^{Page 83 note 1} ‘Participle and Imperative in I Peter’, in The First Epistle of St Peter, by Selwyn, E. G. (London, Macmillan, 1946), pp. 467–88Google Scholar; The New Testament and Rabbinic Judaism (London: Athlone Press: 1956), pp. 90–7Google Scholar.

^{Page 83 note 2} Moulton, J. H., A Grammar of New Testament Greek, I (3rd ed.Edinburgh: T. & T. Clark, reprinted 1957), pp. 180 ff., 232 ff.Google Scholar Daube's criticisms of Moulton'sposition include: (I) Moulton's examples are from the concluding parts of letters; (2) they do not have a ‘rule’ character; (3) no Hellenistic Haustafein contain such imperative participles. Meecham, H. G., ‘The Use of the Participle for the Imperative in the New Testament’, Exp. T. LVIII (1947), 207–8,Google Scholar came to Moulton's defence, offering two new examples from the papyri, and reaffirming that the imperative participle is a genuine, if somewhat rare, development in Hellenistic Greek. Barrett, C. K., ‘The Imperative Participle’, Exp. T. LIX (1948), 165–6,Google Scholar came to Daube's defence. Meecham's two new examples, when examined in thi contexts, support Daube. The strongest evidence for Moulton's position is presented by Mayser (Grammatik der griechischen Papyri, II, 196 n. 3, and 340 f.Google Scholar). His examples are not epistolary formulae. Nevertheless, Mayser shows that they cannot prove an established Hellenistic usage. Anacoluthon is the explanation. Barrett concludes: ‘Even if the examples from the Greek usage were far better than they are, those drawn by Dr Daube from Hebrew would be a hundred times better still. The construction in question can be produced at best in a few Greek papyrus letters, which are certainly exceptional; but in the Mishna it is regular’ (p. 166).

^{Page 83 note 3} Barrett, C. K., The Epistle to the Romans (New York: Harper, 1957), pp. 239–40.Google Scholar

^{Page 83 note 4} Cranfield, C. E. B., A Commentary on Romans 12–13 (Edinburgh: Oliver & Boyd, 1965), p. 40 n. 3.Google Scholar

^{Page 83 note 5} Daube's thesis has not received the attention it deserves. While Moule, C. F. D., An Idiom-Boon of New Testament Greek (Cambridge University Press, 1953), p. 180,Google Scholar concludes that ‘a strong case can be made for tracing at least some of these participles to Semitic influence’, N. Turner in vol. III, of the Moulton-Howard Grammar (1963), p. 343, continues the tendency of speaking of the imperative participle as ‘common in the Koine’ and Blass, F., Debrunner, A., Funk, R. W., A Greek Grammar of the New Testament (University of Chicago Press, 1961), p. 245,Google Scholar show no knowledge of Daube's work. It has been in Great Britain that Daube has received a hearing. Selwyn, by virtue of the inclusion of Daube's essay in his commentary on I Peter, approves; Dodd, C. H., Gospel and Law (New York: Columbia University Press, 1951), p. 19 n. 3,Google Scholar approves; Davies, W. D., Paul and Rabbinic Judaism (2nd ed.London: S.P.C.K. 1955), pp. 130 ff., 329,Google Scholar apparently approves; Hunter, A. M., Paul and his Predecessors (2nd ed.Philadelphia: Westminster, 1961), p. 131 n. I,Google Scholar approves. Within a limited survey I found no American or German references to Daube's work.

^{Page 84 note 1} See below for a full discussion.

^{Page 84 note 2} Rom. xii. 9b/I Thess. V. 21 b-22; Rom. xii. 10a/I Thess. iv. 9; Rom. xii. 11b/I Thess. v. 19; Rom. xii. 12a/I Thess. v. 16; Rom. xii. 12c/I Thess. V. 17; Rom. xii. 17a/I Thess. V. 15; Rom. xii. 18/I Thess. v. 13.

