Barnabas' Isaiah Text and Melito's Pascal Homily

by Robert A. Kraft (University of Pennsylvania).

[Publication History: appeared first as a "Critical Note" in the Journal of Biblical Literature 80 (1961) 371-373 when the author was an Assistant Lecturer at the University of Manchester (England); revised and expanded for electronic distribution in September 1993 (keyboarded by Ellen Shevitz). Subsequently some evidence has been added on the basis of TLG electronic searches (December 2006).]

In JBL 79 (1960), pp. 336-50, I tried to point out some of the peculiarities which are found in the quotations from Isaiah as used by the Epistle of Barnabas.  Since the publication of that study, it has come to my attention that the recently recovered Homily on the Passion (Περι Πασχα) attributed to Melito of Sardis\1/ contains three references to Isaiah which shed added light on the corresponding quotations in Barnabas.  

 \1/ There are two extensive MSS of the Greek text:  (1) Chester Beatty-University of Michigan Papyrus 12, ed. by C. Bonner in Texts and Documents 12 (1940), with photographic facsimiles published by F. G. Kenyon (1941); and (2) Papyrus Bodmer 13, ed. by M. Testuz (1960).  Smaller fragments of the Homily exist also in Greek, Latin, Syriac, and Coptic; for the relevant literature, see the footnotes in Testuz.  For the most part, obvious scribal errors and insignificant orthographic peculiarities in these MSS of the Homily have not been noted below. [[@@ need to update bibliography on editions and translations; e.g. Perler, Hawthorne, Hall]]

1.  Isa 3.10a, with no significant variation from LXX/OG MSS, is found in Barn 6.7 = Peri Pascha 72 (a shorter quotation than in Barn). It is interesting that Barn = Peri Pascha = LXX MSS here read εἰπόντες· Δήσωμεν τὸν δίκαιον...  in contrast to the following variant forms of these words attested by some other patristic quotations:  

(1)  λέγοντες Δήσωμεν ... is found in the Dialogue of Papiscus and Philo (12; ed. McGiffert, p. 69),\2/ while its Latin equivalent ("dicentes alligemus ...") is given by Jerome as the LXX/OG reading in his Commentary on Isaiah, ad loc. Possibly λέγοντες also lies behind the reading of Barn\L/ 6.7 (see below, variant item 3), qui dicunt.

  \2/ Compare the Dialogue of Timothy and Aquila (f. 85v; ed. Conybeare, p. 71) -- λέγοντες· δεῦτε καὶ δήσωμεν ..... Note also that the quotation from Jer 11.19 in Peri Pascha 63 has εἰπόντες where all other witnesses to the OG have λέγοντες -- there may well be a reciprocal influence between Peri Pascha 63 and 72 with their similar (formulaic?) phraseology "reckoning evil upon me, they said...."

(2)  εἰπόντες· Ἄρωμεν ... is attested by Justin Martyr (Dial. 136.2 and 137.3, along with δήσωμεν -- see further below), and the ἄρωμεν also is found in Hegesippus (apud Eusebius, HE 2.23.15) and Clement of Alexandria (Strom. 5.108.2, followed by ἀφ’ ἡμῶν, as also in Eusebius, Prep Evang 13.13.35; cf also Basil of Caesarea, Homilia adversus eos qui irascuntur 31.372.13);\3/ the Latin equivalent to ἄρωμεν, "auferamus," occurs in Tertullian (Adv. Marcion 3.22) and in Rufinus' translation of Eusebius, ad loc.  Later authors who attest the ἄρωμεν reading include Cosmas Vestitor (Orationes de translatione corporis mortui Joannis Chrysostomi 2.217; followed by ἀφ’ ἡμῶν) and George Syncellus (Ecloga chronigraphica 412.17)

\3/ Apparently Clement here is quoting from his hellenistic-Jewish School tradition, which goes back (at least) to Aristobulus.  On this problem, and on early quotations of Isa 3.10 in general, see N. J. Hommes, Het Tesimoniaboek (1935), pp. 178 f. and 187-95.

