §593; 1:437, §608; 2:141, §942; 2:205, §§1124-25, 2:218, §1169; CPJ 1:169, §26; 1:187-88, §41; 2:124, §§179-80.

. This particular element of the genealogy appears without biblical precedent (Rahab and Salmon were as many as two centuries apart) and, as far as we know, Jewish tradition (Davies and Allison 1988: 173; though some later depict her as ancestor of priests and prophets — Ruth Rab. 2.1); she may represent Matthew’s way of drawing attention to and emphasizing his point by working another prominent Gentile woman into the narrative (cf. 15:22).

. Schweizer 1975: 25; Reicke 1974: 118; F. F. Bruce 1972a: 65; Gundry 1982: 15; Garland 1993: 18; Reeves 1993: 93. Nolland 1997: 537-39 is cautious about linking Bathsheba with the other women, because she is mentioned only by her former husband’s name.

. A.-J. Levine (1988: 71-80) and others have raised reasoned objections to the Gentile interpretation, but Bauer 1996: 148-49 provides a sufficient response.

. Brown 1977: 71-74; Stendahl 1995: 74; Stauffer 1960: 17; Tasker 1961: 32; Smith 1978: 26, 66-67; Laws 1980: 137; Harrington 1982: 13-14; Edwards 1985: 11; Freed 1987; Blomberg 1991; idem 1992a: 55-56; cf. A. B. Bruce 1979: 62-63; Lachs 1987: 2; Heil 1991.

. The late tradition in b. Shab. 56a attributes some guilt to Uriah, but to soften David’s; especially after rabbinic patriarchs claimed a Davidic heritage, many sages exonerated David in the affair (Shimoff 1993). The earliest tradition does not emphasize Rahab as a “prostitute” (see Lane 1991: 379; Johnson 1988: 162; Jos. Ant. 5.9).

. The view that these women were sinful or scandalous further fails to explain virtuous Ruth, whose most notable literary characteristic for Jewish readers was simply her ethnic alienation from Israel. Josephus draws a moral for his Gentile readers from this ancestor of David being so lowly (Ant. 5.337). But Jewish readers recalled especially her Moabite ancestry. Although a Moabite was not officially permitted to marry into Israel (Deut 23:3), God accepted Ruth the Moabitess because she followed him at a cost to herself (Ruth 1:16). (Jewish interpreters generally argued instead that Deuteronomy intended to exclude only male Moabites — m. Yebam. 8:3; Sifre Deut. 249.1.1-2; Pesiq. R. 49:2.) Subsequent Jewish tradition praised Ruth, Rahab, and Tamar (e.g., Ps-Philo 9:5; ARN 45, §125 B; Pesiq. R. 40:3/4; 49:2; Bamberger 1968: 195-99; Johnson 1988: 159-66; cf. Heb 11:31; b. B. Bat. 91a; Daube 1973: 33); in some traditions many prophets descended from Rahab (Sifre Deut. 357.2.1; cf. Pesiq. Rab Kah. 13:5).

. One could deride another person by ridiculing his low birth, e.g., Josephus’s employing common anti-Egyptian sentiment (Apion 2.41). Marrying an unconverted Gentile was reprehensible (e.g., Gen. Rab. 65:2).

. Some later rabbis also tried to whitewash Solomon’s intermarriages with pagans by suggesting that they were for proselytization (e.g., p. Sanh. 2:6, §2), but 1 Kings 11:1-13 clearly portrays Solomon’s actions in a negative light; the wives influenced Solomon more than the reverse.

. From Jesus’ baptism to his resurrection, Matthew largely follows Mark’s basic chronological and geographical framework (Schnackenburg 1985: 29).

. Hill 1972: 77 prefers to speak in terms of haggadic elaboration rather than pure midrashic creation. Such haggadic elaboration or deletion could be used to answer questions posed by a narrative (e.g., Demetrius the Chronographer Fr. 5 in Euseb. P.E. 9.29.16; Jub. 4:1, 9; 12:14, 13:11, 27:1, 4-5; p. Ketub. 12:4, §8); to make a better story (Jub. 11:14-15, 13:18, 22), e.g., by adding names (Jub. 11:14-15; Lives of the Prophets 19, Joad [§30 in Schermann]; Ps-Philo 40:1; Rook 1990); to heighten the praise of God or the protagonist (e.g., 2 Macc 2:1-8; Jub. 29:14-20; Test. Job 9-15; Test. Jos. 3:1), albeit sometimes by fanciful midrash (Pesiq. Rab Kah. 4:3; cf. Artapanus on Pharaoh’s behavior toward Moses in light of 1 Sam 18:17, 21-25 in Euseb. P.E. 9.27.7). Similarly, negative incidents could be toned down (Ps-Philo 12:2-3; Test. Job 39:12-13 (OTP)/39:9-10; 40:3/4), omitted (e.g., Jub. 13:17-18; 14:21–16:22; 29:13; Test. Zeb. 1:5-7), or justified (CD 4:20–5:3; Jub. 19:15-16; 27:6-7; 28:6-7; 30:2-17, 41; 1QapGen 20.10-11; Jos. and Asen. 23; Test. Jud. 8-12; Test. Iss. 3:1) in the character’s favor. Viviano 1990a suspects that 1 Tim 1:4 polemicizes against narratives like Matthew’s infancy narratives as “myths and genealogies.” Yet the argument works only on the prior assumption that Matthew’s material is mythical; further, the narratives’ theology as a whole hardly fits the object of that epistle’s polemic (cf. 1 Tim 1:6; 4:3; 6:5).

. Noah’s infant glory (1 Enoch 106:2-3; 1QapGen col. 2; cf. Fitzmyer 1965; idem 1971: 77-79; Brown 1966: 51) may resemble Greek myths in which lightning flashes from a divine or semidivine infant (e.g., Asclepius in Paus. 2.26.5). Later tradition amplified Moses’ birth (Bowman 1977: 284-97; contrast Jub. 47:1-8; Ezek Exag. 12-31), especially claiming that he was born circumcised (Ps-Philo 9:13; ARN 2A; b. Soṭa 12a; Lev. Rab. 20:1), a rare phenomenon (p. Yebam. 8:1, §12; Gen. Rab. 26:3; 46:12; 84:6). Such Moses traditions likely stemmed from exegetical inference from Ex 2:2 (cf. Philo Mos. 1.9; Jos. Ant. 2.230-31), but some rabbis may have polemicized against what they considered undue exaltation of Noah (cf. Gen. Rab. 28:8; 29:1, 3; 26:6; 36:3), also the ancestor of the Gentiles, which may explain the unexpected dearth of haggadic expansion about his birth in these sources.

. Sanders 1993: 88-89 believes that Matthew and Luke mined the OT to fill in incomplete information for the infancy stories (so also Meier 1991a: 213), but acknowledges “limits,” as in other Jewish texts and as evidenced by lack of assimilation to David and limits to assimilation with Moses.

. Von Dobschütz 1995: 34 saw in the quotations evidence of Matthew’s scribal method. Some think the fulfillment quotations, which vary from what they take to be Matthew’s characteristic use of the LXX, derive from scribes in his community (e.g., Luz 1989: 78). But Matthew’s mixed-type citations may reflect his own style and themes (Stanton 1993: 358-63), and the formula quotations function as theological “asides” (Stanton 1993: 346). Some writers’ narrative asides include the fulfillment of prophetic oracles (e.g., Diod. Sic. 31.10.2), but virtually all of Matthew’s explicit asides are fulfillment quotations.

. See Liddell and Scott 1968: 1339; Moulton and Milligan 1976: 494; Delling 1967: 828-30; Ford 1966; cf. Carmignac 1977; P.Oxy. 3177.2-3 in Horsley 1981: 71. Ilan 1996: 98 suggests that the story reflects the Jewish use of virginity as “a symbol of purity and innocence”; her evidence (funerary inscriptions citing virginity), however, is inadequate (the inscriptions invite pathos; see comment on 9:18-26).

. See Craig 1986; cf. Goppelt 1981/82: 1:145; France 1976: 105-7; Borg 1987: 33-34; on antisupernaturalism regarding the infancy narratives in general, see Young 1992; on the historical context of modern skepticism, cf. Benoit 1973/1974: 1:39; Kee 1983: 3-16; Bockmuehl 1988.

. For announcement by a dream, cf., e.g., Plut. Alex. 2.1–3.2 (Boring 1995: 37-38); for miraculous omens, cf., e.g., Suet. Augustus 94 (Boring 1995: 43-44); Aul. Gel. 6.1.2-4. For supernatural births of famous philosophers, Diog. Laert. Lives 3.1-2, 45; Iamblichus Life of Pythagoras 2.3-5; Diod. Sic. Library of History 4.9.1-10 (Boring 1995: 33-35, though every one of these sources for supernatural births derives from at least half a millennium after the persons they depict). Davies 1993: 31-32 cites examples of miraculous births in hellenistic biographies (including Plut. Romulus 2; Alex. 2), but notes that these stories presuppose sexually active deities producing semidivine heroes, and that Matthew drew not on that Greek tradition but on the biblical tradition of God as creator. One might also think of Gen 2:21-22: as God formed Eve from Adam, he could form Jesus from Mary’s ovum.

. Hairdressers were not always associated with immorality, though they catered to people of means; cf. Ilan 1996: 188.

. Klausner 1979b: 23-24, 48-51; cf. Goguel 1948: 74; Stauffer 1960: 16-17; Herford 1966: 35-50; Maier 1978: 198-200; Origen Contra Celsum 1.28, 32, 33, 39. Later rabbinic sources also reserved tortures for a particular Miriam hanging by her breasts in hell (e.g., t. Sukk. 4:28; p. Ḥag. 2:2, §5; Sanh. 6:6, §2). Insulting another’s mother’s sexuality was one of the deepest and most punishable insults (e.g., Jos. Ant. 13.292, 293-95). Hanson and Oakman 1998: 57 thus take too seriously the occurrence of one tombstone of a soldier in Syria named Pantera; even were the story true, the odds of chance preservation of precisely the right tombstone are minimal.

. Cf. also A. B. Bruce 1979: 66; Harrington 1982: 14; Meier 1980: 9; Garland 1993: 21. Ancients did not hesitate to speak of a mortal “father” who raised a child despite its being reportedly conceived by a deity (cf. Diog. Laert. Lives 3.1-2, 45, in Boring 1995: 33-34; Eurip. Madness of Heracles 587, 1192). By his faith in God’s promise concerning a supernaturally born son, Joseph may also act as a “son of Abraham” (1:1), but too few clues in the narrative support such a connection for it to be more than conjectural.

. Proposed parallels to divine “spirits” impregnating humans or organisms (Plut. Numa 4.2-4; Table-Talk 8.1-3; Boring 1995: 38-39) are not close, and such thoughts would have been out of the question for Jewish readers. More helpfully, cf. comments on the Spirit and creation (e.g., Hill 1972: 78), an ancient Israelite connection (Gen 1:2; Job 33:4; Ps 104:30).

. Matthew may have the whole section of Is 7–9 in view. Ellis admits that Matthew lacks evidence for testimonia in the sense of an apologetic book, but he does seem to work from a set of passages (Ellis 1977: 201; cf. Lindars 1961: 13-14; Longenecker 1975: 89-90; Stendahl 1968: 207-17). The Dead Sea Scrolls contain some possibly analogous collections of eschatological texts (cf. Allegro 1956b; Fitzmyer 1974: 59-89; in the rabbis, cf. Urbach 1979: 1:686).

. Although Boring 1995: 89 cites the moral in Aesop Fables 172 to illustrate the way ancients might have approached Jesus’ parables, I suspect it works much better for nonparabolic narratives. (On the special hermeneutic for parables, see under Mt 13.) Eighteenth-century English historiography also followed this Greek and Roman historical tradition (Frei 1989: 56).

. Among Gentiles, cf. Epict. Encheir. 40; for young brides cf. also, e.g., Plut. Bride 2, Mor. 138D; Quint. Preface 4; Ninus Romance fr. 1, A.3; 13 in Ovid Metam. 9.714. This was often 13 to 16 among Greeks and Romans (Friedländer 4:123-31; Lewis 1983: 55; cf. Pleket 10 in Lefkowitz and Fant 1982: 136; Pomeroy 1975: 41, 118, 164; Boer 1979: 39, 269; Verner 1983: 41), though 18 may have been most common (cf. Hesiod W.D. 698). In inscriptions in Gardner 1986: 39, about 40 percent of the women studied married before age 15 and almost 75 percent before age 19; in Egyptian census declarations, however, only 10 percent were married before age 15, suggesting some geographical variation; see more fully Rawson 1986b: 21-22; CIL 6.

. Ilan 1996: 67-69 notes that women usually married in their teens (at ages like 13 or 16), but some married even older than 20.

. Cf. m. ʾAbot 5:21; ARN 16A; Pesiq. Rab Kah. Sup. 3:2; Gen. Rab. 63:10; probably Lk 2:42 and Jos. Ant. 10.50 (not yet adult); cf. adulthood at 12 in 1 Esdr 5:41. Egyptian males entered adulthood at 14 (Lewis 1983: 156), as did Roman males (Gaius Inst. 1.196; 2.113; 3.208; Gardner 1986: 14); cf. also Jos. Life 9. The basic matter was achieving puberty (e.g., Gaius Inst. 1.196; Gen. Rab. 91:3; p. Ter. 1:3), as in many traditional societies (Mbiti 1970: 158-73; Eliade 1958: 41; Dawson 1964: 309). Men could marry as young as puberty or 14 (Gardner 1986: 38); in the Greek world, however, men traditionally married in their 30s (Hesiod W.D. 695-97; Pomeroy 1975: 118). Roman men were generally a decade older than their brides (Shaw 1987).

. M. Ketub. 5:2; Ned. 10:5; Song Rab. 4:8, §1; Safrai 1974/76b: 757; Davies and Allison 1988: 199. From the time of Augustus (who desired to replenish especially the aristocracy), Roman law required marriage within two years (Dio Cassius R.H. 54.16.7).

. The wording in 1:25 may argue against perpetual virginity, i.e., after Jesus’ birth (Davies and Allison 1988: 219), a doctrine probably initially derived from the Protevangelium of James 19:3 (Blomberg 1992a: 57; cf. Lk 2:7, 23). Luz 1989: 124-25 contends that the grammar cannot disprove perpetual virginity here, but that the concept is alien to Matthew’s community and would therefore have had to have been expressed explicitly if Matthew wished to communicate it.