^{Page 84 note 3} Col. iii. 12 ff., iv. 2ff.; Eph. iv. 2 ff., v. 21, 22.

^{Page 84 note 4} Cf. Davies, W. D., Paul and Rabbinic Judaism, pp. 123–8Google Scholar; Hunter, op. cit. pp. 129–30.

^{Page 84 note 5} Such a suggestion is in line with the statement of Conzelmann, H., ‘Paulus und die Weisheit’, N. T.S. XII (1966), 232,Google Scholar that Romans is largely commentary on traditional material. Cf. also the remarks of Reumann, J., ‘The Gospel of the Righteousness of God’, Int. XX (1966), 434.Google Scholar

^{Page 84 note 6} Michel, O., Der Brief an die Römer (Göttingen: Vandenhoeck & Ruprecht, 1961), p. 302, also n. 3.Google Scholar

^{Page 85 note 1} For a summary of the arguments on the textual problem of v. II b see Crafield, op. cit. pp. 43 ff. We agree with Cranfield and follow the reading in Nestle. Either way our argument is unaffected.

^{Page 85 note 2} Charles, R. H., The Greek Versions of the Testaments of the Twelve Patriarchs (Oxford University Press, reprint 1960), p. 225.Google Scholar

^{Page 85 note 3} Charles, R. H., Apocrypha and Pseudepigrapha of the Old Testament (Oxford: Clarendon Press, reprint 1963), II, 696–7.Google Scholar

^{Page 85 note 4} Note that in the Didache i. 2 there is a general statement of the Way of Life: ‘The Way of Life is this…’ This is followed in i. 3 ff. by the explanation: ‘Now, the teaching of these words is…’ It was early Christian practice also to give a general statement in a Two Ways formula and then follow it with specifics.

^{Page 85 note 5} Cranfield, op. cit. p. 39, correctly says that it is a mistake to connect v. 9b closely ith v. 9a. It should not be taken as the explanation of or the complement to the idea expressed by ανυπóκριτоς.

^{Page 85 note 6} As Barrett, , Romans, p. 239,Google Scholar recognizes.

^{Page 85 note 7} Kbsemann, E., ‘Gottesdienst im Alitag der Welt (zu Rm 12)’, in Judentum, Christentum, Kirche, ed. Eltester, W. (Berlin: Toepelmann, 1964), p. 171Google Scholar; Schmidt, H. W., Der Brief des Paulus an die Römer (Berlin: Evangelische Verlaganstalt, 1962), p. 213Google Scholar; Knox, J., I.B. IX, 586–7Google Scholar; Sanday, W. and Headlam, A. C., The Epistle to the Romans (Edinburgh: T. & T. Clark, reprint 1955), p. 360.Google Scholar

^{Page 85 note 8} So Manson, T. W., Studies in the Gospels and Epistles, ed. Black, M. (Philadelphia: Westminster, 1962), p. 241.Google Scholar Approved by Munck, J., Christ and Israel (Philadelphia: Fortress, 1967), p. 7,Google Scholar and Bornkamm, G., ‘The Letter to the Romans as Paul's Last Will and Testament’, Aus. Bib. Rev. XI (1963), 2–14.Google Scholar

^{Page 86 note 1} Cranfield, op. cit. p. 49; Barrett, , Romans, p. 241.Google Scholar

^{Page 86 note 2} Barrett, , Romans, p. 241.Google Scholar

^{Page 86 note 3} Kirk, K. E., The Vision of God (2nd ed.London: Longman s, Green & Co.), pp. 118 ff.Google Scholar; Barnard, L. W., Studies in the Apostolic Fathers and their Background (New York: Schocken Books, 1966), pp. 93–4.Google Scholar Barnard calls attention especially to I QS iii. 13-iv. 26.

