Intertextuality and the Reading of Midrash (18 page)

I have heard the murmurings of the children of Israel. Say to them, at evening
[16:12]: This discourse has already been said by Moses to Israel [above in verse 8], but it is repeated because He said, "I have heard the murmurings of the children of Israel," for in the beginning He said, "I hereby cause bread to

rain down for you,'' as one who does a favor for them, or because of their merit, and now He says that it is accounted for them as a sin, and because of this murmuring He will do thus with them, that they should know that "I am the Lord your God," for until now you do not believe.

Our story of the manna, in Exodus 16 (and indeed all of the texts to a greater or lesser degree) is heterogeneous,
39
On the one hand, there are the ample indices of a story of need, complaint, prayer, and miracle, which dominate the text; on the other, the initial murmuring with its strong overtones of exaggeration and rebellion ("If only we had died by the hand of the Lord in the land of Egypt, when we sat by the fleshpots and ate our fill of bread. For you have brought us out into this wilderness to starve this whole congregation to death"). Looked at from a literarycritical perspective (synchrony), rather than a literaryhistorical perspective (diachrony), the text is polyphonic.
40
Built into its very structure is contradiction and opposition. Here and all through the canon, there is a dialogue of voices evaluating the wilderness period, a voice which proclaims that it was the time of the greatest love of Israel for her God and a voice which cries out that under the very marriage canopy, as it were, Israel was unfaithful to her Groom. Each tanna hears only one of these voices.

The doublevoicedness of the Exodus narrative doubly inscribes itself in two different genres. On the one hand, there is a series of narratives in the Torah, which tells a story of the people's need in the desert, their (or Moses's) prayer, and God's fulfilling of the need. These narratives are ideologically consistent with the view of the Israelites' behavior in the desert as praiseworthy. The use of the language of freewilled gift of the manna in our story seems to place our text within the intertextual code of this genre, on the other hand, there is another collection of narratives which tells a story of unwarranted complaint on the part of the people and God's answering the complaint by ironically fulfilling the false need together with a terrible punishment. Our story seems to belong to the first pattern. As Childs has insightfully remarked, the verb "murmured" in both of our narratives creates a contradictory or ambiguous moment within the narrative by referring intertextually to the blame series. Each of the tannaim may be said, therefore, "to smooth 'the troubled passage between text and reader' by soothing the contradictions within the text."
41
The midrash, however, as an edited text, does not soothe these contradictions. Quite the opposite, by presenting two diametrically opposed views, it manifests a powerful awareness of ambiguity as built into the text itself and not as a pure function of reading.

Each of the tannaim indeed hears only one of the "voices in the text." The Mekilta, on the other hand, hears both voices. We do not have only two antithetical views of what the Torah means, as if in two separate books. What we

have is one text,
Mekilta Wayyassa
, in which both of these views are articulated together. The midrash is thus heterogeneous in the same way as the Torah is. Rabbis Yehoshua and El'azar can be read as hypostases, as personifications of the voices that
we
hear within the Torah itself.
42
It becomes unnecessary to reduce their controversies to polemics and ideological differences grounded in their historical circumstance—even in this most explicitly valueladen of substantive hermeneutic controversies—when we see that these controversies arise from within the text as encoded by its very Author Himself, and we are no more (or even less) justified in such a reduction than we are in reducing the different readings of a great work of secular literature
43
to the political differences of its readers.
44
The interplay between the hermeneutic practice and other cultural/social/political ideologies and practices is complex and dynamic, with each area of practice affecting the others and affected by them.

Patterson seems then to be confuted by this text in his claim that ambiguity is located only in the reader's work. This ambiguity can, it appears, be located in the text. However, it could be argued as well that my analysis
confirms
Patterson's theoretical point, since each of the readings in the Mekilta denies ambiguity. Put in other words, I believe that Patterson is arguing that since "for some readers no text is ambiguous," this puts into question the very idea of a
textual
ambiguity as such, placing my (perhaps by now, our) perception of ambiguity on precisely the same epistemological level as those readings which foreclose that ambiguity. I think that I can counter that argument in two ways. First of all, the energy which the tannaim must expend in order to read the text, the penetration and boldness of the

hermeneutic moves required to rationalize the ambiguity, serve not to hide or deny that ambiguity but rather to dialectically reveal it. Thus Geoffrey Hartman has argued that "the heterogeneity of poem or original text by no means disappears in the older hermeneutics, but it appears only by way of the daring interpretation that is startling and even liberating in its very drive for harmony."
45
Each of the tannaitic interpretations, in order to drive toward a harmonious understanding of the nature of Israel in this time, is forced to distort the local meanings of certain passages. The strength of these harmonistic readings is placed into relief by contrast to the weak harmonistic way in which later midrash deals with these contradictions. The later tradition almost invariably assigns the positive indices within the Torah to the "true Israel" and the negative ones to the ''mixed multitude." The refusal of the Mekilta to adopt this reading is a strong affirmation of the complexity and polyphony of the Torah.
46
Thus R. Yehoshua, usually presented in the Mekilta as a proponent of ''plain" reading, here presents several very daring interpretations in his desire to understand the story in the light of the "honeymoon" tradition of Israel and God at this period, while R. El'azar works to emphasize the evil of their ways, e.g., by turning their murmuring against Moses into a fullscale blasphemy against God. Each of these interpretations, by

the very distortion necessary for its maintenance, forces us to recognize the ambiguity encoded in the narrative. Sternberg has remarked on the function of such ambiguities in the Bible:

