employed maxim. But accurate citation is not characteristic of the compiler, for he apparently considered specific references superfluous in a compendium designed for ecclesiastical purposes.

1

Harnack, who counts in the Teaching 23 direct borrowings from the N. T., attributes 17 of them to Matthew. He recognizes the employment of Lukan material, but explains it by a hypothesis "daß der Verf. der Didache unter dem 'Evangelium des Herrn' ein aus dem Lucas-Ev. bereichertes Matthäus-Evangelium vorausgesetzt und benutzt hat". This writing he conjecture to be the Gospel according to the Egyptians. Krawutzcky, on the other hand, traces the origin of the quotations to the Gospel according to the Hebrews. Others resort to the 'lost Hebrew Logia', which they suppose to form the basis of our canonical Matthew. But none of the above explanations are satisfactory. Harnack gives no evidence for his conjecture; Krawutzcky's hypothesis of the Teaching's Ebionism has long since been exploded; and, as Professor Bacon has shown, the supposed testimony to a lost Hebrew Logia gospel different from our present Matthew has no foundation in fact. The simplest explanation for all the phenomena of citation from the N. T. is that the compiler of the Teaching was familiar with both Matthew and Luke, but naturally used Matthew because of its higher authority and greater popularity. Tatian combines Matthew and Luke in the same way as the Teaching in 1:3 (See TuU, 71).

Other N. T. traces in the Teaching are numerous, but, as in the case of the O. T., they are more obscure and vague than those just referred to. We have seen how looseness and freedom of quotation characterize many of the Fathers. The difficulty in tracing sources may be also explained by the adroitness with which the compiler has done his work. While Keim and others recognize the closest kinship between the Teaching and the Fourth Gospel in their innermost sphere of thought, the parallels, especially in the eucharistic prayers, are so many and so striking that dependence upon the Fourth Gospel is not improbable. Ephesians, Hebrews, and other writings are also used, but only in the Two Ways sections common to Barnabas.

1. TuU, II, 79.

2. See esp. Expositor, 1920, "Why According to Matthew".

3. Cf. Drews, P., "Untersuchungen zur Didache", ZNTW, 1904, 53 ff. 4. Cf. p. 82. See art. by Plummer in The Churchman, July, 1884; art. on the "Didache", DAC; and Zahn, Ztschr. f. Kirchengesch, 1886, 66 ff.

A few examples of the way in which the compendist uses the N. T. sources are important for an understanding of his method:

The results of our discussion are as follows:

1. In the quotation from Malachi the Teaching does not appear to be relying upon the LXX.

2. It is simpler and more reasonable to suppose that the compiler of the Teaching is drawing from the gospels of Matthew and Luke than to suppose that he is employing some hypothetical Ur-Matthew or a lost gospel, such as the Gospel according to the Hebrews, none of whose extant fragments substantiate Krawutzcky's view, or the Gospel according to the Egyptians, which Harnack suggests as a source. The weight of cumulative evidence for the employment of the Fourth Gospel, not to mention the extra-canonical Barnabas or Hermas, makes it natural to suppose that our canonical gospels of Matthew and Luke were familiar to the compendist. When the Teaching refers to 'the gospel', there is no reason for assuming an oral gospel. The way in which the term 'gospel' is employed suggests a written rather than an oral source.

3. Abridgement, variation, and conflation characterize 1:3-2:1 as clearly as they characterize other parts of the Teaching. The

opening sentence of the manual revises the words of Barnabas after the gospel contrast of "a narrow way which leadeth to life" and "a broad way that leadeth to destruction". The next sentence is similarly derived from Barnabas, but the Teaching cuts down the precepts and adapts them again to a gospel setting and contrast. The negative form of the Golden Rule, probably derived from Acts 15:29 (the form represented in Codex Bezae), is again patterned after the Matthean positive form (cf. Robinson, 48 ff.).1 When we come to the 'interpolation', we find the writer continuing in the same manner, and following the Sermon on the Mount, as in 1:1 and 1:2d, as his source. It seems far-fetched to suppose that the passage in question is the work of a later writer when we find exactly the same phenomena here as elsewhere in the Teaching.

4. The writer of the Teaching chooses his materials with great freedom. He makes combinations, abridges sources, picks allusions from the phraseology of the gospel and adapts them to his central purpose. Nor is his process very self-conscious. On the one hand, he has so steeped himself in the language of the gospel that its diction and idiom have become second nature to him; on the other hand, the N. T. phraseology is made subservient to his own plan, so that the Scriptural passages are wrought into new combinations and subjected to a principle of arrangement suitable for a compendium.

