your relationship to your neighbor rests on mutual attested in H, Georgian (without 1.5-6), and indirectly in

your relationship to your neighbor rests on mutual attested in H, Georgian (without 1.5-6), and indirectly in

1.2-3a bor. That means, however, that the fundamental com mandment whose fulfillment leads "to life" is:

2.2-7 (ov ipovevaeiq, ov ixoixevoeig ktA.),22 that is, by a list of prohibitions that, in the context of the source, had the function of expanding and concretizing the basic commandment through special commands.23 The text attested in H, Georgian (without 1.5-6), and indirectly in Constitutions (up to 5a) offers, before this, a rather long insertion (1.3b—2.1) that appears to be specifically

Christian.24 Its great antiquity is demonstrated also by P (for 1.3c-4a).25 There would be further evidence if 1.3d and 5c were really attested by quotations in the Syrian

Didascalia apostolorum l.26 The insertion at Did. 1.3b—2.1 can be described as a "Christian interpolation" or "sectio evangelica." The paired questions of the originality of the sectio and the source of its material are among the crucial problems of the Didache.

1. Love God (demand) who created you (reason).

2. Love your neighbor as yourself (demand), because your relationship to your neighbor rests on mutual avoidance of evil (reason).

The summary may be secondary, but it is, in any case, well thought out. It clearly bears a didactic, catechetical character.

■ 3a The quotation of the "basic commandment" was followed, in the source, by a kind of explication of this demand. It is indicated and introduced by the expression

"This is the teaching20 [that comes] from these words"

(tovtuv de 7W AoytiP rj dtdaxv21 avrr]). In the source, this was followed immediately by the text of commandment and an explication through a series of prohibitions is lacking. Rordorf ("Gebrauch des

Dekalogs," 432-34) has pointed out that in early Christian times there was a tradition of combining the Golden Rule with the second table of the Decalogue. The Didache is the oldest example of this, but we should also note that here the Didache is quoting an originally Jewish source.

24 More precisely: the text fragment (fol. 1) begins with ovx'l xai . . . and ends with oapKetKuip e-ncdvixeiuiP. See above, Introduction, §3b.

25 By contrast, 1.3b—2.1 is lacking in Barnabas, Doctrina apostolorum, Canons of the Apostles, Epitome, and there is nothing corresponding to it in the Life of Shenoute, Syntagma doctrinae, and Ficles patrum.

26 (CSCO 401.41 and 402.12), Latin 2.28-31 (TU 75.4); also Syriac 17 (407.177 and 408.161), Latin

37.25-26 (75.63) and Syriac 17 (407.179 and 408.162), Latin 38.23-25 (75.64): see above,

Introduction, §5.

b. The sectio evangelica (1.3b-2.1) piece of redaction that was of special significance for the Didachist and his writing. A series of good argu ments can be advanced for this second opinion. First, the sectio is well attested in the Didache tradition; it appears already in P (P witnesses to 1.3c-4a), which is the oldest text fragment of the Didache we possess; the Apostolic Constitutions also presumes a Didache text that already contained the sectio. There would be further evi dence of this if 1.3d and 5c were really attested by quota tions in the Didascalia apostolorum. It is true that these arguments are not sufficient by themselves, but an observation regarding the content may be brought for ward in addition: Kara rrju evToXpv (1.5) is a typical redactional expression of the Didachist; dAAo: ku'l Ttepl tovtov de elprjTai, with a quotation following (1.6) also points to his hand; and finally, the significant expression

Kai eat] TeXecoq (1.4) is taken up, as reAetoq eot], in 6.2, which is certainly redactional. Indeed, it is only at 6.2 that a decisive light is shed on what the text of the sectio

General Without any doubt, 1.3b—2.1 represents a subsequent interpolation in the text of the "basic document." The text of the sectio is lacking, as we have just observed, in all the recensions of the Two Ways tractate except for that in the Apostolic Constitutions and its derivatives.1 The sectio is thus found only in the Didache and in writings that directly or indirectly derive from the Didache. From the point of view of content also, it is quite clear that the sectio can never have belonged to the original Jewish Two

Ways tractate; here, in the sectio, we find characteristic motifs from Jesus' preaching, elements of the Synoptic tradition that are otherwise completely absent from the Two Ways tractate. Add to this, finally, that 2.2 follows seamlessly after 1.3a. In 2.1 the redactor had to create a more or less successful transition, a subsequent connec tion designed to restore the direct link in the source that had been destroyed. From this point of view it is sensible to describe 1.3b—2.1 as a Christian interpolation.

The only question remaining is: Who functioned as interpolator? A whole group of interpreters assumes that this section represents a jbostdidachistic interpolation, that is, an insertion that belongs not to the original his tory of the document but to the history of its transmis sion.2 The contrary opinion holds that the sectio comes from the Didachist.3 In that case, what we have before us would not be an interpolation in the strict sense but a is meant to convey in the context of the whole docu ment.4 In conclusion, 1.3b—2.1 goes back to the Dida chist, who interrupts his source at this point and enriches it with a series of reminiscences drawn from the Synoptic tradition. It is therefore better to speak of the sectio evangelical That does not mean that the material of the sectio must have been an ad hoc invention of the Didachist. It may have existed, in whole or in part,

1 Fragmenta Anastasiana (Funk, Didascalia, 2.64); something analogous is true of Isaac the Syrian (ed.

Besson, 51-52). For these two documents, see above, Introduction, §5.

2 Harnack, Apostellehre, 32-33; Streeter, "MuchBelaboured Didache," 369-70; Creed, "Didache,"

374-79; Audet, Didache, 261-80. (For Audet the argument that the text of the sectio differs from its context by being couched in the second person sin gular also plays a role. Against the attempt to draw literary-critical consequences from this, see Nautin,

"Composition," 199-200, and passim.) In addition, Vielhauer (Geschichte, 733) considers the sectio prob ably a subsequent interpolation. Wengst (Didache,

20, and passim) now also speaks in favor of interpo lation.

3 For example, Drews, "Apostellehre," 258; MacLean, "Introduction," ix (probably); Nautin, "Composition," 200-201; Glover, "Quotations," 12;

Koester, Synoptische Uberlieferung, 220-26, 238-41 (the interpolation stems from the compiler of the Didache, who is to be distinguished from the redac

Mattioli, La Didache dottrina dei dodici apostoli:

Introduzione, traduzione e note (Rome: Edizione Paoline, 1969; 3d ed., 1980) 42-43, 130; Rordorf and Tuilier, Doctrine, 100, and frequently; Rordorf,

"Transmission textuelle," 513. The advocates of the fiction hypothesis have also argued for an origin with the Didachist (but with both correct insights and, in part, false presumptions, and sometimes with questionable argumentation): Muilenburg,

Literary Relations, 44-47, 144; Dix, "Didache,"

243-45 (at least for 1.3b, 4, and 5a); Robinson, Didache, 56-58; idem, "Barnabas and the Didache,"

230-38; Connolly, "Streeter on the Didache,"

364-79; Vokes, Riddle, 32-33, and passim. The authenticity of the section 1.3b—2.1 was important for the advocates of the fiction hypothesis, because here in particular it appeared possible to demon strate that the Didache is dependent on the NT and Hermas. But the question of the origins of the sectio is to be kept separate from the question of its pos sible sources.

See below.

The principal difficulty in accepting the authentictor of the material in the sectio). See Umberto

1.3b—2.1

Table 6 Did. 1.3b-d Matt 5:44, 46-47 (44) v .

Luke 6:27-28, 32-33 (27) .;. ayanare rovq exdpovq vpwv dyairdre rovq ex&povq vptip, KaAwq irotetTe roiq pioovotv vpaq, (28) evApyelre (3b) evAoyelre rovq Karapuspevovq vplv Kal irpoaevxeade virep rav exdpav vpav, vparevere be virep tuv biUKOvrup vpaq.

(3c) wota yap x«P<-C ear dyairdre rovq Karapuipevovq vpaq, -irpooevxeade Kal irpoaevxeade virep rap biwKOVTuv vpaq (...) (46) eat' yap ayairpapre rrepl ru>p eirppea^oproip vpaq.

(32) Kal el dyairdre rovq ayairupraq vpaq; Toiiq dyairwpraq vpaq, Tiva ptodop eyere; ovxl Kal oi reAcorat to avro ttolovolp; (47) Kal ear doiraopode rovq ayairapraq vpaq, iroia vplpxaprq earth Kal yap oi apapmAol rovq ayairupraq avrovq dyairaotP.

(33) Kal yap eav ayadoirotfjre rovq abeA(povq vp&p popop, ri ireptaaop iroieire; ovx'l Kal oi edvtKol rovq dyadoiroiovvraq vpaq, iroia vplv xapiq eariv; Kal oi apapraAol ro avro rroiovoiv.

[(27) dyairdre oi)xtKal ra edpp

TOVTO 1TOLOVOIV;

(3d) vpelq be (prAelre TO aVTO ttolovolp; [(44) dyairdre rovq praovpraq vpaq, Kal ovx/e^ere eyidpov. rovq exdpovq vpu>v ...] rovq exdpovq iipuiv, KaAuq irotelre roiq ptaovatp vpaq ...] P. Oxy. 1224, fol. 2r., col. 1 (Grenfell and Hunt, x) = K]at rr[p]oaevxeade virep [tup ex$]par vpup. as fixed oral tradition or written source.6 The apparent occasion for the Didachist's interpretation was the key phrase "love of neighbor," which he—as a Christian—saw as being made concrete and fulfilled in terms of Jesus' commandment of love of enemies. In this way the bap tismal candidates are made acquainted at the very begin ning with a unique characteristic of their new condition.7 these things were already combined in his tradition?) the prohibition against resistance (v. 4b-d). This is followed in vv. 4e-5a, by way of transition, by the admonition to permit oneself to be robbed, and this in turn is followed, in v. 5b-d, by a parenetical statement about right giving.

(There is a parallel in Hermas Man. 2.4-6.) All this con cludes with a scriptural quotation (v. 6). The material of The sequence of the interpolation does not adhere to the dominical saying about love of enemies. The Didachist includes (on his own initiative, or because ity of the section was always its omission in the par allel versions. But this is easily explained if one sup poses that this section was not in the Two Ways tractate and was first introduced by the Didachist in his own text. That explains the literary evidence as a whole.

The expressions of Koester (Synoptische Koester thus distinguishes between the compiler of the Didache and the redactor of the sectio (p. 218).

"The motive for the insertion is clear ... it has Christianized this Jewish morality. This addition to the Duae viae is thus inspired by a desire to adapt it to Christians" (Nautin, "Composition," 201).

Similarly Rordorf, "Tradition apostolique," 108.

Uberlieferung, 217-19) tend in the latter direction.

Table 7 Did. 1.4b-5a Matt 5:39-42 Luke 6:29-30 (4b) eav Ttq oot (39b) aXX ooTLg oe

(29a) TQ TV1TT0PTL 06

da painopa pairi^ec eiri ttjv otayova eig trip de^idp otayopa elg Tr)p de^cdp ocayopa oov, OTpefop avTQ orpefop avra 7rapexc Kai rr}p dXXpp, Kai rr]p dXXrjP .. .

Kai Tpp dXXr\p, Kai ear? TeXetog.

(4c) edp dyyapevorj oe Tig (41) eai doTtg oe ayyapevoer piXiOP ep, pLXiop ep, viraye per avrov dvo. viraye per avrov dvo.

(4d) eav aprj rtg (40) Kai ra deXovTi oot (29b) Kai airo tov atpovToq oov KpLdrjpac Kai to ifxaTtov oov, top xnapa oov Xa/3elp, TO uaTtov dog avra onpeq avTu Kai tov xtT&va Kai tov \tTuva'

Kai TO LfJLaTLOV. pi] KaXvoijg.

(4e) edp AdjSrj tic aiTO OOV TO oov, pr] d'KatTer oiide yap dvpaoat.

(5a) iravTt (30) iravTi Ta airovPTL oe dldov (42) ra aiTovpri oe dog aiTovPTL oe didov, Kai top deXopra airo oov dapioaodal Kai airo tov atpovToq to. aa Kai fir) airaiTet. prj aTtooTpa(prfq pr) airatTet. course, by representatives of the fiction hypothesis,

Robinson,14 Connolly,15 and Vokes.16 Within this tradi tion is also the attempt of Layton,17 but he investigated only the sectio evangelica. For him, the "compositor" of the sectio is thus not homogeneous. Where does it come from?8

The literary data have not been clearly analyzed to date. (1) The simplest assumption is that the Didache is dependent on our Synoptic Gospels, in this case

Matthew and Luke. This position was advocated espe cially (in the wake of some precursors)9 by Massaux.10 It is consistent with this position to suppose that Did. 1.5 shows additional literary dependence on Hennas. This is supported, for example, by Muilenburg,* 11 Goodspeed,12 and Richardson.13 The same position is upheld, of 1.3b—2.1 was a very deliberate writer who took material from Matthew, Luke, and Hermas, and in v. 6 from the Old Testament as well,18 while making stylistic changes in his sources.19 In this the author of the interpolation was guided by an intention to make his text give an archaic, "apostolic" impression, corresponding to the fic tionalized title with its claim to apostolic authority.20

8 Cf. tables 6-8. These tables are adapted from Koester, Synoptische Uberlieferung, 220, 227, and 231.

9 E.g., Funk, Patres apostolici, 1.4-59; Bigg and MacLean, Doctrine, 2.

10 Influence, 3.144-82. This section is primarily depen dent on Matthew, but there are additional influ ences from the Gospel of Luke. Hagner (Use, 280) also held that the Didache had used the Gospel of Matthew. 15

11 Literary Relations, 46,167. 16

12 "Doctrina," 231. Goodspeed sees material from 17 Matthew 5, Luke 6, 1 Peter (!), and Hermas Man. 18

2.4-6 redacted here.

13 "Teaching," 163, 165.

14 Didache, 47-50; "Barnabas and the Didache,"

230-38. The modifications and additions by the Didachist are supposed to serve to cloak this depen dency (p. 230). Beyond this Robinson even consid ered (p. 232) the possibility of influences from 1

Pet 2:11; John 18:22; 19:3, and—with reservations— even 1 Cor 7:7 (p. 236).

"Streeter on the Didache," 367-70.

Riddle, 51-61.

"Sources," 343-72.

This, in turn, has consequences for one's decision about what the "compositor" thought of Hermas: he

1.3b —2:1

Table 8 Did. 1.5

Hermas Man. 2.4-6 Luke 6:30

(5a) iravri tco anovvii oe (4c)

7T OLO LV

(30) navii anovvii oe bibov btbov bibov, Kai and tov aipovroq * koll fxrj aira'crer p,rj ana'nei.

(5b) iraoi yap naoiv yap Doctr. 4.8 deXei biboodat 6 debt; omnibus enim 6 iraTrjp bibooflai deXei dominus €K TUP ibiUP eK tup IbLup dare vult XG:pLOpL(XT(0V. bupppaTup. de donis suis.

(5c) paKapcoc; (6a) 6 biboiiq 6 ovp biboiit; Kara rr)v ePToXpp' a-doioq yap eonp. ocduoq eoTLP ... oval Hermas tui XapfidpoPTv (5a) oi ovp XapfidpoPTec; (5c) oi be ep inroKpioei el \JL6V yap Xap/3aP0PTeg Xpeiap exup Xap/3dpei rig, aduoq carat • 6 be pi) xpctaf exup buoet biKrjv, dTTobuoovoiP Xoyop tu deu TLOOVOLP biK7]P.