. Matthew probably records Joseph’s sexual restraint with Mary before Jesus’ birth to underline Mary’s virginity and Joseph’s self-control, rather than from fear that a nonvirgin birth would contaminate a virgin conception. Although some of his ancient readers may have thought that women could conceive during pregnancy, apparently few felt that this conception mixed new paternal heritage into the first fetus (Gardner 1986: 53, citing Arist. Gen. Anim. 553a). But the couple’s restraint also demanded Mary’s trust in Joseph: the wedding night would provide no proof of her virginity, hence there would be neither current vindication nor subsequent legal protection for her should Joseph ever reject her (cf. Deut 22:13-19; in modern times, see Eickelman 1989: 174). But second-century rabbis required claims against a bride’s virginity either immediately or within a month after the wedding (p. Ketub. 1:4, §4).

. For more endowed homes, see later sources in Safrai 1974/76b: 736. Some Jewish circles (e.g., Essenes) apparently allowed intercourse only for procreation, but other circles (including most later rabbis) allowed married couples to practice it for pleasure if they wished (Ilan 1996: 107-11).

. Later rabbis, undoubtedly following the lenient Pharisaic tradition, invariably ruled in favor of women when they claimed exceptional reasons for a hymen not bleeding when married (Ilan 1996: 98-99). Obviously a pregnancy before marriage would not lend itself to such an explanation.

. Llewelyn 1992: 14 correctly notes (citing m. Qidd. 4:9; t. Yebam. 13:1; DJD 2.19.6-7; but also cf. m. Giṭ. 9:3) that adult women could refuse marriage; Ilan 1996: 84 specifies the decease of the woman’s father. Most Jewish marriages involved young virgins (who were preferred; Jos. Ant. 4.244; cf. Life 414) rather than adults (who were usually divorcées or widows); see Keener 1991a: 72-78; children nevertheless often could influence betrothal plans (Parthenius L.R. 1.2).

. Cohen 1966: 292, 298; on Roman betrothal, cf. Paulus Opinions 2.1-9; Cohen 1966: 279-347; Gardner 1986: 45-47.

. Writers regularly point this out (Argyle 1963: 27; Sandmel 1978b: 194; Gundry 1982: 21; Belkin 1940: 250; Davies and Allison 1988: 199).

. M. Giṭ. 6:2; Ketub. 1:2; 4:2; Yebam. 2:6; b. Giṭ. 26b; Montefiore 1968: 2:5; Davies and Allison 1988: 199.

. The exhortation not to fear means not to fear to do God’s will; cf. 8:26; 9:2; 10:26; 14:27; 17:7; 28:5, 10. Joseph would soon have a different reason to fear (2:22).

. For an example of divorce for adultery, see, e.g., Apul. Metam. 9.27-29.

. Judeans, reportedly less careful (Ilan 1996: 100 even cites a recently discovered marriage contract indicating prior cohabitation), also assumed a greater likelihood that the woman would get pregnant before the wedding. Even today, some Middle Eastern villages assume that any woman alone with a man for over 20 minutes has had intercourse with him (Delaney 1987: 41). The circumstances are thus quite different from those portrayed where a Greek loyally defends his fiancée’s honor even to death (Soph. Antig. 631-765, 1221-43).

. Cf. Allen 1977: 9; Tosato 1979; contrast Romaniuk 1980, who takes the participle as concessive.

. Jos. War 2.117; Sherwin-White 1978: 32-45; Stewart 1975; O’Rourke 1971: 174-75. For debate on appropriate penalties for adultery in Palestinian Jewish sources, see Ilan 1996: 135-41.

. Many traditional societies prohibit or limit adultery, though the behavior of the wife is more at issue than that of the husband (e.g., Firth 1963: 119, 475-77; Stephens 1963: 245; Schapera 1966: 204-7; Barnouw 1955: 23; Wilson 1957: 134, 258, 262; Mbiti 1971: 193, 268-69, 275; for allegedly harsher penalties for adulterous women than men in early Judaism, cf. Swidler 1976: 148-54). Romans executed Vestal Virgins who had intercourse (Plut. R.Q. 96, Mor. 286EF; Dion. Hal. 1.78.5); husbands or fathers who caught adulterers in the act could under specific circumstances kill them (Aul. Gel. Attic Nights 10.23; Quint. 7.1.7; Plut. Solon 23.1; Diog. Laert. 6.1.4; b. Ned. 91b; cf. Char. Chaer. 1.4.7, 11-12), and other legal punishments for adultery existed (Sen. De Benef. 6.32.1; Mart. Epig. 2.60; 2.83; 3.85; Ach. Tat. 6.5.3-4; O’Rourke 1971: 181-82; in classical Greece, Dover 1984: 146; though cf. Epict. Disc. 1.18.5-6, 11-12). What mortals could not avenge, God would (cf. Sib. Or. 3:764-66; 5:430; Test. Abr. 12 B; Acts of John 35), also by bestowing public shame (Sir 23:21; Ps. Sol. 8:8-10; Num. Rab. 9:1).

. “Like the dowry in Ptolemaic Egypt, the kethubah was used to penalise the spouse who was either at fault or initiated the divorce (e.g., m. Ket. 5.7, 7.6, t. Ket. 12.1)” (Llewelyn 1992: 12-13). For the Greek dowry, cf., e.g., Plut. Bride 12, Mor. 139DE; Diod. Sic. 32.10.2; for parallels between the Jewish and pagan customs, see Cohen 1966: 348-76; among the hellenized Herodians, see Hanson 1990. Divorces normally required repayment of the dowry (cf. P. Tebt. 104.12, 31-33, 37-38; P. Oxy. 1273.5-20; B.G.U. 1103.10-14); rather than risk having to repay high dowries, some men preferred to marry poorer women (Ps-Phocyl. 199-200; Jos. Apion 2.200; cf. Plut. Bride 22, Mor. 141AB; Educ. 19, Mor. 13F–14A), but this view was not universal (Ovid Metam. 4.704-5).

. He may have paid a “bride price” if the “reforms” suggested by Ilan 1996: 90-91 (cf. also a rule in m. Soṭa 1:5) were not as pervasive in Galilee, and perhaps not in Judea, in this period as rabbinic tradition might imply (the sources suggest that they were instituted only by the first-century Pharisaic “schools”). After the reforms, a Jewish man might marry young and pay off the bride price only in the case of divorce; he had to continue supporting her until he finished paying off the bride price (p. B. Meṣ. 1:5, §4). In many traditional cultures bride prices represent a significant investment, testifying to how much the groom and bridal family value the wife; sometimes the groom might not even finish paying (cf. Eickelman 1989: 173 on wealthier Middle Eastern families today). In some societies the amount of money involved depended partly on the bride’s physical attractiveness (Ps.-Demosth. Or. 59, Against Neaera 113); Mediterranean peoples often viewed lack of monetary bride price as unusual (Arrian Indica 17.4).

. M. Giṭ. 2:5; 9:3, 4, 8. Cf. most commentators (Hill 1972: 78; Carson 1984b: 75; Tosato 1979; Derrett 1973: 38; France 1985: 77; Davies and Allison 1988: 204-5); pace Gundry 1982: 21-22.

. A git, or certificate of divorce, could provide the husband possession of the dowry (cf. B.G.U. 1103, in Sherk 1988: 246), but the more complex the certificate the more necessary the aid of a village scribe, hence publication of the news.

. The mention of Joseph’s anxiety probably has more to do with the story line than with the frequency of citing mental states when relating dreams (cited by Davies and Allison 1988: 206; Plut. Brut. 36.4; Jos. Ant. 11.334; Philost. V.A. 4.34). We may infer the centrality of the dream to Matthew’s account by the supernatural guidance (dreams — 2:12, 13, 19, 22; cf. Acts 2:17 — or the star — 2:2, 9) represented in every paragraph from 1:18-25 to the end of the second chapter.

. Some ancient readers might also think Joseph regretful for having mistrusted Mary (cf. Ach. Tat. 5.19.6, which may also suggest suspense here), but at any rate readers would recognize that God vindicated Mary in her betrothed husband’s eyes.

. One who denounced his new bride as not being a virgin was normally presumed honest, given the price of the wedding banquet (p. Ketub. 1:1, §7; though cf. differences on the burden of proof in m. Ketub. 1:6); exceptions were made if she had bled earlier (p. Ketub. 1:1, §4).

. Miller 1990 thinks that early Judaism blurred the older distinction between night visions and dreams; Gnuse 1990 favors the OT dream genre here (standard revelations such as those to Jacob, not like Joseph’s or Daniel’s). Josephus hellenizes some of his dream reports (see Gnuse 1989)

. Later rabbis shortened Jesus’ name (but not that of other Yeshuas) to Yeshu, possibly to avoid this meaning of his name (Blinzler 1959: 24-25). The rabbinic tradition of the Messiah’s naming before the world’s creation (Hagner 1993: 19 cites b. Pesaḥ. 54a, bar.; one might add b. Ned. 39b, bar.; Gen. Rab. 1:4; Lev. Rab. 14:1 — his spirit; Pesiq. R. 33:6; Midr. Ps. 72:17; in Lev. Rab. 30:16 the Messiah himself appears preexistent) may not have been prominent enough in Matthew’s day to have been obvious for interpreting this passage (though cf. perhaps 1 Enoch 46:1-2; 48:2-3; 62:7; 4 Ezra 12:32; 13:26); it might even respond to Christian polemic (Montefiore and Loewe 1974: 586). The Tannaitic Mek. Pisha 1.54-56 fails to presuppose the tradition, so even if it existed this early it was probably not prominent even in rabbis’ thinking. More relevant, early tradition reports that Moses’ name fit an Egyptian verb for “save” (Jos. Ant. 2.228; cf. Apion 1.286; Allison 1993b: 144-45).

. 8:25, though not in Mk 4:38; Mt 9:22; 14:36. In the eschatological sense, cf. 10:22; 16:25; 19:25; probably 21:9; the Greek term is broad in meaning. Early Christians continued the Jewish story of salvation in history (Wright 1992a: 458).

. Cf. Ellis 1974: 28; Danker 1992; Sib. Or. 8.456-79; Apoc. Adam 7:9; Test. Sol. 6:8; pace Harris 1992: 256-58. Given the emperor’s role as Jupiter’s agent in Flavian imperial theology, Jesus’ role as the agent of God’s presence and reign (see Carter 1998b) would have implications far beyond Palestine. Allison 1993b: 154-55 observes that Moses acted as a god to Pharaoh (Ex 4:16; 7:1) and tentatively asks if new Moses imagery might be in view; nevertheless, the reference to omnipresence in 18:20 (see comment there) demands a higher Christology than this.

. In view of the unlikelihood that John is unfamiliar with the Bethlehem tradition, Jn 7:42 either represents polemic against it (for its Judean connections) or, more likely in our view, assumes it as common knowledge among his audience, ironically polemicizing against the ignorance of the Judean elite.

. For one hypothetical reconstruction of Matthew’s source, see Brown 1977: 116-17. It is highly unlikely that Matthew has transformed Luke’s shepherds (pace Gundry 1982: 26): although developing traditions could transform other people into rulers (e.g., LXX Job 2:11; Test. Job 3:7/6; later tradition on the Magi), this story has little in common with Luke; even the timing (2:16) is quite different.

. Friedländer 1965: 1:211; O’Toole 1978: 16-17. Overman 1996: 44 points to the broader category of “embassy” of which our examples represent a more specific kind.

. McNamara 1968; Charbel 1983; Custance 1975: 19-20; Davies and Allison 1988: 228. Schnackenburg 1985: 23 is uncommitted. Biblical tradition called Arabia “East,” but this is hardly a frequent location for Magi.

. F. F. Bruce 1951: 184; Hagner 1993: 27; cf. Argyle 1963: 30; Liddell and Scott 1968: 1071; Montefiore 1968: 2:9.

. On the Magi of Persia, see further Olmstead 1959: 28-29, 196, 251, 372, 449, 477-78, 496, 517.

. Magi were not, however, positive in all cases; cf., e.g., Herod. Hist. 3.79-80. Jews may have also viewed their association with dreams as a divine accommodation to pagan inability to hear more clearly (cf. Gnuse 1990b).

. Earlier texts apply “King of kings” to the Babylonian ruler (Ezra 7:12; Ezek 26:7; Dan 2:37), other texts to any great king (Test. Jud. 3:7), some Greeks to Zeus or the supreme deity (Dio Chrys. Second Disc. on Kingship 75; PGM 4.1201-2); magical papyri often employ the expression imprecisely (e.g., PGM 2.53-54; 4.640-44). Jewish texts most often apply the title (sometimes “king of kings of kings”) to the supreme God, whether in hellenistic (3 Macc 5:35; Philo Decal. 41; Spec. 1.18; PGM 13.605; cf. Egyptian Book of the Dead, spell 185Eb, part S-1), nonrabbinic Semitic (1 Enoch 9:4; 84:2; 3 Enoch 22:15; 25:4; Aramaic incant. text 67:2), or rabbinic texts, both tannaitic (m. ʾAbot 3:1; t. Sanh. 8:9; Sifra Sav Mekhilta DeMiluim 98.8.5; ARN 25; 27A; 1, §1; 27, §56; 29, §61B) and later (b. Ber. 28b; 32b-33a; 62b; Sanh. 38a; Gen. Rab. 8:7; 12:1; 14:1; Ex. Rab. 2:2; 6:1; 20:1; Lev. Rab. 18:1; 33:3; Num. Rab. 1:4; 4:1, 20; 8:3; 14:3; 15:3; 18:22; Ruth 2:3; Esth. Rab. 3:15; Qoh. Rab. 2:12, §1; 4:17, §1; 5:10, §2; 9:15, §7; 9:18, §2; 12:1, §1; 12:7, §1; Song Rab. 1:12, §1; 7:5, §3; Lam. Rab. 1:16, §50; Pesiq. R. 13:7; 15, preamble; 23:8); as well as in early Christian texts (1 Tim 6:15; Rev 17:14; 19:16). This application may continue an earlier adaptation of the royal title in “God of gods” (Deut 10:17; Ps 136:3; cf. PGM 5.468-69; 22b.20; 62.24; PDM 14.684), but derivation from secular imagery may have marred the title’s sacredness for a few interpreters (p. Meg. 1:9, §17).

. His audience might, however, associate them with Parthia and hence with opposition to Rome, the power with which Herod was allied and which crushed Jerusalem in 66-70 (see Horsley 1993a: 56).

. E.g., Horace Ode 2.17.17-25; Petron. Sat. 76; Mart. Epig. 9.82.1; Juv. Sat. 6.579; Artem. Oneir. 2.36; Vettius Valens Anthologies 5.9.2; Stobaeus Eclogae 1.5.14; P. Tebt. 276; PGM 4.651; 13.709-11; 36.330-32; 62.52-75; PDM Sup. 183-84. In more detail, see MacMullen 1966: 128-62; Friedländer 1965: 1:185-88; Winslow 1971: 243-44. The Greeks had borrowed the practice from the East (cf. Rochberg-Halton 1984; Rochberg-Halton 1988; Rochberg-Halton 1989 [including comment on nativity omens]; Grant 1953: 60; ANET 333). For skepticism concerning astrology, see, e.g., Epicurus in Long 1974: 41-42; as well as Pliny N.H. 2.6.28-30; Sextus Empiricus Against the Professors 5.