5. The prevalence of the same sources throughout the Epistle of Barnabas, e. g. Ephesians and Hebrews, and the employment of these sources in the Teaching only in the passages paralleled by the Epistle of Barnabas makes it most probable that Barnabas is the borrower of these passages and that the Teaching is drawing these references from the Epistle. In this connection, the employment of Isa. 66: 2, so clear and comparatively exact in Barnabas, whose Isaiah citations are uniformly more exact than any others, is obscured in the Teaching by its characteristic adaptation to a gospel setting.

6. Finally, the abundant use made of the N. T. in the Teaching and the appeals to the words of the gospel stand in contrast to the Epistle of Barnabas, where the substructure is built from O. T. quotations.

1. Mt. Πάντα οὖν ὅσα ἐὰν θέλητε ἵνα ποιῶσιν ὑμῖν δι ἄνθρωποι, οὕτως καὶ ὑμεῖς ποιεῖτε ἀυτοῖς. Codex D: καὶ ὅσα μὴ θέλετε ἑαυτοῖς γείνεσθαι ἑτέρῳ μὴ ποιεῖν. Teaching: πάντα δὲ ὅσα ἐὰν θελήσῃς μὴ γίνεσθαι σοι, καὶ σὺ ἀλλῳ μὴ πόιει.

If Barnabas was acquainted with the Teaching, it is hard to understand why he should expressly omit the gospel sentences not only in 1:3b-2:1, which is disputed, but also in chapter 16. Moreover, knowing the Gospel of Matthew, does it seem likely that he would have adapted the words of Jesus (T. 1: 2) to the setting of an O. T. apocryphal writing? In view of his estimate of Jesus, the new covenant, and the new peoble, such an assumption is audacious. Teaching 3: 7, too, is in all likelihood taken from the N. T. and not from Psalms. But why doesn't Barnabas build up his writing from gospel language as the compiler of the Teaching does? The question is not difficult to answer. Like the writer of the Epistle to the Hebrews, Barnabas has set out to prove a thesis which demands the employment of the O. T. The truth, divinely communicated to him as a Christian, is actually contained in the O. T., and it is the O. T., therefore, which constitutes his source of authority. His purpose is to prove that this truth has been revealed by the coming of Jesus.

It is also possible that the ties of a former affection and loyalty played some part in his recognition of the O. T. as binding and authoritative, as it doubtless did in the case of many of his more eminent forerunners. Similarly, he may not have felt the same degree of binding attachment to records which were comparatively recent. He was himself conscious of a special revelation, and this may have prompted him to rely more on divine ovveois or yvuois than upon writings, especially in a period when not everyone was certain of their authenticity.

Chapter VI.

Jewish Character

A. The Epistle of Barnabas

The results of the foregoing chapters, especially those relating to literary character and style, have demonstrated that Barnabas unfolds his exegesis in the manner characteristic of a Jewish rabbi. Furthermore, Barnabas' discussion of the sacrifices, feasts, and Temple of the Jews, the 'old covenant' and the 'old people', the atonement rites, and circumcision evinces the fanaticism of a convert for his new-found faith and revulsion against a former allegiance. More than once, and in striking passages, does our writer depart from the reading of the LXX and follow the Hebrew. In view of these facts, instances of acquaintance with Jewish rites and practices would naturally be sought, particularly in a letter which seeks to point out the vanity of all the institutions central to Judaism. It is only to be expected that one familiar with the methods of the synagogue should bear witness to its rites and practices.

There has been some disagreement regarding the accuracy of Barnabas' information on Jewish subjects in Part I, but it is noteworthy that those most familiar with Jewish rites have been the best supporters of the trustworthiness of his descriptions. The great Jewish rabbi Güdemann believes that the writer was a converted Jew, and Kohler

1

1. Religionsgeschichtliche Studien, "Zur Erklärung des Barnabasbriefes", 109 ff. He sums up his conclusions in three statements:

den biblischen

I. Die Vertrautheit B.'s mit den jüdischen Opfervorschriften und traditionellen ist eine so eingehende, daß man ihn hiernach nur für einen geborenen Juden und zwar für einen besser unterrichteten, als er scheinen will, halten kann. 104.

II. B. kennt nicht bloß die Hagada und den Urtext, sondern er weiß mit beiden in so selbständiger Weise umzugehen und sie für seine Zwecke auszubeuten, daß dieses Moment für seine jüdische Abstammung und Erziehung entscheidet. 119 III. Für die jüdische Abstammung B.'s zeugt auch seine biblische und zuweilen selbst rabbinische Redenweise. 129.