LparL eXafie Kai etg rtStart eXafiop Kal elq tv Matt 5:25c-26 Luke 12:58c-59 (5d) ev ovvoxxi (25c) . . . Kal elg (pvXaKrjp (58c) . . . Kai ... be yepppepoq pXpdijoy. ae jSaAet eiq (pvXaKi)p. eijeraodrioeTai. rrepl up enpal-e.

(26) apr/p Xeyu aoi, (59) Xeyu ool, Kai oiiK e^eXevaerai oil pr) e^eX'drjc; oil pr) e^eXdrjq eKeldep, eKeWep, eKeidep, pexPiQ ov airobu euc~ dp axobuq €C0£ Kai TO eOX&TOV top eaxoiTOP KobpaPTpp. top eoxawp KobpaPTr/p.

Xemop airobuq.

(2) Among the older opinions that, in my view, have by no means been overcome by Layton is Glover's hypothesis,23 assigning the passages in the Didache that Layton's stylistic judgments are interesting, but whether they are sufficient to demonstrate literary dependence on Matthew, Luke, and Hernias21 is debatable.22

JTS 34 (1933) 347; Audet, Didache, 269-70; Koester, may have regarded it as a canonical work (Layton,

"Sources," 362).

19 "Rhetorical climax, chiasmus, Blessing-and-woe, refrain" ("Sources," 370 n. 51; cf. the analyses on pp. 352-61). Layton explains the differences from the source texts as the consequence of stylistic metamorphosis.

20 "Sources," 371-72.

21 Wengst (Didache, 19) expresses cautious agreement.

22 Out of the question is the idea (previously advanced from time to time: cf. Dix, "Didache," 242-50) that the Didachist was quoting the Diatessaron. To the contrary, see Connolly, "Didache and Diatessaron,"

Synoptische Uberlieferung, 229-30; Layton, "Sources,"

351.

23 "Quotations," 12-29.

analysis is that of Rordorf.33 His results lie somewhere between those of Koester and Audet. Rordorf assesses34 recall the Synoptic material to a tradition parallel to Q.

As regards the material in 1.3b-5a, this is interesting to the extent that virtually all the parallels24 point to the Q tradition.25

1.3b-4 and 1.5-6 as two passages that were brought together even before they were introduced into the Two

Ways tractate. The core of 1.3b-4 is said to go back to old tradition, contemporary with our Synoptic Gospels but independent of them. The independence of the tra ditions is said to be shown by expressions like edvr) in (3) The already venerable hypothesis that posits an otherwise unknown apocryphal sayings collection as the basis for the interpolation goes a step farther: thus, first of all, Drews,26 who thought that this was a written source of dominical sayings bearing the title et'roAr); the material in vv. 1, 3, 4, 5a, and the core of v. 5b-d, as well as v. 6 would have been derived from that source.

Koester's reflections tend, to some degree, in a similar direction.27 He gave this text an extensive analysis and came to the very complicated conclusion that the inter polator's basic source was a "finished collection of say ings"28 produced by an unknown redactor who used texts from Matthew and Luke (expanding and altering them) and inserted a piece of Jewish traditional material and a scriptural quotation, the derivation of which is problematic.29 This source was copied by the compiler

(i.e., the Didachist).30

1.3 and btbovat pamopa in 1.4.35 In addition, according to Rordorf, 1.4 (apart from the expression nai aco/xcmkwv) could go back to this old tradition. The original text of 1.3b-4 has, however, been amplified. Did. 1.5 is an old tradition that recalls Luke without quoting him, but 1.5 has grown in stages and was also amplified at a later time. Thus one must distinguish between the old

(and partly augmented) tradition behind 1.3b-4 and 1.5 and the later didachistic redaction. The old tradition was redacted by the Didachist, from whom come the commandment about fasting (1.3; see later at 8.1), the expressions nai eorj TeXetoq and ovbe yap bvvaaat (1.4; cf. 6.2), as well as the end of v. 5 (ev owoxv kt\.) and v.

6. Important in Rordorf's exposition are above all the references to archaic elements in the sectio and to later, didachistic redaction, by means of which the text as given was integrated into the macrocontext. In the nature of things, some details must remain debatable.

(4) It is also possible, however, to set aside the idea of any written source and to consider the possibility of a purely oral tradition31 that would then—as far as the dominical sayings are concerned—constitute a parallel tradition to the canonical description. This is Audet's position.32 It would then be possible in individual cases to find formulations that, under certain circumstances, might be older than the corresponding expressions in the Matthean and Lukan tradition. That represents more or less the opposite of Layton's position.

(5) The most recent attempt at an extended literary 24 It is true that the only parallel to Did. 1.4c is Matt

5:41. One may at least ask whether the Matthean text also goes back to Q.

25 The parallels to 1.5b-d are a subject in themselves.

[Ed. For dependence on traditions related to Q, see now also Clayton N. Jefford, The Sayings of Jesus in the Teaching of the Twelve Apostles (VCSup 11;

Leiden: Brill, 1989) 38-53; and Andre Tuilier, "La Didache et le probleme synoptique," in Jefford, dependence on a collection of dominical sayings substantially derived from Matthew but expanded from memory (and thus including some Lukan rem iniscences).

31 Wrege (Bergpredigt, 77 n. 1) points to the repetitions (Su pamopa in 1.4b; boq avrqi in 1.4d, and several instances of bibovat in 1.5). "This constant repeti tion of one verb is part of the technique of oral tra dition."

32 Audet, Didache, 166-69, 183-86, 264-65. Cf. "oral style" (p. 264). Still, the interpolator made some additions. Audet counts among these vyorevere be vttep 7w 5lo)k6ptg}v vpaq / Kai ovx e^ere exdpov/nai eoy reXetoq /ovbe yap bvvaoat (p.

271, and passim). Regarding 1.5, Audet is uncertain about the extent of the interpolator's intervention.

Michael Mees ("Die Bedeutung der Sentenzen und ihrer auxesis fur die Formung der Jesusworte nach

Context, 110-30.] 26 "Untersuchungen," 64-67.

27 Synoptische Uberlieferung, 217-19.

28 Synoptische Uberlieferung, 239.

29 Cf. esp. the summary in Synoptische Uberlieferung, 238-39.

30 Aono (Entwicklung, 186-87) supposed for the sectio (differently from the rest of the Didache, in which the author was indirectly dependent on Matthew)

1.3b-4a (1) The Commandment of Love of Enemies (1.3b-4a) 3b-d Bless those who curse you, and pray for your enemies, and fast for those who persecute you.

For what grace do you expect if you (only) love those who love you? Do not even the nations do that? As for you, love those who hate you, and you will not have any enemy.

[4a Avoid the fleshly and bodily passions.] kcu irpooeux^ode virep twv ex^pwv vptivf9 vr)OTev6T€ be virep tuv bupKOVTtpv vpaq.

(3c) iroia yap ear ayanrare rovq dyairavraq ■ 3b The commandment that introduces and, in a sense, dominates the sectio,36 to love one's enemies (3b-d),37 appears (without being expressly designated a saying of the Lord) in a form that is structurally independent, while its content recalls the Synoptic tradition (particu larly the logia in Q):

(3b) evXoyelre rovq Karaptopepovq vpiv,38 vpaq-40 ovxl kozl ra e&vrj tovto41 -kolovolv Didache 1,3b—2,1," VetChr 8 [1971] 55-76) comes in his own way to what is in principle a similar conclu sion: the text as handed down does not stem from Matthew and Luke, but is a new version of motifs from the Jesus tradition composed according to the rhetorical laws of auxesis.

35 "Transmission textuelle," 499-500.

34 "Transmission textuelle," 509-10.

35 The problem of the two variants was seen by Glover ("Quotations," 14), as Rordorf notes ("Transmission textuelle," 502). Glover drew his own conclusions from them. Regarding dcdovou pamopa, Rordorf refers to Wrege, Bergpredigt, 76 n. 2: "Did. 1.4, with its did oven pamcrpa, recalls Isa 50:6, and in any case is independent of Matthew."

36 Theissen ("Nonviolence," 118 n. 5) draws attention to the special place of the commandment of love of enemies in the Matthean and Lukan compositions.

"Something similar is true of the Didache (Did.

1.3ff.). Love of enemies always has a prominent compositional placement."

37 Koester (Synoptische Uberlieferung 223-24) has cor rectly pointed out that the logion first appears as a command to pray for enemies; this, then, represents a characteristic reformulation of the kind experi enced by the tradition in the postapostolic period.

Cf. Koester's examples, p. 221. The Didache's logion is, however, constructed in such a way that, while at first it commands prayer for enemies (evXoyeZre, irpoaevxeaPe), this is done in order to formulate the commandment to love ((pcAeZre) them. Thus (at least as far as the Didache is concerned) it cannot be said that Jesus' original command was softened

(Koester, Synoptische Uberlieferung, 224). On the commandment of love of enemies in the Synoptic tradition, cf. John Piper, "Love Your Enemies": Jesus'

Love Command in the Synoptic Gospels and in the Early Christian Paraenesis: A History of the Tradition and Interpretation of Its Uses (SNTSMS 38; Cambridge:

Cambridge University Press, 1979); Luise Schottroff, "Gewaltverzicht und Feindesliebe in der urchristlichen Jesustradition," in Georg Strecker, ed., Jesus Christus in Historie und Theologie:

Neutestamentliche Festschrift fur Hans Conzelmann zum 60. Geburtstag (Tubingen: Mohr, 1975) 197-221;

Theissen, "Nonviolence," 115-56; Fritz Neugebauer, "Die dargebotene Wange," ThLZ 110 (1985)

865-76.

38 Karapaodai tlvl is Attic (in place of tivgl, as in Luke 6:28; cf. Mark 11:21; Jas 3:9): BDF §152(1);

BDR, §152(1).

39 Cf. Justin 1 Apol. 15.9 (Goodspeed, 35): EvxeoPe vnep tup exftpuv vpuv . . ., where the Didache text agrees with Justin against the Synoptics. Glover

("Quotations," 14) referred to the connection as an indication of the possible use of Q material. The Didache text is, however, even more literally in agreement with P. Oxy. 1224. Georg Strecker ("Eine

Evangelienharmonie bei Justin und Pseudoklemens?" NTS 24 [1978] 310) explained the agree ment between Justin and the Didache on the basis of a statement going back to the written gospel tradi tion in Matthew and Luke, but freely quoted.

40 Audet reads, with the Apostolic Constitutions, <piAr\TC rovq tpiAovvraq vpaq.

41 H has to avTO, probably an assimilation to Matt 5:46 (where a variant reading also offers tovto) and Luke 6:33. The correct text is in P (and Consti tutions).

(3d) vpeiq 5e ipiXeire42 rovq paaodvraq v\uaq The third line, "fast for those who persecute you," vrjoTeveTe de inrep tu>v bmnovTuiv vpotq, has no parallel in the Synoptic tradition of this logion. From the point of view of tradition history it may be, and probably is, a later addition.46 It is parallel to the two preceding clauses (evXoyelre, ... nal npooevxeode). Fasting sup ports and strengthens prayer: Polycarp Phil. 7.2; Act.

Kai ovx eijere ex^pov.

Structurally, v. 3b is presented in three lines with parallelismus membrorum, but one may suppose (see below) that, in the history of the tradition, the third line was added at a later time, and that the whole pericope is marked throughout by a system of paired lines in parallelismus membrorum. The introduction (v. 3b) offers an apparently paradoxical commandment—deliberately for mulated, in fact, as a paradox—for which the reasons will be given in the lines to follow. The structure and content point to wisdom (hokmah) tradition.

Thom. 20, p.131; 145, p.253. What is intended in this passage is probably a special, private fast,47 which has the same petitionary character as prayer (cf. urjarevere virep . . .). Comparable for this subject could be such texts as 2 Sam 12:15-23; Esth 4:16; Didasc. (Syriac) 21

(CSCO 407.209-10; 408.193-94). The point of this logion, however, is that the deed of love is to be done for the persecutor,48 This is the spirituality of the "Sermon on the Mount."49

A saying similar to v. 3b was found in Q, from which it entered Matt 5:44/Luke 6:27-28. Jesus' saying very quickly became proverbial among Christians, however; cf. Rom 12:14; P. Oxy. 1224, fol. 2r., col. 1 (Grenfell and Hunt, x);43 2 Clem. 13.4; Polycarp Phil. 12.3; and so on.44

The paradox of Jesus' eschatological proclamation is clearly evident. 'Ex&poq is the enemy who acts aggres sively against me by cursing me; the "normal" reaction would be to return the curse. But no! For Jesus' followers in such a situation, the commandment was: a blessing for the one who curses, and a prayer for the enemy!45

The paradoxical commandment calls for foundation and explanation (v. 3c-d). These are offered in two forms:

(a) "To love only one's friends would be nothing spe cial: the pagans do that, too!50 It is only the love of ene mies that obtains special favor and a special reward,51 that is, before God!" The content of the logion in v. 3c

42 H has dyairdre. The correct text is in P and the Apostolic Constitutions. Was H influenced by Matt a variant of the free oral tradition or something drawn from free citation.

45 Karlmann Beyschlag describes Jesus' command of love of enemies as the "most radical demand of Jesus for Christians in the first centuries" ("Zur

5:44 and Luke 6:27?

43 For this, see NTApoc, 1.100.

44 The variant irpooevxeoPe virep rav vpoiv (against Q) is also found in P. Oxy. 1224; Justin

Geschichte der Bergpredigt in der Alten Kirche,"

ZThK 74 [1977] 314).

46 Koester {Synoptische Uberlieferung 224) finds here "influence of church custom. . . . Undoubtedly, this reduction to a practical moral admonition is also a sign of a later time." Does the description of this as a "reduction" really fit? Layton ("Sources," 353) calls it "an extra-Synoptic addition by the composi tor of the Didache passage." Thus also Rordorf,

"Transmission textuelle," 501-2, 510 (the redactor).

In fact, one cannot exclude the possibility that the passage is redactional.

47 Audet, Didache, 265: the reference is to fasting from one evening to the next.

48 Rordorf ("Transmission textuelle," 502) sees the addition in light of 8.1 and the Syrian Didascalia and concludes that the reference is to Jeiuish perse cutors, and that the fasting refers to the two fast days in 8.1. Cf. also Rordorf and Tuilier, Doctrine,

1 Apol. 15.9 (Goodspeed, 55); cf. also 1 Apol. 14.3

(Goodspeed, 34); 96.3 (Goodspeed, 211); 133.6

(Goodspeed, 256); Didasc. Syriac) 21 (CSCO 407.209; 408.192): Pray for your enemies. Arthur

Voobus considered "pray for your enemies" an archaic reading of Matt 5:44 (Studies in the History of the Gospel Text in Syriac [CSCO 128; Louvain:

Secretariat du Corpus SCO, 1951] 137-38). He pointed not only to the passage in Justin but also to Passio Ignatii 8.5 (Funk and Diekamp, Patres apostolici, 2.374; Voobus's citation is incorrect): et orare pro inimicis et persequentibus nos ("and to pray for ene mies and those who persecute us'j; also Gregory

Magna moralia 34.19 {PL 76.739A): Pro inimicis vestris orate ("pray for your enemies"); idem, Dial.

