Hostname: page-component-7c8c6479df-5xszh Total loading time: 0 Render date: 2024-03-27T10:29:35.933Z Has data issue: false hasContentIssue false

Textual Evidence of Tatian's Dependence upon Justin's 'AΠOMNHMONEYMATA

Published online by Cambridge University Press:  05 February 2009

William L. Petersen
Affiliation:
University Park, Pennsylvania, USA

Extract

Because of their early date (mid-second century) and interesting variants, the gospel citations of Justin Martyr — he almost always refers to his source as άπομνημονεύματα τν άποστόλων (‘memoirs of the apostles’), and only rarely as a ‘gospel’ — have long attracted scholars’ attention. It is self-evident that the citations contain numerous variant readings and are frequently harmonized; the dispute has been over what best explains this phenomenon. Semisch and Zahn said that the variant readings and harmonizations were due to lapses of Justin's memory. Credner argued that Justin's non-standard citations were the result of his reliance upon extracanonical gospels. Bousset opined that Justin had access to presynoptic traditions; this explained both the similarities and the differences between Justin's text and the later (revised) synoptic text. Von Engelhardt suggested that Justin used a post-synoptic gospel harmony. This position found support in the researches of Bellinzoni, who investigated the ‘sayings of Jesus’ genre in Justin. He noted passages cited more than once in exactly the same harmonized form, suggesting a written document, and passages paralleled in other early Christian sources in very similar harmonized form. Bellinzoni interpreted this as indicating that Justin had used a gospel harmony, which was also used by other early Christian writers. He noted that Justin's variant citations and harmonizations seemed especially close to those found in the Pseudo- Clementine Homilies and Recognitions. Bellinzoni concluded that this harmony was post-synoptic, excluded John, and ‘had great influence on the later manuscript tradition of Matthew, Mark and Luke’. In the last paragraph of his book, Bellinzoni speculated that the harmony used by Justin might have links with the Diatessaron: ‘Tatian was a pupil of Justin, … It is now apparent that the concept of a gospel harmony did not originate with Tatian.’

Type
Articles
Copyright
Copyright © Cambridge University Press 1990

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Justin uses the word εύαγγέλιον only 4 times, twice in the singular, and twice in the plural. By contrast, he uses άπομνημονεύματα 15 times. See Koester, H., ‘From the Kerygma-Gospel to Written Gospels’, NTS 35 (1989) 361–81, esp. 377–80.CrossRefGoogle Scholar

2 Semisch, K., Die apostolischen Denkwürdigkeiten des Märtyrers Justinus (Hamburg, 1848) 389–92Google Scholar; and Zahn, Th., Geschichte des neutestamentlichen Kanons, 1.2 (Erlangen: Andreas Deichert, 1888) 463585.Google Scholar

3 The Gospel of the Hebrews and the Gospel of Peter (so Credner, C. A., Beiträge zur Einleitung in die biblischen Schriften [Halle, 1832] 266–7);Google Scholaror, additionally, the Protevangelium Iacobi (so Hilgenfeld, A., Kritische Untersuchungen über die Evangelien Justin's, der Clementinischen Homilien und Marcion's [Halle, 1850]).Google Scholar

4 Bousset, W., Die Euangeliencitate Justins des Märtyrers in ihrem Wert für die Evangelienkritik (Göttingen: Vandenhoeck & Ruprecht, 1891) 114–16.Google Scholar

5 von Engelhardt, M., Das Christenthum Justins des Märtyrers (Erlangen, 1878) 335–48.Google ScholarThis position was supported by Sanday, W., The Gospels in the Second Century (London: Macmillan, 1876) 136–8;Google Scholarand Lippelt, E., Quae fuerint Justini Martyris AΠOMNHMONEYMATA quaque ratione cum forma Evangeliorum syro-latina cohaeserint (Halle, 1901) 35.Google Scholar

6 An example (Bellinzoni, 28–30) is Dial. 76.4 (= Dial. 120.6 = Dial. 140.4), all of which follow Matt 8. 11–12, save for the fact that each of Justin's three passages begins with the identical opening phrase from Luke 13. 29. Bellinzoni concluded that Justin must have been using a written source to have quoted three times exactly (with the sole exception of the transposition of δυσμν and άνατολν in Dial. 76.4) a passage which runs 29 words; Bellinzoni also concluded that Justin must have been using a source which harmonized Luke 13. 29 with Matt 8. 11–12, for it seems unlikely that one would, spontaneously and from memory, make exactly the same harmonization three times. This is only one of numerous examples adduced by Bellinzoni.Google Scholar

