BARNABAS' ISAIAH TEXT AND THE "TESTIMONY BOOK" HYPOTHESIS ROBERT A. KRAFT JOURNAL OF BIBLICAL LITERATURE 79 (1960) 336-350 The so-called Epistle of Barnabas is in many ways an extraordinary product of early Christianity. It has repeatedly defied attempts of the critic to date it precisely,\1/ to locate its place of origin and destination,\2/ or to suggest what type of person might have been its author.\3/ Nor is there consensus of opinion as to what category of liierature Barnabas represents.\4/ ----- \1/ Suggested dates range from 70-138 C.E. The most recent discussions about Barnabas are the articles by L. W. Barnard: "The Problem of the Epistle of Barnabas," CQR 159 (1958), pp. 211-30; "Judaism in Egypt, A.D. 70-135," CQR 160 (1959), pp. 320-34; and "Barnabas and the Tannaitic Catechism," ATR 41 (1959). See also A. L. Williams, "The Date of the Epistle of Barnabas," JTS 34 (1933), pp. 337-46. \2/ Barnabas is usually considered to be the product of Alexandrian Christianity, probably written to an Egyptian community (see the articles by Barnard). Some commentators, however, suggest that the epistle may have been written to Rome (Volkmar, Lipsius), Asia Minor (Mu%ller), or Syria (Pfleiderer). V. Bartlet, "Papias's 'Exposition': its Date and Contents," in Amicitae Corolla, ed H. G. Wood, p. 20, suggested that Barnabas was written ca. 70-79 C.E. "by an Asian 'teacher' to Christians in Alexandria." \3/ The major problem is, was "Barnabas" a "teacher"? Barnard seems to imply a negative answer (CQR 159), while Barlet explicitly calls the author of the Epistle a teacher (see the preceeding note). \4/ See J. Muilenburg, The Literary Relations between the Epistle of Barnabas the the Teaching of the Twelve Apostles, pp. 48 f., for arguments in support of the true epistolary character of Barnabas in opposition to the views of Wendland and van Manen. W. Bousset, Ju%disch-Christlicher Schulbetrieb in Alexandria und Rom, pp. 312f., sees in Barnabas the ad hoc arrangement of school materials by an early Christian teacher; it is thus like a letter, but is not a true letter. ===== The very fact that Barnabas must antedate the middle of the second century, however, makes it an extremely important document for the student of Christian origins. For the most part, recent investigaions hve connenrated on the Two Ways section in chaps. 18-20,\5/ but this is by no means the only significant material in the epistle. A study of the text and tradition behind Barnabas' formal quotations, for example, [[337]] sheds considerable light on the sources available to early Christian authors, as well as on some aspects of early Christian exegesis.\6/ ----- \5/ See the discussions of F. C. Burkitt, J. A. Robinson, and R. H. Connolly in JTS 33 (1932), 35 (1934), and 37 (1936); H. J. Cadbury in JQR 26 (1936); E. J. Goodspeed in ATR 27 (1945); Muilenburg, Literary Relations; and Barnard, CQR 159 (1958). \6/ This paper contains some partial results from the author's research on "The Text and Tradition of the Explicit Quotations in the Epistle of Barnabas." ===== The Epistle of Barnabas contains approximately 100 explicit quotations from the OT,\7/ one fourth of which can be traced to the book of Isaiah.\8/ In contrast to the quotations from the Pentateuch and the other Prophets, Barnabas' Isaiah references are reasonably close to our LXX texts\9/ and are not greatly influenced by the MT tradition.\10/ A minute comparison of Barnabas' Isaiah variants with known LXX variants, however, fails to indicate any consistent pattern of relationship between the epistle and any one LXX text-type (although agreement with Vaticanus is, as we might expect, less frequent than with Sinaiticus, Alexandrinus, or Marchalianus).\11/ [[338]] ----- \7/ Great difficulty is involved in counting the citations because of composite quotations, unidentified references, repeated quotations, and the like. Almost all of the quotations have a strong OT flavor, but whether they were originally intended to be direct OT citations or are sometimes secondary compositions based on the OT is a separate problem in itself. \8/ Barn 16 3 is probably not a quotation from Isa 49 17 (so also E. Hatch, Essays in Biblical Greek, p. 186), but rather, from some kind of eschatological writing based on OT narratives about the rebuilding of Jerusalem and the temple (compare the final section of "4QTestimonia" in this connection; JBL 75 [1956], pp. 185f.); Barn 9 3b does not seem to be intended as a quotation from Isa 40 3, since the emphasis must fall on the opening words (E)KOU/SATE, TE/KNA, FWNH=S ...), and not on the possible allustion to Isa 40 3 which follows. In the same context in Barnabas, the older commentators habitually identified a citation as from Isa 1 10, an claimed that Barnabas had deliberately modified the ending (e.g., Hatch, Essays, p. 182). There is no need for such an interpretation since the exact words quoted in Barn 9 3 occur in Isa 28 14. \9/ This could also be said for some (but not all) of Barnabas' Psalm references. \10/ H. B. Swete, Introduction to the OT in Greek, p. 412, concurs in general with his judgment (see also Hatch, Essays, pp. 180 ff.). Through an unfortunate misinterpretation of an ambiguous statement by Hatch (Essays, p. 208; see below, n. 88), Muilenberg states that "Barnabas' most conspicuous agreement with the Hebrew appears in his substitution of SINA= for SIW=N [in Barn 11 3], which is found in all MSS of the LXX" (Literary Relations, p. 88). As a matter of fact, all MSS of the MT (and the versions) also read "Zion," not "Sinai." H. Windisch, Der Barnabasbrief (in Lietzmann's Handbuch Supplement), p. 315, states: "Gleichwohl ist an einzelnen Zitaten doch festzustellen, dass der Vf. oder seins Vorlage eine, dem hebra%ischen Text na%herkommende Uebersetzung der LXX vorgezogen hat s. zu 2 5, 7; 6 2; 9 8; 11 2; 15 3." All of these instances are not equally convincing (two of them concern Isaiah quotations). \11/ Swete, Intro., p. 