4.44 {PL 77.404C); Missale Mozarabicum {PL 85.939): cuius (sc. Leocadiae) fiducialis oratio superbum conterruit inimicum ("whose trusting prayer terrified the haughty foe"). He also offers examples from the Syriac textual tradition (pp. 137-38). I suggest that in the beginning this was not a unique reading but 144-45 n. 2. Cremer, Fastenansage, 77: "primarily the Jews"; 78: "By KarapupePOL, exPpo'i, and biunovreq the Didache means the Jews."

1.3b-4a corresponds to the Synoptics: Matt 5:46-47/Luke

6:32-33 Q.5*

(b) "Those who love the ones who hate them will ulti mately have no enemies!" (v. 3d). The beginning of the logion recalls Matt 5:44/Luke 6:27 Q, but there is noth ing corresponding to kocl ovx e|ere ex^pov in the Synoptic tradition.53 Parallel is Didasc. (Syriac) 1 (CSCO

401.14; 402.12): "And again, He says in the Gospel: 'Love those who hate you, and pray over those who curse you, and you shall have no enemy.' "54 This passage in the Didascalia to Didache and Matt 5:44b against Luke 6:28 (rcpoaevXeoOe); the concluding inrep tup 5lukoptup vpaq is found in Matt 5:44, but Matthew refers this expres sion to Trpooevxeode, and Didache to ppareveTe, which is unique to Didache and absent from the Synoptic tradition. The expression icpooevxeaOe vnep tup cxdpup vpup is not in Matthew or Luke, but apparently in the fragment, already cited, from a say ings collection as found in P. Oxy. 1224, fol. 2r., col. 1

(Grenfell and Hunt, x).58 Verse 3c recalls Matt 5:46-47/Luke 6:32-33, but the introductory rroLa yap is reminiscent only of the latter.

Nevertheless, the Didache version is markedly shorter when measured against Matt 5:46-47/Luke 6:32-33.

Finally, v. 3d recalls the beginning of the Synoptic logion (Matt 5:44/Luke 6:27). The Synoptic logion begins with the commandment of love of enemies, and the logion in the Didache ends with it. The last line, kcu oux c^ere exPpop, is again entirely without a Synoptic parallel. For the Didache we may entertain the hypothesis of a parallel oral tradition (without any literary connection), or (perhaps better) the apostolorum may be dependent on the Didache.55 The exact meaning of the Didache passage is not clear.56 We may probably interpret it as: "Those who persist in lov ing their enemies will ultimately paralyze their aggres sion. Love destroys enmity."57

Excursus: Relationship to the Synoptics Concerning relationships to the Synoptic tradition: v. 3b recalls Luke 6:28; kcu irpooevxeode are common

49 In contrast, the reference to fasting for the sake of charity (suggested as an alternative by Knopf, Lehre,

7; cf. Hermas Sim. 5.3.7-8; Aristides Apol. 15.9

[Goodspeed, 21]; Origen Horn. 10.2 In Lev.

[Lommatzsch, 9.372]; Didasc. [Syriac] 21 [end] [CSCO 407.217-18; 408.201]) is a misreading of this passage.

50 The Didache has to: cJ)vq: Matt 5:46 oi reAupac, 55 See above, Introduction, §2c. Something analogous is true of Liber grad. 13.1. Cf. above, Introduction,

§2b.

56 Audet (Didache, 265) wishes to understand it as future imperative. Worth considering as possible distant parallels are Prov 25:21-22; T. Benj. 5.1: 'Eap expire dyadqp biapoiap, tckpcl, kcu ol iropppol apppuTTOL eipripevovoiP vplp ("If you have a good attitude, children, even wicked men deal peacefully with you"). Also Aristides Apol. 15.5 (Goodspeed,

20): rovq abiKoupraq aijTovq 'napanaXovoL nai irpompLXeZq aijTovq eavroZq ttolovol ("They exhort those who do them wrong and make them friendly to themselves"); Justin 1 Apol. 14.3 (Goodspeed, 34);

Clement of Alexandria Strom. 2.19, 102.4 (GCS3, 2.169): ex$PQ eirLKOvprjTeov, ipa pq peipt] ex- §poq' eiuKOupla yap evpoia pep aupbelrat, Auerac be ex'dpa ("One must take care of an enemy, so that he might not remain an enemy. For good will is linked to concern and enmity is thereby dissolved").

Cf. also 7.12.69.4 (GCS2, 3.50).

57 This saying has nothing to do with the Stoic maxim that the wise person has no enemy.

58 See n. 44 above. v. 47 oi efipLKoL; Luke 6:32-33 ol apapTuXoi. Glover ("Quotations," 14) places high value on rd eppr] in the Didache version and suggests that here the Didache may have retained what is (from a traditionhistorical point of view) the more original word.

51 Xaptq, as in the parallels in Luke 6:32-35; 2 Clem.

13.4; Ignatius Polycarp 2.1; cf. 1 Cor 9:16 K* D* F G; more remotely 1 Pet 2:19.

52 In its form, the Didache version is notably shorter than the one in Matthew and Luke.

53 Hemmer (Doctrine, xc) thought of a quotation from a lost gospel. According to Koester, Synoptische Uberlieferung 221, it is probably "a concluding construc tion added by the redactor to the logia in Did. 1.3."

Layton ("Sources," 355) differs: "eal oux e£ere cxdpop seems to be the work of the compositor"

(cf. already at p. 354).

54 The Latin version: Nam iterum in evangelio dicit:

Diligite odientes vos et orate pro maledicentibus vos et inimicum nullum habebitis (2.28-31; TU 75.4). For the agreements between the Didache and Didascalia apostolorum, see above, Introduction, §2c. In addi tion, cf. also Lib. grad. 13.1 (Kmosko, 307); and above, Introduction, §2b.

example, dyairrjTOt, irapaKaXd . . . airexeadai Tav aapKLKav eindvpidv aiTiveq orpaTevovTai Kara Tijq

\pvxr\q ("Beloved, I urge you ... to abstain from the desires of the flesh that wage war agaiflst the soul," suggestion of dependence on an apocryphal sayings collection.59

At this point there have been a number of interpola tions in the original text of the Didache (v. 4a). The text traditions are:

P: aKOve t'i ae del iroiovvTa aaaai oov to irvevpair[p]aTov irdvrav airoaxov Tav aapKe[i]Kav eircdvpeiav.

1 Pet 2:11); apvqadpevoi tt)v dae/3elav Kai Taq KoapiKaq eindvpLaq ("renounce impiety and worldly passions," Titus 2:12). I think, then, that this motif was introduced by means of a gloss (in some such form as *airexov rdv aapKLKav eindvpidv, "avoid fleshly passions").63

H: airexov Tav aapKLKav Kai oapaTiKav eindvpldv.60

Constitutions: airexov Tav aapKLKav Kai KoapiKav eiridvpidv ,6] (2) The archetype of P (or the scribe of P itself?) inserted a further gloss. For this glossator, the admo nition airexov run aapKLKav eindvpidv (which is rather casual) was not sufficient. He introduced it with a phrase intended to point out the significance of the subsequent clause. The result was: aKOve t'i ae

In my opinion, the history of the interpolation can be reconstructed as follows:62 (1) At an early time a glossator intervened in the traditional text, because he missed the motif of resistance to fleshly desires in the previous lines, and this was one of the fundamen tal motifs of early Christian catechesis. Compare, for del irotovvTa adaai oov to irvevpa■ irpdTOV iravTav airoaxov Tav oapKeiKdv eindvpeidv ("hear what you

59 For Drews, see above. Koester (Synoptische Uberlieferung, 226) distinguishes among borrowings from Matthew and from Luke, "influence of free tradi tion," and (predidachistic) redaction. Layton

("Sources," S52-53) makes v. 3b literarily dependent on Luke 6:28 and Matt 5:44, although the author has made deliberate stylistic alterations in the liter ary source. According to Layton, the literary source for v. 3c-d is Luke 6:32 + Matt 5:46 + Matt 5:47b +

Luke 6:27 + (Matt 5:48). For Layton's conjecture about the conclusion of the logion, see below. For Rordorf's analysis, see above and "Transmission textuelle," 501-2, 509-10.

60 Likewise Georgian. Bryennios (Aidaxv> 6) conjec tures Kai KoapiKav (instead of Kai oapaTiKav).

61 Instructive for the compilers' ways of dealing with their source in this area is Isaac the Syrian, 51.2Iff.

M. Besson, "Un recueil de sentences attribue a Isaac le Syrien," OrChr 1 (1901) 46-60, 288-98: Airexov tuv aapKLKav Kai fiiaTiKdv eindvpiav Kai Tav pdovdv tov Koapov ("Abstain from desires of the flesh and of earthly life and from the pleasures of the world"). The text is derived, mediately, from Const. 7.2.4. See above, Introduction, §5.

62 Cf. Niederwimmer, "Textprobleme," 116-18. For the following explication, cf. also Audet, Didache,

55-56, 265-66. He arrives at a conclusion that is effectively similar. Differing is, e.g., Rordorf,

"Transmission textuelle," 503-4, who sees v. 4a as by no means a subsequent interpolation, but instead counts it among the predidachistic material.

Eirvdvplal aapKLKal ("fleshly passions"), as ein&vplai rfjq oapKoq ("passions arising from the flesh"), is to be understood in light of 1 Pet 2:11—

3:17 or Gal 5:13-26. Did. 1.4a is said to be, in this sense, the introduction to 1.4b-e. This can scarcely be correct. Moreover, Rordorf ("Transmission textuelle," 504, 510) also regards Kai. oapaTiKav as a subsequent clarification. Cf. also Rordorf and Tuilier, Doctrine, 145 n. 6.

63 The singular is an accommodation to the clauses that follow. It is difficult to understand the content of the text from its context: cf. the previous note.

Theissen ("Nonviolence," 129) differs:

"Renunciation of self-defense presupposes control of inward aggressive impulses." Aono (Entwicklung,

167) connects this clause with the expression eat] TeXeioq.

64 The expression eirvdvplai a a pan Kai is striking. In 4 Macc 1:31-32 there is reference to eindvpiai fvxiKai, distinguished from eindvplai oapaTiKai: oaippoovvri dr) to'ivvv eafiv eiriKpareia Tav eiridvpiav, rav de eiridvpiav al pev elaiv fvxrKa'i, al de oapaTiKai, Kai tovtuv aptpoTepav eiriKparelv 6 Xoyiapoq ipaivejai ("Temperance, as I understand it, is control over desires, and of desires some relate to the soul and others to the body, over both of which reason obviously holds sway").

65 EapKiKal and KoapiKai are synonymous.

66 Layton ("Sources," 375-78) makes a much more complicated (but hypothetical) reconstruction of the textual history of the gloss in v. 4a. I am simpli fying here: the oldest text form had no gloss, and the interpolation may at first have been only mar ginal, with its most extensive form *aKOve . . . irpatov . . . airexov Tav aapKLKav Kai KoapiKav eindvpiav. Layton then pictures the further devel opment as follows: the marginal gloss was inserted into the text, with the eoeade . . . TeXeioi postulated by Layton at the end of 1.3 (see below) changed to singular and adapted to the grammatical structure of the gloss. The form airoaxov in P, instead of d7reXov, is a corruption of the text. The absence of Kai

1.3b-4a passions").65 Compare Titus 2:12; 2 Clem. 17.3. This passage is instructive for the question of the relia bility of the text of H.66 For this, see above, Intro duction, §3a. must do to save your spirit: first of all avoid fleshly passions")- This formulation is a distant echo of 1 Cor

5:5.

(3) At the same time, however, the text of the short er gloss was handed on and clarified by additions. We find a form like this in H: onrexov tow oapiaKav ncti awpaTLKWV em&vpluv ("avoid fleshly and bodily passions"),64 or in Constitutions: arrexov tup aapKUiuv k<xl KoopiKUV emi%p.La>v ("avoid fleshly and worldly

KoaptKuv in P is explained by homoioteleuton. In the archetype of H, it became a metathesis, koclkaoptKUV [sic], which explains the H reading kox aupaTLKUv. At the same time, aKOve tl ae . . . -navtoh> was excised and the phrase Kai ear] TeXecoq was transferred to the position now given by H. Cf. the stemma on p. 377.

(2) The Renunciation of Violence (1.4b-5a) 4b-e If anyone strikes you on your right cheek, turn your other one to him too, and you will be per fect. If someone presses you into one mile of service, go along with him two. If someone takes your cloak, give him your tunic as well. If someone takes away from you what is yours, do not demand it back —(since you cannot do so anyway).

5a Give to everyone what he asks of you and do not ask for it back . . . of the fourth (or vice versa). Common to the logia of this pericope is the hyperbolic admonition to unre stricted submission: If someone strikes you68 on the face—namely, on the right cheek69—do not strike back, but instead present the other cheek as well (v. 4b; cf.

Matt 5:39b/Luke 6:29a Q).70 If someone forces you to do a menial task, namely, to accompany him or her for a mile,71 do not refuse; instead, go two miles (v. 4c; cf.

Matt 5:41 [Q?]). If someone takes your cloak, do not resist; give your shirt, too (v. 4d; cf. Matt 5:40/Luke

6:29b Q).72 If someone robs you of what is yours, let it be; do not ask for it back (vv. 4e-5a; cf., for v. 5a, Matt

5:42/Luke 6:30 Q).73 The total abandonment of violence and of one's right is documented in v. 4b by the fact that the person who has been insulted freely offers the aggressor another opportunity to mete out disgraceful treatment; in v. 4c and 4d the one oppressed or robbed gives the aggressor even more than was asked. In v. 4e and 5a, he or she refrains from making any demand for The logion about love of enemies is continued in v. 4b-e and v. 5a with a sequence of four or five logia that also have contiguous parallels in the Synoptic tradition, where they are combined, as here, with the command ment of love of enemies:

(4b) eap Tiq ooi 5q5 pamapa eiq tt]p de^iap6' oiayova, OTpe\J/OP avrqj Kai ttjp aXXpp, koll earj reActoq.

(4c) ear ayyapevarj ae tlq pLXiOP ev, throve pef axnov dvo.

(4d) eap apt] Tiq to IpaTiOP aov, 8oq avToi Kai top xiT^va- (4e) eap Xaj3rj Tiq a-no aov to oop, pr) ccTra'iTei• ov8e yap dvpaoai.

(5a) iraPTi t<5 alTovPTi ae didov Kai pr) bcTra'iTei.

The first four logia belong together, both formally and in content (cf. the fourfold eap . . . , OTpexpOP, viraye,

8oq, pr) arra'iTei). The fifth logion is obviously a doublet 67 According to Dix ("Didache," 242-43) be^cav was lacking in the original Didache text. Cf., to the con trary, Connolly, "Didache and Diatessaron,"/71S 34

(1933) 346-47; Koester, Synoptische Uberlieferung, 227 n. 1, 229-30; Rordorf, "Transmission textuelle,"

504 n. 10.

68 Aibovai pamapa. tlvl as in John 18:22; 19:3.

69 It is the right cheek in Didache and Matthew (against Luke), perhaps as a special challenge: "a blow on the right cheek is given with the back of the hand and is regarded as especially insulting" (Wrege,

Bergpredigt, 75-76).

70 For the third line, Kai ear) reXetoq, see below.

71 For ayyapevetv, see BAGD, s.v.; Gerhard Schneider, "ayyapevw," EDNT 1 (1990) 12.

72 Wrege, Bergpredigt, 76-77, and 77 n. 1: Luke and the Didache think of this as robbery (Matthew differs: designed to collect a pledge).

73 For the peculiar expression ovbe yap bvvaaac in v.

4e, see below. Did Johannes Climacus quote Did.