. Cf., e.g., 1 Enoch 6:7, MSS; 8:3; Jub. 8:3; 12:17; 13:16-18; Philo Praem. 58; Syr. Men. Sent. 292-93; Sib. Or. 3:221-22, 227-29; Sifra Qed. pq. 6.203.2.1; Sifre Deut. 171.4.1; Did. 3:4.

. Mek. Pisha 2.44-46; in more detail, see Charlesworth 1977; cf. Allegro 1964; Mayer 1972: 123-25; Siker 1987; Stieglitz 1981; contrast Lehmann 1975. First-century writers even interpreted elements of the temple (Jos. War 5.214, 217; cf. Philo Q. Ex. 2.79) accordingly. Later Judaism became even more open to these influences (Wächter 1969; cf. Goldstein and Pingree 1979), and later synagogues adopted the Greek zodiac to symbolize God’s rule over the cosmos (May 1944: 9; Goodenough 1953/68: 8:167-218; Hachlili 1977; Meyers 1980: 106; Narkiss 1986: 185-86; Shanks 1984; CIJ 2:212-13, 240-42); but astrology permeated the Mediterranean world even more in those centuries, e.g., in relation to Mithraism (cf., e.g., Beck 1976abc; Campbell 1968: 44-180) and Gnosticism (Burkitt 1932: 30-33). Pagans could also accept prediction without causation (e.g., the late writer Plotinus 2.3.1).

. Though Judeans had reason to resent a foreigner’s reign (Deut 17:15), it was known that the Idumeans had converted to Judean customs (Strabo Geog. 16.2.34); in favor of Herod’s claim to be religiously Jewish, see Hadas-Lebel 1993. Nevertheless, it was also known that Herod the Great built pagan temples for Gentiles (e.g., Jos. Ant. 15.298; 16.147; on his attempts to achieve honor through benefaction in general, see Jacobson 1988). For Matthew, Jesus is the rightful king (21:5).

. Some see in the star a symbolic reference to Num. 24:17 (Meier 1980: 11; Harrington 1982: 17; Albright and Mann 1971: 14; cf. Test. Levi 18.3; Hagner 1993: 25); Davies and Allison 1988: 235 suggest that Matthew fails to cite it because his formula quotations in 2:1-23 have a geographical focus. (Far less plausible is the suggestion that this verse was excised from Matthew’s text by an anti-Gnostic redactor — Albright and Mann 1971: 15-16; it lacks textual support, dates Gnosticism too early, and ignores Matthew’s usual practice of citing only one primary text per passage in the infancy narratives.) Further, Bar Kochba provides early second-century evidence for the assumption of star or “son of a star” as a messianic title (Albright and Mann 1971: 15). But in Mt 2:1-12 the star signals Jesus’ coming rather than represents the Messiah (Soares Prabhu 1976: 9).

. Kepler suggested a temporary light from a nova or supernova (Sachs and Walker 1984); some later scholars have suggested in his name that the event was historically a conjunction of Jupiter and Saturn in 7 B.C. (Stauffer 1960: 32-34; Harrington 1982: 16-17), or a conjunction of Jupiter, Saturn, and Mars followed by a temporary bright star in 7-6 B.C. (Ramsay 1898: 215-17), a suggestion Babylonian astronomical texts may disconfirm (Sachs and Walker 1984). Chinese astronomers observed a nova for 70 days in 5/4 B.C. (the 1977 Quarterly Journal of the Royal Astronomy Society, cited in France 1985: 82; see Humphreys 1992). Luz 1989: 132 surveys various proposals, but, given Luke’s failure to mention the star and Jesus’ parents’ apparent lack of knowledge (Mk 3:31-35), doubts even a historical nucleus here. Clear astronomical evidence for a historical nucleus may be questionable here, but the counterarguments are also weaker than they might first appear: Luke’s birth narratives do not extend to Jesus’ second year, as here, and the Gospel tradition gives other examples that miracles were less than coercive (e.g., the probably reliable Q tradition in Mt 11:3-6). Allison 1993c argues that the star represents an angel; given the angelic character of stars in Jewish thought this is reasonable, though it does not fit the imagery of the current narrative.

. Küchler 1989 thinks that Jesus here replaces Augustus as rightful emperor because the star could symbolize Augustus; but such an interpretation would better suit Luke’s presentation of Jesus’ nativity.

. The significance of this portrayal becomes all the starker when we recall the role this particular enemy played in Jewish thought; Egypt is virtually Israel’s only enemy in rabbinic parables (Johnston 1977b: 626). For Philo’s Greek dislike of native Egyptians, cf. Goodenough 1962: 132.

. Micah’s Ephratah is the Ephrath of Gen. 35 with a he directional, i.e., “Bethlehem toward Ephrath.” Matthew probably omits the detail as irrelevant for his readers, substituting the more helpful “in Judah.” The “shepherd” statement alludes to Mic 5:4; Matthew also changes the “least” to “not the least” (see Gundry 1982: 29 for targumic parallels; cf. 20:26). The weaving in of a clause from the Davidic 2 Sam 5:2 is also significant (Gundry 1975: 207). For studies of Matthew’s interpretation of the text here, see France 1981a; Petrotta 1985; idem 1990.

. Later generations may have connected the grotto of Jesus’ birth with the Mithraic cave (Gervers 1979; cf. Campbell 1968: 6-11; for Mithraism borrowing Christian symbolism, cf. Latourette 1970: 247), but the tradition of Jesus’ birth, and certainly Bethlehem’s caves, predates the primary spread of Mithraism in the Roman world in the second century (cf. Grant 1986: 40-41). Pace Boring 1995: 42-43, it is unlikely that the Tammuz grove indicates a pagan cultic site before the Christian site; Bethlehem was a Jewish town, and the supposition of a pre-Hadrianic pagan grove there overstates the paganization of Judea near Jerusalem. (Sacred groves were, however, common; see, e.g., Deut 16:21; Paus. 5.10.1; 8.38.2, 5; Lucan C.W. 3.399-425.) For the tradition of the cave, see also Justin Dial. 78; Origen Contra Celsum 1.51; and later Euseb. Life of Constantine 3.43 (Finegan 1969: 20).

. Early Christian readers sometimes connected the star with a “pillar of light”; see the Syrian Chronicle of Zuqnin, redacted ca. A.D. 775, in Boring 1995: 41. For Matthew’s probable adaptation of this pillar by way of Wis 10:17; 18:3, see Viviano 1996.

. After what could have been as long as two years (cf. 2:16) Jesus was certainly no longer in the manger of Lk 2:7, 13; that accommodation was probably only a temporary measure due to crowding in the family house (see Bishop 1964; Danker 1972: 25; Bailey 1979; Byrne 1982; Kipgen 1983; cf. LaVerdiere 1985; Ottey 1986).

. Of course, these items had other functions in other contexts (e.g., 1 Enoch 29:2). The Egyptians used myrrh for embalming with any spices except frankincense (Herod. Hist. 2.86); the Essenes cleansed clothes for the Sabbath with frankincense (libona, CD 11.4). It grew in Syria (Stern 1974/84: 1:15), among other places. Perhaps more significantly, they were used with some offerings (e.g., Ex 30:23, 34; Lev 2:1-2, 15-16; 5:11; 6:15; 24:7; Num 5:15 LXX; Jub. 3:27; 6:3; 16:24). Because most frankincense and myrrh derived from South Arabia and Somaliland, they were quite expensive (Pliny N.H. 12.65; Virg. Georg. 2.117; P. Tebt. 35; Horsley 1987: 129-31). Babylonian Magi burned frankincense on silver altars (Olmstead 1959: 517 cites Arrian Anab. 3.16.3; Curt. 5.1.19ff; Marmor Parium B, 6).

. Widely held; see, e.g., Brown 1977: 187-88; Argyle 1963: 31; Meier 1980: 12; Davies and Allison 1988: 228, 253; Davies 1993: 36; cf. Montefiore 1968: 2:9.

. Lachs 1987: 10 cites various rabbinic sources (generally late) to illustrate the gifts both Jew and Gentile would offer the Messiah (b. Pesiq. 118b; Esth. Rab. 1:1; Tanchuma Shoftim 19; Midr. Ps. 87.6, 189b; Gen. Rab. 78:16; Ex. Rab. 35:5).

. One would “fall” to worship in the more technical sense but also to make entreaty (8:2; 15:25; 17:6, 14; 18:26; 20:20; 26:39; Jos. Life 138; sometimes with clasping knees, e.g., Eurip. Orestes 382). Respect for a deity could include looking at its feet (PGM 13.704-5).

. If these Magi were Zoroastrian (see, e.g., b. Sanh. 39a; cf. Frye 1996: 67-74), they were dualists rather than traditional polytheists (pure Zoroastrians were monotheistic dualists, but in practice paganism had sometimes filtered into Zoroastrianism over time — see Olmstead 1959: 475); but most of Matthew’s audience would grasp only the narrative’s emphasis that they were pagan astrologers, especially given their probable primary reliance on the Daniel narratives.

. This naiveté might reflect or conflict with popular stereotypes. Those in the Mediterranean world generally thought some Eastern “barbarians,” particularly the Scythians, to be savage and murderous (2 Macc 4:47; 3 Macc 7:5; 4 Macc 10:7; Jos. Apion 2.269; Herod. Hist. 1.15; 4.66, 76; Diog. Laert. 1.102; Hor. Ode 4.5.25; Sen. Dial. 4.15.1; Plut. Fortune of Alex. 1.5, Mor. 328C; Char. Chaer. 2.9.3; Sext. Emp. Pyrrh. 1.149; 3.208, 210); but some viewed the Scythians as noble barbarians (e.g., Anacharsis to Croesus Ep. 9; Strabo 7.3.7; cf. Greek Anth. 7.92). But widely known Greek experience with the Persians had revealed a powerful civilization quite in contrast with the Scythians.

. Luz 1989: 145 thinks the oldest form of the rabbinic traditions does not depend on Matthew here, and thus may attest that Jesus actually stayed in Egypt (certainly not that Matthew seeks to refute the magician libel, pace Lachs 1987: 11). On sorcery in Egypt, see, e.g., ARN 28A; 48, §132B; b. Qidd. 49b; Gen. Rab. 86:5; Exod. Rab. 9:6; 20:19; cf. Achtemeier 1976: 156; the prefect’s circular in Horsley 1981: 47-51; Hom. Od. 4.228-34; Dauphin 1993; Frankfurter 1994. Perhaps the demon in Tob 8:3 flees to remote regions of Egypt to escape civilization, but it may simply be fleeing as far as possible from the fish-liver smoke.

. Herod had no formal extradition treaty with the Roman authorities in Egypt, even if he knew where the family had fled or suspected (as he probably did not, in the logic of the story) that the object of his pursuit had escaped. Further, the family would easily have blended into the large Jewish community in Egypt, and if Jesus was born around 7 B.C., Herod sought his life roughly two years later (2:16), and Herod died in 4 B.C., then Jesus probably remained there little more than a year. Popular arguments concerning Jesus’ “missing years” neglect that his contemporaries called him Jesus “of Nazareth,” ignore his carpentry work (cf. on 13:53-58), and so on.

. Arguments against the account based on treating summaries as detailed observations, however, are liable to stray from the mark. Thus one should no more press the grammar of Lk 2:39 to contradict the historical plausibility of Matthew’s account (pace Gundry 1982: 36-37) than one should read Acts 9:19-26 against Gal 1:17-18; Luke has no more reason than other writers, ancient or modern, to detail points that do not concern him.

. Cf., e.g., Brawley 1987: 57; Ladouceur 1980; Aune 1987: 130; Char. Chaer. 3.4.9-10, 18; 8.4.2.

. Besides repeating Goulder’s examples of earlier Moses typology (1 Macc 2:1, 14, 26, 49ff.; 3:4, 8; Allison 1993b: 309), Allison 1993b: 11-95 provides numerous other examples of earlier Moses typology (some of them, including those in the Hebrew Bible, strong); for the prophet like Moses, see 73-84; for the Mosaic Messiah, 85-90; for other parallels between Jesus and Moses, cf. Kensky 1993. For Matthean echoes of the Exodus narrative, see also Goldberg 1989: 355-56, though he disagrees with Matthew’s understanding. Such “comparison” was a typical rhetorical technique (Stanton 1993: 77-80). Wright 1992a: 388 views the new exodus as a central Matthean theme.

. For a thorough chart providing comparisons to the story of the persecuted royal child (Moses, Abraham, Rev. 12, Cypselus, Mithridates, Romulus and Remus, Augustus, Nero, Gilgamesh, Sargon I, Cyrus, Zoroaster, Frêdun, Krishna), some fuller than others, plus still remoter parallels, see Luz 1989: 152-55. Cf. also Soph. Oed. Rex 1022-34. Nevertheless, cultural tendencies dominant in various cultures should not mute the extent to which Matthew’s story may actually critique similar hero stories (see Horsley 1993a: 162-72).

. Noted in Crossan 1986; Soares Prabhu 1976: 289-90; Brown 1977: 114-15; France 1985: 85; Meier 1980: 13; Overman 1990b: 77-78. In rabbinic tradition astrologers foretell Moses’ birth (b. Sanh. 101a; Soṭa 12b; Ex. Rab. on 1:22), and the Jerusalem Targ. Ex. 1:15 calls them “magicians” (Allison 1993b: 145).

. Also Davies 1966b: 15; Gundry 1982: 34; Teeple 1957: 74. In some Jewish traditions, Pharaoh went so far as to slay even Egyptian children to protect his throne (Lachs 1987: 12-13 cites b. Soṭa 12a; Ex. Rab. 1:18), but these are probably too late to prove relevant here.

. Stanton 1993: 201-2 curiously accepts the flight of the Magi as paradigmatic for the readers but rejects the flight of Jesus as such. Yet Jesus’ sufferings in the Gospel typically foreshadow those of the disciples (e.g., 10:22-25; 16:24).

. Soares Prabhu 1976: 227-28, who thinks the narrative may contain historical reminiscences, suggests that Hos 11:1, in contrast to the other formula quotations in the infancy narratives, may even have been in Matthew’s source. Whether or not the text is preredactional, it certainly fits Matthew’s Israel typology well.

. It is true that they are not leaving Egypt here (as in 2:21; Davies and Allison 1993: 261); yet this divergence is intentional, for by presenting Herod as a new Pharaoh Matthew has purposely and ironically inverted the ethnic perspective of the tradition.

. A frequent view; see, e.g., Davies 1966b: 15; Longenecker 1975: 144-45; Meier 1979: 55; F. F. Bruce 1980a: 41; Patte 1987: 37; Gundry 1982: 34; Blomberg 1992a: 67; Hagner 1993: 36-37.

. That two citations address the locations indicates the centrality of these four place names for Matthew (Stendahl 1995: 71-72).

. Good 1990, citing also 2:21-23; 4:12; 12:14-15; 14:12-13; 15:21 and following the tradition of Wisdom’s withdrawal in 1 Enoch 42 and elsewhere. For earlier biblical precedent, cf., e.g., Gen 26:17, 21.

. Macrob. Sat. 2.4.11 (cf., e.g., in Van der Horst 1973: 222; Reinhold 1983: 100) probably reflects tradition springing from the Gospel itself (Smallwood 1976: 104n.156), although it may follow a standard style of rhetorical insult (Diog. Laert. 6.2.41); Ass. Mos. 6:6, which compares Herod to the Egyptians, may also reflect Christian influence, and Gk. Ezra 4:11 clearly does. For relations in the Herodian family, see Hanson 1989.

. Some view both Matthew’s account and part of the material in Josephus as tyrant-legends (Smallwood 1976: 103-4; cf. Mason 1992: 97); first-century-B.C. posthumous criticism of Herod certainly abounded (cf. Test. Mos., ch. 6). But Josephus reports such misdeeds of Herod in his War as well as in his Antiquities, though the former tends to be more positive toward Herod (cf. Mason 1992: 90-96).

. For later Jewish traditions on this passage, see Zatelli 1991. In Pesiq. Rab Kah. Sup. 5:2, many ancestors lamented from their graves over the exile. Matthew adapts the language of the quotation slightly (see Charlesworth 1985: 72); but this was customary in Jewish citations.

. Cf. also Davies and Allison 1988: 269, more tentatively, and citing in support Mek. Pisḥa 1 (Laut. 1:5-6) on Ex 12:1.

. If Jesus was born around 6 or 7 B.C., as many scholars think (cf. Stauffer 1960:32-34; Harrington 1982: 16-17, and discussion above; Dionysius Exiguus was slightly off on his sixth-century calendar), his family may have brought him back to the land of Israel when he was about 3, since Herod died in 4 B.C. (Against Thorley 1981, we cannot date Herod’s death after 4 B.C.; cf. Bernegger 1983.) Since Jesus had been less than two years old when his family left Palestine (2:16), they may have remained in Egypt for only a year.

. It is doubtful that the plural (“those who sought”) alludes specifically to Antipater as well as Herod (pace Wojciechowski 1988); though they died only five days apart, the former is not mentioned in the narrative and would probably not have been prominent in the tradition.

. Recent estimates suggest a population around 15,000 for Sepphoris, halving some previous estimates (Horsley 1995: 166).

. Meyers and Strange 1981: 27, 56; cf. Finkelstein 1962: 1:41; Stauffer 1960: 54.

. Malina and Rohrbaugh 1992: 37, by contrast, place the size at “a hundred or so,” which is probably too small. Egyptian tax records indicate that hamlets that small did exist; at the same time, however, large villages could hold 3,000 to 6,000 inhabitants (Lewis 1983: 67-68, citing W. Chr. 63; P. Mich. 224; P.S.I. 101-2). Nathanael’s tone is dismissive (Jn 1:46), but at least he has heard of the place (though it is also true that the likeliest site for Cana — cf. Jn 21:2 — was only about nine miles from Nazareth).

. Cf. Walker 1963: 392-93; Black 1967: 198; Fenton 1977: 51; Pritz 1988: 12; idem 1991; Filson 1960: 62; Lachs 1987: 14; Hagner 1993: 41-42.

. Meyers and Strange 1981: 27; Finegan 1969: 29; Lachs 1987: 14; cf. perhaps Vermes 1973: 72. Pottery fragments also suggest an earlier history as part of ancient Israel (Horsley 1995: 193). For a later dating of priestly settlement in Galilee, however, see Trifon 1989.

. The limited trade connections (for agriculture, cooking ware, and storage containers, see Adan-Bayewitz and Perlman 1990) do not suggest a religiocultural continuum (see Freyne 1994; Horsley 1995: 250-51). Sepphoris was the most pro-Roman city in Galilee (Jos. Life 346-48; cf. 30, 38, 124, 232); for rural Galilean hatred toward these cities, see Jos. Life 373-75, 382-84.

. Note also careful observation of the Sabbath in the other hellenized city, Tiberias (Jos. Life 275, 279).

. Some material from chs. 3, 4, and 12, though originally prepared for this commentary (in late 1994), was used in earlier form in Keener 1997b: 91-134.

. Vengeance and consolation remain central to “apocalyptic eschatology,” including that of Matthew’s possibly sectarian community in conflict after 70 (Sim 1996: 227-35).

. Because John’s agricultural images suit the period in and shortly after April (Ramsay 1908: 227) and because the Jordan was deeper and warmer after the rainy season (Malina and Rohrbaugh 1992: 39), some scholars seek to pinpoint the season as well as the location, but this is more problematic. One is not restricted to citing metaphors current only at the season; likewise, the Jordan’s depth varies by location and the basin can remain warmer than the Judean hills (recall, e.g., nearby Jericho’s palms).

. Although Mark freely blends texts (Is 40:3; Mal 3:1) in the standard Jewish manner (linking them by means of the common phrase “prepare the way,” Longenecker 1975: 138), naming the more well-known prophet in his composite quotation, Matthew simplifies Mark’s citation by including only the quotation actually from Isaiah (cf. Meier 1980: 23; composite quotations provide one primary reason for postulating the early Christian use of testimonia — Fitzmyer 1974: 65). The formula in 3:3 fits contemporary Scripture citations (Davies and Allison 1988: 292 compare CD 10.16; 16.15; 11QMelch 2.14). Fitzmyer 1961 argues that in general early Christian quotations are closer in form to Qumran than rabbinic citations; still, fulfillment formulas may appear less in the Qumran Scrolls (Fitzmyer 1974: 55; idem 1966: 253), though they become common in Targumim (McNamara 1983: 220).

. Josephus also portrays the Essenes as ascetics (War 2.120), and evidence suggests that John baptized near them (see Jeremias 1971: 43). But if John was ever an Essene (Betz 1990), it was not during his public ministry — the Essenes did not accept those who differed from them in any respect (see Pryke 1964). That John historically baptized in the wilderness (Mk 1:4-5) is clearly early tradition, presupposing a geographical situation (the narrow Jordan Valley in the wilderness) that obtained only in Palestine (Theissen 1991: 39). Theologically, Israel was “initiated” into the land through the Jordan (Josh 3–4); prophets in Elisha’s day also gathered near the Jordan (2 Kings 6:2, 4; cf. “baptism” in 5:10).

. For Matthew’s various literary parallels between John and Jesus, see Allison 1993b: 137-38; Stanton 1993: 81. On the historical level, Jesus’ baptism by John presumably implies Jesus’ agreement with John’s message (Sanders 1993: 94; pace Crossan 1991a: 230-38).

. E.g., Gnilka 1997: 75. Cf. the vegetable clothing of Bannus in Jos. Life 11. Davies and Allison 1988: 295 cite Jos. War 1.480 (though it lacks direct bearing on desert nomads) and modern bedouins. The Judean wilderness today is even sparser than then, rarely providing even locusts and honey (Hepper 1992: 54-55).

. E.g., Jerome Homily 91, on the Exodus (ACCS, 7); Hooker 1983: 9; Lane 1974a: 51; Ladd 1974b: 35; Hengel 1981b: 36. The Synoptic miracle traditions and passages like Lk 9:61-62 (cf. 1 Kings 19:20; Lk 10:4 — cf. 2 Kings 4:29) transferred some Elijah images to Jesus, but for Jesus these were inadequate (cf. Lk 9:8, 19-20, 33-35). The antiquity of the eschatological Elijah tradition is guaranteed, though scholars still dispute the antiquity of his messianic forerunner role; see, e.g., Milikowsky 1982-83; Schneider 1962: 169; Faierstein 1981: 86; Fitzmyer 1985; Allison 1984.

. Cf. 4 Ezra 6:26; t. ʿEd. 3:4; Sifre Deut. 41.4.3; 342.5.2; see Teeple 1957: 4-8; Lane 1974a: 324.

. See A. B. Bruce 1979: 80; Argyle 1963: 36; Davies and Allison 1988: 296 cite Lev 11:20-23; Pliny N.H. 6.35; 7.2; 11QTemple 4; m. Ḥul. 3:7; Ter. 10:9; and modern bedouin practice as well as that of the poor in “Arabia, Africa, and Syria.” Cf. also Qoh. Rab. 10:8, §1 (different in Esth. Rab. 3:7); etc.

. E.g., Ex 3:8; 13:5; Prov 24:13; 25:16, 27; 27:7; Jub. 1:7; Sir. 11:3; 24:20; Ep. Arist. 112; t. B. Bat. 1:9; Bek. 1:8.

. Ancients thought that bees recycled flowers into honeycombs — Ovid Metam. 13.928; Arist. De An. Hist. 5.22.4. Though some have thought of vegetable honey (A. B. Bruce 1979: 80; cf. Kraeling 1951: 195n.11; Jos. War 4.468; Aelian Animals 5.42), this proposal is improbable here (Davies and Allison 1988: 296, citing Diod. Sic. 19.94; b. Ber. 38a; Ketub. 111b; cf. Jos. War 4.469); since date honey is more obscure, it would need to be explicitly indicated. The spiritual honey of Jos. and Asen. 16.14/7-8 and the abundance of honey in the new era (Sib. Or. 3.622) are probably also irrelevant.

. Jos. Ant. 18.12 praises the Pharisees for their simple diet; John’s diet is far simpler, and an ancient audience that understood Josephus’s praise would also understand that of Mark and Matthew.

. The Q material in 3:7-10 (Meier 1994: 27-31) and 3:11-12 (Meier 1994: 32-33), none of which bears marks of Christian christological revision, has strong claim to authenticity; both Jesus (as is often suggested; see, e.g., Meier 1994: 116-30) and some of his disciples may once have followed John. Although others knew of John the prophet (Jos. Ant. 18.116-19), it was especially his adherents who retained his teachings (see Theissen 1991: 60). That during his ministry Jesus did not “baptize” anyone “in the Spirit” may support the authenticity of the prophecy as well (Witherington 1990: 38, 41-42).

. 1 Kings 8:46; 1 Esdr 4:37-38; Sir 8:5; Test. Abr. 9A; 1 Enoch 40:9; 4 Ezra 8:35; see further Moore 1971: 467-69; Bonsirven 1964: 114. One might repent one day before death, hence should do it daily (m. ʾAbot 2:10; ARN 15A; 29, §62B; Shab. 153a; Midr. Ps. 90:12). Some later Amoraim observed that repentance even one day before death could spare the wicked hell (Ruth Rab. 3:3; Qoh. Rab. 1:15, §1).

. Vermes 1993: 192 suggests that John’s and Jesus’ repentance was personal as opposed to the corporate repentance found in Qumran liturgy (e.g., 1QS 1.24-26), but the personal decision on entrance into that community (e.g., 1QS 3.9-11; 5.8, 13-14; acknowleged in 1993: 192n.8) is actually more analogous. For eschatological associations with repentance in some circles and probably John’s proclamation, see Sanders 1985: 92, 106-7.

. Ancient Near Eastern societies from Egypt (e.g., Book of the Dead, spell 20) and Mesopotamia (Moyer 1969: 130) to the Hittites (Moyer 1969: 132; ANET 207, 209) required ritual washings; indeed, many unrelated cultures do the same (Fallaize 1918: 455-66; Mbiti 1970: 169, 172; Fry et al. 1984: 61). Various Greek cults employed lustrations as preliminary purifications in their initiatory rites (e.g., Apul. Metam. 11.1; Mylonas 1961: 248; Wild 1981: 129-48), although scholars have rightly pointed out that such acts were simply preliminary washings, and never initiatory of themselves (Wagner 1967: 71-72, 102-3; Burkert 1987: 101; Nock 1964a: 60-62, 133). Jewish people had long used pools for ritual immersion (see, e.g., Yadin 1966: 164; Pearlman 1867: 179; Avigad 1980: 139-43; Reich 1984). Some have suggested immersions at the Qumran community as background for John’s baptism and/or Christian baptism (e.g., Thiering 1980; idem 1981b; Smith 1982; Taylor 1997: 76-88, though cf. also 99). But while Qumran may have practiced an initial washing preceded by repentance (1QS 5:8-23; Black 1961: 94, 100-101), this washing, too, was simply the first among many washings for a monastic community following a reformed priesthood (cf., e.g., Jos. War 2.150; Driver 1965: 496-506; Ringgren 1963: 221; Milik 1959: 102-3; Pryke 1964; Sutcliffe 1960a). Just as most Jews would assume a commitment to purity among those immersing themselves for a ritual bath (cf. b. Taʿan. 16a; Pesiq. R. 44:1), initiatory immersion demanded sincerity to function efficaciously (cf. 1QS 3.4-9; 5.13-14). Although some argue against proselyte baptism as a source (Beasley-Murray 1962: 18-31), scholarly opinion is increasingly shifting in the direction of recognizing it as a source, with whatever modifications (e.g., Abrahams 1917: 42; Montefiore 1968: 1:8; Kraeling 1951: 99-100; White 1960: 78-79, 320; F. F. Bruce 1972b: 156; Ladd 1974: 41; Goppelt 1981/82: 1:37; Meeks 1983: 150; LaSor 1987).

. One may note several differences between proselyte baptism as known from the sources and John’s baptism, especially the latter’s eschatological character and application to Jews (Rowley 1940: 333-34), but these differences do not undercut the likelihood of proselyte baptism as the basic model that John has adapted (Rowley 1940: 313-14).

. E.g., m. Pesaḥ. 8:8; t. ʿAbod. Zar. 3:11; Pusey 1984; contrast Taylor 1956; Légasse 1976. See further below.

. According to a tradition stemming from second-century Palestinian rabbis, idolaters (most Gentiles) were impure enough to render common contact with them or their homes contaminating (e.g., m. Pesaḥ. 8:8; ARN 8 A; b. ʿAbod. Zar. 8b; Pesaḥ. 9a). Other forms of washing were common even for minor impurities (e.g., Jos. War 2.129; Ant. 6.235; 18.19; m. Miqw.; Para 11:6; t. Miqw.; ARN 8, 25A), even of objects (e.g., CD 11.3-4; m. Maksh. 4:6; Miqw. 9:5-7, 10; Sifra Shir. pq. 9.115.1.6-8; p. Ḥag. 3:8, §§1-3; cf. Mart. Epig. 3.47); well-to-do families on the Temple Mount even had their own mikvaot (Avigad 1980: 142; Sanders 1992: 223-25). That ritually impure Gentiles could enter Judaism without cleansing is inconceivable.