^{Page 86 note 4} Cf. Grundmann, W., ‘άγαθóζ’, Theological Dictionary of the New Testament (Grand Rapids: Eerdmans, 1964), I, 15–16.Google Scholar

^{Page 86 note 5} Seen. p. 84 n. 2. One should also note Rom. xii. 10b/ Rom. xii. 3, Phil. ii. 3 Rom. xii. 18/II Cor. xiii. 11; Rom. xii. 12c/ Col. iv. 2a.

^{Page 87 note 1} Michel, op. cit. p. 305; Selby, D. J., Toward the Understanding of St Paul (Englewood Cliffs, N.J.: Prentice.Hall, 1962),Google Scholar Appendix II, ‘Paul and the Synoptic Tradition’, p. 340.

^{Page 87 note 2} Selby, op. cit. p. 340; Barrett, , Romans, p. 242.Google Scholar

^{Page 87 note 3} Selby, op. cit. p. 340; Michel, op. cit. pp. 308 f.

^{Page 87 note 4} Selby, op. cit. p. 340.

^{Page 87 note 5} Verses 9 b-13 are Christian if πνεμαƬƖ in u. I I b is a reference to God's Spirit, as seems likely.

^{Page 87 note 6} Daube, , New Testament and Rabbinic Judaism, p. 91.Google Scholar

^{Page 87 note 7} Cf. Bornkamm, G., Jesus of. Wazareth (New York: Harper, 1960), pp. 98–9Google Scholar; Davies, W. D., ‘Ethics in the New Testament’, I.D.B. II, 174–5.Google Scholar

^{Page 87 note 8} Kraft, R. A., The Apostolic Fathers, III. Barnabas and Didache (New York, Nelson, 1965), 187,Google Scholar points Out that the Didache echoes any number of sayings ofJesus in its section on the Two Ways. (1) Matt. v. 5/Did. iii.7; (2) Matt. v. 25–6/Did. i. 5c; (3) Matt. v. 38–48/Did. i. 3b-5a; (4) Matt. vii. 12/Did. i.2 c; (5) Mk. xii. 30-I/Did. i. 2a, b; (6) Lk. vi. 27–35/Did. i. 3b5a. It is significant that these sayings are echoed not only in the section, i. 3b-ii. 1, which Kraft (p. 7) regards as a late, Christian addition to the basic Two Ways tradition but also in iii. 7, which presumably is early.

^{Page 88 note 1} In Selwyn, op. cit. pp. 473 f.

^{Page 88 note 2} Sanday and Headlam, op. cit. p. 365.

^{Page 88 note 3} Ellis, E. E., Paul's Use of the Old Testament (Grand Rapids: Eerdmans, 1957), pp. 107 ff.Google Scholar

^{Page 88 note 4} ibid. pp. 160, 170, 172, 174, 180, 182.

^{Page 88 note 5} See a similar remark about the situation in Acts in Lindars, B., New Testament Apologetic (Philadelphia: Westminster, 1967), pp. 35 f.Google Scholar

^{Page 88 note 6} Cf. Bearcislee, W. A., ‘The Wisdom Tradition and the Synoptic Gospels’, J.A.A.R. XXXV (1967), 233Google Scholar; Robinson, J. M., ‘ΛΟ⌈Ο│ ΣΟϕωΝ: Zur Gattung der Spruchquelle Q’, in Zeit, und Geschichte, , ed. Dinkler, E. (Tubingen: J. C. B. Mohr, 1964), pp. 92–3.Google Scholar

^{Page 89 note 1} Ellis, op. cit. pp. 48–9.

^{Page 89 note 2} See below.

^{Page 89 note 3} Brown, F., Driver, S. R., Briggs, C. A., A Hebrew and English Lexicon of the Old Testament (Oxford: Clarendon Press, 1957), p. 922.Google Scholar

^{Page 90 note 1} Two objections could be raised in addition to these four. (I) The form of the commands in vv. 10–13, it could be objected, is different from that in vv. 14 ff. Two types of response can be made to such an objection. First, there is no necessity to say that vv. 10–13 and vv. 14 ff., if they do go back to a Semitic source(s), are from the same source. The two sections could just as easily represent fragments of different codes. Second, in early Christian ethical codes a group of injunctions with one form may be set beside a group of rules with quite a different form. Cf. the Didache i-v. (2) It could also be objected that some of these injunctions have Hellenistic parallels. This is sometimes suggested for vv. 15, 18, 19 a. In each case, however, a careful reading of the context in which the alleged parallel occurs reveals it to be no parallel at all. See n. a below, p. 91 nn. 3, 4.