413, has a very brief analysis of Barnabas' textual affinities in quotations from Psalms and Isaiah. He concludes that "the leaning in Isaiah towards the text of Q [Marchalianus], expecially when found in company with A or Hebrew letterA is noteworthy." In the light of the variants noted by J. Ziegler, Isaias (Vol. 15 of the Go%ttingen LXX), however, Swete's etimate seems less noteworthy. It is true that Barnabas' Isaiah text disagrees less withg Q readings than with other MSS like B, A, S (Hebrew letter), and V, but this is because Q is usually in agreement with the majority of better MSS on a given reading -- that is, it seldom has "peculiar" readings. Thus we find that Barnabas' "peculiar" readings are almost never supported by Q. The exceptions are (1) Barn 3 1 (Isa 58 5) where Q, 198, 239\1, and 407 also have the second LE/GEI KU/RIOS, but in this Q adds PANTOKRA/TWR, against Barnabas and the others; (2) Barn 14 8 (Isa 49 7a) reads QEO/S instead of O( QEO/S, in agreement with A, S\c/, B. and many minor MSS. In many instances the orthography of the original scribe of Q differs from that of existing Barnabas MSS. Barnabas has a few "peculiar" variants which are in agreement with B (Vaticanus): (1) in 6 2a (Isa 50 9) the second W(S is omitted as in B, S\*/, Bohairic, 109, and 736; (2) in 14 9 (Isa 61 1 ) is the variant TH\N KARDI/AN (instead of TH=| KARDI/A|) which also occurs in B, some minor "Alexandrian" MSS, many "Lucianic" MSS, an two Catena MSS; (3) the QEO/S reading (already noted in connection with Q) is in B. Many of Barnabas' variants are unique or find support only in other Church Fathers. Windisch, Barnabasbrief, p. 315, concludes: "Hieraus schon ergibt sich, ... dass Abweichungen vom LXXtext also nicht in jedem Falle auf eine besondere Uebersetzung oder Textform zuru%ckzuru%hren sine." If there is any tendency in Barnabas' Isaiah variants to agree with extant MSS of the LXX, it seems to be towards such "Alexandrian" MSS as 544, 198, and 407; buy such agreement is not consistently demonstrable. ===== Certain peculiarities in Barnabas' citations have often led critics to point to an ancient catena of OT texts rather than to direct use of the OT itself as the source of these references. The similarity between the composite quotation in Barnabas 11.2-3 and a citation in Justin's Dialogue with Trypho 114 (see the discussion below) was one of the points at issue when Edwin Hatch made his classical suggestion that "Greek-speaking Jews" probably had produced propaganda manuals, some of which consisted "of extracts from the OT."\12/ Hatch's words were published in 1889; men like Vollmer\13/ and Burkitt\14/ advanced similar theories until in 1920 Rendel Harris sent his thoroughgoing testimony book hypothesis onto the scholarly battlefield.\15/ Readction to Harris' form of th theory was not lacking,\16/ but few critics have cared to deny the possibility that some sort of testimony literature may have existed quite early in the Christian tradition. Harris' claim that there was originally one anti-Judaic handbook of collected OT extracts (which book Harris equated with the cryptic "logia of Matthew" mentioned by Papias). however, has been largely abandoned in subsequent years.\17/ [[339]] ----- \12/ Hatch, Essays, p. 203 ("On Composite Quotations from the LXX"). \13/ Die alttestamentlichen Citate bei Paulus. Vollmer thought that the testimony liberature existed in Hebrew for dogmatic purposes. \14/ The Gospel History and its Transmission. Burkitt seems to have coined the term "testimonia" in this connection. \15/ R. Harris and V. Burch, Testimonies. A more recent and in some ways even more thoroughgoing testimony book hypothesis is presented by Stather Hunt, Primitive Gospel Sources. \16/ See especially N. J. Hommes, Het Testimoniaboek, which deals primarily with the words of Harris and D. Plooij (Studies in the Testimony-Book). \17/ Not even Stather Hunt argues for only one testimony book: "In speaking of 'The Testimony Book', I do not usually mean one single volume. I use the term as a short and convenient one to cover that cycle of writings which embodied the testimony principle, and was in existence ["though probably in different recensions," p. xi] before the writing of the first Gospel" (Sources, p. 14). The conclusions of A. L. Williams, Adversus Judaos, p. 7, are relevant for the present study: "There is every probablility that there was not only one Book of Testimonies, but several. It is not likely that at first the OT passages were ven written down at all, for they would be but few, and very esily remembered. But as time went on ... many would begin to make written memoranda of the chief texts for their own use. It is not probable that such notes would always be alike. To one Evangelist certain texts would appeal, to other Evangelists others. There would thus be many little Books of Testimonies, as we many call them. But in the process of time there would arise ... something more elaborate and more complete ... [but] never ... anything quite complete.... There would inevitably be a large measure of matter that was common to all such books. And, in fact, several of such little books have survived to our own time." The conclusions of this paper, which were already in embryo form before I read Williams' treatment, substantially agree with his analysis of the development of the testimony literature. ===== Interest in such "testimony literature" has been stimulated recently by the publicaion of some Qumran Cave 4 fragments under the provisional titles of "4Q Tesimonia" and "4Q Florilegium."\18/ It is no longer possible to deny the early use of collected scriptural excerps, not i i easy to maintain the Christian origin o he testimony lieterature.\19/ Judaism already had produced testimony collectsions in Hebrew before the end of the first century C.E., and it would be peculiar if Greek-speaking Jews of the same period in history had not entered into a similar enterprise. There should be, therefore, no a priori objection to the suffestion that early Christian authors, perhaps even some NT authors themselves, may have used testimony literature. ----- \18/ J. M. Allegro, "Further Messianic References in Qumran Literature," JBL 75 (1956), pp. 182-87; and "Fragments of a Qumran Scroll of Eschatological MIDRA-S~I^M," JBL 77 (1958), pp. 350-54. See also J. A. Fitzmyer, " '4QTestimonia' and the New Testament," TS 18 (1957), pp. 513-37. There are some noteworthy differences between "4QFlorilegium" and "4QTestimonia"; the former contains "Eschatological Midrashim" organized around certain OT passages (especially II Sam 7 10 ff.), while the latter includes few editorial comments and no real introductory formulas. See the recent note by W. R. Lane, "A New Commentary Structure in 4QFlorilegium," JBL 78 (1959), pp. 343-46. Sections of Barnabas differ in ways which might be explained by the suggestion s that the author used several different types of sources like those discovered at Qumran. \19/ As O. Michel, Paulus und seine Bibel, seemed to imply: "Es fehlt jede Spur spa%tju%discher Florilegien" (p. 43); and again: "Es finden sich keine Spuren vorchristlicher Florilegien, sei es spa%tju%disch-hellistischer Art (Hatch), sei es spa%tju%disch-rabbinischer Art (Vollmer)" (p. 52). ===== Is there strong evidence that Barnabas took some or all of his Isaiah quotations from a widely used testimony book? The peculiar variaions, occasional composite quotations, noncontextual exegesis and topical grouping of some references seem to suggest that Barnabas may have used such a source.\20/ On the other hand, the epistle is in many ways [[340]] strikingly different from the later anti-Judaic Christian writings which argue from scriptural testimonies.\21/ It i neither in the "dialogue" form, nor does it entirely follow th topical arrangements reflected by the testimony books of Cyprian\22/ and Pseudo-Gregory of Nyssa.\23/ The author of Barnabas does not adhere to one primay method of using his OT citations:\24/ (1) sometimes he presents a lit of shoot quotaions, apparently selected by means of the "Stichwort" principles;\25/ (2) sometimes he gives a running midrash on ann OT section of text;\26/ (3) occassionally there seems to be evidence of live discussion (anticipated objections, etc.) behind Barnabas' argument;\27/ (4) frequently he diverges into allegorical interpretaions:\28/ and (5) occasionally he quotes several consecutive verses from the OT, although shorter prooftext references are more usual.\29/ ----- \20/ The major pieces of evidence on which the testimony book hypothesis traditionally rests are: (1) composite quotations, (2) ascription of quotations to wrong authors, (3) peculiar variations which are common to severl authors who seem to be independent from each other, (4) the same combinations of quotations in different independent authors, and (5) anti-Judaic polemic. Barnabas does not exhibit all these features consistently. Muilenburg, Literary Relations, p. 110, goes so far as to say (in answer to Windisch): "Of an original Testimonienstoff ... and Didachestoff there is no evidence, at least so far as Barnabas is concerned." Obviously there is some direct editorial work in Barnabas' quotations, but it has been greatly exaggerated. Many of the variants which are traditionally blamed on "faulty memory" or "willful adaptation" probably came from the sources behind the epistle. As we shall see, the problem is no longer whether there were sources, but rather, what was the nature of the sources. Some of them certainly must be called "Testimonienstoff." \21/ As Stather Hunt puts it: "Though 'Barnabas' is not written in the usual testimony style, it is in effect a testimony writing" (Sources, p. 210). In isolation from Hunt's over-all thesis, this particular conclusion seems accurate. \22/ Ad Quirinum: Testimoniorum Libri Tres, ed. Hartel. \23/ Delecta Testimonia Adversus Judaeos (Migne, PG 46). \24/ See von Ungern-Sternberg, Der traditionelle alttestamentliche Schriftbeweis 'De Christo' und 'De Evangelio' in der alten Kirche bis zur Zeit Eusebs von Ca%sarea, pp. 272 f. On the basis of the diversity of materials (especially typological and allegorical) in in Barnabas, he argues that "hierdurch ist in den einfachen Traditionsstoff ein fremdartiges Element eingedrungen, welches die Einheitlichkeit der Darstellung in hohem masse sto%rt." \25/ See especially the string of short quotations in 9 1-3, which exhort men to "hear," and the longer refereences to "circumcision" which follow in 9 4-9. \26/ See 10 1-9, 11-12 on the Mosaic food laws; 10 10 on Ps 1 1; chap. 15 on the sabbath rest. \27/ See especially 6 2b; 14 1; 16 6 f.; et passim. \28/ Numerous examples may be found in chaps. 7-8, 10, and 12. \29/ See 3 1-5 (Isa 58 4-10); 11 6-7 (Ps 1 3-6); 2 5 (Isa 1 11-13); 13 1-5 and 14 2-3 (several long Pentateuch citations). ===== In most instances the quotations from Isaiah are given without explanatory editorial comment concerning their interpretation and application -- they seem to stand as independent, out-of-context prooftexts for the arguments of Barnabas. They do not seem to be used as [[341]] "pointers to a larger context," and thus do not add support to Dodd's theory about NT methods of using the OT.\30/ Certain characteristic features sometimes recur in these quotations: (1) there is a definite tendency for the closing words of the citations (and the transition words in composite quotations) to vary from the LXX wording;\31/ and (2) in some instances, the texts apparently have been made to fit better into the parenetic context of the epistle by the standardization of their pronouns and verbs to thh second person plural,\32/ and by emphasizing such phrases as "behold," "woe to you," and "the Lord says."