. Meier 1994: 52 objects that proselyte baptism does not appear in Josephus, Philo, the NT, or Joseph and Aseneth, whereas circumcision appears in Jos. Ant. 20.34-48 and the NT. But initiatory washings were common in many cultures (e.g., IG 2 [2].1366 in Horsley 1983: 23; PGM 1.54; Epict. Disc. 3.21.14; Plut. Isis 75, Mor. 381D; Iamb. V.P. 71-74; Diog. Laert. 7.1.119; 8.1.33; Ach. Tat. 8.3.2; Apul. Metam. 11.1, 23; cf. Wild 1981: 129-48) and left no physical marks, hence they would not be controversial and not require mention; Philo and the NT do not describe conversions to Judaism, and as a post-OT practice tebillah may not be universal, though it was surely used in Palestine (see more fully Keener 1997b: 63-64, 146-49).

. Kraeling 1951: 103 thinks that John merely requires Abraham’s descendants to live like Abraham, rather than that he expects them to approach God on the same terms as Gentiles; but this disallows for John the “originality” Kraeling elsewhere allows in some measure for the nature of his baptism (p. 110), and thus misses Qumran evidence (and cf. Amos 9:7).

. This was probably the same sort of fanciful biology in which Egyptian cats jumped into crocodiles’ mouths, then clawed their way out unscathed, killing their host (Strabo 17.1.39; Diod. Sic. 1.87.5).

. In the fourth century B.C., Aristotle declared that vipers produce eggs internally but bring forth their young alive; he made no comments on the nature of their delivery, however (Arist. Gen. Anim. 1.10, 718b-30). Likewise Strabo altogether omits discussions of Indian vipers’ birth (Strabo Geog. 15.1.45). Not all writers in subsequent centuries agreed with Herodotus; in the second or third century A.D. Aelian, following a work of Aristotle’s successor Theophrastus no longer extant, at one point agrees that the mother’s womb bursts, but argues that it is through no malice of the offspring (Aelian On Animals 15.16); but he elsewhere concurs with and elaborates on Herodotus’s account, as noted above. Clawing out of the womb expresses evil in Gen. Rab. 34:10.

. Aelian regards the hippopotamus as the most ungodly of creatures because it eats its father (On Animals 7.19). Ancient Mediterranean peoples considered parent-murder one of the most hideous crimes conceivable (e.g., Soph. Oed. Rex 1440-41; Corn. Nep. 15 [Epaminondas], 6.2; Diod. Sic. 1.77.8; Dion. Hal. 8.80.1; Seneca De Clementia 1.23.1; Epict. Disc. 1.7.31-33; Plutarch Romulus 22.4; Marc. Aur. Med. 6.34; Apuleius Metam. 10.8; Diog. Laert. Lives 1.59; 1 Tim 1:9), despite occasional just cause (Hom. Il. 9.461; Soph. Electra 1411-16), and continued to recall Nero’s matricide (Mart. Epig. 4.63.3-4; Sib. Or. 4:121, 5:30, 8:71). Most people believed that even children who justly avenged their fathers (cf. Orestes in Hom. Od. 3.196-98, 310) erred in carrying out revenge against their mothers, and thus the Furies would torment them (e.g., Eurip. Electra 1238-91; Orestes 531-32, 549-63; Soph. Oed. Col. 1299, 1434; Ovid Metam. 9.407-10).

. One could seek to flee from wrath rightly (Is 26:20-21; Zeph 2:3; 1 Thess 1:10) or wrongly (Is 2:10, 19-22; Hos 10:8; Rom 2:3; 4 Macc 9:32). John’s opponents’ attempt to flee would prove vain.

. The tradition that Arabian vipers nested especially under balsam trees (e.g., Paus. 9.28.3-4) might have smoothed the connection between the images of vipers and trees in Mt 3:7-10, though this is at most speculation.

. Q follows the literal Semitic and LXX idiom “make fruit” here (Black 1967: 138-39). Fruit appears as moral behavior or acts in Jewish texts, e.g., Is 5:2; Hos 10:1, 12-13; 4 Ezra 3:20; cf. Prov 11:30; Jas 3:18; Apoc. Sedr. 12:5 (Davies and Allison 1988: 305 also cite Ps 1:3; Prov 1:31; Is 3:10; Sir 23:25; Jos. Ant. 20.48; 2 Bar. 32:1; Apoc. Adam 6:1; b. Qidd. 40a; and even the genetically unrelated Buddhist text Dhammapada 5); fruit appears in various other senses as well, a naturally multivalent image in an agrarian society.

. Rabbinic parables use trees to symbolize “contrasting types of men” (m. ʾAbot 3:18; Qidd. 40b; ARN 39:1; Johnston 1977b: 596). Jewish teachers often portrayed Israel (Jub. 1:16; 36:6; 1QS 8.5; 11.8; CD 1.7; 4 Ezra 5:23; Ps-Philo 12:8-9; 23:12; 28:4; b. Ḥul. 92a; Gen. Rab. 88:5; 98:9; Lev. Rab. 30:12; Esth. Rab. 9:2) or the patriarchs (1QapGen 19.14-23; Pesiq. Rab Kah. 15:5; 27:9; Gen. Rab. 53:3) as trees or plants; cf. also references in Mussner 1960; Gaster 1976: 352; Fujita 1976; Wirgin 1965: 22-26. In view of Mt 3:9, it may be noteworty that the Gentiles might be “en-grafted” (Rom 11:16-17; b. Yebam. 63a; cf. Riggans 1987), a practice often noted in ancient texts (e.g., Plut. Table-Talk 2.6.1-3, Mor. 640B-641A; Marc. Aur. 11.8; m. Sheb. 2:6; Sifra Qed. par. 3.202.1.5; b. Shab. 73b; Gen. Rab. 41:1; Num. Rab. 3:1; Ramsay 1906: 221-29). Cf. the church as a tree in Hermas 3.8.

. That this passage repeats the Targumic image of Abraham making converts (Delcor 1971) seems less likely. Jewish people regularly viewed themselves as “the children of Abraham” (e.g., 4 Macc 6:17, 22; 18:1; b. Ber. 6b). Recognizing Abraham’s intercessory power (e.g., Gk. Ezra 2:5; Test. Abr. 14A; Gen. Rab. 35:2), third-century traditions attribute to his intercession the protection of Israelites from Gehenna (b. ʿErub. 19a; Gen. Rab. 48:8 — third-century tradition; cf. 2 Macc 1:2; Justin Dial. 44.1; m. Sanh. 10:1; Song Rab. 8:9, §3; contrast Lk 16:24-26!).

. Manson 1979: 40; Argyle 1963: 36; Fenton 1977: 56; Gundry 1982: 47; Mounce 1985: 22; Hagner 1993: 50.

. One may also note the specific and different merit of the “matriarchs” in some late texts (Pesiq. Rab Kah. 11:6; Ex. Rab. 1:12; Lev. Rab. 21:11; 36:5; Num. Rab. 11:2; Pesiq. R. 12:5; 15:9; cf. Pesiq. Rab Kah. 2:5).

. Evidence for works-righteousness in early Judaism (e.g., Scott 1995: 274-76) should therefore be used to qualify Sanders’s thesis (1977), not to deny its corrective value vis-à-vis earlier distortions (cf. the balance in Boccaccini 1991: 218-20). Just as modern Christian soteriology on a popular level is rarely consistent, one need not expect full consistency in the diversity of early Judaism.

. The Qumran Scrolls also link the eschatological outpouring of the Spirit and end-time judgment (1QS 4.12-13, 21; Gnilka 1997: 74). Some think that John’s original proclamation addressed only the threshing floor image of wind blowing the chaff so that it could be separated from the wheat and burned (cf. Lev. Rab. 28:2; Qoh. Rab. 5:15, §1). In this case, John would have thought not of the “Holy Spirit” but of a “purifying wind,” which would be translated much the same way (A. B. Bruce 1979: 84; Flowers 1953; cf. Dunn 1978b: 695; Is 4:4; Mal 3:2; Barnard 1957: 107 suggests the image of a fiery stream). But four reasons militate against taking John’s original words as merely “wind and fire”: Jewish people usually understood “holy spirit” as a reference to God’s Spirit; wind, like fire, can represent God’s purifying Spirit in the OT; many — especially John’s contemporaries in the wilderness — associated the Spirit with purification (see Keener 1991b: 65-69); and all extant traditions apply the saying to God’s Spirit (cf. F. F. Bruce 1966: 50; Aune 1983: 132). Thus it is more likely that John offered another wordplay (cf. 3:9); he announced the promised outpouring (a water image) of God’s Spirit for the end time (e.g., Joel 2:28-29; Is 44:3; Ezek 39:29) as an agent separating the wicked from the righteous for judgment; and only water could douse fire. Meier 1994: 32-33, conversely, thinks “Spirit” but not “fire” is authentic, but I think that Mark abbreviates Q for his preface (on Mark’s use of Q, cf., e.g., Catchpole 1992) and that the Fourth Gospel’s author omits “fire” because his eschatology is primarily realized. Since Jesus did not historically baptize with fire, the prophecy is hardly a “prophecy after the event.” The combination of “fire” and “wind/spirit” in PGM 4.618 (where “fire” relates to stars); 62.24 is probably coincidence.

. On Matthew’s assimilation of John to Jesus, see, e.g., Wink 1968: 33-35 (for radical distinctions, see 36-39); for further examination of the content of the proclamation, see comment on 4:17.

. Some hellenized Jews like Philo stressed judgment far more rarely; e.g., Cong. 171-72 says: God, being good, would not literally send famine or death (but contrast Ep. Arist. 131; Sib. Or. 3.34).

. According to a common belief, “all” Israel would be saved, with specific exceptions (m. Sanh. 10:1; b. Ḥag. 15b; Sanh. 110b, bar.; Num. Rab. 14:1; Song Rab. 1:1, §5; cf. b. Sanh. 103a, where even Manasseh, or Pesiq. R. 1:5, where even Jeroboam ben Nebat, may be saved). Justin in the second century contends that only the repentant remnant of Israel will be saved (Dial. 26, 28, 118); he is apparently more skeptical than Paul (Rom 11:25-26) about a mass turning of Israel. Cf. also 1QM, where the righteous remnant of Israel is arrayed against the wicked nations and the vast number of apostate Jews.

. E.g., Ps 1:4; Is 17:5-6, 13; 29:5-6; 33:11; 41:15-16; Jer 15:7; Hos 13:3; Joel 3:13; Zeph 2:2; cf. Ex 15:7; Jer 4:11-13; 13:24; Amos 8:1-2.

. Compare further 1QM 14.1; Test. Abr. 12A; Sib. Or. 3.760-61; 5.274; Num. Rab. 4:1; cf. Sifre Deut. 312.1.1; 343.5.2.

. In this context, the Tannaitic tradition that Ishmael and Esau were chaff from Abraham and Isaac, but that Jacob produced no chaff (Sifre Deut. 312.1.1; 343.5.2), may be relevant; but if influence exists either way, the rabbinic tradition might indicate polemic against the Christian teaching (cf. Rom 9:7-13) rather than the reverse.

. Cf. also the references in Davies and Allison 1988: 310; Keener 1993: 599. For the conflation of water (“baptism”) and fire metaphors, see the many references in Davies and Allison 1988: 316, who also cite texts linking fire and water in judgment (Ps 66:10-12; Is 30:27-28; 43:2; 66:15-16; Sib. Or. 3.689-92; Jos. Ant. 1.70) and as a river or flood of fire (e.g., Dan 7:10; 1QH 3.29-36; 1 Enoch 67.13; Sib. Or. 2.196-205, 252-54; 3.54, 84-87; 4 Ezra 13:10-11). The river of fire image is also Greek and Roman (e.g., Virg. Aen. 6.735-42), where it can depict torment in Tartarus (e.g., Virg. Aen. 6.551). At the very time salvation comes for some (Israel), judgment (often fiery hell) comes for others (e.g., 1 Enoch 96:1-3; Sifre Deut. 307.3.2; 324.1.5; Ex. Rab. 30:24; cf. Sib. Or. 4.42).

. Many believed that it was eternal for at least the worst sinners (4 Macc 9:9; 12:12; t. Sanh. 13:5; probably 1 Enoch 103:8; 108:5-6; Ps-Philo 38:4; Asc. Isa. 1:2; 3 Enoch 44:3; t. Ber. 5:31; b. Rosh Hash. 17a; p. Ḥag. 2:2, §5; Sanh. 6:6, §2; cf. Diod. Sic. 4.69.5; Plut. Divine Vengeance 31, Mor. 567DE; for Gehenna’s vast size, note b. Pesaḥ. 94a; Taʿan. 10a; Song Rab. 6:9, §3; cf. Virg. Aen. 6.577-79). But in what may be the most common early Jewish view, most sinners endure hell at most only temporarily and are then destroyed (cf. 1QS 4.13-14; Ps. Sol. 13:11; Gen. Rab. 6:6; most sinners in t. Sanh. 13:3, 4; Pesiq. Rab Kah. 10:4; Pesiq. R. 11:5; cf. 2 Macc 12:43-45) or released (Num. Rab. 18:20; other texts are unclear, e.g., Sir 7:16; Sifre Num. 40.1.9; Sifre Deut. 311.3.1; 357.6.7; ARN 16 A; 32, §69 B; 37, §95 B); twelve months is a familiar duration (b. Shab. 33b; Lam. Rab. 1:11-12, §40). Many Jewish storytellers conflated Gehenna with the Greek Tartarus (e.g., Sib. Or. 1:10, 101-3, 119; 4.186; 5.178; 11.138; cf. Gk. Ezra 4:22; b. Giṭ. 56b-57a; p. Ḥag. 2:2, §5; Sanh. 6:6, §2; Apoc. Pet. 5-12). Although Luke does not reject future eschatology in his effort to contextualize for Greek readers (Acts 17:31-32; 23:6; 24:15; contrast, e.g., Jos. Ant. 18.14, 18; War 2.163; Philo Sacr. Abel. et Cain. 5, 8), Matthew’s emphases retain more of their original Jewish flavor (cf. Milikowsky 1988; Goulder 1974: 63).

. In John’s image, the judge burns chaff already separated from the wheat; thus Webb 1991 suggests that John saw his ministry as already having separated the two.