1.4? For this, see above. Introduction, §2b.

74 This is true even though the key phrase "reign of God" is not found in the logia themselves.

75 The Didachist (as we will see) calls attention in his own way to these connections by means of his own inserted Kai ear) reXetoq. See below.

76 It is true that the same addition is found in Lib. grad. 2.2 (Kmosko, 27) and 22.15 (671)—dependent on the Didache? See above, Introduction, §2b(l).

77 This addition is thus an interpretation. Kai ear) reXeioq in 1.4 recalls reXeioq ear) in 6.2, where the Didachist is also speaking (see below); cf. Rordorf and Tuilier, Doctrine, 32-33; and Rordorf,

"Transmission textuelle," 504-5, 510. It will become clearer in the context of 6.2 what the Didachist's this version contemplates a court proceeding

1.4b-5a restitution. The Didache version of the dominical sayings expresses no fundamental motive different from that in the Q version: in the renunciation of a violent enforce ment of one's own right the law of a higher, although paradoxical, order becomes visible: the law of the reign of God.74 At the same time the attitude of those who are insulted, oppressed, and robbed (which is grotesque according to ordinary and customary standards) reveals the peculiar powerlessness of aggression. Those who live according to these precepts have removed themselves from the ordinary norms and rules of society and set themselves on a road that (as far as this world is con cerned) leads to the discipleship of the cross and (in light of what is to come) to the reign of God.75 attempts to interpret the existing tradition in a partic ular direction.77 Verse 4c is almost identical to Matt

5:41 (for which there is no Lukan parallel); however, the succession of the logia (4c/4d) is the reverse of that in Matthew (5:41/40). Verse 4d is again close to Luke (6:29b); the Matthean KpLdrjvou is lacking and the succession of garments is the reverse of that in Matthew: IpaTLOv—xltuv, which causes us to think of a robbery (differently from Matthew).78 Verses 4e and 5a are doublets,79 and yet v. 5a is more strongly remi niscent of the Synoptic tradition (and particularly

Luke 6:30), while v. 4e only distantly resembles the Synoptic parallels. In v. 4e the expression ov5e yap bvvaaai (which has no parallel in the Synoptic tradi tion) stands out. This little clause has confused inter preters from the beginning.80 The most obvious interpretation is: "Let yourself be robbed, because you cannot really defend yourself, no matter what!"

The clause, thus understood, would make obvious the Essentially (except for the possibly important variant dcbovca pamopa), v. 4b agrees with Matt 5:39b. The last line (Kai eaij reXeioq) falls outside the parallelismus membrorum and is, in my view, a redactional addi tion, namely, a gloss by the Didachist,76 who thus more precise intention in the use of this phrase is.

Koester (Synoptische Uberlieferung 222) finds no motive that could have impelled the compiler to make such an insertion. "But even if the compiler is

/ identical with the redactor, what interest could he 226-38; Layton ("Sources," 355-61) sees Matt 5:39b

+ 4:41 + Luke 6:29 + Matt 5:40 + Matt 5:42 + Luke 6:30 as having been used, from a literary point of view, and makes an effort to explain the differences from the sources as deliberate stylistic alterations.

See Rordorf, "Transmission textuelle," 504-5, 509-10.

79 Koester (Synoptische Uberlieferung, 228) argues that the redactor of the source has "separated [Luke

6:30] into two commandments . . . presumably in order to introduce the admonitions about giving, which follow in Did. 1.5, with a dominical saying."

The clumsy repetition of pr/ arm ret is said (pp.

228-29) to show that the author is not formulating freely (according to Koester, he is here dependent on the Synoptic tradition). There is a protest from Wrege, Bergpredigt, 78 n. 2: "But what Koester con siders literary reworking is really a mnemonic tech nique for simplifying the clauses in community parenesis." The expression in v. 4e is, instead, assim ilated to the preceding logion in v. 4d. In both cases, it is robbery that is contemplated.

80 Cf. Harnack, Lehre, 6, who suspected a corruption of the text but later retracted his emendation

(Kainep bwapevoq) in "Prolegomena," 51 n. 6.

Even the Georgian text sought to correct the text through a paraphrase. For ovde yap bvvaoai, the Georgian reads: "and you cannot do this even for the sake of the faith." have had in introducing such a clause just here, in the middle of Did. 1.4, if not to replace, as quickly as possible, something in the source that had been overlooked, because it may have appeared impor tant to him?'.' (ibid.). Koester thus comes to a com plicated hypothesis: the passage belongs to the source but is now in the wrong place; originally (in the form eoeade ovv vpelq reXeiOL according to Matt 5:48) it would have constituted the end of the logia series in 1.3. The redactor of the logion collec tion (who is not the Didachist) would have omitted it there (replacing it with Kai oi>x eljere exdpou) and now inserted it, in the wrong place and with grammatical alterations. Layton has allowed himself to be persuaded by Koester's inferences to go even farther: cf. "Sources," 348-49, 354-55, 374-75. He emends v. 4e to <3<5e ydp bvvaoai (reAetog eivou), attributes the expression Kai ovx e|ere exdpov in v.

3d to the compositor, and conjectures immediately after that phrase (following Matt 5:48) eoeode ovv vpeiq reXeLOi cog 6 rrarrip t/pcov 6 ovpavioq reAetog eoTLP ("you will be perfect as your heavenly Father is perfect") vet simile. Layton ("Sources," 375-76) associates the origin and placement of the present text, Kai ear/ reAetog, with the textual history of the gloss in v. 4a (see above).

78 Koester's analysis is in Synoptische Uberlieferung,

to suppose that we have before us an oral tradition parallel to that of the Synoptics, or (better) the use of the same apocryphal sayings collection that was already suggested for 1.3b-5a. The Didachist ex panded and commented on his tradition with the clause nai eon] reXeioq. social milieu of those who handed on this text: they are exploited and helpless people who cannot and will not defend themselves.81 Other attempts to understand the text82 remain artificial; it seemed nec essary to suppose a corruption of the text (and this has happened) and to emend it.83 I consider this unnecessary. For the whole pericope it is again easy

81 It is true that Schaff (Church Manual, 165) considers this interpretation trivial.

82 Knopf (Lehre, 9) considers not social pressure but ethical mind-set: "it is impossible for him, in spirit, to take back what is his with counterforce or by legal means." Cf. also Koester (Synoptische Uberlieferung, 229), who regards this clause as part of the redaction of the source; cf. in addition the interpre tation in Rordorf ("Transmission textuelle," 505,

510), who connects this text with 6.2 and attributes it to the didachistic redaction.

83 For Harnack's original suggestions, see n. 80 above.

Layton ("Sources," 348-49) emends to ube yap bvvaaat (reAetog elvai). The phrase reAetog eivai is said to have fallen out because of homoioteleuton

(elvai/bvvaoai), and then dbe was wrongly written as oiibe.

1.5b—2.1

(3) Almsgiving (1.5b—6) andTransition (2.1) 5b-d For the Father wants people to give to everyone from the gifts that have been freely granted to them. Blessed is the person who gives accord ing to the commandment, for he is guiltless.

Woe to the person who takes. If someone takes something because he is in need, he is guiltless, but if he is not in need he shall have to give an account of why he took and for what purpose; if he is imprisoned he shall be interrogated about what he has done, and shall not go free until he has paid back the last penny.

6 But then about this sort of thing it has also been said, "Let your charitable gift sweat into your hands until you know to whom you are giving it."

2.1 The second commandment of the teaching:

(namely, a gloss by the Didachist).87 The saying in v.

5c ends with an eschatological warning. With this (additionally in the Didache, but no longer in Hermas) is associated another saying that is also known from the Synoptic tradition: Verse 5c corresponds to Matt

5:25-26 (cf. Luke 12:58-59), but the contexts of the Q saying (judicial process) and the Didache version

(unjustified acceptance of alms) are different.88 In summary, the aphorism about almsgiving, also found in Hermas, has in its Didache version been framed at beginning and end with reminiscences from the Jesus tradition (and thus Christianized). The form of the aphorism in Hermas may well be the older from the point of view of the history of the tradition.89 It still shows a purely Jewish aspect.

Analysis On the saying navr'i tu ocItovvtl ae 5l5ov (v. 5a "to everyone who asks of you, give") follows now a little reflection on right giving (of alms) in v. 5b-d. The mate rial in v. 5b and 5c is no longer taken from the Jesus tra dition but from ajewish or Jewish-Christian parenesis that appears also in Hermas Man. 2.4-6.84 The transition from one p-adition to the other is not marked in the Didache in any way.

Excursus: Hermas and the Didache Verse 5a is the connecting phrase between the Jesus tradition quoted in what went before and the parenetic section that follows. The tie between v. 5a and 5b is purely associative (Travri tu> olLtovvtl ae bibov—iraoi, .

. . deAec biboadac). Verse 5b agrees essentially with Hermas Man. 2.4c (6 Ttarrip/d debq and xapcaparau/ boipripLixToiu are unimportant variants).85 In v. 5c the agreement with Hermas is looser. What Hermas pre sents as an aphorism appears in the Didache version in the form of a macarism (or woe).86 Hermas Man.

2.6a and Did. I.5a/5c are reversed, while Man. 2.5b is completely missing. The Didache version avoids the doublet in Hermas Man. 2.5a and 5c. In Did. 1.5c the added Kara rrjv evroXpv is certainly redactional

84 Didasc. (Syriac) 17 (CSCO 407.177-79; 408.161-62)/Latin 37.25-27 (TU 75.63) and 38.23-25 (75.64) may come from the Didache. See above, Introduction, §2c. Const. 4.3.1-2 is depen dent on the Didascalia apostolorum.

85 For the parallel in Doctr. 4.8, see below.

86 Layton ("Sources," 363) considers it possible that the author of the interpolation is formally orienting the text to Luke 6:20-26.

87 Layton's reference ("Sources," 360) to Hermas Man.

2.7a is misleading.

88 Rordorf and Tuilier, Doctrine, 147 n. 4.

89 Koester, Synoptische Uberlieferung, 230, 232.

This is a very Jewish idea: God's creative gifts are given for all;96 therefore poor people have a right to alms, and the wealthy are obligated in turn to give them.

Moreover, it should be noted that the gifts given by the rich to the poor are really God's gifts.97 Thus the one who gives alms is only a manager, one who distributes the divine gifts.98 This fundamental idea is then (v. 5c) clarified in the form of a macarism (to be followed by a woe): panapioq 6 <5tdovq" Kara rqv evToXqv adwoq yap eoTLV ("blessed is the person who gives according to the commandment, for he is guiltless").100 The expres sion Kara rrjv evToXqv ("according to the command ment"), typically, has no parallel in Hermas Man. 2.6;101 I consider this phrase an addition by the Didachist (cf.

Did. 13.5, 7). It probably refers to a command of the Literary dependence on Hermas90 is improbable.91

It is more likely that the aphorism about right alms giving attested in both Hermas and the Didache goes back to a common tradition. It is uncertain whether the combination of the Jesus tradition with the Jewish parenetical material was the work of the Didachist or whether it already existed beforehand. In the former case (which I regard as more probable) the Didachist would have proceeded in a similar man ner to that adopted later at 16.1-2, where he also joined Jesus tradition with parenetical material.92

Comment ■ 5b-d The reflection begins with the aphorism Ttaot yap deXeu didootiat 6 irarrip ck tu>v L5'lup x^ptaparcot'

(5b).93 narrjp = God, and the Idea xotpiapara (Hermas Man. 2.4c: 5aipp/xara)94 are the gifts of God's creation.95

90 See above, Introduction, §7a(3). According to Layton ("Sources," 361-72) the Didache is depen dent on Hermas from v. 5a onward, and in v. 5d on Matt 5:25-26.

91 Cf. also Glover, "Quotations," 15-16; Koester, Synoptische Uberlieferung, 230-31; Audet, Didache, the fvxq in Sib. Or. 2.54: ndoa re yap fvxv pepottup lie oil eon yapiapa ("For every soul of mortals is a gracious gift of God"). But Philo Leg. all. 3.78

(to which Audet refers, Didache, 274) can scarcely be a parallel, because there xdpeapa = xdpcq (besides which, xd-pcapa in both places in §78 is text-critically problematic: cf. Ludwig Cohn, "Philo von

Alexandria," NJKA 1 [1898] 539 n. 1). Other nonChristian instances of xdptcipa are late: see BAGD, s.v.; Hans Conzelmann, "xaLpu ktX.," ID NT 9

163-66, 271-75; Kraft, Didache, 140-41; Rordorf, "Transmission textuelle," 507.

92 Is it possible that he also intervened redactionally in the stylization of the given material in v. 5bc

(beyond Kara rqv evToXpv)?

93 Irritatingly, this single sentence from the sectio Chris tiana (1.3b—2.1) has a parallel in Doctrina, but in a very different place—in 4.8: omnibus enim dominus dare vult de donis suis ("The Lord wishes to give to all from his gifts"). One may suppose that this sen tence was also in the Didache's source, at the same place (i.e., after 4.8). The Didachist omitted it there

(at 4.8), understandably enough, in order to avoid having a doublet. Cf. Koester, Synoptische Uberliefe rung 233 n. 2. On the one hand, if this considera tion is correct, it would also be an argument for assigning the sectio to the Didachist. On the other hand, if we interpret the sectio as a postdidachistic interpolation, the omission of the text correspond ing to Doctr. 4.8 (end) remains inexplicable. (It is true that the hypothesis is slightly hampered by the absence of the text in question also in the Canons of the Apostles and the Epitome. Canons completely breaks off its quotation from the tractate at this point, however, and Epitome has other omissions in its version of the text corresponding to Did.

4.3-11.) 94 Hermas Sim. 2.7: bupqpara rov KvpLov.

95 This (unusual) sense of xapiapara is secured by the parallels in Hermas. There is special reference to (1974) 402-3.

96 Cf. Ps.-Phocyl. 29-30 (Sib. Or. 2.89-90^); also T. Zeb.

6.7; 7.2.

97 Deut 15:14b; m. 'Abot 3.7a: Rabbi Eleazer b. Yehuda from Bartota said: "Give to [God] from what is

God's; for you and what are yours belong to God."

(See the translation and commentary by Marti and Beer, 71.) 98 Cf. Lev. Rab. 34.2, on Lev 25:25 (see Str-B 4.540).

99 Acts 20:35 appears to me only a Christian variant on the Jewish aphorism. Const. 4.3.1-2 has com bined the saying about giving (copied from Didascalia apostolorum) with Acts 20:35, and also declares the woe saying to be a word of the Lord.

For the parallel with the woe in the Nicetas catena, cf. Clement of Alexandria GCS2, 3.xxvii-xxviii; and see above, Introduction, §2b.

100 For a possible quotation in Didascalia apostolorum, see above, Introduction, §2c.

101 Moreover, in Hermas the text is not a macarism.

102 Glover ("Quotations," 15-16) thinks of the logion quoted in Acts 20:35. Drews ("Untersuchungen,"

64-65) believes that it means panctpioq 6 bibovq, and that evroAr] is the title of the collection of com mandments being quoted here. 1 Clem. 2.1, rjbt-ov bibovreq rj Xapfiavovreq ("giving more gladly than

1.5b—2.1

Lord, but it is not clear which concrete commandment the Didachist has in mind.102 Is there a reference here to the word of the Lord quoted just before (in v. 5a)?103 Or does it simply mean "in the right way, as the Lord has commanded us," without any thought of a specific say ing? 'Aficpoq ("guiltless") describes the giver as a person righteous and blameless before God.104 As the rich are enjoined to give, however, so the poor are warned not to exploit the obligation of the devout to give alms (cf. again Hermas Man. 2.5): oval t<5 AapftavovTi ("woe to the person who takes")—and this is immediately clari fied: those who are in want105 may, of course, accept alms. But woe to those who are not in need and never theless take alms! They must expect to be called on to explain106 why and for what reason (lvoltl . . . Kai eig tl)107 they have accepted alms.108 The accounting will be unavoidable—and here the text again slips into the tone of the Synoptic tradition (v. 5d):109 the one who accepts alms will be thrown into prison (avvoxv)110 like a thief, and there will be grilled about this behavior.111 Such a one will not be released from prison until the last penny is paid. These last words recall Matt 5:25c-26 (cf. Luke

12:58c-59).112 The underlying tradition in Did. 1.5d can scarcely be understood except as eschatological; owoxv probably refers to an eschatological place of punishment or purification.113

■ 6 At this point, I believe, the Didachist introduces a quotation from Sacred Scripture (which of course meant the "Old" Testament),114 to serve as a warning against too-hasty giving and also as a scriptural foundation for such a warning (v. 6).115 The introduction to the quota tion, ocAXa Kai irepi tovtov 5e116 ctprjrat (cf. the similar introduction to quotations in 9.5 and 16.7), betrays the hand of the Didachist.117 The subsequent quotation is a crux interpretum, because to date it has not been possible receiving"), seems to be an idea related to Acts

20:35, but the saying in 1 Clement is not declared a word of the Lord.

103 This is the interpretation of Knopf, Lehre, 9; and Layton, "Sources," 365-66; Rordorf ("Transmission textuelle," 507) also considers it probable.

104 'Atfaog (dap) does not have the usual meaning of "guiltless" here, but rather means "not blamewor thy." The one in question has acted rightly, as the commandment asks. Hermas Man. 2.6 has the same sense.

105 Xpeiav cxclp, as in Mark 2:25; Acts 2:45; 4:35; Eph 4:28; 1 John 3:17.

106 AlKpp didopae is identical with biupp tlvclp (.Hermas Man. 2.5). According to Layton ("Sources,"

366) the expression in the Didache is compounded from Hermas Man. 2.5a, d'Kobaaovatv Aoyop . . .

Start, and 5c, tloovolp btKpp.

107 IpoctL is made up of tpa tl (sc. yepprca): BDF,

§299(4); BDR §299(3).

108 For a possible quotation in Didascalia apostolorum (including certain contacts with Hermas), see above,

Introduction, §2c.

109 There is no parallel to v. 5d in Hermas.

110 P. Land. II, 354.24. Here csvpoxp does not mean "danger, fear" but "prison," as the parallel in Matt

5:26 shows. See Helmut Koester, ''avpoxp,'' TDNT 7 (1971) 887.

111 T^erdfetf is the technical term for examination under torture. See BAGD, s.v.

112 According to Koester (Synoptische Uberlieferung 237) the author may have used Matt 5:26, but Did. 1.5d appears "no longer to be part of the piece of tradi tion . . . used by the redactor of Did. 1.3ff. at 1.5"

(p. 237). Accordingly, for Koester v. 5b (part of the Jewish tradition) and the subsequent quotation in v.

6 belong to the predidachistic source (p. 238).

According to Layton ("Sources," 366-67) Did. 1.5c is dependent on Matthew.

113 Cf. Rordorf ("Transmission textuelle," 508 and n.

9), pointing out the eschatological interpretation of the corresponding tradition in Tertullian De anima

35.2-3 (CChrSL 2.836-37); 58.8 (869); cf. idem, De oral. 7 (CChrSL 1.261-62); and Sextus Sentences 39

(Chadwick, 16).

114 This is not a word of the Lord, as Drews supposed ("Untersuchungen," 65); cf. also Alfred Resch,

Agrapha (2d ed.; Leipzig: Hinrichs, 1906; repr.

Darmstadt: Wissenschaftliche Buchgesellschaft, 1967) 91-92. The sense of the quotation is intended to be identical with Eph 4:28. "But both the linguis tic relationship with Eph 4:28 and the Synoptic character of the logion are—apart from Idpaodra— undeniable" (p. 92). Glover ("Quotations," 16-17) calls it a Jesus logion.

115 If the Didachist really had a scripturally formulated source before him for vv. 3-4 or for v. 5, in any case v. 6 was not part of it. Drews ("Untersuchungen,"

65) differs, as does Koester (Synoptische Uberlieferung, 238). For the connection between the ideas in 1.4-5 and 1.6, cf. below.

116 Bryennios (Aibaxp, 8) conjectured dp.

117 Thus v. 6 is not a gloss (as Peterson supposed, "Probleme," 147-49).

dicit: Sudet eleemosyna in manu tua, quousque invenias ius tum cui earn tradas. Alius est qui te quaerit, alium tu debes quaerere. Nec eum qui te quaerit relinquas inanem; Omni enim petenti te da: sed alius est quern tu debes quaerere; Sudet eleemosyna in manu tua quousque invenias iustum cui des

("My brethren, it is both said and read, 'To everyone who asks give,' and in another place Scripture says, 'Let alms sweat in your hand until you find a righteous per son to whom to give.' One person begs of you, another you ought to seek out. But do not leave empty the per son who begs of you. 'To everyone who begs of you, give,' but there is another whom you ought to seek out.

'Let alms sweat in your hand until you find the just per son to whom to give'"). Cassiodorus Expositiones in Psalterium 40 {PL 70.295D): Legitur enim: Omni petenti te tribue. Scriptum est etiam: Desudet eleemosyna in manu tua, donee invenias justum, cui earn tradas ("The text is read,

'Provide to everyone who asks of you.' It is written, 'Let alms sweat in your hand until you find a just person to whom to give'"); Expositio in Psalterium 103 {PL

70.733B): Herbam vero servituti hominum, hoc est, ut illi necessaria tribuantur, de quo scriptum est: Desudet eleemosyna in manu tua, donee invenias justum cui earn tradas ("Grass is for the service of people, that is, so that what is neces sary might be provided to him of whom it is written,

'Let alms sweat in your hand, until you find a just person to whom to give them'"). Gregory the Great Regula

Pastoralis 3.20 {PL 77.84CD): Ne sub obtentu largitatis ea quae possident inutiliter spargant, audiant quod scriptum est:

Sudet eleemosyna in manu tua ("Lest they squander use lessly what they possess under the guise of generosity, let them listen to what is written, 'Let alms sweat in your hand'").121 Vita Chrodegangi episcopi Mettensium 11.27

{Monumenta Germaniae historica 10.568, 28ff., ed. Pertz):

Memoratus est scriptum: Sudet elemosina in manu tua, donee to give a secure demonstration of its source: tSpaxrara)118

17 eAeriiJLOOvvri oov eig rag x^P<xg oov, fiexpcg av yvtpg, tlpl 8uig.

Among the texts suggested are Sir 12:1: Eav ev iroirjg, yvcofii. Tivi iroteZg, nod carat x&pig rolg otyocdoZg oov

(Vg: Si benefeceris, scito cui feceris, et erit gratia in bonis tuis multa; "if you do a favor, know the person to whom you are doing it, and there will be favor for your good deeds"). But a derivation from the LXX version as we have it is not really possible. Ps.-Phocyl. 23,7rAr/pcoaei119 oeo xetp\ c'Aeov xPR^omi irapaoxov ("you must fill your hand. Give alms to the needy"), is not a genuine parallel; however, it seems that Sib. Or. 2.79 p (from Ps.Phocylides) is a distant echo of this saying, Ldpwori oraxvon> XPVfovTL -irapaoxov ("with perspiring hand give a portion of corn to one who is in need"). Strikingly enough, Latin writers from Augustine onward know the saying in a form closely related to the Didache version; moreover, they frequently quote it as "Scripture."120

Augustine Enar. in Ps. 102:12 (CCSL 40.1462): De alio enim dictum est: Omni petenti te da; et de alio dictum est:

Desudet. eleemosyna in manu tua, donee invenias iustum, cui earn tradas ("About one it is said, 'Give to everyone who asks of you'; about another it is said, 'Let alms sweat in your hand until you find a righteous person to whom to give'"); Enar. in Ps. 103; Sermo 3.10 (CCSL 40.1509):

Sicut enim de illo qui te quaeret dictum est: Omni petenti te, da, sic de illo quern tu debes quaerere dictum est: Sudet eleemosyna in manu tua, donee invenias iustum cui earn tradas ("For so it is said about him who asks of you, 'To everyone who asks of you, give.' So about the one whom you ought to ask, 'Let alms sweat in your hand until you find a righteous person to whom to give'"); Enar. in Ps.

146:17 (CCSL 40.2135): Utrumque dictum est, fratres mei, et: Omni petenti te da, modo lectum est; et alio loco scriptura

118 H reads Ibpmdm. Bryennios (Atbaxri, 8-9) had already conjectured Lbpuoam.

119 On this difficult irAppoiaei, cf. Nikolaus Walter, "Pseudo-Phokylides," in Ernst Vogt and Nikolaus

Walter, Poetische Schriften (JSHRZ 4.3; Giitersloh:

Mohn, 1983) 199, n. 23a.

120 For the following list of attestations, cf. Ehrhard, Altchristliche Litteratur, 48 n. 5; Resch, Agrapha,

91-92 (history of the findings); Kneller, "Zum 'schwitzenden Almosen,'" 779-80; Cuthbert H.

Turner, "Adversaria patristica," /7'.S' 7 (1906) 593-95; Charles Taylor, "Traces of a Saying of the Didache,"JTS 8 (1907) 115-17 (history of the find ings); Anselm Manser, "Zur Didache 1,6 aus der Vita

Chrodegangi," ThR 8 (1909) 459-60; Altaner, "Doctrina," 340-41; Audet, Didache, 276-80.

121 This passage was copied in Taio Caesaraugustomus Sententiae 3.34 {PL 80.892B).

122 Charles Taylor, "Traces of a Saying of the Didache,"

JTS 8 (1907) 115: the text of Piers Plowman may be corrupt. The correct text might be Sudet elemosina tua in manus tuas, donee scias cui des ("Let your alms sweat in your hands until you know to whom to give it").

The instances cited in Kneller ("Zum 'schwitzen den Almosen,'" 779-80) and elsewhere from a fast-

1.5b—2.1 invenias iustum cui earn tradas ("Recall the Scripture, 'Let alms sweat in your hand until you find a just person to whom to give them'"). Abelard De Eleemosyna Sermo

{Opera, ed. Cousin, 1.552): Sed sudet, sicut scriptum est, eleemosyna in manu tua, donee invenias qui dignus sit ("As is written, 'Let alms sweat in your hand until you find one who is worthy'"). Bernard of Clairvaux Ep. 95 {PL

"Desudet eleemosyna in manu tua, donee invenias justum cui des. " Hinc alibi dicit Scriptura: "si benefeceris, scito cui feceris, et erit gratia in bonis tuis multa" ("It does not say, in a gen eral way, 'who receives a person,' but specifically, 'who receives a just person.' Likewise it is said elsewhere, 'Let alms sweat in your hand until you find a just person to whom to give them.' Elsewhere Scripture says, 'If you give a benefaction, know the one to whom you do so and there will be much grace in your good deeds'").123

182.228C): Ideo ait: Desudet eleemosyna in manu tua, donee invenias {v.l. videas) justum cui des ("Therefore it says, 'Let alms sweat in your hand until you find [u./. see] a just man to whom to give them'"). Peter Comestor Historia

Scholastica; Liber Deuteronomii 5 {PL 198.1251C):

Desudet eleemosyna in manu tua, donee invenias justum, cui Finally, the following example from the "Postille" of Hugh of St. Cher (Venetian ed. [1600], 3.194) is especial ly important: Scito cui feceris, idest antequam des, attende cui des. Unde infra eodem secundum aliam translationem.

Desudet eleemosyna in manu tua, donee invenias cui des, vel cui dare debes ("Know the person to whom you would do something, that is, before you give, pay attention to the one to whom you give. The same point is made accord ing to another translation, 'Let alms sweat in your hand until you find one to whom to give, or one to whom you ought to give'").124 Kneller had already pointed emphati cally to this instance125 because it could provide an indi cation of the supposed scriptural citation: Hugh of St.

Cher quotes the aphorism in connection with his explica tion of Sir 12:1. Audet discussed this question in depth.126 He took up this suggestion and, emphasizing the note that this saying represents an alia translatio, he proposed the hypothesis that the aphorism is the Latin translation (deviating from the ordinary Vg text) of a dif ferent version of the Greek text of Sirach than the one attested in the great uncials.127 "Did. 1.6 uses, with a cer tain freedom, a Greek Ecclesiasticus related to, if not iden tical with, the one that served as the basis for at least one of the ancient Latin versions or revisions of the des, id est ilium cui debes ("Let alms sweat in your hand until you find a just person to whom to give, that is, one to whom you owe something'"). Piers Plowman, B,

Passus 7.75 (in Resch, Agrapha, 92): Sit elemosina tua in manu tua, donee studes cui des ("Let your alms remain in your hand while you ponder to whom to give them'").122

Peter Cantor Verbum abbreviatum 47 {PL 205.150B): Cui etiam des, considerandum, juxta illud poetae: Videto cui des.

Et iterum: Desudet eleemosyna in manu tua etc. ("You must give thought to the person to whom you would give, as the poet says, 'Look to the one to whom you give,' and again, 'Let alms sweat in your hand, etc.'"). Guntherus

Cisterciensis De Oratione, Ieiunio et Eleemosyna 13.1 {PL 212.211BC): Ne dederis indigno, si dignus aeque indigens valeat reperiri, quemadmodum scriptum est: Sudet eleemosyna in manu tua, donee invenias justum cui des ("Do not give to one who is not worthy, if it might be possible to find someone who is both worthy and poor, as is written, 'Let alms sweat in your hand until you find a just person to whom to give'"). Innocent III De Eleemos. 5 {PL

217.756B): Non dicit indiffinite, qui recipit hominem, sed determinate, "qui recipit justum" quemadmodum alibi legitur:

Ecclesiae tit. S. Sabinae Cardinalis primi ordinis praedicatorum (Venice, 1600) vol. 3.

125 Kneller, "Zum 'schwitzenden Almosen,"' 780.

126 Audet, Didache, 276-80.

127 Audet, Didache, 277. The existence of a second Greek translation of the book of Sirach is undisput ed. On this question, cf. simply J. Ziegler, ed.,

Sapientia Jesu Filii Sirach (Septuaginta 12.1; 2d ed.; Gottingen: Vandenhoeck & Ruprecht, 198) 73-74.

According to Ziegler, Sirach was translated into Greek several times (p. 74). For the alternative translations, of special importance are the variant readings of the Vetus Latina. ing poem from Ivrea (p. xi), and from B. Jonas of Orleans, are not quotations but only reminscences of this aphorism.

123 This passage is especially interesting because it com bines our aphorism with the familiar Vg text of Sir

12:1. Note also: "Hinc alibi dicit Scriptura. ..."