\33/ ----- \30/ C. H. Dodd, According to the Scriptures, p. 126. \31/ See Barn 3 5 in Hebrew letter (Isa 58 10); 11 2b in Hebrew letter and C (Jer 2 13 is united with Isa 16 1); 12 4 (Isa 65 2); 14 8 (Isa 49 7a); and 16 2b (Isa 40 2 united with 66 1). This tendency may account for the strange phrase, TAU=TA EI)S MARTU/RION, which occurs in the Greek MSS (but not in the ancient Latin translation) of 9 3 after a quotation from Isa 1 2a, but it seems more probable that the words are intended to emphasize that the appeal of Isa 1 2 to the heavens and earth constitutes a legal witness against the deafness of Israel and for the rightous judgments of the Lord (compare similar comments in Barnabas 8 3-4). It is unlikely that the phrase means "these citations are found in the testimony book" or something to that effect. Pergaps 6 2b (Isa 28 16b) and 5 14 (Isa 50 6) are also examples of this editorial tendency. \32/ 11 3 in the Greek MSS (Isa 16 2) and 3 1-2 (Isa 58 4-5) exhibit this tendency, which may also come from Barnbabas' sources. Compare the changes in 11 4 to "they" (Isa 45 3 ["thou"]) and in 11 5 to "thou" (Isa 33 16 ["he"]). \33/ See especially 3 1 ff.; 5 14 6 2; 16 2b. ===== There are many minor variants in Barnabas' twenty-odd Isaiah quotaions, and there are also several very interesting major deviations from the best extant LXX texts. We shall now turn to an examination of some of the latter passages which seem especially revlevant for an evaluation of the testimony book hypothesis.\34/ ----- \34/ Additional "major variants" in Barnabas' Isaiah quotations which are not discussed in this paper may be found in 3 1-5 (Isa 58 4-10); 6 3b (Isa 50 7); 11 4 (Isa 45 2); 12 4 (Isa 65 2); 14 8 (Isa 49 7a); and 14 9 (Isa 61 1). ===== (1) At the end of chap. 12, the author of Barnabas argues that Jesus is "Son of God" and not "Son of David" on thh basis of Ps 109 (110) 1 ("The Lord said to my Lord, sit at my right hand ....") and Isa 45 1 ("The Lord said to my Messiah the Lord, whose right hand I held....").\35/ In the latter passage, the MT and LXX read, "Thus [[342]] and the Lod (God) to my Anointed one Cyrus ...." When some ancient (scribe or) expositor found in his Greek text the word KU/RW| (the Greek transliteraion of Hebrew Hebrew text), he must have suspected that an iota had passed out of his text, for in some traditions\36/ KU/RW| (Cyrus) became KURI/W| (Lord), as is evidenced by the use of Isa 45 1 with his same peculiar variant in Irenaeus,\37/ Tertullian,\38/ Cyprian,\39/ Novatian,\40/ Lactantius,\41/ Tyconius,\42/ Evagrius,\43/ et al.\44/ Jerome noted and corrected the reading in his Commentary on Isaiah,\45/ and the testimony book of Ps-Gregory (which may come from about the time o Jerome) quotes the verse correctly but proceeds to argue that what was said about "Cyrus" was really only fulfilled in Christ!\46/ This tesimony from Isa 45 1 apparently had gained a revered spot in the apologetic of early Christianity. ----- \35/ There is some textual difficulty here within the MSS of Barnabas which must be noted: Hebrew letter\*/ and C present what appears to be a harmonization of the opening words with Ps 109 (110) 1 (EI)=PEN KU/RIOS TW=| KURI/W| MOU OU(= E)KRA/THSEN TH=S DECIA=S AU)TOU=...). Two of the MSS of family G (which consists of 8 defective Greek MSS descended from one archetype), b\*/ and f, read EI)=PEN KU/RIOS TW=| XRISTW=| MOU KU/RW|... with Hebrew letter\c/. The Latin and remaining MSS of family G have what has been accepted as the original reading (...TW=| SRISTW=| MOU KURI/W ...), although the Latin corrects the opening words to Sic dicit dominus (LXX, OU(/?TWS LE/GEI KU/RIOS). There is similar corruption at the conclusion of the argument in Barn 12, but it is not directly relevant to this paper. \36/ The fact that the KURI/W| variant has no known LXX MS attestation is a strong argument that it is an exegetical rather than a purely textual) tradition. \37/ Ap. Pr. 49 (twice). \38/ Adv. Prax. and 28; Adv. Jud. 7. \39/ Test. i, 21. \40/ De Trin. 26 (21) (Migne. PL 3:937). \41/ Div. Inst. iv, 12. \42/ Liber Regularum I, ed. Burkitt, Texts and Studies III (1894), pp. 3 f. \43/ Altercatio Simonis et Theophili, ed E. Bratke, pp. 13 f. \44/ Ziegler's critical apparatus also includes Augustine as a witness to this unusual reading. \45/ Migne, PL 24:456f.; so also Jerome, Commentary in Abacuc on 3 13 (PL 25:1389). \46/ Test. 16; see also Isidore, De Fide Cath. contra Jud. i, 3, 2 f. (PL 83). ===== (2) Another relatively popular reading found in Barnabas (3 4) and corrected by Jerome\47/ is from Isa 58 8 -- "Then shall your light break loose ealy, and your garments shall rise quickly...." The best LXX MSS read "healings" (I(A/MATA; MT, "healing") in place of "garments" (I*MA/TIA),\48/ and it is easy to see how the textual tradition in Greek became confused. It is doubtful that any theological prejudices originally influenced this variant -- Barnabas builds no argument upon it -- but it is interesting that Tertullian (and others) laer used it as a support for the bodily resurrection ("your garments," that is, your flesh, "shall speedily arise" after death).\49/ Justin,\50/ Cyprian,\51/ and one major MS of Clement of Alxandria\52/ also have the reading "garments" in quotations [[343]] from Isa 58 8. It is perhaps significant, however, that Irenaeus does not preserve this unique variant, but reads instead (like the LXX), "your healings (sanitates) shall speedily arise."