^{Page 90 note 2} See below, p. 93 n. 5. The parallel with Epictetus II. v. 23, however, is too loose to be significant.

^{Page 90 note 3} It is important to note that v. 21 is not excluded by our criterion. This is sobecause, outside of the fact that it has no imperative participle, no other serious objection can be raised against it. Moreover, if Test. XII Patr. can still be regarded as a pre-Christian Jewish source with Hebrew roots, as I think it can, then the fact that v. 21 has a parallel in Test. Benj. iv. 3 supports our refusal to eliminate v. 21 merely because it does not have an imperative participle. (κάν μή βούλωνται περί αύτοũ εις καλά οτος τò άηαθòν пοιν νικã τò κακòν: Charles, R. H., Greek Versions, p. 219.Google Scholar) Cf. also Test. Jos. xviii. 2.

^{Page 91 note 1} With v. 16b cf. Pirke Aboth i. 10, a saying attributed to Shemaiah (from whom Shammai and Hillel received the Law). Also, cf. Pirke Aboth vixsxs. 5. Cheyne, T. K., ‘The Rendering of Rom. xii. 16’, Exp. Second Series VI (1883), 469–72,Google Scholar argues that the difficulty in rendering συναпάηομαι is due to a mistranslation from the Hebrew. If he is correct, this may add further evidence to Daube's thesis.

^{Page 91 note 2} With v. 17 cf. I Pet. iii. 9, where an imperative participle is also used.

^{Page 91 note 3} With v. 18 cf Aboth. i. 12, a saying of Hillel. The supposed parallel with Epictetus iv. v. 24 is really no parallel at all when it is read in Context. It poses no problem to a theory of a Semitic source.

^{Page 91 note 4} With v. 19a cf. Test. Gad vi. 7; IQS x. 18; C.D. ix. 2–5; Sirach xxviii. 1–7. The supposed parallel with Seneca, Dc Ira, III. 12. 39, is really no parallel at all. It poses no problem for a Semitic source. Cf. Clemcn, C., Primitive Christianity and its Juan-Jewish Sources (Edinburgh, T. & T. Clark, 1912), p. 72.Google Scholar

^{Page 91 note 5} With v. 21 cf. Test. Benj. iv. 3 and Test. Jos. xviii. 2. The echoes of Jesus' sayings in vu. 17 a 18, and 19 a point to a Christian context for the code.

^{Page 91 note 6} The question of word usage must be raised. Is the language in the proposed fragment non-Pauline? Two words (συναпάηεσθαι éκδιετν) occur only twice in Paul. (So Morgenthaler, R., Statistik des Neutestamentlichen Wortschatzes, Zürich: Gotthelf Verlag, 1958, pp. 145, 92.Google Scholar) Otherwise the language is found elsewhere in Pauls' letters. This can be explained by two factors: (I) it would have been almost impossible for Paul to avoid using the terms found here; (2) Paul uses parallel injunctions elsewhere in his letters, so it is possible that he is using traditional ethical material repeatedly.

^{Page 91 note 7} Because it does not follow the LXX text.

^{Page 92 note 1} Filson, F. V., ‘The Christian Teacher in the First Century’, J.B.L. LX (1941), 326,Google Scholar says: ‘The theory of a rigid, severely controlled oral tradition is too mechanical to meet the requirements of a Spirit-guided teaching group. On the other hand, the lack of a rigid fixity should not be pressed to such a degree as to hold that there was no framework at all.’.

^{Page 92 note 2} Davies, Paul and Rabbinic Judaism, pp. 123–8Google Scholar; Hunter, op. cit. pp. 129–30.

