^{page 174 note 2} See the writer's The New Testament and Rabbinic Judaism (1956), pp. 158ff., 183ff., 278ff., 412ff.Google Scholar

^{page 175 note 1} The merely annual character of Passover was no hindrance: in those times, the liturgy would not be forgotten about during the year, to be rediscovered, as a surprise, when the festival came round again.Google Scholar

^{page 175 note 2} See Goldschmidt, E. D., Die Pessach-Haggada (1936), pp. 68f. Not all his emendations, however, seem inevitable.Google Scholar

^{page 176 note 1} Nearly everything controversial. Finkelstein, L., Harv. Theol. Rev. XXXI (1938), 291 ff.; XXXV (1942), 291 ff.; XXXV, (1943), 1ff.Google Scholar, favours very early datings, Stein, S., J.J.S. VIII (1957), 15, very late ones. While some of the former's conclusions are precarious, especially when he tries to be too precise, the latter's argument is from silence—extremely weak. Does Philo give a detailed account of the philosophical importance of, say, baptismal catechisms or the Eighteen Benedictions? Stein would leave far too little time for the complicated developments which must have preceded what we find in Mishnah, Tosephta, Mekhilta and Siphre. A great deal, of course, depends on what we understand by a ‘literary’ fixing of the liturgy, which he makes late. One can agree that in New Testament times there was no written liturgy, and not even a single authoritative oral one, without denying that certain orders of service had become traditional—identical only as to a few essentials, otherwise different in different regions and groups, with constant interchange between them and much fluidity even within each. Stein himself admits that ‘Kiddush, Hallel and some very elementary questions and answers in connection with the rites of the festival’ were established long before the New Testament period. In view of the evidence of the gospels, it would be impossible to concede less: but does Finkelstein assume so much more? Incidentally, Stein's principal thesis, that the Passover eve liturgy is indebted to the Symposia literature, gains in plausibility—in fact, is plausible only—if that influence can have been exerted in pre-Christian centuries.Google Scholar

^{page 176 note 2} See Strack, L. and Billerbeck, P., Kommentar zum Neuen Testament aus Talmud and Midrasch, IV (1928), pt. 1, p. 61.Google Scholar

^{page 177 note 1} See the important article by Petuchowski, J. J., J.B.L. LXXVI (1957), 293ff.;Google ScholarGoldschmidt, pp. 21, 70f.;Google ScholarStrack and Billerbeck, pp. 70f.Google Scholar The final compromise—reached between the second and fourth century A.D.—is first to take bread and herbs separately (the Passover having dropped out) and with blessings, and then to take them together, according to Hillel, but without blessings. No doubt in Hillel's circle the Passover, bread and herbs had been taken only in his way, and with blessings, as a memorial rite. At that stage, the designation of the rite must have been zekher lehillel, ‘in remembrance of Hillel’, and the formula must have run as quoted in the text. Though this mode spread, however, the traditional mode—the separate taking—remained prevalent and came to be regarded as more correct by the majority of scholars. How early they arrived at this decision is not clear from Bab. Pesahim 115a; hardly in Hillel's lifetime or even before the destruction of the Temple. At any rate, from the destruction, there was a strong, new consideration against Hillel: the bitter herbs—which the Bible mentions only in connexion with the Passover, now gone–assumed the character of a mere rabbinic duty, whereas the unleavened bread–which the Bible mentions also independently of the Passover–continued as a scriptural duty. Hence, in the final compromise, the separate taking plays the major role. Hence also, the title of the memorial rite is expanded into zekher lammiqdash kehillel, ‘in remembrance of the Temple according to Hillel’: strangely involved, but the reference to the Temple is considered necessary to make it clear that it was only then that the combined taking was a proper fulfilment of the Law, all three duties being still scriptural. Similarly, a few words are added to the opening of the formula: ‘Thus did Hillel— in the time when the Temple was standing.’ (The same addition—pace Stein, p. 42—in an interpretation of the Passover going back to Gamaliel I: Goldschmidt, pp. 63f.) One more remark. The fact that Hillel, in the usual manner, invokes a verse for his peculiar way of taking the three foods does not mean that he did not choose that way because it had a certain significance. Exactly what the combined taking signified it would lead too far afield here to inquire.