124 Audet {Didache, 277) quotes the Venetian ed. 2.194.

For the transmission of the exegetical work of Hugh of St. Cher (S. Caro), cf. Friedrich Stegmuller,

Repertorium Biblicum Medii Aevi (11 vols.; Madrid, 1940-58) 3.114-73. The editio princeps of Venice

(1487) was unfortunately inaccessible to me. I am quoting above from Ugonis de S. Scharo S. Romanae

the unworthy. One receives the impression that the Didachist wanted, by quoting this aphorism, to restrict the appeal to unconditional almsgiving that existed in his tradition. He affirms the tradition. Indeed, blessed is the one who gives! But at the same time he warns against giving too quickly, and therefore challenges the hearer not to forget the scriptural saying he quotes in v. same writing."128 He posits129 that this Greek form of the text must have been known in Syria-Palestine from the first half of the first century; the Greek text is said to go back to a Hebrew text form that had already distanced itself from the original. Patrick W. Skehan130 (taking up

Audet's suggestions) has pointed to the textual form of Sir 12:1 in a manuscript from the Cairo Geniza (arising out of a scribal error in the Hebrew text tradition). This form is said, in turn, to be a precondition for the form of the text proposed by Audet in the Greek tradition.

6.134

■ 2.1 With the quotation in 1.6 the sectio in the narrower sense is complete. There follows, in 2.1, a brief remark that bears the marks of tradition: bevrepa be evroXi] rrjg btbaxVS ("the second commandment of the teach ing"). We recall that the compiler of the Didache inter rupted the text of his source (the "basic document") at Audet's (and Skehan's) suggestions are impressive but not compelling. Altaner131 has asserted with assurance that the Latin authors after Augustine who quote this aphorism "have acquired their knowledge of the Didache quotation directly or indirectly from Augustine."132 Here

Altaner refers to the broad circulation of Augustine's Enarrationes in Psalmos. Indeed, Augustine alone could have been both the direct and indirect source of the Latin quotations. In that case, the following is possible:

With the authority of the African church father, people quoted the saying received from him as Scripture and ultimately sought a source for it in the Bible; Innocent

III is demonstrably the first to have connected the apho rism with Sir 12:1 (although only dubiously), and Hugh of St. Cher's "Postille"133 then fabricated the idea that our aphorism stems from the alia translatio of Sir 12:1.

The result is that the origin of the quotation in Did. 1.6 seems to me uncertain, despite the extensive proofs pre sented by Audet.

1.3a in order to introduce the sectio evangelica, as we call it. Now, having completed the interpolation, he saw him self faced with the task of creating a transition to the source. This is accomplished by 2.1. The Didachist was apparently inspired by the final clause of the source

(toutccp be tuv Aoyuv 17 btbaxv €otlp uimy. 1.3a). In the source this refers to the series of prohibitions that fol lows (2.2-7). It now occurred to the compiler that he could describe the commandments in the sectio and the series of prohibitions in the basic document as two suc cessive concretions of the fundamental commandment

(1.2). Therefore he writes bevrepa be evroXi] rrjg bcbaxvs (2-1) and thus implicitly makes the command ments in the sectio the TCpwvq evToXr), and the following commandments the bevrepa evroXr] (but without any thought of such a thing as the double love command ment).135 One cannot call this arrangement particularly

The aphorism (wherever it comes from) uses a witty image to warn against too hastily squandering alms on 133 Or a tradition to which Hugh of St. Cher refers?

134 In the tradition as given, only the one who unjustly receives was threatened. Now, through the scriptur al saying, the one who gives is also warned. There is a certain tension between 1.5 and 1.6. It is explained by the situation of the Didachist, who wanted to preserve the old tradition but at the same time to protect it from misuse.

135 Correctly Drews, "Apostellehre," 259; Rordorf, "Transmission textuelle," 509. It would be mistaken to interpret (with Harnack, "Prolegomena," 45-47)

1.3-6 as interpretation of the irpaTp ePToXrj (love of God), and 2.2-4.14 as interpretation of the bevrepa evroXr) (love of neighbor). Ulricus Hiintemann divides as follows: 1.3-6 is an exposition of the 128 Audet, Didache, 279-80, with agreement from Layton, "Sources," 368; Rordorf, "Transmission textuelle," 508-9.

129 Audet, Didache, 280.

130 "Didache 1,6 and Sirach 12,1," Bib 44 (1963) 533-36, with agreement from Layton, "Sources," 368 n. 49;

Rordorf, "Transmission textuelle," 508-9.

131 "Doctrina," 342.

132 Augustine himself (according to Altaner, "Doctrina,"

340-41) could scarcely have derived the quotation from a Latin Didache. Rather, the aphorism would have "come to him perhaps as a familiar saying . . . by an avenue unknown to us" (p. 341). Similarly

Adam, "Herkunft," 42-43, who, however (in accord with his interest in combinations), suggests with an "uncertain perhaps" (p. 43) the Abelonii as vehicles of the tradition, and accordingly makes both Doctr.

4.8 and Did. 1.6 dependent on Sir 12:1 (pp. 56-57).

■npwTov dycnvqaeiq, chap. 2 an exposition of the bevrepov ... pi) irotet ("Ad cap. 1 Doctrinae XII apostolorum," Antonianum 6 [1931] 195-96).

1.5b—2.1 skillful. The division thus created remains an emergency construction behind which there is not a deeply reflec tive theological conception but simply a literary principle of order. The construct is not significant except as a reconnection to the quotation of the source. The (emer gency) transition in 2.1 is then followed immediately by the continuation of the text of the source (2.2ff.). bias: (a) He wished to give a stronger Christian character to the existing Two Ways tractate; this he accomplished through quotation of dominical sayings that adjure the Christian to pray for his or her enemies and demand submission to opponents. In this way candidates for bap tism are presented with some of the special characteris tics of Christian life and teaching, (b) With the addition of nai ear] reXetoq, the Didachist gives a special inter pretation to the sayings of the Lord that have been quot ed: fulfilling these commands is a matter of the one who But let us pause for a moment before leaving the sectio, and attempt to review and summarize the results of our observations. (1) The text section 1.3b—2.1 was inserted by the hand of the Didachist. This section is redactional, not a postdidachistic interpolation. (2) The Didachist is not formulating freely here, but is making use of a vari ety of materials whose origins are uncertain. In vv. 3,

4b-5a, and 5d there is reliance on logia from the Jesus tradition, but we cannot point with certainty to the source from which the Didachist derived them; in v. 5b and 5c he makes use of a parenetic aphorism from Jewish or Jewish-Christian tradition that is also to be is reXecoq ("perfect"). In this way the Didachist express es, in his own fashion, the eschatological character of the lex Christi, and raises the instruction of neophytes to the level of eschatological motivation.136 (c) The additional logion about almsgiving recalls a further unique feature of the new community into which the neophyte is enter ing. It was characteristic of early Christians that they accepted the Old Testament and Jewish obligation to care for the poor, and when possible even intensified it.

(In later chapters, we will learn of Christian hospitality and charity.) At the same time, the Didachist is already aware of some bad experiences: the Christian obligation to give support can also be misused and exploited.

Through the addition of v. 6 the Didachist seeks to pro tect the Christian willingness to give alms (which he affirms, and to which the neophytes are invited) against exploitative misuse. (5) Only after the completion of the work was the text in v. 4a interpolated, in a series of stages, by means of a subsequent gloss. found in an earlier form in Hermas Man. 2.4-6. Probably this represents mutual dependence on some other source; it is unlikely that the Didachist is borrowing from Hermas. (3) The Didachist redacted the tradition, how ever; the phrase nal ear] reXetog ("and you will be per fect," v. 4b) is his, as are the references to the evToXr]

(v. 5c) and to the scriptural saying whose origin is unclear (v. 6). If the coupling of traditional logia and parenesis also comes from his hand, one may also sur mise redactional interventions in "v. 5b-c (even though their precise extent cannot be demonstrated). (4) If this analysis is correct we can also determine the Didachist's

136 Vokes (Riddle, 143) wished to see in the emphasis on the TeXeiorriq (here and in 6.2) an indication of the Montanist character of the Didache. But in the Didache it is a question of the problem of fulfilling the "Sermon on the Mount," a problem that existed from the beginning. It is not an issue of heretical perfecti.

c. A List of Prohibitions (2.2-7) 2 You shall not murder. You shall not commit adultery. You shall not corrupt children. You shall not fornicate. You shall not steal. You shall not practice magic. You shall not mix poison. You shall not murder a child, whether by abortion or by killing it once it is born. You shall not covet what belongs to your neigh bor.

3 You shall not swear falsely. You shall not bear false witness. You shall not speak evil of any one. You shall not harbor resentment.

4 You shall not equivocate, either in what you think or in what you say, for equivocation is a mortal snare.

5 Your word shall not be false or empty [but shall be fulfilled by deed].

6 You shall not be given to greed, or robbery, or wickedness, or malice, or pride. You shall not plot evil against your neighbor.

7 You shall not hate anyone. Some people, though, you shall call to task; for others you shall pray. Others you shall love more than yourself. ov pvr)OLKaKr)oeiq, ovk eor\ btyvcopcov ov8e diyXuioooq (irayiq yap davaAnalysis

At 2.2-7 the Didachist returns to quoting the basic docu ment. Originally this section followed immediately after tov rj diyXuoo'ia), ovk curat 6 Xoyoq oov Kevoq, ov \pev5riq. ovk eorj TrXeoveKTpq 1.3a, and thus was the 8i8axv of the fundamental com mandment. Now (after the wooden transitional verse

2.1) it forms the beginning of the devrepa evToXrj. De facto, we are dealing with a piece of parenesis in the form of a catalog, namely, a list of prohibitions arranged as follows: ov (povevoecq, ov poLX^voeuq, ov8e apna^ ovbe vTtoKpiTpq ov8e KaKopdriq ov8e virepri<pavoqov Xpipij j3ovXrjv 7Tovr/pav Kara tov TrXrjoLov oov. ov piopoeiq imvra avdpunov (general conclusion) aAAd ovq pev ... virep ttjv xf/vxw oov.

If necessary, one may further acknowledge the second table of the Decalogue as funishing the basic structure for the whole.1 The author of the tractate used this structure with a great deal of freedom.2 3 As a rule, the prohibitions are expanded by additional prohibitions of related content. Ov piorjoetq ktX. constitutes the sum mary conclusion. The catalog of vices in 5.1-2 is (not in ov Trcabo<pdopr)06L(;, ov iropvevoeiq, ov KXeipecq, ov payevoetq, ov (pappaKevoetq, ov ipovevoeiq reKVov ev <pdopq, ov5e yevvpdev airoKTevelq, ovk eTU$vp.r)oeiq ra tov irApoLov, ovk eTUOpurioeiq, ov \pevbopapTvppoeiq, ov KaKoXoyrioeiq, 1 Cf. Audet, Didache, 283.

2 The independent ov payevoea; and its entire series is striking. Oi)K eimJvprjaeiq ktX. is misplaced. See below.

3 Koester (Synoptische Uberlieferung, 162) judged "that Barnabas abbreviates and rearranges, whereas the Didache has preserved this series in its more original form" (with reference to Windisch, Barnabasbrief

404-5). On this, see above, Introduction §6b. Still, we may count on finding some exceptions. Thus ovk eTTidvprioeiq in Didache may not be in the right place. Audet (Didache, 283) surmised that 5.1 origi nally followed immediately after 2.2-7, and thus regarded 3.1—4.14 as a later addition (within the predidachistic history of the Two Ways tractate).

According to him, the additional growth consisted

2.2-7 form but certainly in content) a doublet of this one. It and 'ncaboip'dopeiv can only refer to unnatural sexual acts of all kinds.8 Ov nXe^eiq ("You shall not steal") again follows the Decalogue (Exod 20:15, MT; Deut

5:19, MT).9

is difficult to form a judgment about the traditionhistorical relationship between Did. 2.2-7 and Barn.

19.4-7. It seems to me, however, that here the Didache generally (although not in every detail) presents a form of the tractate tradition that is closer to the original than that of Barnabas.3

Comment A new series of prohibitions begins with ov ptayevoecq. Forbidding the use of magic is as characteristic of Old Testament Jewish religion10 as is the prohibition of sexual offenses. Both vices, taken together, are especially abhorrent11 when practiced by Gentiles.12 Ov <papp.aKevoetq specifically prohibits compounding poisons.13

■ 2 The sequence begins with a quotation of the fifth and sixth commandments of the Old Testament Deca logue. Ov tpovevoeiq, ov p.oixevoetq ("You shall not mur der . . . commit adultery"):4 cf. Exod 20:13-14 (MT);

Deut 5:17-185 Mark 10:19 par.6 The final prohibition is expanded by oi) 'KOtibotpdoprioeLq ("you shall not corrupt children")7 and ov Tropnevoecq. IIopveveLV, poix^vcw,

Ov (povevoevq renvoi) ev <p#opq14 forbids abortion: "you shall not kill a child by [procuring] miscarriage."15 This passage, derived from the Jewish tractate, offers the old est explicit Christian instance of the prohibition of of two parts: 3.1-6 and 3.7—4.14. Thus Audet arrived at three parts (in addition to the introduc tion) in the Via vitae, namely: "the instruction for the Gentiles" (2.2-7), "the instruction of the wise"

(3.1-6), and "the instruction for the poor" (3.7— 8 4.14). The second and third instruction are said to have been added subsequently (Didache, 308-22, esp. 311). In my opinion, all that is definitely estab lished in this analysis is that 3.1-6 is a later addi9 tion. The other conjectures remain uncertain. Also, a clearer division should be made between 3.7-10 and 4.1-14.

In Doctrina this pair is reversed and a third member is inserted: non falsum testimonium dices.

The rearrangement of the commandments in LXX with respect to the MT (with influence on Luke

Compare also the prohibition on the Philadelphia stele (SIG 3.985,11.27-28, p. 117): pp pdepefiv ppbe rralba pride]itapbevov ("corrupt neither boy nor virgin").

On the whole subject, cf. Ps.-Phocyl. ppre yaponAoTteeLV ppr apoeva Kvirptv bpiveiv ("neither commit adultery nor rouse homosexual passion"); and van der Horst, Sentences, 110-11.

Oi) KXeipet, ov payevoetq, ov pappaKevoecq has no parallel in Barnabas. Oi) KXefeiq has no parallel in Doctrina. Have they been omitted through homoio- / teleuton? (Schlecht, Doctrina, 47).

Cf. Exod 22:17 LXX; Lev 19:26, 31; Deut 18:9-14;

1 Sam 15:23; 2 Kgs 9:22; Sib. Or. 3.224-30.

The association of oi) payevctetq with the preceding prohibitions is to be understood in this way.

That there was also a rich fund of Jewish magical lore is quite a different subject. Cf. van der Horst,

Sentences, 213.

Cf. Exod 22:17 LXX; Deut 18:10 LXX; Wis 12:4; Ps.Phocyl. 149: $dppaKa pp revx^tv, paytKuv fiLfiXav ontexecrdcn ("Make no potions, keep away from magical books"). For Christian polemic against the vice of pappaKeia, see Gal 5:20; Rev 9:21; 18:23;

21:8; 22:15; Hermas Vis. 3.9.7. The warning against payeia and pappaK(e)ia returns once more in the vice list in Did. 5.1; cf. Barn. 20.1. The association of payevetv and pappaKevetv is obvious; cf. the parallels in van der Horst, Sentences, 212—13. In the prohibitions on the Philadelphia stele, the magical material is called <pd.ppa.Kov itovppov, which is made concrete in what follows: cirostbai novppaL and piXrpov (SIG 3.985,11. 18-20, p. 117). Cf. Weinreich,

Privatheiligtums in Philadelphia, 55.

Cf. Barn. 19.5.