\53/ ----- \47/ Comm. on Isaiah on 58 8. \48/ Ziegler lists the following LXX MSS witnesses for "garments": "Alexandrian" MSS 26, 86\*/, 106\*/, 239, 407, 534, 538, 544; "Hexaplaric" MSS 109, 736 (oII); "Lucianic" MSS S\ca/, 147, 311; and the Catena MS 91\*/. Barnabas' MS Hebrew letter has the (corrected) I)A/MATA reading. \49/ De Carnis Res. 27. \50/ Dialogue 15. \51/ Test. iii, i; De Dom. Orat. 33; De Op. et El. 4. \52/ Paed. iii, 12, 89 (MS P\*/). \53/ Adv. Haer. iv, 17, 3. Theophilus, Ad Autol. iii, 12 (one of the few times he quotes a passage from Isaiah which is also in Barnabas), also has I)A/MATA in his quotation of Isa 58 8 (so PG 6; G. Bardy's edition in Sources Chre/tiennes 20 reads TA\ I)MA/TIA [sic], but the accompanying translation by J. Sanders correctly has "ta gue/rison"). ===== (3) Thus far we have noticed single word variants which obviously have arisen within the Greek tradition, and hich frequently have been used as evidence to support a testimony book hypothesis. Of a somewhat different nature is another Barnabean variant found in the same quotation from Isa 58 which preserves the I(MA/TIA reading. The LXX of Isa 58 6a is: OU(XI\ TOIAU/THN NHSTEI/AN (E(GW\) E)DELECA/MHN; while Barn 3 3 has: I(DOU\ AU(/TH (H()\54/ NHSTEI/A H(\N E(GW\ E(CELECA/MHN. Justin\55/ and Cyprian\56/ are in essential agreement with the LXX. Irenaeus, however, reads: Hoc est ieiunium, quof ego elegi; and Clement of Alexandria (who admittedly knew the Epistle of Barnabas) concurs with Barnabas. ----- \54/ Hebrew letter omits the H(, which is found in C. \55/ Dial. 15 (exact LXX including E)GW/. \56/ Test. iii, 1. It is also possible that Tertullian, De Ieinui 2, is alluding to our passage (non tale ieiunium dominus elegit), but lack of context precludes assurance that Isa 58 5a (LXX, OU) TAU/THN TH\N NHSTEI/AN [E)GW\] E)CELECA/MHN) was not in mind instead. Cyprian's version of 58 6a is non tale ieiunium elegi (or accepi), while he has for 58 5a non hoc ieiunium (quod) ego elegi. ===== These examples create a confusing picture of the early Christian use of testimonia. Ps-Gregory reads "Cyrus" instead of KURI/W|. Irenaeus agrees more with Barnabas than with the LXX in the beginning of the quotation from Isa 58 6-10, although a few words later he does not have the I(MA/TIA variant found in Barnabas\57/ -- yet Irenaues does attest the KURI/W| tradition. Both Tertullian and Cyprian have the KURI/W| and I(MA/TIA readings, but neither agree with Barnabas' text of Isa 58 6a. ----- \57/ Of course, it is not impossible that the extant Latin versions of Irenaeus (or the Greek textual tradition behind it) has corrected Irenaeus' original text at this point. ===== (4) Another brief example will serve to confirm this confusing lack of pattern among our sources. In a quotation from Isa 42 6-7, which is otherwise almost identical with known LXX texts, Barn 14 7 preserved the variant PEPEDHME/NOUS where the LXX has th equivalent word DEDEME/NOUS (" ... to open the eyes of the blind, to lead from bonds those who have been bound/shackled ...."). Although Ziegler lists no LXX MSS attestation for Barnabas' reading, PEPEDHME/NOUS is also found in quotations of Isa 42 7 by Justin\58/ and Ps-Gregory.\59/ [[344]] Its Latin translation, alligatos (or adligatos), occurs in Lactantius\60/ and Evagrius.\61/ On the other hand, Tertullian\62/ and Cyprian\63/ seem to agree with the LXX. Having sufficiently illustrated the complex nature of the situation, let us now turn to one of the crucial passages in the development of testimony book theories -- the "stone testimony" of Isa 28 16. ----- \58/ Dial. 26, 65, 122. \59/ Test. 16. \60/ Div. Inst. 20 (Epit. 43 [48]); his phrase is ex vinculis alligatos. \61/ Sm. et Theoph. (p. 47): e (or de) vinculis vinctos). \62/ Adv. Marc. iii, 20 and Adv. Jud. 12 ( de vinculis vinctos). \63/ Test. iii, 7 (a vinculis vinctos). ===== (5) Isa 28 16 was an extremely popular Christian text. It is well known to NT critics because of its use in Rom 9 33 and I Pet 2 6. Barn (6 2b-3a) quotes it in connection with Ps 117(118) 22. Irenaeus,\64/ Tertullian,\65/ Hippolytus,\66/ Cyprian,\67/ Methodius,\68/ Aphrahat,\69/ ps-Gregory,\70/ and the anonymous Dialogue of Athanasius and Zacchaeus\71/ all have it in one form or another. I must emphasize the phrase "in one form or another," for it is quite important at this stage of the argument. ----- \64/ Adv. Haer. iii, 21, 7. \65/ Adv. Marc. v, 5 and 6; Adv. Jud. 14 (see also 10). \66/ Refut. Omn. Haer. v, 7, 35. This is more an allusion than a formal quotation of Isa 28 16. \67/ Test. ii, 16. \68/ Sermo de Simeone et Anna 6; allusion, not formal quotation. \69/ Demonstratio I: De Fide 6. \70/ Test. 8. \71/ Dial. AZ 111. ===== In the history of tesimony book speculation it was the strange form of Isa 28 16 in the NT which helped suggest that Paul and the author of I Peter may have used a scriptural anthology rather than the OT text itself as the immediate souuce for their quotations.\72/ Depenence of Romans on I Pet 2 6 seemed chronologically unlikely, and dependence of I Peter on Rom 9 33 seemed improbable because of certain unique differences in the form of the quotaion. That is, both I Pet 2 and Rom 9 read, "Behold, I am placing in Zion a stone...," as opposed to the LXX reading, "Behold, I will insert into hh foundaions of Zion a stone...." Thus far, both NT texts preserved a unique reading which seemed nearer to the MT than to the LXX. But Rom 9 conflates the [[345]] rest of the verse with Isa 8 14, while I Pet 2 refers to Isa 8 14 only after finishing the Isa 28 16 referene with some minor variation from the LXX and after noting Ps 117(118) 22. Since it is unlikely that the author of I Peter unraveled Paul's conflate quotation and proceeded to quote Isa 28 18 and 8 14 separately, the argument ran, both I Peter and Romans must reflect the use of a common souce which had Semitic peculiarities -- that is, a testimony book. Further evidence was addued from Ps-Gregory's quotaion of this passage, which is essentially the saame as that in I Pet 2 6, but has a few minor variations. Aphrahat's citation again has the main features of I Pet 2 6, but also includes some unique variations.