The coexistence of two (or more) mutually exclusive hypotheses—concerning action, motive, character—always enables the author to kill two birds with one stone, using the same materials for different ends. Above all, it enables him to base sequence and effect on the tensions between the two possibilities. Each reading may serve to balance and ironize the other. The emergence of such a hypothesis in a text that equally validates its contrary renders each quite unlike a similar hypothesis appearing unchallenged.
47

If this is true of the literary text itself, then it is true, as well, of the midrashic text, which doubles the ambiguity of the Torah's narrative with its own dialectic. Thus, each of the tannaitic interpretations in the Mekilta can be said to balance and ironize the other, and it is certainly the case that the presentation of one interpretation in a commentary that equally validates its contrary renders each quite unlike a similar interpretation appearing unchallenged. The effect of the midrashic text as a whole is to present a view of textuality that occupies neither the extreme of assuming a univocal "correct" reading of the text, nor the extreme of "whoever finds a lesson there useful to the building of charity, even though he has not said what the author may be shown to have intended in that place, has not been deceived, nor is he lying in any way." Rather, the midrash seems to present the view of an ancient reader who perceives ambiguity encoded in the text itself with various dialectical possibilities for reducing that ambiguity, each contributing to but not exhausting its meaning(s). Our midrashic text here is, in effect, occupying the position of Steinberg the critic himself, observing and commenting on the too singleminded resolutions of the text's ambiguity. It is in that sense that I have referred to it as "metacommentary." Moreover, the Mekilta does not speak discursively and abstractly in metalanguage
about
the ambiguity of the Torah. It represents the tension and inner dialogue of the biblical narrative by tension and inner dialogue of its own, providing us with an elegant precursor of Harold Bloom's famous remark, "
All interpretation depends

upon the antithetical relation between meanings, and not on the supposed relation between a text and its meaning
.''
48
However, intertexuality is a much more complex notion than this. It is also a matter of the deepseated ideological codes which determine and allow for any cultural practice. In the next chapters, I will begin to locate the workings of the intertext in this latter sense in the Mekilta's reading.

Up to this point, I have tried to show how midrash can be understood as an intertextual reading practice, founded on the intuition that the Bible is a selfglossing text. The Mekilta uses other parts of the Bible to read the Exodus in several different ways. I have suggested that these structures can be read as manifesting the two basic organizing principles of language—paradigms and syntagms. In this and the next chapters, I will be reading the major genre of syntagmatic midrash. This kind of midrash, the mashal, does its hermeneutic work by recasting diverse texts into a narrative, which then frames and contextualizes the verse to be interpreted.

The term "mashal" is often glossed "parable", and indeed there is good reason for this. First of all, as has been conclusively shown by David Flusser, the parables of the Gospels belong certainly to the same tradition as the mashal in rabbinic literature.
1
Second, the very term
parabole
translates the Hebrew in the Septuagint and the Gospels. However, there are also very great differences in semiotic structure between the parable of the Gospel and the mashal of the midrash. In order to show better the structure of the mashal, I will begin by discussing its differences from the parable.

Susan Wittig has written a very dear analysis of the semiosis of the Gospel parable and its congeners in later Western literature.
2
Her thesis is that in normal denotation a sign vehicle refers to a
denotatum
, "which has or could have physical existence in the extralinguistic world (e.g., 'The householder went out early in the morning to hire workers for his vineyard')." In the parabolic system, however, the denotatum, in turn, becomes a sign in itself, "designating an
unstated designatum
, a conceptual referent which exists only in a world of moral and physiological abstraction, which cannot be perceived in the extralinguistic world, and which must be supplied by the perceiver of the sign (e.g., 'The spirit of God at the beginning of time began to seek out righteous men')":
3

The structure of this duplex semiotic gains its energy and effectiveness from the nature of the semantic relationships which link its components. The first order linguistic sign vehicle is linked
conventionally
and
arbitrarily
to its de

notatum, as are nearly ail linguistic signs; the second order material sign vehicle, however, is linked
iconically
to its object in the same way that the structure of a diagram formally exemplifies and exhibits the structure of its object.
4

In Wittig's description the parable is allegorical in its structure. A structure is created in language which is interpreted through the reference of its signified to another signified that resembles it in some way or other. Moreover, Wittig argues that the parable is, in the phrase of Wolfgang Iser, an indeterminate text, one marked by "
hiatus
—the lack of syntactic or semantic connections—and 'indeterminacy'—the omission of detail—and which invite(s) the reader to establish his own connections between the lines of the text, so to speak, and to create his own significant detail, when the text does not offer it."
5
The parables of the Gospels are in Wittig's view just such texts. The "indeterminacy" of meaning is suggested by the fact that the "second order denotatum" is not stated in the text. (It should be made clear, incidentally, that this analysis is governed by the assumption—common but nevertheless an assumption—that the applications of the parables in the Gospels are not an original part of the text. If the application is an original part of the parable, then it can hardly be described as an open text.) The reader, by being forced to interpret such texts, comes up against his/her own "expectations and preconceptions—his own meaningsystem.''
6
Then:

In semiotic terms, such texts are selfreflexive in a metacommunicative dimension, calling to our attention not their syntactic or semantic structures, but the variety of ways in which those structures are actualized in our minds [and] are made to yield their potential meanings. When we read a text characterized by the quality of indeterminacy, as the parables are, we are reading ourselves as well as the text, and are being forced to an awareness of the creation of meaning in our own minds, as well as to an awareness of the meaning itself.
7

Now, whether or not Wittig's characterization of the Gospel parables as "indeterminate" texts is successful
8
in any case the midrashic mashal manifests a completely different structure—which is not, of course, a criticism of her work. Let us go back to the example that I have discussed briefly in chapter 2 and see the differences which this text shows from Wittig's description (and from the parables, if she is correct). I will quote the text again: