Intertextuality and the Reading of Midrash (13 page)

Nahmanides struggles to render the two stories a noncontradictory narrative of

two different events, in spite of the extraordinary difficulties such an interpretation raises. The answers which Nahmanides offers, indeed the very fact that he gives several, indicate how much of a difficulty there is in assuming that the two texts on the giving of the quail relate two different occurrences. It is, accordingly, not surprising at all that modem critical scholarship considers these two accounts as parallel stories of the same events deriving from different sources. George Coats, a scholar who has closely read the "murmuring" passages, details this position:

Yet the outline of the two is similar enough to suggest that at least the accounts of the quail are parallel: (1) The people remember the varied foods which they had in Egypt. (2) Moses hears their reminiscences. (3) The case is in some manner presented to . . . [God]. (4) . . . [God] responds with the gift of quail. (5) This process is associated with the murmuring motif. Unless we are to assume that two different traditions of the quail existed without contact but nevertheless narrated their material in the same manner, we must conclude that we have parallel accounts of the same tradition. The differences in form and content could then be explained as the peculiar emphases of the two sources.
27

Coats's solution is to accept the sourcecritical option, i.e., to claim that we have the
same story
in two different versions. The midrash assumes as well that only one account is preserved, but will not, of course, simply leave it at the level of two accounts of the same events which partly contradict each other. The Book has only one Author. A strong reading is required, then, to bridge the gaps and explain the repetition of the two texts.

In the Mekilta, the discourse in Exodus is made to tell the same story in tropes and narrative figures which is told in Numbers in mimetic order. The very brief mention of the quail in the Exodus passage only exists for the sake of the contrast with the giving of the manna. The event does not take place here at all, but only in the future narrative to be related in Numbers. Since at the time of the narrative of Exodus, the events regarding the quail are only
about
to come, what the midrash implies here is a baroque version of the biblical structure of forecast fulfillment.
28
The forecast element is made explicit in the Mekilta via the use of predictive language: "R. Yehoshua says, God said to Moses: Before me is revealed what the Israelites have said and what they will say."
29

All of the elements of the midrashic story are related explicitly in Numbers: full bellies, complaint about the manna, and God's ill will. There is more evidence that the Mekilta is indeed reading the Numbers text together with the Exodus text to generate its meanings. In the next pericope of the Mekilta we read, as a gloss on the verse; "At evening you will eat meat": "You contradict yourselves. You asked for bread. Since flesh and blood cannot live without bread, I gave it to you. Then you went and asked for meat, with full bellies.

Behold, I will give it to you, so you won't say that I do not have the ability to give it you, but behold I give it you, and in the end I will collect from you," a dreadful hint of the plague to come. Where? At the end of Numbers 11. And if that were not enough, the Mekilta continues its interpretation with the next sequence of verses from Numbers, as a simple continuation of the interpretation of Exodus! and as if to say that those verses are part and parcel of the same story. We have, therefore, practically explicit evidence that the rabbis do indeed see these two discourses as telling the same story.
30
Since, as I have shown, there is strong textual evidence that these are indeed two reports of the same events, we do have here a structure of repetition, albeit somewhat attenuated. We could locate this repetition at a redactorial rather than authorial level to account for the attenuation of the repetitive structures, but nevertheless a synchronic reading of these texts practically requires that they be seen as repetition. The solution of the midrash is a strong way out of an apparently impassable exegetical dilemma creating the opportunity for the release of meaning (almost the explosion of meaning) at the axiological level.

We come now to the very climax of the reading, the interpretation of evening and morning in verse 8. Since they cannot be read mimetically, iconically, and literally, as ordered signs signifying the timing and order of events to take place, they must be read as symbols, whose meaning is figurative. They are, indeed, understood (as are all symbols or arbitrary signs)
31
as deriving meaning from their binary opposition. This binary opposition of night and day is homologous to an opposition in values.

Why is the giving of meat mentioned in Exodus at all, since it will not actually happen until much later in Numbers, when bellies are full with bread? The answer is that only through the relation of opposition to the giving of quail in the evening does the giving of manna by morning get its meaning. The midrashic text seems to repeat itself, saying: He gave manna to Israel with a bright countenance. The quail, which they asked for with full bellies, He gave them with a dark countenance, but the Manna, which they asked for appropriately, He gave them with a bright countenance. Now the clause "but the manna" is otiose, a repeat of "He gave manna," etc., unless we understand it in the way I have posited, namely, that first the simple statement is made to the effect that from the manna being given by morning we know that it was given with a bright countenance. Then the midrash shows how we know this, namely, because of the opposition in the next verse. We have here, then, a very early adumbration of structuralist semiotic, in which signs are read through their opposition to others. The traditional commentator on the Mekilta, R. Yehuda Najara, has shown a fine awareness by saying, "Since
this
was given in the evening but
that
in the morning, it teaches that this was given with a darkened countenance and that with a shining one." In other words, one signifier would not signify and only the two together in their opposition mean anything. Now it is clear why the giving of the quail is mentioned

here even though it will not happen until later; it is in order to set up the binary opposition.