\73/ ----- \72/ The texts compare as follows: [[column 1]] LXX I)DOU\ E)GW\ E)MBALW= EI)S TA\ QEME/LIA SIW\N LI/QON POLUTELH=, E)KLEKTO\N A)KROGWNIAI=ON, E)/NTIMON EI)S TA\ QEME/LIA AU)TH=S KAI\ O( PISTEU/WN E)P' AU)TW=| OU) MH\ KATAISXUNQH=| [[column 2]] Rom 9 33 I)DOU\ TI/QHMI E)N SIW\N LI/WON PROSKO/MMATOS KAI\ PE/TRAN SKANDA/LOU (omit) KAI\<\Gr> (PA=S) O( PISTEU/WN E)P' AU)TW=| OU) (MH\) KATAISXUNQH/SETAI(-NQH=|) [[column 3]] I Pet 2 6 (as in Romans) (as in Romans) LI/QON E)KLEKTO\N (as in LXX9 (omit) (as in LXX) (as in LXX) \73/ D. Plooij, Studies, pp. 15-20, uses evidence from Aphrahat to build a theory that there was an Aramaic testimony book used by both Paul and Aphrahat. See also A. Bakker, "Testimony-Influence in the Old-Latin Gospels," Amicitiae Corolla, esp. pp. 9 ff. ===== The anonymous fourth-century Dialogue of Athanasius and Zacchaeus witnesses to the Pauline form of Isa 28 16 conflated with 8 14, as does Tertullian in two places.\74/ But the matter becomes even more complicated by the third Tertullian quotation of this testimony -- this time more like the usual LXX form. Cyprian and Irenaues also give the LXX form with minor variations, while Barnabas begins his Isa 28 16 quotation in accord with the LXX, but then makes a (Johanninelike) paraphrase of the final clause: "and whoever hopes on him will live forever."\75/ The difficulties which these facts present for any testimony book hypothesis are not removed by the observation that Ps 117(118) 22 frequently (but not always) occurs in close connection with Isa 28 16;\76/ nor does Isa 8 14 consistently occur in these contexts.\77/ In shot, it is impossible that all these authors had used essentially the same common source for their quotations of Isa 28 16.\78/ [[346]] ----- \74/ Adv. Marc. v, 5 and Adv. Jud. 14 (see notes 64-71 for the remaining patristic references in this paragraph). \75/ So family G MSS read (KAI\ O(\S E)LPI/SEI E)P' AU)TO\N ZH/SETAI EI)S TO\N AIW=NA); Hebrew letter and C have KAI\ O( PISTEU/WN EI)S AU)TO\N (etc. as in G), while the Latin has the ambiguous phrase, et qui crediderit in illum, followed by the correct Isa 28 16 ending, mon confundetur (compare Barn 8 5; 11 10 f.). \76/ Barnabas, I Peter, Cyprian, Ps-Gregory, and Dial. AZ. \77/ Romans, I Peter, Tertullian (references in n. 74), and Dial. AZ. Harris and other try to identify traces of Isa 8 14 (MT) in Barnabas' introduction to the quotation from Isa 28 16. The evidence is inconclusive (despite Hommes agreement with Harris on this point; Testimoniaboek, pp. 89 f.). Another OT passage which frequently is to be found among the "stone testimonia" is Dan 2 34 (and context). It is interesting to note that both Justin (see Dial. 34, 86, 100, 113, 114) and Tertullian (Adv. Jud. 9) claim that Christ is called "Stone" in many scriptural passages. \78/ After a detailed examination of Harris' "Stone-Testimonia" evidence, Hommes concludes: "Harris' argumenten zijn waardeloos" (Testimoniaboek, p. 91). Dodd, in somewhat more restrained language, seems to concur with our conclusion (Scriptures, pp. 42f.). Although the Dead Sea literature sometimes alludes to the OT "stone" passages, its evidence does not contribute directly to the present study. ===== (6) All of the Isaiah quotaions in Barnabas were note equally popular with ealy Christian authors. Immediatey preceding the reference to Isa 28 16, Barn (6 i f.) quotes from Isa 50 8-9 in a form which is strikingly different from the LXX. Among Ante-Nicene writers, the full form of this "testimony" is found elsewhere only in irenaus, and has it both in his Apostolic Preaching and in Against Heresies\79/ with the same over-all peculiarities as we have seen in Barnabas, but with significant differences as well. The texts compare in the following way: ----- \79/ Ap. Pr. 88, Adv. Haer. iv, 33, 13. Clement of Alexandria, Strom. iii, 12, 86, gives a phrase from Isa 50 9b which is too brief to be of relevance for our argument. ===== [[column 1]] LXX TI/S O( KRINO/MENO/S MOI; A(NTISTH/TW MOI A(/MA: KAI\ TI/S O( KRINO/MENO/S MOI; E)GGISA/TW MOI. D)DOU\ KU/RIOS BOHQEI= MOI: TI/S KAKW/SEI ME; I)DOU\ PA/NTES U(MEI=S W(S I(MA/TION PALAIWQH/SESQE, KAI\ W(S SH\S KATAFA/GETAI U(MA=S. [[column 22]] Barnabas (as in LXX) A(NTISTH/TW MOI: H)\ TI/S O( DIKAIOU/MENO/S MOI; E(GGISA/TW TW=| PAIDI\ KURI/OU. } } (omit) OU(AI\ U(MI=N O(/TI U(MEI=S PA/NTES (as in LXX) KAI\ SH\S (as in LXX) [[column 3]] Irenaeus Quisquis iudicatur? Ex adverso adstet. Et quisquis iustificatur? Appropinquet puero Dei (Ap. Pr., "Lord") Et (omit in Ap. Pr.) Vae vobis, quoniam omnes veterascetis sicut vestimentum, et tinea comedet vos. (Isa 2 17 follows in composite quotation.) Although it is not impossible that Irenaeus or his translators exhibit a knowledge of Barnabas in this quotation, it is highly improbable. It seems that here Barnabas and Irenaeus reflect a common source which had the "servant of the Lord" addition,\80/ the omission of the line which followed in the LXX, and the "woe to you because" modification\81/ (and perhaps the composite Isa 2 17 ending?). But if this source were a well-known and widely used testimony book, why are there no other references to this passage which is so adaptable to anti-Judaic polemics? ----- \80/ Notice the mention of the "servant" in the immediate context (Isa 50 10 -- compare Barn 9 2b); in fact, it is the "servant" who seems to be speaking in 50 4-9. \81/ This phrase is used similarly by Justin, Dial. 114, in his conflated quotation based on jer 2 13 (see below). It is an infrequent idiom in the LXX; see Sirach 41 8. ===== (7) Possibly Barnabas' most provocative and unique Isaiah reading occurs in the composite quotation from Jer 2 12 f. and Isa 16 1 f. in Barn 11 2-3. The only other reference to Isa 16 1 that I have been able [[347]] to find in alleged testimony book literature prior to Isidore of Seville (6th-7th cent.)