to the "Two Ways" Tractate precisely 1639C-1643C). If (3) and (5) are each counted
or the related Fides patrum (PC 28.1637-1643C, more precisely 1639C-1643C). If (3) and (5) are each counted as two, and the Didache added, we have altogether eight different but somehow related recensions of the Two
6. The Relationship of the Didache to the "Two Ways" Tractate a.The Findings The Didache begins, in chaps. 1-6 (more precisely, in 1.1—6.1), with a tractate on the two ways the human being can go: the way to (eternal) life, or the way to (eternal) death. There are numerous parallels to this motif of the two ways in the literature of popular philos ophy and religion.1 But there are also (and this is what interests us here) striking parallels to the Didache trac tate itself in the literature of earliest Christianity that must stand in a close literary relationship to the Didache.
The only question is, What is that relationship? The writ ings in question are (1) the Epistle of Barnabas 18-20,2
(2) the Doctrina apostolorum, (3) the first part of the socalled Apostolic Church Order or Canons of the Apostles
(more precisely, chaps. 4-13),3 and the related Epitome, (4) part of the Arabic Life of Shenoute, and finally (5) sec tions in Ps.-Athanasius Syntagma doctrinae (ed. Batiffol),
Ways tractate before us.4
1. While the Didache begins with a version of the Two Ways tractate, the analogous tractate in the Epistle of Barnabas is near the end of the document (chaps.
18-21). Almost the whole of the material included in these chapters of the Epistle of Barnabas recurs in the Didache, with the exception that in Barnabas 18 and 19.1-2 the material is only about half equivalent; at 20.1
Barnabas has a somewhat different introduction, but beginning with the catalog of vices, the sections are again equivalent, with but few exceptions, and from 20.2 onward even their order is the same.5 There is a strong likelihood of literary relationship. But how should we think of it? Is the Didache dependent on Barnabas,6 or Barnabas on the Didache?7 If the Didache were dependent
1 On this, see below.
2 I will later express the opinion that parts of the Two Ways tractate are also concealed in chap. 21.
3 On the traces of the tractate in chap. 14, see below.
4 The number increases if we include Const. 7.1.2-19 or the parallels in the Fragmenta Anastasiana and in Isaac of Syria (see above). To count them would, however, distort the picture, because the Apostolic
Constitutions is dependent on the Didache, and the Fragmenta Anastasiana and Isaac of Syria are directly or indirectly dependent on the Apostolic Consti tutions.
5 The absence of the "teknon sayings" (Did. 3.1-6) in Barnabas is striking. The omission of the sectio evangelica (Did. 1.3b—2.1) should in my opinion be con nabas and the Didache," 166-67; idem, "Streeter on the Didache," 374, 376-77. Cf. then the new version of Robinson's work in idem, "Barnabas and the Didache," 119-20, 146, 238-40; Vokes, Riddle,
27-51 (p. 44: "There seems no legitimate reason for denying that the 'Two Ways' is the original composi tion of Barnabas, used by the Didache"). John
Lawson (A Theological and Historical Introduction to the Apostolic Fathers [New York: Macmillan, 1961] 75)
"inclines to the view that Barnabas adopted 'The Two Ways' from a Jewish source, and that the writer of Didache took it over from him."
7 E.g., Zahn, Geschichte, 3.310-12; Schaff, Church Manual, 20, 121 (but Schaff also suggests the possi bility of a common source); etc. Other early advo cates of this hypothesis are listed in Ehrhard,
Altchristliche Litteratur, 49-50. Funk, in turn, wrote frequently in favor of this hypothesis; cf. Funk,
Abhandlungen, 2.109-11, 117-24, 131-34; idem, "Didache unci Barnabasbrief," ThQ 87 (1905)
161-79; later Bosio, Padri apostolici, 1.3.
8 Also striking is the agreement of Barn. 18.1c with Doctr. 1.1b against Did. 1.1, quite apart from the mutual absence of the sectio evangelica in Barnabas and Doctrina apostolorum.
Muilenburg (Literary Relation, 161-64) differs, see ing further contacts between Didache and Barnabas outside the Two Ways tractate. But the parallels he adduces are quite far-fetched. Cf. also the list in Vokes, Riddle, 27-28. (Of course, the strength of proof in the individual passages is very differently evaluated by Vokes himself.) In my opinion, howsidered in another context.
6 Thus Bryennios, Aibaxf ith'-c'. Other early advo cates of this position are listed in Ehrhard,
Altchristliche Litteratur, 49-50. Later advocates were Robinson (Didache, esp. 69-70), supposing that the Two Ways tractate stems from Barnabas, and that the Didache is dependent on Barnabas; also, in detail, Muilenburg (Literary Relations, 130), who considers Barnabas to be the originator of the Two
Ways tractate, and attempts to show (pp. 140-58) that the Didache is dependent on Barnabas. Thus the 9 compiler of the Didache may be using a Jewish apocryphon in 3.1-6 (p. 149; there was a similar sugges tion earlier by Robinson, Didache, 62). Arguments along the same line appear in Francis C. Burkitt,
"Barnabas and the Didache," JTS 33 (1932) 25-27; cf. also Connolly, "Relation," 237-53; idem, "Bar