7 One of the many examples is Apol. 16.13 (= Dial. 35.3), which Bellinzoni (44–7) concludes harmonizes Matt 24. 5 with Matt 7.15–16/19; this same harmonization appears in the Ap.Con. 6.13.3 (Les Constitutions apostoliques, III–VI, tome II, ed. Metzger, M. [SC 329; Paris: Cerf, 1986] 336–8)Google Scholarand the Pseudo-Clementine Homilies 11.35 (ed. Rehm, B. & Irmscher, J. [GCS 42; Berlin: Akademie, 1953] 172).Google Scholar

8 Bellinzoni, 140. One must remember, however, that Justin quite probably had multiple sources – some written and some oral – at his disposal; hence, each of the suggestions of earlier scholars might be correct for certain passages.Google Scholar

9 Reported by Irenaeus, haer. 1.28.1 (SC 264.354–56); cp. Eusebius, h.e. IV.29.1 (SC 31. 213).Google Scholar

10 Bellinzoni, 142. It should be noted that there is abundant evidence for pre-Tatianic gospel harmonies: see the examples given in the text infra, at n. 85.Google Scholar

11 Kline, L. L., The Sayings of Jesus in the Pseudo-Clementine Homilies (SBL Dissertation Series 14; Missoula, Montana; Scholars Press, 1975).Google Scholar

12 E.g.: Hom. 19.2.5 = Justin, Dial. 76.5 (Kline, 14–15; cf. Bellinzoni, 114–16); Hom. 3.55.1 = Justin, Ap. 16.5 (Kline, 86–7; cf. Bellinzoni, 64–7); Hom. 11.26.2 = Justin, Ap. 61.4 (Kline, 134–7; cf. Bellinzoni, 135–8).Google Scholar

13 Kline, 175.Google Scholar

14 Ibid.

15 TU 70.(Berlin: Akademie, 1958).Google Scholar

16 NTS 24 (1977–8) 297316.Google Scholar

17 Ibid., 316.

18 Ibid., 315.

19 Using the edition of Goodspeed, E. J., Die ältesten Apologeten (Göttingen: Vandenhoeck & Ruprecht, 1984MGoogle Scholar; photomechanical reprint of the 1914 edition).

20 The main witnesses are usually divided into Eastern and Western, on the basis of language and provenance. For the purposes of our study, the following list will suffice:Google ScholarAn introduction to these witnesses is to be found in my The Diatessaron and Ephrem Syrus as Sources of Romanos the Melodist (CSCO 475; Louvain: Peeters, 1985) 2051Google Scholar; Metzger's, B. M.The Early Versions of the New Testament (Oxford: Oxford, 1977) 1036, offers a similar introduction and listing.CrossRefGoogle Scholar

21 Since we have no exemplar of the Diatessaron, we must reconstruct its text from ‘witnesses’, often sequentially rearranged and ‘Vulgatized’ to remove non-standard (when compared with the ‘standard’ canonical text of the locale and period) readings. For a discussion of the problems and methods for ascertaining the reading of the Diatessaron, see my The Diatessaron and Ephrem Syrus 5567Google Scholar; or my Romanos and the Diatessaron: Readings and Method’, NTS 29 (1983) 484507.CrossRefGoogle Scholar

22 Goodspeed, 202.Google Scholar

23 Saint Éphrem, Commentaire de l'Évangile concordant, version armènienne, ed. Leloir, L. (CSCO 145; Louvain: Peeters, 1954) 36.Google Scholar

24 The Commentaries of Ishoad of Merv, ed. Gibson, M. D. (HSem 5; Cambridge: Cambridge, 1911)I.27.Google Scholar

25 The Pepysian Harmony, ed. Goates, M. (EETS o.s. 157; London: Oxford, 1922) 10.Google Scholar

26 Vita Beate Virginis Marie et Salvatoris Rhythmica, ed. Vögtlin, A. (BLVS 180; Tübingen, 1888) 129Google Scholar. Apropos the relationship of this document to the Diatessaronic tradition, see van den Broek, R., ‘A Latin Diatessaron in the “Vita Beate Virginis Marie et Salvatoris Rhythmica”’, NTS 21 (1974) 109–32.CrossRefGoogle Scholar