(3)  A further alternative form of this text in early Christianity rests on the use of Isa 3.10 by Wisd 2.12 -- ἐνεδρεύσωμεν (δὲ) τὸν δίκαιον... -- and is found in Hippolytus (Adv. Judaeos 9; ed. Lagarde, p. 67), Eusebius (Eclogae propheticae 174.23), Epiphanius (Testimonia 46.2), Paschal Chronicle (170.12), Dialogue of Timothy and Aquila (10.32),  George Cedrenus (Compendium 1.385.21), and George the Monk (Chronicon 390.26 = breve 110.460.49; see also Philotheus Coccinus, Vita Sabae iunioris 20.45);  while the Latin equivalent ("circumveniamus ...") is attested by the ancient Latin version of Barn 6.7, by Cyprian (Test. 2.14), Lactantius (Div. Inst. 4.16 = Epitome 40[45], and Evagrius' Dialogue of Simon and Theophilus (2.4; ed. Bratke, p. 36). [[p.372]]

Of special interest is the fact that Justin Martyr explicitly identifies ἄρωμεν as the reading of "the LXX," but claims that δήσωμεν was preferred by the Jews whom Trypho represents (Dial. 137.3; Justin uses δήσωμεν in Dial. 17.2 and 133.2). Nevertheless, no existing MS of the LXX reads ἄρωμεν!

2.  Section 101 of the Homily contains a noncanonical hymnodic speech by the resurrected Christ which begins with an allusion to Isa 50.8 -- τίς ὁ κρινόμενός πρός ()με; ἀντιστήτω μοι.  As I noted in the previous article, p. 346, patristic references to Isa 50.8-9 are extremely rare, and the quotations of this passage in Barn = Irenaeus\4/ differ strikingly from all other known texts.  Melito's allusion, brief though it is, provides a hitherto unattested variant in reading πρός ()με\5/ instead of μοι (as in all other witnesses except LXX/OG MS 36 and Iren., which lack μοι).  In lacking ἅμα after ἀντιστήτω μοι, however, Peri Pascha agrees with Barn = Irenaeus and the Syro-Palestinian version (Irenaeus and Cyril\lemma/ also lack μοι here).  Unfortunately, there is no way of knowing whether the author of Peri Pascha also was familiar with the interesting variations which occur in the remainder of the quotations in Barn = Irenaeus.  

\4/ For the texts, see JBL, 79, p. 346.  This and two other striking correspondences between quotations in Barn and Iren also are treated by L.-M. Froidevaux, "Sur trois textes cite/s par Saint Ire/ne/e," Recherches de Science Religieuse 44 (1956), pp. 408-21.

\5/ Bonner's restoration of the lacuna here in the Chester Beatty text, [νός πρός] με, has been proved correct by the Bodmer MS (πρός με), and is, indeed, a normal LXX/OG construction with κρίνεσθαι -- cf. Judg 21.22; Job 31.13; Jer 2.35 and 32.31; Sir 42.8. On the use of ὁ δικαιούμενός μοι (Barn = Iren) instead of κρινόμενός μοι in the last part of Isa 50.8, see also "Dissertatio contra Judaeos" (TLG) 1.463, 499, 6.159 and LXX/OG MSS 26 407 410 (see also 88).

3.  Most significant, however, is the minute variation of word order in Isa 53.7b which is attested by Peri Pascha 64, Barn 5.2, Acts of Philip 78, and Ps-Chrysostom's Peri Nesteias 6 (twice):

 Barn/Melito/Act Phil/Ps-Chrys
  LXX MSS of Isa 53.7b

ὡς πρόβατον (same)
  ἐπὶ\6/ σφαγὴν ἤχθη (same)
καὶ ὡς ἀμνὸς ἄφωνος\7/

καὶ ὡς ἀμνὸς
  ἐναντίον   ἐναντί
    τοῦ κείραντος\8/ αὐτόν     τοῦ κείροντος αὐτόν\12/


οὗτος\9/ οὐκ ἀνοίγει\11/ (same)
  τὸ στόμα αὐτοῦ (same)

  \6/ Instead of ἐπὶ σφαγὴν, the Chester Beatty MS has ΕΙΣΦΑΓΗΝ, which  Bonner corrected to εἰ<ς> σφαγὴν.  Since the initial ΕΙ (LXX/OG ἐπὶ) could originate as well from the Π dropping out of ἐπὶ as from copying only one of two consecutive sigmas, there is no necessity to accept Bonner's correction.  The Bodmer text has ἐπὶ σφαγὴ [sic] ἤχθην [sic]. Note, however, that the parallel material in Peri Pascha 4 (Bodmer) has εἰς σφαγὴν (Latin ad ...).