$i)opd (or anopdopd) = abortus: SIG 3.1042,1. 7, p. 196; Clement of Alexandria Paed.2.10.96.1 (GCSS

18:20 and some ms. variants of Mark 10:19) cannot be discussed here. Nevertheless, cf. Doctrina: Non mechaberis (moechaberis g), non homicidium facies

("You shall not commit adultery, you shall not com mit murder").

Cf. the prohibitions for participants in the private cult in Philadelphia: prj povov, "not to [commit] murder" (SIG 3.985,1.21, p. 117) and to avoid sexu al sins (11. 25-46, pp. 117-18); also the commentary by Weinreich, Privatheiligtums in Philadelphia, 60.

Ilaibopdopeiv appears to be a constructed word that first appears in the Jewish-Christian tradition, replacing TracbepaoTeZv; so Audet, Didache, 286; and Rordorf and Tuilier, Doctrine, 149 n. 4. In Justin

Dial. 95.1 (Goodspeed, 210) it appears with eibuXoXarpelv; cf. also Tatian Apology 8.1 (Goodspeed,

274); Clement of Alexandria Protr. 10.108.5 (GCS3 1.77); Paed. 3.12.89.1 (GCS3 1.285). The verb is probably derived from ttaibopdopoq, T.Lev. 17.11

(in a vice list); Barn. 10.6. On this subject, cf. also Lev 18:22; 20:13; Rom 1:27; 1 Cor 6:9; 1 Tim 1:10.

neighbor"),21 following Exod 20:17; Deut 5:21 (MT), really belongs with the series ovk eor] 'nXeoveKTrjq kt\„ and has been wrongly placed here.22 In contrast, Barn. abortion.16 But it is certain that from the beginning,

Christians, following Old Testament and Jewish custom, rejected abortion. That there is no specific prohibition of it in the New Testament is accidental. As in Did. 5.2

(and as we find frequently in Jewish and Christian texts)17 the prohibitions on abortion and exposing of infants are connected. Thus the next command is ov8e yevvri&ev18 dnroKTevelq,19 prohibiting the killing of new born children and thus implicitly forbidding that they be exposed.20 The following command, ovk e'ni'dvp.rjoeiq ra tou TtXrjOiOV ("You shall not covet what belongs to your

19.6 preserves the correct order, although only in the case of the two phrases ov yu.77 yevr} erutivnaiv and ov ptr) yevyj ttXeoveKTqq.

■ 3-5 Now follows a series of prohibitions23 related to "sins of the tongue."24 Ovk 6TU0pKr)O6Lc; ("You shall not swear falsely") forbids either false oaths or the breaking of an oath or vow.25 One should note that here (differ ently from Matt 5:33-37; Jas 5:12) it is peijury or the 1.215); cf. Plutarch De tuenda sanitate praecepta

(.Moralia 134F; Babbitt, 278), where it is said of the aKoXaaraL yvvalKeq eKfioXlou; xp^peval mi

(pdopioLq ("licentious women who employ drugs and instruments to produce abortion"). A prohibi tion for fellow members of a pagan cult to make use of ipdopeiov a.TOK6lov is found in the Philadelphia stele (SIG 3.985, 1. 20, p. 117), although here this may be, as Weinreich (Privatheiligtums in Philadelphia, 5) surmises, a reference to magical practices. The stele does not refer to the ritual impurity caused by abortion but "places all the emphasis on the ethical question as such" (ibid.).

For abortion in antiquity generally, see Jan Hendrik Waszink, "Abtreibung," RAC 1 (1950) 55-59. The broad use of abortion in the imperial period is well known (ibid., 57). It was especially common among the more well-to-do classes in that society (p. 59).

"Medicine was strictly opposed to abortion, from the hippocratic oath . . . until the end of antiquity"

(p. 57). It is equally well known that Jewish piety rejected abortion: cf., e.g., Philo Spec. leg. 3.108-9, between Barnabas and the Didache.

23 There is no parallel to the first three members in Barnabas, and Doctrina lacks the parallel to ov pevdopaprvprioeLg.

24 Strictly speaking, ov pv-qaLKaKrioecq (v. 3) is an exception.

25 Cf. BAGD, s.v.; Lev 19:12; Zech 8:17 {Barn. 2.8); Sir 23:11; Wis 14:28-29; 3 Bar. 4.17; 13.4; Ps.-Phocyl.

16-17: pr) 8' emopKrjoriLq p-qf ayvcoq pqre eKOvripevdopKov orvyeei dedc dpflpoToc ootlq 6 poo apt

("And do not commit peijury, neither ignorantly nor willingly. The immortal God hates a perjurer, whosoever it is who has sworn"; cf. Sib. Or. 2.68-69

T); 1 Tim 1:10; and further examples in van der Horst, Sentences, 123-24.

26 Knopf, Lehre, 11. The material in the prohibition series (stemming from the tractate) is pre-Christian

Jewish. It lacks Jesus' eschatological motivation. Cf., in contrast, 1.3b—2.1.

27 Mark 10:19 par.; Rom 13:9 v.l. Further: Lev 19:16; Deut 19:16-19; Prov 21:28; 25:18; Ps.-Phocyl. 12: paprvpLqv pevdrj tpevyetv ("Flee false witness," cf.

Sib. Or. 2.64 T); Apoc. Pet. 29 (Akhmim) (NTApoc, 2.631). The parallel in the vice list of the way of death has pevdopaprvpLcu {Did. 5.1; cf. Matt 15:19;

26:59; Polycarp Phil. 2.2; 4.3; Hernias Man. 8.5).

28 Contrast KaKoXoyeZv nva: Exod 21:16 (cf. Prov 20:9a; Ezek 22:7; Mark 7:10/Matt 15:4); Mark 9:39;

Acts 19:9.

29 As in Hernias Man. 9.3; Diogn. 9.2.

30 Barn. 19.4: ov pvqaiKaKqaeiq m adeXipa oov ("Do not bear malice against your brother").

31 Zech 7:10: mi KaKtav eKaoroq rov adeXipov avrov pr) pvrioiKaKeLToi ev ralq Kapdicaq vpuv ("and do

117; Josephus Ap. 2.202; Ps.-Phocyl. 184: pride yvvr) (pdeipriL /3pe<poq epflpvov evbofh yaorpoq ("Do not let a woman destroy the unborn babe in her belly").

See Sib. Or. 2.281; Waszink, "Abtreibung," 58; van der Horst, Sentences, 233-34.

16 This prohibition, and the entire context, derives from the Jewish Two Ways tractate.

17 Van der Horst, Sentences, 233-34.

18 Read yevvr]dev with Canons, Constitutions, Barnabas, against yewq-devra in H.

19 Cf. Barn. 19.5.

20 See the Jewish and Christian examples in van der Horst, Sentences, 233-34.

21 Emih/pelv with accusative rather than the classical usage with the genitive: BDF, §171(1); BDR,

§171(2).

22 We find the same wrong order in the version in Doctrina, Canons, and Epitome. In v. 6, however,

Doctrina has non eris cupidus ("you shall not covet"), and thus offers a version that stands genetically not devise evil in your hearts against one another"; cf. Barn. 2.8); 1 Clem. 2.5: eiXtKpLveZq mi. (XKepouoL rjre mi dpvqoLnaKoi eiq aXXriXovq ("you were sin cere and innocent and bore no malice to one anoth er"); 1 Clem. 62.2: opovoovvroiq dpvrpnKaKW!; ev aybnnj mi elpqvq ("to live in concord, bearing no malice, in love and peace").

32 Cf. Barn. 19.7.

2.2-7 breaking of oaths that is forbidden, not swearing itself.26 deceptive chatter:39 ovk eorat 6 Xoyoq oov Kevoq, ov

\pev5riq.40

'J'evbopapTvpeiv means to make a false statement (before a judge), to tell a lie when giving testimony (cf.

Exod 20:16; Dent 5:20, MT).27 KanoXoyelv ("to speak evil") is used here absolutely.28 To pppoiKOtKelv ("to har bor resentment" in the absolute)29 should be added tlpl tl kcxkovX0 to be unable to forget the evil one has under gone; to be irreconcilable.31

Excursus: TheText of 2.5

The text is uncertain at this point (and has been expanded by a later gloss).41 The witnesses are:

H: pevbpq, ov Kevog, dAAd pepeorupepoq 7rpd£ei.

Canons: aevoq ovbe pevbriq.

Doctrina: vacuum, nec mendax Epitome: nevoq, Constitutions: Kevoq ... ov \pevoi).

The result is that (1) the clause dAAd pepeoroipepog itpa^ei ("but fulfilled by deed")42 represents a later gloss,43 which does not speak for the quality of the H text, and suggests the possibility of glosses in other

■ 4 This verse continues the prohibitions against sins of the tongue:32 ovk ear) diyvupwv™ ov5e 8LyXtooaoq.

Atypwptpp (or diypwp,oq)?'4 is the false person who says one thing and thinks another. In fact, diyXwoooq has the same meaning here: "double-tongued."35 The warning against false, double-tongued speech is supported by an aphorism (breaking the schema of the context): 7ra7tg yocp 'da.vaTov rj dLyXcoooia ("for equivocation is a mor tal snare").36 The style37 and content38 resemble the Old

Testament.

■ 5 Finally, there comes a warning against empty and 33 Doctrina interprets with duplex in consilium dandum.

34 Text uncertain. Wengst decides for blympoq. H and Barnabas read biyvupuv. In the parallel passages,

Contitutions, Canons, and Epitome read biyuapoq (and the word is also in Const. 2.6.1). Synt. doctr. 1.9: p.17 elvac biXoyov, pr) bivapov. Fides patr. 4 (cf. PC

28.1639D): pi) elvat bLyXaooop prj (PC has the vari ant reading 17) biypupou. Atyvapuiv is found in a Greek scholion on Euripides 633 (Dindorf, 1.172); bLyvwpoq in-Simplicius on Epictetus (Diibner,

28:13-26; cf. Jas 3:1-12. Closer to the OT tradition is the version in Barn. 19.8:^0:71.g yap to ctropa

■davctTov ("the mouth is a snare of death"). That is probably more original than the version in the Didache. Cf. also Doctrina: tendiculum enim mortis est lingua.

39 This is lacking in Barnabas.

40 Cf. T.Naph. 3.1.

41 For what follows, see Niederwimmer, "Textprobleme," 118-19.

42 Cf. Jas 1:22; 1 John 3:18; 7 Clem. 38.2; 2 Clem. 4.1-5; Justin 1 Apol. 14.5 (Goodspeed, 35); 16.8 (37).

/ 134.53); Hippolytus Ref. 5.26.1 (GCS 3.127); 10.15.2 (276); Maximus Confessor Opuscula (PC

91.56B); Etymol. magn. (ed. Gaisford) on dAAoitpooaXXoq.

35 BAGD, s.v. AiyXaoooq does not mean "bilingual" here but instead "false, dishonest." Cf. esp. Prov 11:13;

Sir 5:9, 14-15; 28:13; Philo Sacr. 32 (vice list); Sib. Or.

3.37 (vice list); Constitutions also, at 2.6.1 (vice list).

36 The word biyXaooia occurs only here and in the parallels in Barn. 19.7 and Canons. In Constitutions we read Ttaylq yap ioxvpa aubpl rd ibca xetAr? ("a man's own lips are a strong snare" = Prov 6:2).

AiyXwoo'ia was probably constructed from biyXaoooq and represents a characteristic expression of the Two Ways tractate, perhaps an ad hoc construc tion by the author.

37 Ilayiq davctrow. Prov (13:14); 14:27; 21:6; Tob 14:10; Ps 17:6. Cf. the similar expression in 1QH 2.21; CD

Knopf (Lehre, 12), however, also points to Corp.

Herm. 16.2 (Nock and Festugiere, 2.232): 'EXXrjpeq yap ... Xoyovg exovoc Kevovq aTrobeL^euv euepyqTLKOvq, xai avrr] eoTtv EXXpvwv <pi,Xo<JO<pLa

Xoyav xj/otpoq rjpelg be 01) XoyoLg dAAd tpiovaig peoTalq ruu epyav ("For the Greeks have empty speeches, O king, that are energetic only in what they demonstrate, and this is the philosophy of the Greeks, an inane foolosophy of speeches. We, by contrast, use not speeches but sounds that are full of action"; trans. Brian Copenhaver, Hermetica

[Cambridge: Cambridge University Press, 1992] 58); and see Theognis 979 (Young, 60): Mr) pot avpp elr] yXuoorp iplXoq, dAAd Kai epyui ("I would have no man my friend with lips only, but also in deed").

43 This was already suggested by Schlecht, Doctrina, 48; Knopf, Lehre, 12: "probably to be regarded as an addi tion." Drews ("Apostellehre," 260) considers it prob able that this is "an addition by our redactor," as does

Giet, LEnigme, 94. This can scarcely be the case.

14.2.

38 The fool is caught by his or her own speech: Prov 6:2; the mouth or tongue as TTayiq: Prov (12:13);

18:7; Sir 51:2. The danger of the tongue: Sir

passages as well. (2) Following the indirect witnesses, I conjecture the original text as nevoq, ov \pevbr)q.u

("robber").47 The sequence of the following parts is sug gested merely by association: Ovde viroKpiT-qq (here not "hypocrite," but "wicked, far from God godless"),48 ovde

KCtKoridriq,49 ovde vireprppavoq,59 ov Xpif/ri fiovXrjv irovppav Kara tov TrXpoLov oov51 forbid evil and hostile intentions toward one's neighbor.52

■ 6-7 The next verses begin with prohibitions recalling the ninth and tenth commandments of the Decalogue, but then move to phraseology opposing offenses against love of neighbor. At the beginning is the TrXeoveurpq

("one given to greed").45 The content probably reflects Exod 20:17; Deut 5:21 (MT), but this is no longer clear because of the rearrangements that have occurred in the text.46 The TrXeoueKTrjq is then followed by the ap7ra|

■ 7 The series concludes with a general warning: ov pLopoeiq iravra53 avdpunrov, ctAAd oDg pev eXey^eiq, irepl wv 6c54 irpooev^r), ovq de ayairrioeLq virep rpv ipvxw oov.55 The meaning of this overloaded expression

(alluding to Lev 19:17?),56 which also gives an impres50 In vice lists: Rom 1:30; 2 Tim 3:2; cf. also the quota tion of Prov 3:34 in Jas 4:6; 1 Pet 5:5; 1 Clem. 30.2; also Luke 1:51; 1 Clem. 49.5; 57.2; 59.3 (= Isa 13:11);

Ignatius Eph. 5.3. 'Tireprppuveiv: Ignatius Eph. 5.3; Srnyrn. 10.2; Pol. 4.3; vireprppotv'ia in the list in Mark

7:22 and Hermas Man. 8.3; cf. also Man. 6.2.5; Sim.

8.9.1; 1 Clem. 16.2; 30.1; 35.5. The word family plays a special role in 1 Clement. For the sequence, cf. esp.

44 Wengst, Didache, 68: xevdq ovbe \pevbqq.

45 This was originally introduced by a prohibition against coveting the neighbor's goods, as still evi dent in Barn. 19.6 and Doctrina (non eris cupidus).

Barnabas lacks the parts of the sentence beginning with irXeoveKTqq.