27 Jülicher, A., Itala, Vol. 1 (Berlin: W. de Gruyter, 1938) 14.Google Scholar

28 Idem.

29 Panarion in Epiphanius II, Panarion haer. 34–64, edd. Holl, K. and Dummer, J.CGCS 66; Berlin: Akademie, 1980 2) 350–51.Google Scholar

30 S. Thasci Caecili Cypriani, Opera omnia, ed. Hartel, G. (CSEL 3.3; Vindobonae: Geroldi, 1871) 90.Google Scholar

31 The difference between Justin's reading πρ, ‘fire’, and the reading of the other witnesses, which speak of φς, ‘light’ (the sole exception is the Pepysian Harmony: ‘brightness’), is usually ascribed to internal variations within the same non-received variant tradition - that is, once the original variant was introduced (either ‘light’ or ‘fire’), the alternate term later appeared (as a translational variant, adaptation, etc.). Noteworthy is the recent suggestion of Drijvers, H. J. W. and Reinink, G. J. (‘Taufe und Licht: Tatian, Ebionäerevangelium und Thomasakten’, in Text and Testimony, Essays in Honour of A. F. J. Klijn, ed. Baarda, T. et al. [Kampen: Kok, 1988] 91110) that the original term was Justin's ‘fire’, which Tatian adopted, but changed to ‘light’, which had greater theological significance for him.Google Scholar

32 This is confirmed by the reading in Codex Vercellensis (a), which dates from the fourth century.Google Scholar

33 Goodspeed, 202.Google Scholar

34 Ibid., 203.

35 Leloir, 85; 111 (with minor variations).Google Scholar

36 Marmardji, A.-S., Diatessaron de Tatien (Beyrouth: Imprimerie Catholique, 1935) 37.Google Scholar

37 Burkitt, F. C., Evangelion da-Mepharreshe (Cambridge: Cambridge, 1904), 1.13; Syrc reads: ‘came down as a dove’.Google Scholar

38 The Liège Diatessaron, edd. Plooij, D., Phillips, C. A. et al. (VNAW 19 & 21; Amsterdam: Koninklijke akademie van wetenschappen, 1929–70) 46.Google Scholar

39 Heliand und Genesis, ed. Behaghel, O. (ADTB 4; Halle: Max Niemeyer, 1948 6) 36Google Scholar. Apropos its relationship with the Diatessaronic tradition, the classic source is Grein, C. W. M., Die Quellen des Heliand. Nebst einem Anhang: Tatians Evangelienharmonie herausgegeben nach dem Codex Cassellanus (Cassel, 1869)Google Scholar; and recently, Weringha, Juw fon (also spelt J. von Weringh), Heliand and Diatessaron (Stadia Germanica 5; Assen: van Gorcum, 1965).Google Scholar

40 Epiphanius (ed. Holl, ) 350.Google Scholar

41 Origène, Contre Celse I, ed. Borret, M. (SC 132; Paris: Cerf, 1967) 184.Google Scholar

42 Jerome's remark (vir. inl. 2) that ‘Origen frequently uses’ the Gospel according to the Hebrews lends weight to this contention. Origen displays no direct knowledge of the Diatessaron.Google Scholar

43 Haer. 46.1.8–9: ‘It is said that from him (Tatian) there comes the Diatessaron, which is also called the Gospel according to the Hebrews‘ (GCS 66.204–5).Google Scholar

44 Goodspeed, 35.Google Scholar

45 Bellinzoni, 88, n. 1.Google Scholar

46 Diatessaron Persiano, ed. Messina, G. (BibOr 14; Roma: Pontificio Istituto Biblico, 1951) 63.Google Scholar

47 Described as a student of Ephrem's; cf. Bardenhewer, O., Geschichte der altkirchlichen Literatur (Freiburg: Herder, 1924 2) IV.374Google Scholar; or Duval, R., La littérature syriaque (Paris: Lecoffre, 1907; photomechanical reprint: Amsterdam, 1970) 313–14.Google Scholar