^{page 178 note 1} See The New Testament and Rabbinic Judaism, pp. 326f.Google ScholarGoldschmidt, pp. 54f.Google Scholar

^{page 178 note 2} Cf. Petuchowski, pp. 259f., who says that there must be no human protagonist—God himself being the hero of this drama. In Finkeistein's opinion, XXXV, 329; XXXVI, 6, Moses is left out because the high priests were against exalting a non-priestly Levite (and Aaron is left out as a non-Zadokite); this seems far-fetched. We think it probable that versions mentioning Moses once existed but fell victim to the trend which culminated in the substitution (already referred to) of ‘not through the messenger’ for ‘not through a messenger’. Many questions of detail are unsettled. Finkeistein, XXXVI, 3, draws attention to a eulogy recounting a series of benefits: according to the Haggadah they were conferred on Israel by God (maqom, Place), according to Midrash Siphre on Deuteronomy they—or most of them—were conferred by Moses. Haggadah (Goldschmidt, pp. 60ff.): ‘God who brought us forth from Egypt and executed judgment against them…and gave us the manna to eat’, Siphre: ‘Moses, the man that brought us forth from Egypt…and brought us down the manna.’ The Midrash, Finkeistein infers, protests against the high-priestly, anti-Mosaic group responsible for the Haggadah. But he himself points out, pp. 4f., that the eulogy is based on biblical forerunners like Neh. ix. 6ff.; Ps. CXXXVF. As in these God is the benefactor, surely the natural conclusion is that the Haggadah simply follows them, without any political bias; and that the Midrash represents a sort of popular variety, in which ‘God brought us forth’ would readily become ‘Moses brought us forth’. That no such popular variety is preserved in our Haggadah is, indeed, significant; it is due to the gradual elimination of anything smacking of a godlike Messiah. What place, if any, is to be assigned to John vi. 32 in all this (‘Moses gave you not that bread from heaven, but my Father giveth you the true bread from heaven’) we shall not here try to decide. Finkelstein, incidentally, XXXVI, 35f., appears to regard the two versions of the eulogy given in the Haggadah as equally old. There is something to be said, perhaps, for the prose version being earlier than the poem. It is the last words of the former—‘to atone for all our transgressions’—which speak decisively for a date prior to the destruction of the Temple.Google Scholar

^{page 178 note 3} See The New Testament and Rabbinic Judaism, pp. 192ff.Google Scholar

^{page 178 note 4} See Goldschmidt, p. 37.Google Scholar

^{page 178 note 5} Mishnah Pesahim X. 4;Google Scholar see The New Testament and Rabbinic Judaism, pp. 413ff.Google Scholar

^{page 179 note 1} See The New Testament.and Rabbinic Judaism, loc. cit. and pp. 279 ff.Google Scholar

^{page 179 note 2} vi. 21 (roughly);Google Scholar see Goldschmidt, pp. 36f.Google Scholar

^{page 179 note 3} xxvi. 5ff.;Google Scholar see Goldschmidt, pp. 46ff.Google Scholar

^{page 179 note 4} The biblical text reads 'obedh, ‘A Syrian ready to perish was my father’. The Haggadah, however, understands the word as if it were vocalized 'ibbedh. This daring reinterpretation, Finkelstein rightly holds, XXXI, 300, must date either from the Maccabean era or from an even earlier one, when Palestine was under Egyptian suzerainty and it was advisable to denounce Syria. He argues in favour of the latter alternative; we much prefer the former.Google Scholar