However, this opposition of bright countenance/darkened countenance is not merely an arbitrary opposition, but in fact a highly motivated one. What is the nature of these linguistic signs "bright countenance," "dark countenance"? How are they signified in the biblical text, and how do they signify in the midrashic text—aside from the bare relation of binariness? The first question has been partially answered. The text tells us that the morning is when the manna was given and the evening the time of the quail. Morning, when the day is bright, is a highly motivated sign for the sign, "bright countenance." Evening, when the day darkens, is a sign of the sign, "dark countenance." But what manner of signs are these signified signs? We could say—were the rabbis pantheists—that they are signs of themselves; i.e., that the morning

is
God's bright countenance, as it were, and the evening HIS dark countenance. (We would still be left, however, with the problem of knowing how these signs further signify God's mood.) Knowing, however, that the rabbis are not pantheists, we are induced to seek a more complex semiotic structure. One possibility is to see the giving of gifts at morning and evening as indices of God's mood because of something intrinsic to these times. This is the strategy of Rashi, who writes that since the manna was given at a convenient time, viz., the morning, when it is easy to prepare and eat, we may understand that it was given in good cheer, as opposed to the quail which is given at a most inconvenient time.
32
However, this interpretation leaves the linguistic signs ''with bright countenance,'' "with dark countenance" without any significance, except as dead metaphors for good and ill will. Put another way, Rashi's reading seems to be: since the manna was given in the morning, when it is easy to prepare, we can see that God gave it with a good will. The phrase for good will is "bright countenance." However, the connection that we feel so clearly between the
phrase
"bright countenance" and the morning is unaccounted for in Rashi's reading. In fact, Rashi cannot be faulted for his interpretation, because the actual text which Rashi is commenting on, namely, a parallel to our midrash found in the Talmud, seems not to have known what to do with this connection.
33
It reads, "The manna, which they asked for appropriately, was given to them appropriately," removing the very connection between the sign of giving manna in the morning and the linguistic sign of "bright countenance," for which I am trying to account.

The most satisfactory reading takes the events referred to in the words "morning" or "evening" giving, as iconic signs of God's mood; i.e., the giving of manna in the morning, when it is bright, is a motivated sign of God's bright face and the giving of quail in the dark is a motivated sign of God's dark face. Better still, the fact that the Torah tells us that the manna was given by morning is a sign of God's bright face. But having reached this point, we still encounter a difficulty, for "bright" and "dark" countenance are metaphors. A face

does not literally brighten with joy or darken with sadness. These metaphors are being radically literalized, as if to say that God's actions in the world can be interpreted through the metaphorical code of human speech. In other words, the rabbis make a theosophical inference based on their intuitive understanding of human language—indeed, a radical new meaning for the proposition that the Torah speaks as in the language of men! God gave the manna in the morning and the quail in the evening,
because we say
"He gave it with a bright countenance" to mean "He gave it willingly." Another way of putting this would be to say that such verses as "May God shine His countenance upon you" are being activated intertextually to render ideological meaning to a "historical'' event, the giving of manna in the morning.
34

In fine, then, we see that what at first glance are sharp distortions in the Mekilta's reading of Exodus 16 are in fact the results of its authors' attempts to interpret a seemingly contradictory repetition of the narrative in Numbers 11. A repetition (especially, but not exclusively, a contradictory one) creates a gap, then, just as surely as do details left out of a story. The repetition of the narrative in the Torah sets up an interpretive gap; this gap must be read, if the unity of the text is to be maintained. What is perhaps so special about midrash is the way that value is insistently
implied
through the filling of the gap, without, therefore, there being a need to explicitly paraphrase or translate the values of the story into a nonnarrative discourse.

If we reexamine now the structure of reading in the two examples discussed here, we can perceive the similarities between them in spite of differences. In both cases the midrashic reading is a gapfilling. In the first case, the gap arises from the ambivalent evaluation of the Israelites' behavior in the wilderness together with the unexplained anomaly of the Torah's narrative indicating that Moses initiated the move in but this one instance. The struggle for the text generates two mutually exclusive readings of the narrative, actually two mutually exclusive reconstructions of the very events. In the second case, the gap arises from the fact that what is apparently the same story is told twice in the Torah and the two narratives seem to contradict one another. In both situations, the interpreter who is not willing to adopt a diachronic reading strategy of dissolution of the text is forced to interpret—strongly or weakly. My claim is that midrash (as practiced at least in the Mekilta) is a method of strong reading of the gaps of the text, filling them, as it were, by inserting the intertext into them. In the next chapter, I shall further explore the specific workings of ambiguity

in the Torah's narrative and how they are represented and dealt with in the Mekilta.

The previous chapter dealt with the Mekilta as a reading of the gaps and heterogeneity in the Torah's narrative. In this chapter, I would like to continue that discussion, focusing on the precise semantics and semiotics of ambiguity and the way that the midrash responds to it. I will here read closely the Mekilta's interpretation of a particular narrative of the Torah, in order to show that the midrash seems to be responding to ambiguity in the discourse itself. Moreover, I will try to show that the resolutions of the ambiguities which the midrash presents are not merely appropriations or preemptions of the text but rather choices made from interpretive options that the canon itself offers.

Michael Riffaterre has provided precise analyses of ambiguity and intertextuality in literary production. His most powerful notion is that of the "ungrammaticality," the awkwardness of a textual moment, at any linguistic or discourse level, which by its awkwardness points semiotically to another text which provides a key to its decoding. This notion takes us one step beyond "gaps" as the intertextuality of the production of the narrative text and into the possibilities of reading that text and its intertextuality. I am interested here in a particular kind of such ungrammaticality, the "dual sign,'' which by calling up two intertextual codes generates two possible decodings of the text:

. . . the dual sign works like a pun. We will see that the pun in poetic discourse grows out of textual "roots." It is first apprehended as a mere ungrammaticality, until the discovery is made that there is mother text in which the word is grammatical; the moment the other text is identified, the dual sign becomes significant purely because of its shape, which alone alludes to that other code.
1

The dual sign is thus the lexical equivalent (and because of its ungrammaticality, the hyperbolic equivalent) of two simultaneous, synchronized expansions generated parallel to each other, but separately. . . . Separated in the text, they are again united in the portmanteau word, whose hybrid morphology arouses the reader's curiosity so as to better guide him to the significance.
2