48 Reading #451 in Biblia Polyglotta Matritensia, VI, Vetus Evangelium Syrorum, Diatessaron Tatiani, ed. Ortiz de Urbina, I. (Matriti, 1967) 36Google Scholar, citing Nau, F., ‘Aba, comment, in Diatessaron’, Revue de l'Orient Chrétien 17 (1912) 6973.Google Scholar

49 Plooij, 70.Google Scholar

50 This lone element of the reading is paralleled by Epiphanius, haer. 58.2.1 (GCS 66.359): ε;ίς βασιλε;ίαν ούρανν.Google Scholar

51 Clemens Alexandrinus III, Stromata Buck VII und VIII …, edd. Stählin, O. and Früchtel, L. (GCS; Berlin: Akademie, 1970 2) 175.Google Scholar

52 Strom. I.1 is often so interpreted; e.g. Whittaker, M., Tatian, Oratio ad Graecos (Oxford, 1982), ix.Google Scholar

53 Goodspeed, 216.Google Scholar

54 Syriac: Saint Éphrem, Commentaire de l'Évangile concordant ou Diatessaron, ed. Leloir, L. (CBM 8; Dublin: Hodges Figgis, 1963) 148, 149; Armenian: Leloir 151 (supra, n. 23).Google Scholar

55 Jülicher, 1.136.Google Scholar

56 Jülicher, A., Itala, Vol. 3 (Berlin: W. de Gruyter, 1976 2) 205.Google Scholar

57 Irenée de Lyon, Contre les hérésies, Livre I, edd. Rousseau, A. & Doutreleau, L. (SC 264; Paris: Cerf, 1979) 290.Google Scholar

58 Hippolytus, Refutatio omnium haeresium, ed. Marcovich, M. (PTS 25; Berlin: de Gruyter, 1986) 149.Google Scholar

59 Clément d'Alexandrie, Les Stromates, Livre V, tome 1, ed. le Boulluec, A. (SC 278; Paris: Cerf, 1981) 130. In this edition, the editor's placement of quotation marks around only the first two words, είς άγαθóς, obscures the parallel!Google Scholar

60 Die Pseudoklementinen, I. Homilien, edd. Rhem, B. & Irmscher, J. (GCS 42; Berlin: Akademie, 1953) 242.Google Scholar

61 Goodspeed, 208.Google Scholar

62 ibid.

63 Goodspeed, 37.Google Scholar

64 Evangelion da-Mepharreshe, ed. Burkitt, F. C. (Cambridge: Cambridge, 1904) 1.315.Google Scholar

65 Plooij, 477–9.Google Scholar

66 Jülicher, A., Itala, Vol. 2 (Berlin: de Gruyter, 1970 2) 115–16.Google Scholar

67 See other examples in my The Diatessaron and Ephrem Syrus, 64–7 (Venetian Harmony); 80–3 (Liège); 95–104 (Venetian).Google Scholar

68 Goodspeed, 110.Google Scholar

69 Marmardji, 385.Google Scholar

70 Liège: Plooij, 580; Theodiscum: Das Leben Jhesu, ed. Gerhardt, C. (CSSN series minor, tome I, Vol. 5; Leiden: Brill, 1970) 126;Google ScholarHaaren: Het Haarense Diatessaron, ed. de Bruin, C. C. (CSSN series minor, tome I, Vol. 2; Leiden: Brill, 1970) 90;Google ScholarCambridge: Het Diatessaron van Cambridge, ed. de Bruin, C. C. (CSSN series minor, tome I, Vol. 3; Leiden: Brill, 1970) 32.Google Scholar