^{page 179 note 5} The last verse, ‘and he has brought us into this place…’, was removed after the destruction of the Temple; see Goldschmidt, p. 9.Google Scholar

^{page 179 note 6} See Goldschmidt, pp. 45f.Google Scholar

^{page 179 note 7} xxiv. 2ff.Google Scholar

^{page 180 note 1} On some points concerning a change of master, see The New Testament and Rabbinic Judaism, pp. 278ff. The Matthean Passion-narrative certainly ends with glory, xxviii. 18: ‘All power is given unto me’; in the others the pattern is perhaps a little obscured by the details of the Christophanies. The provenance of some speeches in Acts—vii. 1 ff. and the like–is worth investigating.Google Scholar

^{page 180 note 2} See The New Testament and Rabbinic Judaism, pp. 158ff.Google Scholar

^{page 180 note 3} cx (LXX: cix). 1;Google ScholarDavid's son, for example in Isa. xi. 1, 10;Google ScholarJer. xxiii. 5.Google Scholar

^{page 180 note 4} Joshua ben Hananiah, Bab. Niddah 69b ff.Google Scholar

^{page 180 note 5} The famous Midrash of the four sons; Goldschmidt, pp. 39ff. Commentators, not realizing the classification on which the Midrash is based, have had recourse to artificial explanations and critical remedies; see, for example, Goldschmidt's reason, p. 44, why the reply the Haggadah has for the wise son is more suitable for and must be transferred to the simple. Finkelstein, XXXVI, 8ff., is driven to postulate a highly composite character. Even if this were correct, our thesis would stand: the composition would have been undertaken (in its final stages at least) with a view to achieving a whole in conformity with the classification. But more probably the Midrash is not composite at all. (In trying to make out his case, Finkelstein introduces several arbitrary assumptions, such as that a Midrash, when it employs a question put in the Bible—say, Exod. xii. 26, ‘What mean ye by this service?’—must also employ the reply provided there—in this case, ‘It is the sacrifice’ and so on; and that if it does not do so, we have proof of subsequent interference. Goldschmidt, pp. 41, 43, approaches the same position. But this is to allow the Midrash—certainly the early Midrash—far too little freedom.) Stein, pp. 35f., draws attention to Philo's four types of children: that obedient to both father, reason, and mother, worldly education, that obedient to the father only (Philo sometimes seems to regard the former as the best, sometimes the latter), that obedient to the mother only and (the worst) that obedient to neither. But Stein himself recognizes that there could only be the faintest alliance with the four of the Haggadah: the wise, the wicked, the simply pious and the one who does not know how to ask. Perhaps a fourfold division was generally popular. Perhaps some division like Philo's, current before, had encouraged the author of the Midrash to create one of his down.Google Scholar

^{page 181 note 1} In the anecdote of R. Joshua the sequence is wisdom, interpretation, vulgarity, way of the land. Even in the version of the Passover eve Midrash offered by the Mekhilta the second and third questions are switched round—and, incidentally, both in this version and that of the Palestinian Talmud (Pesahim 37d) the son of simple piety is ununderstandingly turned into a half-wit.Google Scholar

^{page 181 note 2} Exod. xii. 26f., xiii. 8, xiii. 14;Google ScholarDeut. vi. 20f.Google Scholar