\82/ is a passing allusion by Justin in Dial. 114. Justin claims to be quoting from Jeremiah, and begins by giving the content of Jer 2 13,\83/ which is followed by the words: MH\ E)/RHMON H|0= OU)= E)STI TO\ O(/ROS SIW/N, O(/TI I(EROUSALH\M BIBLI/ON A)POSTASI/OU E)/DWKA E)/MPROSQEN U(MW=N; these last thoughts seem to come from a conflation of Jer 2 31 (MH\ E(/RHMOS E)GENO/MHN TW|= I)SRAH/L;), Isa 16 1 (MH\ PE/TRA E(/RHMO/S E)STIN TO\ O(/ROS SIW/N;) and Jer 3 8 (KAI\ E)/DWKA AU)TH|= [I)SRAH/L] BIBLI/ON A)POSTASI/OU). ----- \82/ Contra Judaeos, i, 9; see also later authors such as Peter Damiani, Antil. contra Jud. 4 (PL 145:54), and Peter of Blois, Contra Perfid. Jud. 10 (PL 207:837), who present a fully Christianized argument which can be traced back as early as Jerome's Comm. on Isaiah (on 16 1). In these authors, the desert rock represent the once-barren Moabitess, Ruth, from whom a lamb (Christ) is sent to Zion. For such an argument the MT tradition is needed ("lamb" where the LXX has "serpents," etc.). \83/ I)EREMI/AS ME\N GA\R OU(/TW BOA=|: OU)AI\ U(MI=N, O(/TI E)GKATELI/PETE PHGH\N ZW=SAN KAI\ W)RU/CATE E)AUTOI=S LA/KKOUS SUNTETRIMME/NOUS OI(\ OU) DUNH/SONTAI SUNE/XEIN U(/DWR. ===== Barnabas gives a more extended quotation from Jer 2 12-13,\84/ followed without a break by Isa 16 1b in this strange form: MH\ PE/TRA E(/RHMO/S E(STIN TO\ O)/ROS TO\ A(/GIO/N MOU SINA=;\85/ the striking difference here is, of course, "My holy mount Sinai" instead of the LXX (and Justin's) "Mt. Zion," or the LXX variant (with the MT) "Mount of the daughter of Zion." This fact alone makes it unlikely that Justin took his quotation from Barnabas, but it also seems dificult to isolate a commonsource which was used directly by both Justin and Barnabas. The hypothetical common source, which could have been some sort of midrashic commentary on jer 2-3 (to account for Justin's form of the reference) and which must have included the whole of Isa 16 1-2 (perhaps by association with Jer 2 31\86/). would already have had to be modified into two textual traditions by the time it was used by Barnabas and Justin respectively. ----- \84/ Some commentators (e.g., Hatch and Muilenburg) think that Barnabas' text has been revised towards the MT in this quotation -- note the imperatives at the beginning. Barnabas' BO/QRON QANA/TOU is unique. \85/ The remainder of the quotation, from Isa 16 2, is relatively close to the LXX wording (note changes from singular to plural in Barnabas). \86/ Notice the similarity between Jer 2 14a (MH\ DOU=LO/S E)STIN I)SRAH/L...), 2 31 (see above), and Isa 16 1b. In a commentary on Jer 2-3, it would be easy to employ Isa 16 1 f. as complementary or explanatory material (such a method was common practice). ===== Could it be that the author of Barnabas willfully emended "Mt. Zion" into "my holy mount Sinai"? This is at least unlikely, for why should any anti-Judaic writer make Sinai into a holy mountain? One might so speak about Zion, which is continually referred to as "the holy mount" in scripture and in tradition; but Sinai is seldom if ever regarded [[348]] in such a way.\87/ Thus, one could understand it if Barnabas had edited his reference to "Sinai" in place of "Zion," or read "my holy mount Zion" for "Mt. Zion," but it is hard to imagine "my holy mount Sinai" as the deliberate emendation of the author of the epistle. It would seem, rather, that Barnabas is in this instance painfully faithful to his source, which had somehow acquired this unique reading.\88/ And if this is true, Justin did not directly use the same source. ----- \87/ In the OT, Ps 67 18 (68 17) comes closest to Barnabas' phrase -- the LXX has: O( KU/RI/OS E)N AU)TOI=S E)N SINA= E)N TW=| A(GI/W|. Even here, A(GI/W| need not refer to SINA=. Notice how Irenaeus, Ap. Pr. 83, interprets this passage: "The Lord ... from Zion ascended up on high" -- even though he has quoted the text correctly: "The Lord (is) among them in Sinai, in (his) sanctuary. He ascended up on high..." (Robinson's translation). Elsewhere the OT speaks of "consecrating" Sinai (see Exod 19 23 LXX -- A(GI/ASAI AU)TO/), but not of Sinai as a "holy mount." For Zion as the "holy mount," see such passages as Isa 11 9; 56 7; 57 13; 65 9-25; Jer 38 (31) 23; etc. \88/ Hatch, Essays, p. 208, suggested that Barnabas may provide "an important correction of the LXX text, for whereas all the MSS of the LXX have SIW=N, the context and the Hebrew require SINA= (see n. 10 above). It is true that "Zion" does not fit well into the MT context, but "Sinai" seems to fit no better. Isa 15-16 presents an oracle against Moab, and thus 16 1 would most logically require a mountain in the immediate vicinity of Edom-Moab (note the reference to Hebrew word or PE/TRA; is it a place name here?). Perhaps the passage originally spoke of a mount near Kadesh and Sin ("mount of Kadesh-Sin"?) in the desert. If so, "Kadesh" could easily become "holy" (Hebrew text) in some traditions (see Deut 33 2 LXX for a similar problem), and "Sin" (Hebrew word, which the LXX transliterates as SEIN or SIN; MS A has SINA in Num 27 14!) could become either Zion or Sinai. In any case, Isa 16 1, as we have it in extant texts, is "exceedingly difficult and obscure"; in fact it is "most uncertain" concludes G. B. Gray, ICC Isaiah I, pp. 285ff. ===== There are other examples of Barnabean Isaiah passages which (1) are popular in our ealy Christian authors but are quoted in widely different textual forms,\89/ or (2) are not very popular in the same lierature but seem to point to a source behind Barnabas which was not a complete Isaiah text.\90/ With regard to h popular passages, there seems to be no consistent pattern among the Fathers -- occasionally Barnabas' variants are also found in some other Father(s), at other times Barnabas and the same Father(s) preserve conray traditions on a crucial variant.\91/ [[349]] As for the less popular passages, we can only wonder why they were not more widely used if there really were a widely-known testimony souce from which Barnabas drew his citations.