71 In addition to the readings that will be presented in this section, two other readings in Justin have been noted as paralleling the Diatessaron, but their status is dubious: (1) Justin, Dial. 78.7 = Matt 2.16 (?) A reading found at Matt 2. 16 in the Liège, Pepysian and Persian Harmonies, in Ludolph of Saxony's Vita Jesu Christi, and in The Heliand and Saelden Hort, ‘the Magi did not return to him (Herod)’, is clearly Diatessaronic. Justin, however, is drawing on Matt 2. 12. The Diatessaron gives a doublet of this instruction, having it at Matt 2. 12, as well as interpolating it into Matt 2. 16. Hence, Justin is almost certainly not a parallel for this reading. (2) Justin, Dial. 103.6 = Matt 4.1 The Liege, Venetian and Pepysian Harmonies, and Jakob van Maerlant's Rijmbijbel all interpolate at the beginning of Matt 4. 1 ‘after Jesus was baptized’. The reading lacks Eastern Diatessaronic support, but clearly is part of the Western harmonized tradition. Justin makes the identical interpolation in Dial. 103. 6. However, when read in context, Justin's reference leaves doubt as to whether it is a quotation of Matt 4.1. And where there is doubt, we have chosen not to adduce the passage as evidence. In this instance, the phrase is probably an introduction to set up the temptation account necessitated by the fact that Justin is concluding a digression; furthermore, it is followed by a repetition of the statement of the voice from heaven at the baptism - another indication that we are probably dealing with literary stage-setting. Both of these readings are from Quispel, G., Tatian and the Gospel of Thomas (Leiden: Brill, 1975) 145–7, who, additionally, logged the agreement between Diatessaronic witnesses and Justin in #2 of our Secondary Evidence, and Exhibits 1 and 2 of our Primary Evidence (pp. 180–1; 164–6).Google Scholar

72 Leloir, 189 (supra, n. 23).Google Scholar

73 At Luke 10. 22; Tetraevangelium Sanctum iuxta simplicem Syrorum versionem, edd. Pusey, P. E. and Gwilliam, G. H. (Oxford, 1901) 386–7.Google Scholar

74 Migne, PL 198.1572.Google Scholar

75 See the list in the apparatuses of Plooij, 154, or The New Testament in Greek, The Gospel according to St. Luke, III.l (Oxford: Oxford, 1984) 232.Google Scholar

76 Cf. The New Testament in Greek, III.l, p. 268.Google Scholar

77 See my ‘New Evidence for the Original Language of the Diatessaron’, in Studien zum Text und zur Ethik des neuen Testaments. Festschrift zum 80. Geburtstag von Heinrich Greeven (BZNW 47; Berlin: W. de Gruyter, 1986) 325–43.Google Scholar

78 The same confusion may account for the reading ‘sins’ in certain Diatessaronic witnesses at John 1. 29.Google Scholar

79 Plooij, 73.Google Scholar

80 ibid., 154, in the apparatus.

81 Tertulliani Opera, Pars I (CChr.SL 1; Tvrnholti, 1954) 612.Google Scholar

82 Plooij, 81.Google Scholar

83 Excepting, once again, Marcion's omission of βούληται.Google Scholar

84 The sole exception is Exhibit no. 4, whose western support comes from what is a secondary witness, the Vetus Latina, whose genre is not harmonized, and whose reading, ‘Father’, is only an echo of the full variant, ‘my Father in the heavens’. The Vetus Latina is, however, known to contain numerous Diatessaronic readings; see infra, n. 91.Google Scholar

85 On ‘Vulgatization’, see supra, n. 21.Google Scholar

86 Bell, H. I. and Skeat, T. C., Fragments of an Unknown Gospel (London: Trustees of the British Museum, 1935) 8–41; they date the papyrus to the middle of the second century (2–7).Google Scholar

87 See, for example, the passages quoted by Epiphanius at haer. 30.13.2 or 30.13.7, and the discussion ofVielhauer, P., ‘Jewish Christian Gospels’, in The New Testament Apocrypha, ed. Hennecke, E. and Schneemelcher, W. (Philadelphia: Westminster, 1963) 1. 117–65.Google ScholarBertrand, D., ‘L'Évangile des Ebionites: Une harmonie évangéelique antéVieure au Diatessaron’, NTS 26 (1980) 548–63, has argued that this harmony preceded the Diatessaron.CrossRefGoogle Scholar

88 Cf. Koester, H., ‘The Synoptic Gospels in the Second Century’, in Gospel Traditions in the Second Century: Origins, Recensions, Text and Transmission, ed. Petersen, W. L. (Notre Dame, Indiana: Univ. of Notre Dame, 1989) 1937.Google Scholar

89 Ap. ad Algasiam, (121) 6, in Saint Jérôme, Lettres, ed. Labourt, J., tome 7 (Paris: Belles lettres, 1961) 860.Google Scholar