^{page 181 note 3} Exod. xiii. 8. In The New Testament and Rabbinic Judaism, p. 165, I thought that the author of the Midrash, forced by the biblical material on which he worked, had to sacrifice the category of contradiction and substitute the person who cannot ask. It now seems to me that, while introducing the person who cannot ask—because Exod. xiii. 8, in contrast with the other three passages, contains no question—he managed to retain contradiction. The teaching of the verse, ‘Because of that which the Lord did unto me when I came forth out of Egypt’, is one of the most difficult ones (from the rabbinic point of view) when confronted with other references to this event. It speaks of ‘me’, whereas elsewhere we find ‘us’ (for example, Deut. xxvi. 8) or even ‘your fathers’ (for example, Josh. xxiv. 6). The careful formulation of the ‘redemption’ quoted above (Goldschmidt, p. 69) is significant: ‘Blessed art thou, who redeemed us and redeemed our fathers from Egypt.’ The relation between ‘I’ and ‘we’ and ‘our fathers’ and indeed ‘our children’ is dealt with in many sections of the Haggadah; cf. for example the rationalizing solution, ‘and if the Holy one had not brought out our fathers, we and our children and their children would still be enslaved to Pharaoh’ (Goldschmidt, pp. 36f.). No doubt Exod. xiii. 8 was a familiar example of apparent contradiction, to be solved by assigning to each statement its appropriate domain.Google Scholar

^{page 182 note 1} Professor Kingsley Barrett kindly suggests to me that the figure of the disciple who does not ask, yet receives the appropriate instruction, may have been made use of by John: xvi. 5, 30.Google Scholar

^{page 183 note 1} See Goldschmidt, pp. 39f.Google Scholar

^{page 184 note 1} Unless we count the fruit of the vine.Google Scholar

^{page 184 note 2} See The New Testament and Rabbinic Judaism, pp. 189ff.Google Scholar

^{page 184 note 3} See Stauffer, E., Jesus (1957), pp. 36ff., 153f.Google Scholar

^{page 184 note 4} Josephus, Ant. 2. 9. 2. 205;Google ScholarRabba, Exodus on 1. 22.Google Scholar

^{page 184 note 5} What has contributed to the misinterpretation is the pericope Matt. ii. 19ff., which does allude to the episode of Moses. But it has not the same history as the report of the Massacre.Google Scholar

^{page 184 note 6} xi. 1.Google Scholar

^{page 184 note 7} See Goldschmidt, pp. 47f.Google Scholar

^{page 184 note 8} For example, Rabba, Genesis on xxv. 20.Google Scholar

^{page 184 note 9} Jer. Targum on Num. xxii. 5. Balaam occurs also in the farewell speech of Josh. xxiv. 9, which is part of the third scheme noted above. In our present Haggadah, only vv. 2–4 are quoted, but the full scheme requires the continuation at least as far as v. 13 (Goldschmidt, p. 11 n. 2).Google Scholar

^{page 185 note 1} Gen. xlvi. 4.Google Scholar

^{page 185 note 2} See, for example, Rabba, Exodus on iii. 12.Google Scholar

^{page 185 note 3} For example, Josephus, Ant. 14. 11. 2. 274, 15. 2. 3. 20f.Google Scholar

^{page 185 note 4} Luke xiii. 32.Google Scholar

^{page 185 note 5} ‘rmi—’;dmi: ℸ—ℸ.Google Scholar

^{page 185 note 6} xxxi (LXX: xxxviii). 15.Google Scholar

^{page 186 note 1} As most components of the Passover eve liturgy are mentioned also in sources other than the Haggadah—Mishnah, Mekhilta and so forth—it might be asked whether the gospel parallels may not derive from the general pool of rabbinic material rather than directly from the service. Suffice it to say, as for the Markan questions, that not only is their dramatization by an evangelist on the basis of an ordinary Midrash improbable in the extreme, but—as indicated above (p. 181 n. 1)—the versions of the Midrash of the four sons to be found in Mekhilta and Palestinian Talmud are both of them less close to the gospels than is the Haggadah; and as for the Massacre, that the full story of Laban and Jacob is given exclusively in the Haggadah—Siphre on Deuteronomy omits the comparison with Pharaoh, and even in the portion it does not omit notes only deviations, taking it for granted that the reader is familiar with the Passover eve exposition (see Goldschmidt, p. 11).Google Scholar

^{page 186 note 2} Our Haggadah dwells on the ten plagues; Goldschmidt, pp. 57ff. Other versions presumably related ten miracles: cf. Mishnah Aboth v. 4.Google Scholar