. For the prophets in general, see, e.g., 2 Kings 9:7, 36; 10:10; 14:25; 17:13, 23; 21:10; 24:2; Ezra 9:11; Is 20:3; Jer 7:25; 25:4; 26:5; 29:19; 35:15; 44:4; Dan 3:28; 6:20; 9:6, 10; Amos 3:7; Zech 1:6; cf. ARN 37, §95 B; for David, see 2 Sam 3:18; 7:5, 8, 19-21, 25-29; 1 Kings 3:6; 8:24-26, 66; 11:13, 32, 34, 36, 38; 14:8; 2 Kings 8:19; 19:34; 20:6; 1 Chron 17:4, 7, 17-19, 23-27; 2 Chron 6:15-21, 42; Ps 78:70; 89:3, 20; 132:10; 144:10; Is 37:35; Jer 33:21-22, 26; Ezek 34:23-24; 37:24-25; cf. ARN 43, §121 B; for Moses, see Ex 14:31; Num 12:7-8; Deut 34:5; Josh 1:1-2, 7, 13, 15; 8:31, 33; 9:24; 11:12, 15; 12:6; 13:8; 14:7; 18:7; 22:2, 4-5; 1 Kings 8:53, 56; 2 Kings 18:12; 21:8; 1 Chron 6:49; 2 Chron 1:3; 24:6, 9; Neh 1:7-8; 9:14; 10:29; Ps 105:26; Dan 9:11; Mal 4:4; cf. Ps-Philo 30:2, famulum; ARN 43, §121 B. Cf. also the patriarchs (Gen 26:24; Ex 32:13; Deut 9:27; Ps 105:6; 2 Macc 1:2; Jub. 31:25; 45:3; Test. Abr. 9 A; 2 Bar. 4:4; ARN 43, §121 B) and Israel as a whole (Lev 25:42, 55; Deut 32:43; Is 41:8-9; 42:1, 19; 43:10; 44:1-2, 21; 45:4; 48:20; 49:3; Jer 30:10; 46:27-28; Ezek 28:25; 37:25; 2 Bar. 44:4; t. B. Qam. 7:5; ARN 43, §121 B; Gen. Rab. 96 NV; p. Qidd. 1:2, §24; cf. Tob 4:14 MSS). Being the slave of a deity represented high status; see the inscription in Grant 1953: 122; Martin 1990: xiv-xvi (citing Soph. Oed. Rex 410; Plato Phaedo 85B; Apul. Metam. 11.15; inscriptions), 46, 49; cf. Rom 1:1. Slaves of rulers exercised high status; see, e.g., Epict. Disc. 1.19.19; 4.7.23; Sherk 1988: 89-90; Deissmann 1978: 325ff. passim; Suet. Gramm. 21 (in Dixon 1988: 19); P. Oxy. 3312.99-100 (in Horsley 1983: 7-9); cf. Char. Chaer. 5.2.2.

. Davies and Allison 1988: 315 cite also b. Qidd. 22b, bar.; Pesaḥ. 4a; Sifre Num. 15:41; Plaut. Trin. 2.1; Lachs 1987: 45 adds Mek. Ex. 21:2; b. B. Meṣ. 41a; ʿErub. 27b; Ketub. 96a; p. B. Meṣ. 7:9, plus Roman custom.

. E.g., Ex 24:13; 33:11; Deut 1:38; 2 Kings 2:2-6; 3:11; 4:12, 43; 5:20; 6:3, 15, 17; 8:4; Diog. Laert. Lives 7.1.12; 7.5.170; t. B. Meṣ. 2:30; ARN 25A; b. Ber. 7b; Gen. Rab. 22:2; 100:10.

. Isaiah 40:3-5 depict readying roads for the king’s arrival, hence the coming of God’s kingship (cf. 52:7); heralds would go before him announcing the good news as he led his people forth (60:6; 61:1).

. Although Matthew and Luke could draw from Mark, this narrative is probably also in Q (Meier 1994: 103), and Mark may well have drawn from it.

. Matthew describes John’s “consent” in the same language as the devil’s departure (4:11), suggesting that John unwittingly had taken the devil’s side in seeking to deter Jesus’ mission (Blomberg 1992a: 86n.82; cf. similarly 4:10; 16:23).

. E.g., Lampe 1951: 39; cf. Barth 1963: 138; Uprichard 1981; cf. Mk 10:38-39; Robinson 1962: 160-61; Lane 1974a: 380.

. Davies and Allison 1988: 329 cite further references for eschatological expectation (Job 14:12 LXX; Ps 102:26; Is 64:1; Hag 2:6, 21; Sib. Or. 3.82; 8.233, 413; Mt 24:29; 2 Pet 3:10; Rev 6:14), which I take as more influential here, although they prefer the more frequently Christian reading of personal revelation (e.g., Ezek 1:1; Jn 1:51; Acts 7:56; 10:11; Rev 11:19; 19:11 [should be eschatological]; 2 Bar. 22:1; Test. Abr. 7; Herm. 5.1.1.4; Apocr. Jn 1). Taking Jesus’ “going up” from the water (3:16) as a type of his ascension (Robinson 1962: 162-63n.7) probably reads too much into the phrase, especially outside Luke, since Matthew never so much as narrates the ascension.

. Although some alternative flying creatures that could land on people (such as bats or locusts) would be inappropriate symbols of the Spirit, one must ask why God uses a dove rather than some other birds. Matthew 10:16 (contrast Lk 9:3) does not seem helpful here (though patristic sources viewed this dove as a symbol of innocence — Origen Homilies on Luke, Homily 27; Tert. On Baptism 8; or gentleness — Chrys. Matthew 12.3, ACCS, 13). Jewish use of the dove to symbolize God’s Spirit (Abrahams 1917: 48-49; Barrett 1966: 38; cf. Odes Sol. 24:1; 28:1) is both rare and late; so also the rabbinic comparison with the dove of the brooding Spirit in Gen 1 (Taylor 1952: 160-61; cf. Davies and Allison 1988: 331-34; Allison 1992) and the divine voice (b. Ber. 3a); the eagle comparison (t. Ḥag. 2:5; p. Ḥag. 2:1; Lachs 1987: 47) is probably irrelevant. Judaism more frequently used the dove to represent Israel (e.g., Ps-Philo 39:5; b. Ber. 53b; Shab. 49a; 130a; Ex. Rab. 20:6; Song Rab. 1:15, §2; 2:14, §§1-2; 4:1, §2; cf. Goodenough 1953/68: 8:41-46; Johnston 1977b: 595 finds this figure in Mek. Beshallah 3.86ff.; 7.27ff., but doubts whether it is frequent enough to constitute a standard metaphor). Some commentators think that this last source reinforces the image of Jesus as the ideal Israelite (Lane 1974a: 57). But the Gospel passages portray the Spirit, not Jesus, as a dove. Associations with Aphrodite (Plut. Isis, Mor. 379D; Ovid Metam. 13.673-74; cf. Sophia in Schroer 1986; perhaps Lucian Syrian Goddess 54) or other deities (Hom. Od. 3.371-72; 22.239-40; cf. Il. 7.58-59) or comparisons with deities’ speed (Hom. Il. 21.493) or inconspicuousness (Hom. Il. 5.778) are even farther from the mark.

. Gregory Thaumaturgus (Fourth Homily, On the Holy Theophany, or Of Christ’s Baptism, ACCS, 12) felt that the dove identified Jesus as a new Noah. Though elsewhere as well doves brought objects (Hom. Od. 12.62-63), Garnet 1980 rightly links the dove with Noah; but his linking of Noah, Enoch, and the Son of Man are too tenuous to make the case. One could note that Sib. Or. 1.242-52 uses for Noah’s doves a term related to the prophetic doves of Dodona (cf. Liddell and Scott; Herod. Hist. 2.57), but this connection, too, is less than obvious (since Matthew applies a different term). The parallel in Test. Jud. 24:2-3 is too close, probably representing a Christian interpolation.

. Rabbinic literature frequently mentions the Bath Qol, or “daughter of a voice” (e.g., b. ʾAbot 6:2; B. Bat. 73b, 85b; Mak. 23b; ʿErub. 54b; Shab. 33b; 88a; Soṭa 33a; p. Soṭa 7:5, §5; Pesiq. Rab Kah. 15:5; Lev. Rab. 19:5-6; Deut. Rab. 11:10; Lam. Rab. Proem 2, 23; Lam. Rab. 1:16, §50; Ruth Rab. 6:4; Qoh. Rab. 7:12, §1; Pesiq. Rab Kah. 11:16).

. See nonrabbinic conceptual parallels in Dan 4:31; Jos. Ant. 13.282-83; Artapanus in Euseb. P.E. 9.27.36; Sib. Or. 1.127, 267, 275; Dion. Hal. 1.56.3; 5.16.2-3; 8.56.2-3; Arrian Alex. 3.3.5; Lucan C.W. 1.569-70; Plutarch Isis 12, Mor. 355E; Mart. Pol. 9; cf. Johnson 1948: 62-63.

. E.g., Sifre Deut. 357.10.3; b. Ber. 61b; B. Bat. 58a; Ketub. 104a; Soṭa 21a, 48b; p. ʿAbod. Zar. 3:1, §2; Hor. 3:5, §3; Soṭa 9:16, §2; Pesiq. Rab Kah. 9:2.

. E.g., Kingsbury 1975: 14; Hill 1979: 59; Kelber 1979: 18-19; Cranfield 1955: 58; Borg 1987: 41, 53n.19; Witherington 1990: 150-51.

. This functions as an annunciation formula (cf. 27:37; its “this is” is lacking in Mk 15:26). On the form as a recognition oracle, see Aune 1983: 68-70, 272.

. Some contend that an original ambiguous pais (“servant” or “son”) underlies the Gospels’ huios (“son”), referring to the servant rather than to the “Son” (Marshall 1969: 327 summarizes the arguments, which he rejects). But Mark, emphasizing suffering, would not have dismissed potential servant language (Mk 10:45; 14:24), and Q apparently carries the same sense. The word order of Ps 2:7 LXX differs from that of Mark (Cranfield 1955: 61), but the voice may place “Son” later to keep the added “beloved” with “in whom I am well pleased”; the early Christians used Ps 2 for Jesus regularly (e.g., Acts 13:32-33 and Heb. 1:5); see Marshall 1969: 332-33. More scholars find Is 42 in Mk 1:11 than find Gen 22:2 there, but in Mark the Genesis allusion is probably paramount (cf. Best 1983: 81; Stegner 1985; Suder 1982); the Greek version sometimes uses “beloved” to translate yaḥid (an only son), among other places in Gen 22 (Dodd 1961: 130n.1), increasing the pathos of the sacrifice; cf. Mk 14:36; 15:39. God’s voice also spoke “from heaven” at the sacrifice of Isaac (Gen 22:11-12; cf. the Christian interpretation in Test. Levi 18:4-12), though one could also use the dove to appeal to a voice in Noah’s time (cf. Gen 9:1; Sib. Or. 1:127, 267). At the binding of Isaac, see also Isaac’s vision of heavenly angels attesting his obedience in Targ. Ps-J. Gen 22:10 (Boring 1995: 51).

. For proponents see, e.g., Marshall 1969: 335; Jeremias 1971: 53-54; Kingsbury 1983: 40, 65; Schweizer 1975: 37.

. See Jeremias 1972: 123; idem 1971: 71-72; Manson 1961: 55; Dupont 1966; Riesenfeld 1970: 75-76; Feuillet 1979; Michaels 1981: 63. Gerhardsson 1966: 11 views the Q testing narrative as “haggadic midrash,” expanding its core in relation to Deut 6–8 (71).

. Although the number has symbolic significance, I think that it is likely that Jesus literally fasted 40 days to provide the symbol rather than offered a shorter fast that he then portrayed symbolically as 40 days. This may seem more plausible to me in part because I have devout friends who engage in 40-day fasts.

. Some of Theissen’s arguments in favor of a date in the 40s (1991: 207-15) are speculative; the figure he finds most relevant in the reign of Caligula appears in Daniel and Psalms of Solomon, and the imperial cult was an issue both before and after Caligula. But given some of his stronger arguments for the antiquity of Q (Theissen 1991: 203-34, passim), of which I believe the testing narrative a part (rather than merely a late addition to Q, as if the written scroll was undergoing regular redaction), his date in the 40s is very likely correct.

. Trade connections existed with India in this period (e.g., Paus. 3.12.4; Petron. Sat. 38; Mart. Epig. 4.28.4; cf. Sib. Or. 11:299; Wheeler 1971: 115-71; Charlesworth 1970: 57-73; for Sri Lanka, Pliny N.H. 6.84-85), and some Greek thinkers interacted with Indian thought (Juv. Sat. 6.585; Diog. Laert. 9.11.61; Philost. V.A. bks. 2 and 3; Finegan 1952: 149), some of it likely Buddhist in this period (cf. Scott 1986). But such connections are certainly negligible (cf. Delaygue 1995) compared with closer Greco-Roman and particularly Jewish biblical contexts, so some proposed echoes of Buddhist sources in Jewish texts are unlikely (e.g., Stehly 1975; probably likewise other Indian sources in Flusser 1986).

. E.g., Teeple 1957: 75-76; Dunn 1970a: 30; Riesenfeld 1970: 76; Albright and Mann 1971: 36; Collins 1974; Meier 1979: 59-61; Gundry 1982: 53; France 1985: 98; Stegner 1990. Gerhardsson 1966 also finds in the threefold test an allusion to the Jewish interpretation of the three demands for loving God in the Shema, but the links are too speculative (see Davies and Allison 1988: 353).

. God led his people (e.g., Ps 80:1; 143:10), especially in the exodus: Ex 13:18, 21; 15:13, 22; Deut 8:2; Neh 9:12; Ps 77:20; 78:14, 52; 106:9; 136:16; Is 48:21; Jer 2:6, 17; 23:8; Hos 11:2-4; Amos 2:10; Acts 13:17; Bar 2:11; 1 Enoch 89:22; 4 Ezra 14:4; Sib. Or. 3.255; Pesiq. R. 12:8; for his leading in the new exodus, cf., e.g., Rom 8:14; Ps. Sol. 17:40-41; Pesiq. R. 30:2. The Bible also connected the Spirit’s presence with the exodus (Is 63:10-11, 14; Hag 2:5). Buse 1956 even sees Is 63 in the background of Mk 1’s baptism narrative, though he probably goes too far.

. Some also believed the wilderness to be a special haunt of demons (see comment on 12:43; cf. Is 13:21; 24:14 LXX; 1 Enoch 10:4; 4 Macc 18:8; Alexander 1980: 29; Jeremias 1971: 69; Davies and Allison 1988: 354 cite also Targ. Yer. I to Deut 32:10). Apart from a few rugged people like John who made the “wilderness” between the Jordan Valley and Judean hills their home, it represented a dangerous and inhospitable setting. One had to return to the Jordan Valley for food and water, and the rugged terrain made injury easy (Sanders 1993: 113).

. Luz 1989: 185-86 thinks that the christological interpretation is more central in Matthew than a parenetic one. While the christological interpretation is probably central, the recurrent motif of believers’ tests in this Gospel prohibits us from rejecting the parenetic focus altogether, however.