90 Ep. ad Carpianus. The text is most readily available in N-A26, p. 73*.Google Scholar

91 We ignore the Vetus Latina and Vetus Syra, which betray the impress of the Diatessaron. On the Diatessaron's influence upon the Vetus Syra, see Black, M., ‘The Syriac Versional Tradition’, in Aland, K., Die alten Übersetzungen des neuen Testaments, die Kirchenväterzitate und Lektionare (ANTT 5; Berlin: W. de Gruyter, 1972) 120–59;Google Scholarupon the Vetus Latina, see Vogels, H. J., Beiträge zur Geschichte des Diatessaron im Abendland (NTA 8.1; Münister, 1919).Google Scholar

92 See my The Diatessaron and Ephrem Syrus, 154; it is interesting to note that 1 of 6 is 17%, and 5 of 28 is 18%.Google Scholar

93 Our evidence and this observation contradict Bellinzoni's statement (supra, at n. 8) that Justin's gospel, which he concluded was a harmony, ‘had great influence on the later manuscript tradition of Matthew, Mark and Luke’Google Scholar

94 The MSS are: Paris graec. 450, dated 11th September 1364 [Goodspeed's ‘A’]; and Cod. Claromont. Mediomont, dated 1541 [Goodspeed's ‘B’]. In Goodspeed's opinion, the two MSS come from the same archetype, and thus represent a single witness (A History of Early Christian Literature [Chicago: U. of Chicago, 1942] 142).Google Scholarvon Harnack, A. (Die griechischen Apologeten des 2. Jahrhunderts in der kirchlichen Ueberlieferung) (TU 1.1–2; Leipzig, 1882] 77) described ‘A’ as ‘ein sehr sorglos und liiderlich geschriebener Codex’.Google Scholar

95 Bousset, W. (Die Euangeliencitate, 21–31) and A. Rahlfs (Der Text des Septuaginta-Psalters Septuaginta Studien 2; Göttingen: Vandenhoeck & Ruprecht, 1907] 203–6)Google Scholar investigated Justin's citations of, respectively, the prophets (Amos and Isaiah) and the Psalms. Just as Justin's gospel citations differ from the ‘standard’ gospel text, so these OT citations differ from the ‘standard’ LXX text. Bousset and Rahlfs concluded that the citations in our two MSS of Justin had been tampered with, and were unreliable transmitters of Justin's text. This appraisal reinforced the sceptical opinion of scholars regarding Justin's gospel citations. However, in 1953, D. Barthelmey published a Greek fragment of Micah found near Qumran (‘Redécouverte d'un chaînon manquant de l'histoire de la Septante’, RB 60 [1953] 18–29). It dated from the end of the first century, and disclosed a text which agreed closely with the Hebrew Micah, Symmachus, Aquila, and the Quinta known to Origen, as well as Justin's citations in Dial. 119. This demonstrated that Justin's non-standard LXX citations were neither corruptions of his tradition, nor his own invention. An independent investigation by Sibinga, J. Smit, (The Old Testament Text of Justin Martyr, I. The Pentateuch [Leiden: Brill, 1963], esp. 149–50) reached similar conclusions for the Pentateuch.Google Scholar

96 That a ‘fifth source’ was used by Tatian was first noted by Hugo Grotius, Annotationes in Libros Evangeliorum (Amsterdami, 1641) 7. Studies include: C. A. Phillips, ‘Diatessaron – Diapente’, Bulletin of the Bezan Club 9 (February 1931) 6–8;Google ScholarPeters, C., ‘Nachhall ausserkanonischer Evangelienüberlieferung in Tatians Diatessaron’, AcOr 16 (1937) 258–94;Google ScholarCharlesworth, J. H., ‘Tatian's Dependence upon Apocryphal Traditions’, HeyJ 15 (1974) 517.Google Scholar

97 Burkitt, F. C., ‘Tatian's Diatessaron and the Dutch Harmonies’, JThS 25 (1924) 113–30; the idea of a Latin original is now rejected.Google Scholar

98 A convenient ‘synopsis’ of the proximate sequence of pericopes in the major Diatessaronic witnesses is found in Leloir's, L.Le Diatessaron de Tatian’, OrSyr 1 (1956) 208–31; 313–34.Google Scholar