^{Page 92 note 3} Barnard, op. cit. p. 99.

^{Page 92 note 4} Post-Pauline circles remembered Paul as a teacher. Cf. II Tim. i. ii; Acts xiii. I.

^{Page 92 note 5} Taylor, V., The Gospel According to St Mark (London: Macmillan, 1955), pp. 408–9.Google Scholar For Jewish examples, cf. Pirke Aboth v. 1–6, 7–8, 10–15, etc.

^{Page 92 note 6} ibid. pp. 93 ff., 249.

^{Page 92 note 7} ibid. pp. 91–2. Cf. Pirke Aboth iii. 2–3, etc.

^{Page 92 note 8} Michel, op. cit. p. 305. Ashe, R. P., ‘Rom. 12: 13–14’, Exp. T. XXXIX (1927–1928), 46,Google Scholar because of the link-word tries to connect the two rules into one thought. A knowledge of the link-word principle in late Judaism would have prevented his reading too much into what is only a formal connexion.

^{Page 93 note 1} See n. 4 below.

^{Page 93 note 2} Michel, op. cit. p. 306.

^{Page 93 note 3} ibid. p. 308.

^{Page 93 note 4} Verse 15 does not seem to be integral to the finished structure. It was probably attached to v. 14 before the final redaction took place. If so, this would explain both why it is there and why it is irrelevant in the finished structure. The phrase at the beginning of v. 18 (eἰ δυνατν) could have been a part of the injunction in the Semitic code or it could have been added by the redactor. No certain decision is possible.

^{Page 93 note 5} The final form of vv. 14–21 could, of course, have been given to it by some Hellenistic Christian before Paul, but there is no need to posit another hand. Since the introductory formulae are Pauline the probabilities are that Paul was the final redactor.

^{Page 93 note 6} Turner, N., Grammatical Insights into the New Testament (Edinburgh: T. & T. Clark, 1965), pp. 165–8,Google Scholar gives four reasons for not accepting Daube's thesis. We are now in a position to respond to them. (1) In the German edition of Moulton's work (1911), Daube would have found an eighth example. Turner, himself, however, says that Mayser (op. cit. II, 196 n.3) had already rejected this example, just as he had the other seven. Moreover, this example does not possess ‘rule’ character any more than the other seven and so is irrelevant. (2) The date of Daube's Tannaitic examples may well be later than the apostolic period. However, because of the conservative nature of the oral tradition, if a saying is Tannaitic it is generally regarded as legitimate to use it in New Testament studies. Moreover, Daube's case does not rest on the specific Tannaitic sayings which have parallels in Rom. xii, but in the claim that the imperative participle was typical of the Tannaitic period even as early as the begnning of the Christian era. (3) C. L. Mitton has shown conclusively (!) that the author of I Pet, is indebted to Ephesians in general, and in particular that I Pet. ii. 18 (which has the imperative participle) depends upon Eph. vi. 5 (which has the same injunction but with a straightforward imperative). However, , Kümmel, W. G., Introduction to the New Testament (Nashville: Abingdon, 1966), p. 297,Google Scholar says: ‘The repeatedly advocated supposition of a literary dependence of I Peter upon Romans (and Ephesians) (McNeil–Williams, Beare, Feine–Behm, Mitton) is, of course, improbable, because the linguistic parallels can be explained out of a common catechetical tradition (see the proof by Selwyn, 365 ff.).’ (4) Biblical Greek may already have absorbed the imperatival participle into its system, as it had absorbed numerous ther Semitisms, and so would have been in the air, rendering a Semitic source an unnecessary hypothesis. However, if the imperatival participle was a part of spoken Jewish Greek in the first century C.E. we have no way of knowing it, apart from Turners supposition. Moreover, the fact that the peeling away of what appears to be Hellenistic material in Roin. XII. 14–21 results in a two-strophe code (see above) argues for precisely the Semitic source Turner decries.