. Jewish literature in the time of Jesus and succeeding centuries still recognized his roles as accuser (Rev 12:10; Jub. 1:20; 48:18; 1 Enoch 40:7; 3 Enoch 14:2; 26:12; b. Yoma 20a; Gen. Rab. 38:7; 84:2; Lev. Rab. 21:4; Num. Rab. 18:21; Qoh. Rab. 3:2, §2; Pesiq. R. 47:4), deceiver (Jn 8:44; CD 4.15-17; 1 Enoch 54:6; Test. Job 3:6/5; 6:4/3; 17:2/1; 23:1; Test. Abr. 16 A; Apoc. Mos. 17:1; cf. Pesiq. Rab Kah. 26:2), and tempter (Jub. 17:15-18; b. Qidd. 81ab; Ex. Rab. 19:2; Best 1965: 53; Garrett 1998: 32-42), sometimes to turn biblical heroes from their God-appointed tasks (Pesiq. R. 40:6).

. Contemporary and later Jewish literature also recognized that God sent testing (e.g., Jub. 17:17; Sir 2:1; 36:1; Wis 3:5-6; 10:10; 11:9; Jdt 8:25; 3 Macc 2:6; Ps. Sol. 16:12-15; Gen. Rab. 55:2; Num. Rab. 17:2; cf. Jub. 19:3), but viewed as evil testing whose goal was stumbling (4 Macc 9:7; 1QS 11.9-22). Jewish sources praised those who had endured testing, such as Abraham (e.g., m. ʾAbot 5:3).

. Although Jewish tradition sometimes conjoined fasting with visions or revelations (2 Bar. 20:5; 43:3; cf. Col 2:18; Herm. 1.2.2; 1.3.1, 10, 13; 3.5.1), no such connection is likely here (pace Hagner 1993: 63). It more frequently accompanied repentance or mourning (Sir 31:26; Bar 1:5; 2 Bar. 5:7; 9:2; 12:5; Test. Zeb. 4:1-3; Apul. Metam. 2.24; Did. 7.4) and prayer (Tob 12:8; Bar 1:5; 2 Macc 13:12; Test. Benj. 1:4-5; cf. Acts 13:2; Did. 1.3; 8.1-2; Test. Jacob 7:17), often with weeping (Bar 1:5; 2 Macc 13:12). Whereas Greeks could use fasting for purificatory initiation (Mylonas 1961: 241-43, 258-59), Judaism forbade ascetic or unhealthy fasting (t. Kip. 4:1-2; Taʿan. 2:12, 14; p. Taʿan. 3:11, §3), though many of the pious engaged in much fasting (p. Ned. 8:1, §1; Taʿan. 4:3, §2). To the earliest Gospel audiences, Satan was literally capable of the sorts of tests mounted here (Job 1:6– 2:10; cf. Sanders 1993: 114).

. Many recognize the Moses allusion (e.g., Gundry 1982: 55; France 1985: 98; Allison 1993b: 165-72; cf. 2:16, 20), although the Scripture citations allude to Israel as a whole. Although normal fasts were typically from one sunset to the next (e.g., p. Ned. 8:1, §6), later Jewish literature had others repeat Moses’ 40-day fast, e.g., Adam in the Life of Adam and Eve 6:1. Cf. Zosimus’s fast from wine and bread for 40 years in the desert in History of the Rechabites 1:1 (possibly dependent on Christian desert monasticism); contrast Abaris’s continual magical fast (Blackburn 1986: 191) and Baruch’s seven-day fast that did not make him hungry (2 Bar. 21:1; contrast Mt 4:2). Ancients also recognized that 40 days without food could precipitate death by starvation (Diog. Laert. 8.1.40).

. Biographies often recounted the protagonist’s refusal to turn from a heroic vocation for other temptations (Shuler 1982: 96). Jewish teachers sometimes told stories of their predecessors’ in-vulnerability to sexual or other temptations (ARN 16 A), although other stories illustrated their predecessors’ vulnerability to temptation (e.g., b. Qidd. 81a in Urbach 1979: 1:480)

. Magicians also sometimes metamorphosed themselves or others (Apul. Metam. 1.9; 2.1, 5, 30; 3.21-25; 6.22; Blackburn 1986: 190, 193; in modern times, cf. Mbiti 1970: 220, 256-58; Nanan 1994), and were known for attempting invisibility (e.g., PGM 1.222-31, 247-62, 256-57; cf. Smith 1978: 120, though this feat sometimes simply represented divine help — see, e.g., Hom. Il. 20.321, 443-46; 21.597-98; Eurip. Helen 44-45; Iph. Taur. 27-30; Or. 1629-36). Some may have been skeptical of such alleged metamorphoses (Paus. 1.41.9), and some later Jewish teachers denied that demons could create (Alexander 1980: 33 cites b. Ḥul. 105b, probably following Edersheim), but some rabbis believed that they themselves had harnessed God’s creative power (e.g., ARN 25 A), and in later texts some even understood God’s activity in analogy with that of magicians (Hayman 1989); some Jews felt they could harness demons to do their bidding (e.g., Test. Sol. 7:8). Magic was widespread, yet not only in Palestinian Judaism (e.g., Wis 17:7; m. Sanh. 6:4; 7:11; Ps-Phocyl. 149), but even among Greeks and Romans, the charge of sorcery was a serious one (cf., e.g., Apuleius’s defense; Kee 1983: 213; Smith 1978: 75-76; Theissen 1983: 239-42).

. Interestingly, Young 1995: 31-32 cites a tradition in which the Messiah would announce redemption from atop the temple (Pesiq. R. 36; also Lachs 1987: 51); a link is possible, but the tradition is unfortunately not demonstrably early (nor is Jos. War 7.27-30, which he also cites, necessarily relevant). Smith 1978: 105 and others point out that claims to be able to fly also characterized magicians; early Christianity’s closest parallel to this claim (Acts 8:39) is not analogous (cf. Keener 1993: 346-47). Cf. also the first-century false prophets who falsely promised the pious Jerusalemites the deliverance they wanted to hear (e.g., Jos. War 6.285-87). Albert Schweitzer in effect revived the deluded visionary hypothesis for the twentieth century by applying it to Jesus, albeit more sympathetically.

. Matthew has “holy city” (4:5) for Luke’s “Jerusalem” (Lk 4:9; Luke emphasizes Jerusalem, possibly suggesting that Matthew altered Q here); Matthew seems to regard the city as holy on account of its heritage (5:35; 27:53), though it has been defiled (23:32-38; 24:15).

. In hellenistic circles, being a son of Zeus could also entitle one to rule humankind, as in the account in which the oracle of Ammon addressed Alexander of Macedon (cf. Diod. Sic. 17.51.1-2; Olmstead 1959: 511). Alexander yearned to rule the entire world (Dio Chrys. 4th Discourse on Kingship 49).

. Manson 1979: 44, cites 1 Enoch 24–25; 2 Bar. 76:3. Cf. Meier 1980: 30; France 1985: 99; Blomberg 1992a: 85n.81; Lucian Charon. Cf. the mountaintop between earth, sky, and sea, where all is visible, in Ovid Metam. 12.39-43.

. Mount Nebo allowed Moses to survey the land (Deut 34:1-4) and look “every direction” (Deut 3:27), which the rabbis understood as viewing the whole earth (Hagner 1993: 68). Jub. 4:26 lists Eden, the “mountain of the East,” Sinai, and Zion as God’s holy places. Theissen 1991: 233 suspects that Mount Hermon provided the literal basis for the mythical mountain.

. E.g., Jub. 15:21-32, 35:17, 49:2-4; 1 Enoch 40:9; 61:10; 2 Enoch 20:1 (longer version); Mek. Shir. 2:112ff.; Sifre Deut. 315.2.1; b. Ber. 16b-17a; Yoma 77a; 3 Enoch 26:12; 29:1; 30:1-2; Test. Levi 3:8; Test. Job 49:2; Test. Sol. 6:4, 8:10; Asc. Isa. 1:3; 2:2 (but 1:4 sounds suspiciously Christianized); 3 Bar. 12:3; in early Gnostic texts, e.g., Apoc. Adam 1:4; Hypostasis of the Archons; on LXX Deut, see, e.g., Dodd 1935: 18-19; cf. the sun and moon in Gen. Rab. 6:9. Cf. more fully Keener 1992: 64-65.

. Although the NT and some Jewish tradition portray Satan as the ruler of this age (Jn 12:31; 14:30; Eph 2:2), Jewish tradition rightly recognized the biblical notion that God was king of the universe, ruler of the world (2 Macc 7:9; Sifre Deut. 27.2.1; ARN 24, §51B); though angels could be “princes” (the “prince of light,” CD 5.18; 1QM 13.10) and the present age was evil, the “ruler of the world” could be a positive title (3 Enoch 1:4; 30:2); but cf. also the prince who rules over the evil ones (1QS 2.5; CD 12.2; 1QM 1.15; 13.2, 4; 15.2-3; 17.5-6; 18.6; Asc. Isa. 2:4; Test. Benj. 6:1; Pesiq. R. 20:2; 53:2; “Prince Mastema” in Jub. 11:11; 17:16; 18:9, 12; 48:2, 9, 12, 15), which is what the NT means. On the title, cf. most thoroughly Segal 1981.

. “Get away from me!” indicates disgust; cf. Is 30:22 MT; m. ʿAbod. Zar. 3:6; Shab. 9:1; God thus addresses Satan in Pesiq. R. 20:2. Cf. also “Go!” in the exorcism in 8:32. The relationship, hence the sense, is quite different in the magician’s polite dismissal of an angel once finished with it (PGM 1.185). The story line of a deity offering a mortal reward (e.g., Lucian Judgment of the Goddesses, LCL 3:400-401, Hera’s promise to Paris) is no closer than that of kings promising their subjects great rewards (e.g., Mk 6:23). Serving God “alone” recalls the LXX and some Jewish literature (Flusser 1989).

. Many scholars, however, suggest that Luke changes the original order here (e.g., Gundry 1982: 56; Carson 1984b: 111; France 1985: 97; Theissen 1991: 207; Blomberg 1992a: 84); Luke climaxes with the temple, which is admittedly characteristically Lukan, and Matthew’s climax may repeat the original point in Q.

. Some later texts playfully recount such debates with learned demons (Alexander 1980: 36; cf. b. Sanh. 89b; Schechter 1961: 252; Edersheim: 645; the conflict in Test. Job in Nickelsburg 1981: 243). The more frequent challenge-riposte sequence of an honor-shame culture is undoubtedly more relevant, however (see Malina and Rohrbaugh 1992: 41-42). Later rabbis recommended use of Torah to overcome temptation (Gen. Rab. 22:6).

. For the formula “It is written” elsewhere, cf., e.g., 2:5; CD 1.13; 5.1; 7.10-11; 11.18, 20; 1QS 5.15, 17; 8.14; m. Giṭ. 9:10; Sanh. 10:1; Mek. Pisḥa 1.76; Sifre Deut. 56.1.2; p. Meg. 1:5, §1; Sukk. 2:10, §1; 3:5, §1; Taʿan. 3:11, §5; Gen. Rab. 1:4; Num. Rab. 19:8; 3 Enoch 2:4 MSS; 5:14; 18:7, 18; 28:4, 9, 10; 31:2 (perhaps a nonscripture formula in 1QM 15.5-6, but the text is incomplete); in hellenistic papyri, see Deissmann 1923: 249-50; cf. “as it is said” (CD 19.15, B; m. ʾAbot 1:18; 2:13; ARN 36A); “it/he says” (CD 4.19-20; 7.14; 9.8-9; 10.16; 1QpHab 6.2; Mek. Pisha 1.70-71; cf. CD 7.8); God “says” (CD 4.13; 6.13; 8.9; cf. Ep. Arist. 155; Mt 15:4; 19:4-5); Moses “said” (CD 5.8; 8.14); the prophet “said” (CD 6.7-8); God “taught” (1QM 11.5-6; cf. CD 9.7). Scripture itself spoke, adapted expressions, etc. (Mek. Shir. 6 in Laut. 2:43-44). Ancient writers could also cite passages directly without noting they were quoting (Plut. Brutus 23.3), and pagans, like Jews, cited past experts as authorities (e.g., Quint. 4.1.12).

. The Gospels suggest the same central texts found among some of their contemporaries: if Cave 4 provides a fair cross-section of the most popular biblical books for the Qumran sectarians, they are first Deuteronomy, then Isaiah, then Psalms (Cross 1980: 43).

. Estimates have ranged as high as 12,000 (Davies and Allison 1988: 378, following Meyers and Strange 1981: 58), but the current estimate is closer to 1,000-2,000 people (Sanders 1993: 103; Stanton 1995a: 115; Horsley 1995: 194), hence not much larger than higher, older estimates for Nazareth. The more current, lower estimates appear to be more likely (this village is no more well known to Josephus’ readers in Life 403 than any other in the vicinity), but the debate has not yet closed.

. Häfner 1993 suggests an inclusio with Nazareth in 2:23, thereby connecting the passages and including both in Matthew’s “prologue”; a similar inclusio may connect the proclamations of John and Jesus in 3:2; 4:17.

. See Saunders 1977; Meyers 1988: 69, 71; cf. Julian in Stern 1974/84: 502-72; Epict. Disc. 4.7.6.

. Malinowski 1980; cf. Mayer 1981; Meyers and Strange 1981: 37-38, 45.

. Whereas Upper Galilee had quite strong commercial ties with Tyre (Vale 1987), Lower Galilee was almost completely Jewish by Jesus’ day (Goodman 1983: 31-32). Although hellenized art and speech may especially have impacted the latter through its two large cities (cf. Meyers 1976), in many respects it was probably no more hellenized (cf. Goodman 1983: 88-89). Literary and archaeological sources both suggest a cultural distinction between upper Galilee (the Golan) and lower Galilee. The latter included larger and more culturally mixed urban areas; although most of its inhabitants lived in villages, Galilee’s cities, which reflected a larger cultural continuum, influenced the villages continuously (Freyne 1988: 171). Nevertheless, archaeological and literary evidence confirms that the heavy population of lower Galilee was primarily rural and agricultural (Freyne 1988: 144-45; cf. idem 1980), and villages, despite cultural influences, were mainly autonomous politically and economically (Goodman 1983: 120). Differences and hence misunderstandings between rural and urban culture were pronounced in antiquity (Finley 1973: 123-49; MacMullen 1974: 15, 30, 32; cf. idem 1966: 163-91; Applebaum 1974/76a: 663-64), despite the influence of the latter on the former (cf. Millar 1981). This clash between urban and rural life undoubtedly obtained in Galilee as well (cf. Freyne 1988: 146-47; Eddy 1996: 465).

. Those who wish to place Jesus in a specifically hellenistic context have inappropriately exaggerated the cultural influence of these two cities on the surrounding countryside; see more correctly Horsley 1995: 174-81, who rightly cites Jos. Life 374-75, 384 and other evidence.

. Avi-Yonah 1974/76b: 110; Beitzel 1991; cf. Meyers and Strange 1981: 43; Reicke 1974: 117; Hagner 1993: 73.