46 The warning against greed recurs in the vice list at 5.1 (= Barn. 20.1). For 7rAeove^La and 7tAeoveKrqq in lists, cf. Mark 7:22; Rom 1:29; 1 Cor 5:10-11;

6:10; Eph 5:3; Col 3:5; 1 Clem. 35.5; Polycarp Phil 2.2; Hermas Man. 8.5; Sim. 6.5.5. See also Luke

12:15; 2 Cor 7:2; 9:5 (12:17-18); Eph 4:19; 5:5; 1 Clem. 35.5: irXeove^iav... KaKopPeLaq .. . virepr]ipavLav. The vice list of the way of death in Did. 5.1

(= Barn. 20.1) has VTreprppavLa.

51 d. Barn. 19.3.

52 Cf. BAGD,s.u./3ouArj.

53 Ov—irdvra is a Hebraism (cf. Schaff, Church Manual, 170).

54 lie p i &v Oe in P, Canons, Epitome; Wengst, Didache, 68; against trepi be u>v in H. For the text of Doctrina and Constitutions, see below.

55 Only the last clause has a parallel in Barnabas (19.5): dyairpoecq tov irXqoLov oov virep Tr\v pvxqv oov ("You shall love your neighbor beyond your self"; cf. ayairdv vpdc virep rrjv pvxqv pov, 1.4; iravTaq ayairav virep ti)v pvxqv pov, 4.6). Barn.

1 Thess 2:5 (4:6); 2 Pet 2:3, 14; Hermas Man. 6.2.5; Barn. 10.4.

47 For dp7ra£, cf. also 1 Cor 5:10-11; 6:10; Luke 18:11

(in vice lists); also on this theme: Matt 12:29; 23:35 par.; Barn. 10.4. The sin of theft appears again in the vice list of the way of death: Did. 5.1 (= Barn. 20.1).

48 Behind the meaning "wicked" for viroKpirriq is qjn, which Audet (Didache, 293-94) has pointed out, cor rectly in my opinion. This meaning (remarkable in relation to classical and other Hellenistic usage) begins with the translation of qjn as viroKpcrriq

("godless") in LXX (Job 34:30; 36:13), and was con tinued in later translations (Audet, Didache, 293;

Ulrich Wilckens "viroKpivopcn ktX.," TDNT 8 [1972] 564). For general information on this linguis tic usage by diaspora Judaism, see Wilckens,

"viroKpcvopcu," 563-66, with reflections on how the usage may have arisen. Of course, the equation of viroKpLTqq with rpn is still not a sure indication of the origins of the tractate in diaspora Judaism.

Nota bene: NT usage of inroKpiTrtq also largely derives from that of diaspora Judaism: cf. Wilckens,

"viroKpivopou," 566-69:'TiroKpLou; in Did. 4.12 should probably be interpreted similarly; while the viroKpLoetq in 5.1 are probably "hypocrisies." That Wilckens himself (p. 570) interprets the word differ ently in Did. 2.6 is a separate issue.

49 Sib. Or. 3.37: fevbav btyXuoouiv avdpdirav Kai 19.5 has certainly preserved the older version of the tractate. The version we now read in the Didache is secondary from a tradition-historical standpoint.

Doctrina has a shorter text (without a climax): neminem hominum oderis, quosdam amabis super animam tuam ("you shall hate no human beings, some you will love more than your own soul"). This leads to the suspicion that the source of the Didache may have been similar (i.e., without a climax), so that the climax found here (clumsy as it is) may be a clarifi cation by the Didachist. But the climax is also found in the Canons and Epitome. It must therefore have been in the same source, and there is simply an omission in Doctrina. "Quosdam indicates that L is not original" (Drews, "Apostellehre," 260). The for mally related climax in Jude 22-23 is striking, see

Sakae Kubo, "Jude 22-23," in Eldon J. Epp and Gordon D. Fee, eds., New Testament Textual Criti cism: Its Significance for Exegesis: Essays in Honour of KCLKopdihv ("false, double-tongued men and immoral").

2.2-7 sion of being secondary,57 is uncertain. There appear to be two or three58 classes of people:59 a first class who have committed faults-they are to be reproved and set straight (ovg p'ev eXey^ecg);60 a second class, perhaps those who show themselves resistant to correction—one should pray for them (irepi av 5e TTpoaev^rj)-, finally, a third class: we should probably assume that these are members of one's own group. For them the command ment is: dyairr/oeig virep rqv ipvxw oov ("You shall love more than yourself").61 'i'vxp here probably means one's own "life," or one's own "person": one is one's own

\f/uxV-G2 To love someone more than one's own life means to love that one more than oneself.63 Thus what is probably being said is that the demand for love, the "love commandment," applies to everyone, but the con crete realization of the love commandment is different as regards the different groups of human beings. The commandment of love is most stringent toward the members of one's own group. Here it is not merely "love your neighbor as yourself," but even "love your neighbor more than yourself."

Bruce M. Metzger (New York: Oxford University Press, 1981) 239-53; Knopf, Lehre, 13: "There appears to be some kind of relationship between [Did. 3],7 and Jude 22-23." True, but what kind? In addition, our Didache passage may be quoted in Lib. grad. 16.4 (Kmosko, 395): see above Introduction, ed as probable by Drews, "Apostellehre," 260).

Knopf {Lehre, 13) explained the addition in light of Jude 23. Epitome, however, lacks ovq be eAepoeiq.

58 Harnack, Lehre, 10; Knopf, Lehre, 13; Kleist, Didache, 155; Wengst, Didache, 93.

59 The following interpretation follows Harnack. Is it correct? Knopf {Lehre, 13) differs very little: "The first [group] contains those people who can still be saved, even though they have been affected by sin; they can be found inside or outside the commu nity. . . . The second type of people encompasses those who, according to human opinion, are lost, the unapproachable. . . . Finally, the third type are the full-fledged members of the community." Cf. also Wengst, Didache, 93.

60 Did. 4.3; 15.3: eAeyyeiv.

61 Gesta ap. Zenophilum (CSEL 26.192) quotes this pas sage, either from the Didache or from the tractate: see above, Introduction, §2b.

62 Audet, Didache, 294: "more than one's self." Wengst, Didache, 94 n. 13 (with reference to BAGD, s.v.

§2b.

56 Lev 19:17: ov piopoeu~ top abeAtpov oov rrj biavoia oov, eAeypu eAeyl-eiq rov -KArjoLou oov mi ov

Aripp-q biavrov apapriav ("You shall not hate in your heart any one of your kin; you shall reprove your neighbor, or you will incur guilt yourself"). Cf.

A. de Vogue, "'Ne hair personne,'" RThAM 44 (1968) 3-9. The biblical rov abeAtpov oov (or rov it Apatov oov) has been replaced by the universal irdvra duPparrov (de Vogiie, 4). Apostolic

Constitutions has the maxim ov piopoeiq irdvra dvdpuncov three times (in 7.5.3 from Didache, with the Constitutor having expanded the text by using

Lev 19:17b; then in 7.2.3 in the middle of a review of the sectio, and finally in 2.53.6 as a gloss in a review of the Didascalia, where the Constitutor this time includes the whole of Lev 19:17). Later instances can be found in de Vogue, 7-9. The Nullum odire in Regula Magistri 3.71 (SC 105.370) and Rul. Ben. 4.65 (CSEL2 75.36) is important.

There are possible distant relationships to some ver sion of the Two Ways tractate. See de Vogiie, 8. See above, Introduction, §2c.

57 Canons apparently read the version offered by the Didache. The compiler of the Canons inserted, after eAey^eiq: ovq be eAepoeiq (Funk, Patres apostolici,

/ 'I'VXV)- 63 For this whole subject, cf. also Rul. Ben. 2.25 (CSEL2 75.25), where, of course, the reference is to mem bers of one's own group, the monastic community.

1.8 considered this version original; it is also regard

d. The "Teknon" Sayings (3.1-6) 1 My child, flee from all evil and from everything like it.

2 Do not be an angry person, for anger leads to murder; nor should you be a zealot or a quar relsome or hot-tempered person, for from all of these [traits of character] flow murderous acts.

3 My child, do not be a person given to passion, because passion leads to fornication; nor should you be given to obscene speech or to bold gazes, for from all of these [actions] flow acts of adultery.

4 My child, do not practice augury, because this leads to idolatry; nor should you be an enchanter, or an astrologer, or a person who performs purificatory rituals; you should not even want to see cor hear> such things, for from all of these [activities] idolatry is spawned.

5 My child, do not be a liar, because lying leads to theft; nor should you be given to avarice or to vainglory, for from all of these [traits of character] theft is spawned.

6 My child, do not be a grumbler (against God), because this leads to blasphemy; nor should you be presumptuous or disposed to invent evil, for from all these [attitudes] blasphemies are spawned. offers a parallel text to 3.1-6 (and for 3.7-10), but 3.1-6

Analysis The next group of sayings is characterized by a constant address from the fatherly teacher to his student, who is called T6KV0V ("child") (3.1, 3, 4, 5, 6). We can call this group the "teknon sayings." They also come from the source, the tractate. Noteworthy is that there are no par allels to the teknon sayings as such (3.1-6) in Barnabas.

They therefore must not have been present, as yet, in that (probably older) recension of the Two Ways tractate that served as a source for Barnabas.1 Parallels in Barnabas then reappear at (3.7 and) 3.8c-10.2 Doctrina is not characterized by the constant address "my child," which instead is found only at the beginning, in 3.1:

"Fili, fuge. . . ." In addition, there is no parallel to 3.3 and 4a. (This can be interpreted either as an omission,3 or— though this is less likely—3.3, 4a were missing from Doctrina from the beginning.) The teknon sayings also return in Canons, including the material corresponding to Did. 3.3, 4a (Canons also has a parallel to Did. 3.7-10); however, the apostrophe in the second saying is doubled

(something that is certainly secondary and explained by 1 Muilenburg (Literary Relations, 34, 149) has expressed the opinion that Did. 3.1-6 was derived from a Jewish apocryphon. This notion had been expressed earlier by Robinson, Didache, 62; idem,

"Barnabas and the Didache," 242 (a Jewish or early Christian apocryphon as the source); Vokes, Riddle,

76: "a fragment of an earlier writing incorporated bodily into the Didache"; Connolly, "Relation," 242:

"To my mind there can be no doubt in the world that this piece was found by the Didachist in some earlier writing—possibly some 'Testament,' or part ing admonitions of a father to his son—and inserted by him in his version of the Two Ways." The correct version is in Audet, Didache, 311, who recognizes

3.1-6 as a later addition to the original Two Ways tractate. The other hypotheses associated with this can be set aside here.

2 The parallels to 3.7 are only rudimentary.

3 Drews ("Apostellehre," 261) explained the absence of vv. 3 and 4a in Doctrina as "probably only the result of a mistake."

4 The respective sources of the Life of Shenoute and of Syntagma doctrinae and Rides patrum also contained the teknon sayings. In the Life, the whole text of Did.

2.6—4.8 has been revised into teknon sayings. (This is certainly secondary.) By contrast, in Syntagma doctri-

3.1-6 the rather long insertion found in Canons at this point).

Finally, Epitome also contains the material of the teknon sayings, although it is not stylized in the same way (the apostrophe is lacking in Epitome).4 These observations can be explained if we adopt the following model: On the one hand, Barnabas rests on a recension of the Two

Ways tractate in which the teknon sayings were yet absent (B recension of the tractate). On the other hand,

Doctrina, Didache, Canons, and Epitome go back to a (later) recension that already contained the teknon say ings (C recension).5 More precisely: Didache and Doctrina rest on sub-archetype C1, in which the teknon sayings were retained in the form found in C. (Doctrina reveals some omissions that are secondary and may have arisen only in the Latin textual tradition.) Canons and Epitome rest on sub-archetype C2, which is characterized by a styl ization of the entirety of the material, and thus also of the teknon sayings, into apostolic sayings (the omission of the apostrophe in the material of the teknon sayings in Epitome is secondary). For the whole, see above, pp.

35-41, where the corresponding hypotheses are dis cussed.

Third saying: reading omens, magic, and the like lead to idolatry (3.4).

Fourth saying: lying leads to theft (3.5).

Fifth saying: grumbling leads to blasphemy (3.6).

This is sapiential material6 in a psychologized form.

The primary intention of the series is to reveal apparent ly minor vices as, in fact, dangerous temptations, which (if one surrenders to them) lead to serious sins. The warning appeals—in the style of hokmah (divine

Wisdom)—to the understanding and assent of reason.7

The student (the T6KP0P instructed by the experienced teacher) is to recognize and learn the danger in time: principiis obstal In the nature of things, this section belongs within the context of early Jewish rigoristic morality. The same motif of sharpening the moral sense through understanding is inherent in the psychological contexts that we recognize in other contemporary Jewish sources, and that (although with eschatological intensifi cation) also shape some elements of the Jesus tradition

(Matt 5:21-26). Still, the tractate lacks the eschatological inexorability that characterizes the Jesus tradition.8

Closest to this mentality are certain passages in the Testaments of the Twelve Patriarchs.9 Formally the teknon sayings are relatively well organized. We find five say ings, each of them (except for the first) introduced by an address, followed by a first warning, a first reason, a sec ond warning, and a second reason. The third saying is

The disposition in the Didache is now as follows:

First spying: general introduction; wrath leads to mur der (3.1-2).

Second saying: lust leads to fornication and adultery (3.3). nae and hides patrum (where only rudiments of the material of the teknon sayings are present) the apos trophe has disappeared completely. In addition, the apostrophe is lacking (naturally enough) in the Apostolic Constitutions, which derived this section directly from the Didache.

5 The sources of the Life of Shenoute and of Syntagma doctrinae and Fides patrum must also, ultimately, go back in some fashion to this recension (C). Because they are unacquainted with the stylization of the material into apostolic sayings, we can think more precisely of a C1, but nothing more certain can be said about it.

6 Audet, Didache, 301: "a sapiential adaptation of the Decalogue," may go too far.

7 Audet speaks [Didache, 305) very accurately of a "didactic imperative, rather than juridical or legal."

8 For the motif of radicalizing in general, see Rordorf, "Gebrauch des Dekalogs," 437-38. The Jesus tradition is more radical than the Didache: in text within which Jesus' interpretation of the law stands.

Cf., e.g., T. Jud. 14.1: Kal vvv, reeva pov, /it) pedvoKeode olvu> - on 6 oluog dtaoTpeipei top poop azo Tijg aApdeiag, nai eppdAAet opypp ezidvpiag, Kai odpyei eig zAaprjp Tovg 6<pdaApovg

("And now, my children, I tell you, Do not be drunk with wine, because wine perverts the mind from the truth, arouses the impulses of desire, and leads the eyes into the path of error"); 19.1: Tenva pov, r)

<piAapyvpia irpog e'idaAa odpyel (!), on ep zAdpij 8l apyvpLov Toiig pr) oprag deovg dpopa^ovotp, Kai ttol€l toii exopra avTrpi eig eKoraoLP epzeoeip

("My children, love of money leads to idolatry, because once they are led astray by money they des ignate as gods those who are not gods. It makes any one who has it go out of his mind"); T. Reub. 2.8-9

(in the context of the teaching on the seven vices): e]Qdopop zpevpa ozopag Kai ovPovoLag, pedf/g ovveLGepxcraL did rr]g (piArjdopiag r) dpapria: (9) did tovto eoxarov eon rf/g KTioeug Kai zpaTop rfjg veoTTjTog, on dypoLag zezArjpunai Kai avrrj top vewTepov odr]yei cog mpAop ezi fiodpop Kai cog it, wrath already is murder. In the tractate, by con trast, wrath leads to murder. Cf. Knopf, Lehre, 14.

Moreover, the tractate lacks the eschatological con

↑ Back to top