  \7/ The ancient Latin version of Barn has no equivalent for ἄφωνος ("sine voce" or "mutus" in the Latin fathers and versions), and the abridged Latin versions of the Homily (see H.  Chadwick, "A Latin Epitome of Melito's Homily on the Pascha," JTS 11 [1960], pp. 76-82) places "sine voce" in its usual LXX position. But the related passage in Peri Pascha 4 (see also above, n.6) seems to have the possibly formulaic ἀμνὸς ἄφωνος (see the similar forms in 44 and 71), which as a standard designation might well have influenced the order of the wording in the quotation -- see also this terminology in Gregory Nazianzenus Peri Huiou/Orat 29 (PG 20.20f). A similar conflating of the terminology of Jer 11.19 ἀρνίον ἄκακον with Isa 53.7 πρόβατον ... ἀμνὸς ... ἄφωνος appears (in often homiletic contexts) in a number of Christian authors including Origen, Eusebius, and Ps-Ignatius (based on a search of the TLG electronic texts).

  \8/ Act Phil and Ps-Chrys, with many LXX/OG witnesses, have κείροντος, and the second quote in Ps-Chrys has ἐνώπιον for ἐναντίον -- a frequent variant pattern in the Old Greek materials.  The reading of Barn/Peri Pascha also has extensive LXX/OG and patristic support (see Ziegler's edition of Isaiah in the "Goettingen LXX" on this and other LXX/OG variants mentioned).

  \9/ οὗτος in some LXX/OG witnesses and also in the Bodmer MS of Melito.

  \10/ The Greek MSS of Barn [cp. Act Phil] end the quotation with αὐτόν, but the ancient Latin version agrees with Peri Pascha in including the words "sic non operuit os suam."

  \11/ αὐτόν is lacking in several LXX/OG MSS and in the quotations in Clement of Rome (16.7); Justin M.  (Dial. 114; but cf. Apol. 50); Tertullian (De Carn. Res. 20; but cf. Adv. Jud. 9 and 13); Cyprian (Test. 2, 15; but cf. Epistle 4.4 [to Rogatian]; Lactantius (Div. Inst. 4, 18\var/); Origen (Homily 10.1 in Jer.; but cf. Comm. on John 6.273 and 28.166), Eusebius (Dem. 1.10.15; 9.4.4; 10 [Proem.4]; 10.8.32); and Jerome (Comm. on Isa, ad loc.).

In Barn, this section forms the conclusion to a longer quotation which begins with phrases from Isa 53.5 (as in LXX/OG); in Melito and Act Phil (ed. Lipsius and Bonnet II, 2, p. 30; cf. pp. 96f., where the material from Isa 53 is lacking), this begins a quotation which is completed by a phrase which comes from the middle of Isa 53.8 -- τὴν (δὲ) γενεὰν αὐτοῦ τίς διηγήσεται.\12/

\12/ The abridged Latin version of Peri Pascha also includes the initial phrase of Isa 53.8.

The unique position in which  ἄφωνος is found in Barn/Melito/Act Phil adds strong support to the hypothesis that popular quotations ("testimonies") circulated among early Christian authors in various textual forms.  It is not impossible that such "peculiar" variations ultimately may have been derived from Greek MSS of Jewish scriptures which now are lost to us; nevertheless, it is reasonably clear from the respective contexts that none of the writings in question took the quotation directly from a MS of Isaiah (notice the short, "proof text" quotations associated with Isa 53.7 in Peri Pascha and Act Phil).  On the other hand, there is no reason to suggest that there was any direct, literary relationship between Barn, Melito, and Act. Phil.  One thing is clear -- their agreement in such a detail cannot be coincidental, and serves to illustrate the complexity of the situation which is liable to be encounted in any examination of the use of Jewish scriptures in early Christian literature.