. Sanders doubts that repentance was as central to Jesus’ teaching as John’s (1993: 231-32; cf. idem 1985: 109, 203, 206), but the same word statistics could deny the centrality of love in his message despite 22:37-40; although Jesus may have used the term less than John, examples of radical conversion show the centrality of the concept in his ministry (e.g., Mk 2:14-17), and Luke’s emphasis does not negate the occasional use of “repentance” earlier in the tradition (Mt 11:20-21 Q); and the idea that Jesus did not set out many “stipulations and conditions” for the kingdom (1993: 234) needs serious nuancing (cf., e.g., Mt 5:32). John undoubtedly influenced Jesus (see Michaels 1981: 1-24), and Sanders himself accepts that Jesus must have mostly agreed with the message of the prophet who baptized him (1993: 94; see further Chilton 1988).

. Cf. also Lachs 1987: 58. Wilkins 1995: 43-91 finds characteristics of master-disciple relationships in some OT relationships, most convincingly the prophetic groups (Samuel’s and Elisha’s mentorship) and, less demonstrably, the scribal tradition (for prophets, cf. CD 8.20-21; Mek. Pisḥa 1.150-53; ARN 11, §28B). The Greek term mathētēs originally meant both “learner” and, in more technical contexts, “adherent” of a master or school, but by Jesus’ day it meant especially the latter (Wilkins 1995: 42).

. E.g., Kümmel 1957: 19-25; Black 1967: 208-11; Nineham 1977: 69; cf. Carson 1984b: 117.

. E.g., Dodd 1961: 44; idem 1961: 29; Argyle 1963: 42; though cf. Mt 21:34.

. Schweizer 1955: 88; cf. Ladd 1974b: 107; Betz 1968: 71; see Mt 23:8-10.

. Davies 1966b: 133; Kingsbury 1975: 62; Gundry 1982: 62; Garland 1993: 48. Examples of teachers seeking disciples appear more frequently in the Greek tradition (e.g., Aul. Gel. 5.3.6); but these examples are of “radical” teachers, exceptions to the customary Greek as well as Jewish tradition. Robbins 1992: 89-94 cites helpful Greek examples (Plato Theaetetus 144D; Charmides 155A-58E; Alcibiades Major 135D; Aristophanes Clouds 497-517; Diog. Laert. 2.48). Finding little rabbinic precedent for summoning disciples or promising to make disciples into something they are not (1992: 104), Robbins locates the Jewish component of those elements’ origin here especially in the divine call in biblical tradition (e.g., 115, citing Gen 12:1-2), alongside the Socratic model in Greek tradition (1992: 117-18). Philo (1992: 94-95) and Josephus (1992: 97, closer to the Gospels) apply typical Greek school language to the biblical tradition, suggesting the influence of such hellenistic models by this period. It is said that Hillel pursued laborers to get them to study Torah (not just as disciples to become rabbis, however) and receive eternal life (ARN 26, §54B); the earliest sages probably also had more independence than their later counterparts (cf., e.g., p. Sanh. 1:2, §13).

. Some scholars favor an OT background for “fishers of people” (Hab 1:15; Jer 16:16; Mt 13:47; cf. Ezek 47:10; later, CD 4.15-17; 1QH 5.7-8; Ps-Philo 3:11; Strauss 1973); see Jeremias 1971: 132-33; Fenton 1977: 73; Gundry 1982: 62; Lane 1974a: 67-68; MacLaurin 1978b; Derrett 1980. But the OT use is a judgment metaphor (Jos. and Asen. 21:21 MSS would be closer), so the image is questionable unless Jesus provocatively pictured those who should “trap” people for the kingdom (as suggested by Davies and Allison 1988: 398) or intended the allusion by way of contrast (cf. Vermes 1993: 102n.27). Jesus more likely called them to be “fishers of people” primarily because fishing was their prior profession, as the text indicates. A bilingual milieu may also help explain Jesus’ use of the figure, since “catch” (Heb. ṣud, Aram. tzadē) could apply both to physical catching and to winning others by deception or debate (Lachs 1987: 58-59); the image also appears in Greek (Boring 1995: 55).

. Greek sources suggest that fishermen may not always have been of the highest reputation (Horsley 1989: 100), although around the Lake of Galilee the demands of the local economy probably would outweigh any influence from imported perspectives.

. Cf. ILS 7486; Wilkinson 1978: 29-30; Hengel 1974b: 27; Stanton 1989a: 186.

. Cf. fishing implements found in Bethsaida (Arav 1991). Cf. also the abundance of small boats available for crossing the sea from one town to another (Jos. Life 163-64); on the recovered Galilean fishing boat (26.5 by 7.5 by 4.5 feet), see Riesner 1987; Wachsmann 1988; Stone 1989; Andiñach 1990; Peachey 1990.

. Among the poor elsewhere, smoked fish could rank “the most popular item” in a general market’s sales for a day (P. Oxy. 520; Lewis 1983: 136). Rarer, luxury fish (Dupont 1992: 277) and the complex market system in second-century Roman legislation cited by Malina and Rohrbaugh 1992: 44-45 are probably less relevant to the towns of lakeside Galilee (excepting urban Sepphoris and Tiberias), where the market was not far from the industry. A custom of eating fish on the Sabbath (Safrai 1974/76b: 747) may have obtained this early, though Galileans near the lake surely ate fish much more regularly. For fishermen’s livelihoods depending on the sea, see, e.g., Ps-Theocritus The Fishermen (Greek Bucolic Poets, LCL 246-53).

. Jewish Palestine may not have employed economic structures more complex than fishing cooperatives, but other parts of the Empire, such as Ephesus, knew influential fishing guilds (see Horsley 1989: 101-7; cf. P. Oxy. 3270.5, 8 in Horsley 1983: 18).

. Traveling teachers were not the norm in Judea and, though attested, apparently remained rare even in Galilee (cf. Vermes 1993: 46, citing b. Ḥul. 27b; Sanh. 70a), but also not as uncommon (and Greek) as Robbins 1992: xxi, 101-5 and others have supposed (Greek examples may be significant but exceptional as well; cf., e.g., Hock 1980: 27-28); see the evidence for perambulatory or traveling lectures in Koenig 1985: 17; Young 1989: 234n.87 (e.g., m. ʾAbot 3:7; Sifre Deut. 305; ARN 4A).

. Boring 1995: 54-55 adds Aristotle On Philosophy 1.53. Davies and Allison 1988: 400 also cite among Jewish accounts Jos. and Asen. 13 and parallels in the sacrifices of the Essenes and Therapeutae (1QS 1.11-13; 6.20-23) found praiseworthy in hellenistic Jewish apologetic (Philo Omn. Prob. Lib. 76-77, 85; Vit. Con. 25-31; Jos. War 2.122; Ant. 18.20); Davies and Allison 1997: 46 add Philost. V.A. 1.13; Porphyry Vit. Plot. 7. Ancient sages commended those who paid a price for being disciples of the sages, including, apparently, the disdain of some common people more interested in other pursuits (ARN 11A).

. On Aqiba, Sandmel 1978b: 246-47; Witherington 1984: 10; others, Gen. Rab. 95 MSV. Later teachers assumed such long separations to be consensual (p. Ketub. 5:7, §2; cf. Qoh. Rab. 7:6, §2); they also had to explain what happened to Aqiba’s additional disciples acquired during this period (e.g., Qoh. Rab. 11:6, §1).

. See also Safrai 1974/76a: 965; Sifre Deut. 43.3.7.

. Strangely, Gerd Theissen contends that wandering Palestinian charismatics known from the Jesus tradition were Paul’s opponents; yet Paul himself constitutes our best concrete evidence for a literal practice of this lifestyle. Many arguments for continuing itinerant prophets in the Gospels themselves are tenuous (cf., e.g., Tannehill 1975: 139-40; Hill 1979: 154; Aune 1983: 211-16, 243; Stambaugh and Balch 1986: 105).

. On summary reports in ancient literature, see Aune 1987: 54; in the biblical Kings narratives, see, e.g., 1 Kings 14:19-20, 31; 15:8, 24; Rosner 1993: 76.

. Grassi 1989 suggests that Matthew saw his own work as a “gospel of the kingdom,” the way Deuteronomy repeated the earlier Torah; this view may be correct even if not demonstrable from 4:23. “Gospel” had a widespread secular usage (see Diogenes Ep. 23 to Lacydes; Milligan 1908: 141-44; Horsley 1983: 10-15, citing P. Oxy. 3313; Longus 3.33 for wedding announcements; in the LXX, e.g., Jer 20:15), but the early Christian usage echoes particularly LXX Isaiah (e.g., Is 40:9; 52:7; 60:6; 61:1; cf. Ps. Sol. 11:1). Some older authors wrongly associated the term too particularly with hellenistic cults (Nock 1933b: 132; e.g., Epict. Disc. 3.21.13).

. At least one other ancient Jewish source invoked God’s reign in exorcism (4Q510 1.4; Vermes 1993: 130), though this practice makes good sense; for more general observations on exorcism, see comment on 8:17.

. Although skeptical of the story of the Syro-Phoenician woman (which he surprisingly thinks created to illustrate the early Christian view of missions), Meier 1994: 646-77 decides in favor of the authenticity of some specific exorcism accounts, thus also accepting the category as a whole as probable.

. Although some attributed sickness, especially epilepsy, to demonic sources (e.g., Aramaic Incantation Text 53:12; Jub. 10:12; cf. Yamauchi 1986: 100-113), Matthew distinguishes the two in 4:24 (cf. Alexander 1980: 32, 63). He may think demonization could at times produce epileptic behavior (17:15), but his term for epilepsy did have a broader usage than epilepsy alone (cf. Ross 1978; Yamauchi 1986: 129).

. Though some objected to this use of “all” (Rhet. ad Herenn. 2.20.32-33, 21.33), the hyperbolic “all” appears in Josephus, for whom “all Galilee” may refer to the Galilean aristocracy (War 1.291; Ant. 14.295; Horsley 1995: 56), and for whom “all” Balaam’s predictions were fulfilled, but the others that have not been will be (Ant. 4.125).

. For later examples, see, e.g., CIJ 1:476, §662; 1:495, §683; 1:497, §684; 2:367, §1440; 2:369, §1442; perhaps 1:525, §726; 2:360, §1432; 2:361, §1433. Jos. Life 277 also freely applies the hellenistic title to a Galilean structure; the favored title seems to have varied geographically (Applebaum 1974/76b: 490), perhaps until rabbinic influence became more widespread (Schubert 1981).

. From the earliest period (well before A.D. 70) writers associated the synagogues with the Jerusalem temple (Cohen 1986: 163). The associations grew after 70, however, and probably still further under later Byzantine influence (Levine 1996: 446-47).

. No form was mandatory; various architectural types existed through the medieval period (Meyers 1986: 128-32); on pre-70 Judean types, see Chen 1980.

. Well-to-do persons sometimes donated synagogues (Lk 7:5; CIJ 2:8, §738; Theodotus inscription). Some synagogues probably met in well-to-do patrons’ homes, as house churches later did (Meyers and Strange 1981: 141); a private home lies beneath the later Capernaum synagogue (Strange and Shanks 1982: 29-30).

. Cf. also Jos. War 2.289; Ant. 16.43. As Sanders 1992: 199 points out, Philo recognizes a “house of prayer” (Embassy 132), and Jews learned Torah in an assembly on the Sabbath (Hypothetica 7.12-13; cf. Spec. Leg. 2.62-63; Every Good Man Is Free 81). First-century Jews believed that Moses required this (Philo Hypothetica 7.12-13; Jos. Apion 2.175; Ps-Philo 11:8; Sanders 1990: 78).

. Synagogues nevertheless regularly appear as positive places of worship in rabbinic texts; see, e.g., t. Sukk. 4:6; b. Meg. 28ab; p. Taʿan. 3:11, §4. Later rabbis both projected the institution anachronistically into the distant past (e.g., Lam. Rab. Proem 2) and exaggerated their numbers in the more recent past (e.g., Lam. Rab. 3:51, §9).

. On Capernaum’s synagogue, see Strange and Shanks 1983, although the remains of this structure may be later than they think (cf. Chen 1980: 38; idem 1986); less recently, Goodenough 1953-68: 1:181-92. Although the third-century synagogue was 65 feet long and two stories high (resembling some other magnificent synagogues of the period), its first-century predecessor was likely smaller (Sanders 1993: 103)

. See, e.g., Jos. War 2.614; 4.11; Life 85; Pliny N.H. 5.15.71; Pesiq. Rab Kah. 11:16; Qoh. Rab. 10:8, §1.

. Jos. War 1.657; Pliny N.H. 2.95.208; 5.15.72; Hirschfeld and Solar 1980; idem 1984; Schürer 1961: 157.

. The springs of Tiberias were well known (e.g., b. Sanh. 93a; 108a; Shab. 40b; 147a; p. Ned. 6:1, §2; Gen. Rab. 76:5; other references in Urbach 1979: 1:393; Sandmel 1978b: 201; for the synagogue there, see Goodenough 1953/68: 1:214-16; Shanks 1984). Prayer for healing was common (cf. b. Ber. 60b), and appears in the ʾAmida (p. Taʿan. 2:2, §7).

. For aristocratic disdain of the masses and their demagogues, see Isoc. To Nic. 48-49, Or. 2; Areopagiticus; Arist. Pol. 3.6.4-13, 1281q-82b; 3.7.7-9, 1283ab; 3.12.1, 1288a; 4.4.4-7, 1292a; 5.4.1-2, 1304b; 5.5.4-5, 1305b; 6.2.10-12, 1319b; Arius Didymus Epitome (Malherbe 1986: 145); Polyb. 6.3-4; Plut. Praising Inoffensivly 16, Mor. 545C; Dion. Hal. 7.8.1; 7.31.1; 7.56.2; 8.31.4; 10.18.3; Diod. Sic. 10.7.3; 15.58.3; Diog. Laert. 6.42; Diog. Ep. 11; Philo Creation 171; cf. Gardner 1974: 27-59. The masses’ support, however, could prove critical in some situations (Dion. Hal. 7.45.4).

. So Davies and Allison 1988: 417, citing mishnaic usage: m. Dem. 6:11; B. Qam. 7:7; Rosh Hash. 1:4; ʿEd. 7:7; ʿAbod. Zar. 1:8; Sheb. 6:2, 5, 6; Maʿaś. Sh. 5:5; Ḥal. 4:7, 11; ʿOr. 3:9; Ohol. 18:7.

. Cf., e.g., Parker 1975; Smith, McNicoll, and Hennessey 1981; Beare 1981: 122.

. Some early Jewish Christians had experienced this temporary acceptability among their contemporaries; cf. Acts 21:20-21; Jos. Ant. 20.200-203. Possibly the general acceptability continued at the time of writing, but Matthew finds in Pharisaic opposition an ominous warning that the future may prove different.

↑ Back to top