Considerations of Method: A Response to Ezra Fleischer

With his response, Prof. Fleischer himself sketches out the central argument of his series of articles. With this, he furthers one of the primary factors motivating my original review essay: to provide a summary of this significant discussion of early Jewish liturgy for those many students of the field who are unable themselves to read Fleischer's elegant, poetic Hebrew prose. We should all be grateful. As he acknowledges, it was obviously impossible in reviewing a book's worth of articles to present any one argument at the level of detail allowed by Prof. Fleischer's more specific focus here. It is also apparent to me, having studied Prof. Fleischer's briefer public and lengthy private responses, that his articles (which now include "The Kedushah of the 'Amidah [and other Kedushot]: Historical, Liturgical, and Ideological Aspects," Tarbiẓ 67, no. 3 [1998]: 301-50) should be reworked into a book. A decade of dialogue over these issues has brought him to the clearer articulation of his methodology evident in his response.

Most of the points that Prof. Fleischer raises here already receive adequate response in my original review, and my interpretations need not be reiterated. The critical point on which he and I do indeed disagree is not whether there was a revolutionary call for the institution of obligatory verbal worship of God at Yavneh that (gradually, I think) found a home in the preexistent institution of the synagogue, but rather whether it is demonstrable (or even plausible) that this revolution included the dictation of a precisely defined prayer text. Prof. Fleischer's argument is attractive; it certainly fits the eventual state of Jewish prayer and the way that Jews of the past millennium (who knew only authoritatively fixed prayer) have assumed that they should read the rabbinic sources. But do we have reliable evidence for such a text?

Rabbinic traditions after Yavneh were oral for generations-many of them for centuries. If we know that medieval scribes did not hesitate to insert their own liturgical texts into received and "authoritative" manuscripts, is there any reason to suppose that talmudic and midrashic texts in the period of their formulation were any less fluid, that those repeating these texts did not either deliberately or accidentally insert their own preferred formulations? The presence of recognizable liturgical texts in received rabbinic literature teaches only two things. First, these were, at the earliest, the active prayers at the point of the redaction of the larger text, hundreds of years after Yavneh. Second, our familiar prayers often match those of the Talmud because they were shaped by it. As, under the geonim and rishonim, the Babylonian Talmud gained its status as the authoritative and holy embodiment of the Oral Law, local liturgical customs were brought into gradual conformity with it. The best-documented example of this, studied extensively by Prof. Fleischer and others, is the Palestinian rite. Thus our earliest rabbinic prayer texts of any historical value date from half a millennium and more after Yavneh! Even in the following five hundred years, we have only very limited evidence for how Jews prayed. That evidence, coming primarily from the Cairo geniza, still displays enough diversity (and a diversity that dwindles as Babylonian influence grows) that it led Fleischer's predecessors to question the validity of positing an Urtext for Jewish prayer. We simply cannot know with any certainty what was decreed at Yavneh or how Jews fulfilled its decrees in late antiquity.

We can give much more credibility to the records of deviance than to transmissions of what does become authoritative text because the person transmitting the tradition would not have had a familiar liturgy to insert. These are among the rabbinic traditions that Prof. Fleischer de-emphasizes in his studies. The stories of prayer leaders erring in the opening blessing of the 'amidah (and the rabbinic objections to their texts) indicate that it was not obvious to these men that their texts were fixed! (B. Berakhot 33b; B. Megillah 25a; P. Berakhot 9:1, 12d; Midrash Tehillim 19:2). Not a single text suggests that these were incidents of deliberate challenges to...