\92/ ----- \89/ See Isa 1 11-13 (Barn 2 5 f.); 58 6-10 (3 3 ff.); 50 6 f. (5 14); 65 2 (12 4); 61 1 f. (14 9); and 66 1 (16 2b). \90/ See Isa 33 13 and 28 14 (Barn 9 1-3); 33 16-18 (11 5). \91/ Brief statistics at this point are nearly meaningless. Each variant must be examined in detail really to point up the complexity of the situation. For a rough idea, however, the following figures should suffice: about half of Barnabas' Isaiah quotations appear (in whole or in part) in Justin, Irenaeus, Tertullian, Clement of Alexandria, Cyprian, and Ps-Gregory respectively. Barnabas is more often in agreement with Irenaeus (about 67% of the time), than with the others and is most frequently in disagreement with Justin (about 60%) and Ps-Gregory (about 67%). With the other authors, agreements seem to balance the variations in test cases. It must be emphasized that these general statements may be highly deceptive when the actual details are more closely viewed. \92/ The only Isaiah quotation found in Barnabas and in all of the six authors mentioned in the preceding note is 65 2, which is also found in Rom 10 21. ===== Harris' oversimplified testimony book hypothesis cannot account fot these peculiarities. It seems that if Barnabas had used such a testimony book, it was greatly different both in precise content and in textual detail rom other testimony books used by other early Christian authors, and it apparently was based more on a Greek than on a Semitic textual tradition.\93/ On the other hand, the following factors cannot be ignored: (1) Barnabas obviously used sources of some sort; (2) collections of testimonies were already in existence inthh first century C.E. (as the Qumran fragments testify); and (3) there seems to be some sort of lierary relaionship between individual passages quoted with the same peculiar variations in otherwise independent Christian writings. ----- \93/ We must leave open the possibility that Barnabas may reflect some Greek readins which originated in independence from (or prior to) the "LXX," even in Isaiah. This possibility becomes even greater in Barnabas' quotations from the other Prophets and the Pentateuch. One could, of course, argue that the Christian authors who used "the testimony books" did not use it for all of their quotations, but arbitrarily selected references from it, while useing other sources at the same time. THis is always possible, but if it is true, the testimony book hypothesis looses its excitements as a possible solution to the problems of OT quotations. ===== It seem, therefore, that Barnabas may represent one ealy stage inthe adaptation and modificationof lte Jewish testimony literature by Christian authors, and in the transition of that literature towards its later, more developed anti-Judaic forms. Many of Barnabas' prooftexts need not presuppose a specifically Christian collection.\94/ In fact, the attitudes of certain branches of sectarian Judaism towards the temple cultus easily could have motivated the collection of some of the anti-Judaic Barnabas citations.\95/ But Barnabas also contains some specifically Christian prooftexts and interpretaions, as Isa 53 5 and 7.\96/ [[350]] His souces, then, seem to lie womewhere between Jewish testimony literature (as seen at Qumran) and full-fledged Christian testimony books (Cyprian, et al.). Moreover, the complex relationship between Barnabas' quotations and later Christian quotaions could be more easily explained on the supposition that some of Barnabas' sources did not survive in the Christian tradition, while others did. ----- \94/ The Isa 16 1 passage, for example, is more satisfactorily explained if it were culled from a Jewish source. \95/ The Qumran sectarians and some factions in Alexandrian Judaism known to use through Philo's arguments are two of the most obvious possibilities; on the altter, see M. J. Shroyer, "Alexandrian Jewish Literalists," JBL 55 (1936), pp. 261-84, and Philo, Migration of Abraham 91 ff. \96/ Other texts could be mentioned, especially those which deal with Christ's fleshly suffering: isa 3 9 f. (Barn 6 7); Isa 50 6 f. (5 14). It is not impossible (although improbable) that even these texts could have been used in a similar way by some elements of sectarian Judaism. ===== Long ago, Wilhelm Bousset suggested that brief "Schulbetrieb" notes lay behind much late Jewish and early Christian literary endeavor in Alexandria and Rome (see n. 4, above). It may be significant that the "4QTestimonia" fragment is "clealy not part of a scroll," not is it "inscribed on the reverse."\97/ Could it be that such testimony note sheets are the primitive ancestor of he later testimony books? As the codex form gradually replaced the scroll in popular use, so these individual note pages gradually were gathered and revised into larger units, and some note pages disappeared by chance or by design as the course of the Christian apologetic became modified. Similar testimony pages may have been transmitted by Semitic and Greek communities which were similarly oriented, and thus we sometimes find the same testimony text preserved in different recnesions, one of which may seem more Semitic than the other. Some of the testimony pages known to Barnabas would have found their way into the sources used by later writers, others would not. This results in the untold confusion which one finds when he attempts to examine the relationships between these authors. In short, it seems unlikely that the author of Barnabas took his Isaiah quotaions from any one source -- certainly not from a testimony book of the type Harris proposed. He may have used a complete scroll of Isaiah for some of his references; he may have he may have used some midrashic commentary material also; but many of his citations must have come from a late Jewish and early Christian "Schulbetrieb" background which produced short, independent documents of a testimony page nature, each of which had its own organizational theme, yet all of which were related by means of their similar exegetical mentality. It is difficult to imagine any less complex explanaion for the perplexing situation which is evident as one examines the relationship of Barnabas to the "testimony tradition" reflected in the pages of other early Christian writings. ----- \97/ Allegro, JBL 75 (1956), p. 182; the "Nash papyrus" may be a related type of literature -- see the comments of G. E. Wright, Biblical Archaeology, p. 213. ===== //end//