Midrash as Law and Literature

by Geoffrey H. Hartman
Midrash as Law and Literature
Geoffrey H. Hartman
The Journal of Religion
Start Page: 
End Page: 
Select license: 
Select License

Midrash as Law and Literature

Geoffrey H. Hartman / Yale University

My motives in studying Midrash are not pure. I am a raider of the lost ark looking for treasure. It is not for the sake of heaven I study but to bring back voices and types of interpretation of which that ark is as full as Noah's was of beasts. In an era of restitutions, such a restocking of identity through a historical and compensatory search is common enough. But in the case of Midrash-by which I always mean a method of exegesis as well as the collections formed by it-another motive enters. I cannot forget how these writings were slandered, and how public igno- rance abetted such slander in the Nazi era. Jews were demonized at a time when Talmud and Midrash were available yet remained a closed book even to the educated. And for centuries before that, theological anti- Semitism had misrepresented the spirit of Jewish law: non-Jews were taught to see only a crass and stubborn literalism, a mean-spirited, mate- rialistic frame of mind, rather than what David Weiss Halivni has called the predilection of Midrash for justified law, which heaps interpretation upon interpretation.' That era of prejudice and ignorance should be ap- proaching its end.

Yet to make sure of its demise will take a concerted effort. Those who live within the walled garden are often so absorbed by its pleasures and duties that they do not open it to others. And those like myself who sneak through the wall like a thief in the night cannot emerge with more than fragments wrongly detached from a living environment. I am not dis- couraged, however, because the need is so great-the need of those in my generation who "have the Bible and their great poets, but no longer understand how to use their imaginations in reading them." These are Northrop Frye's words forty years ago, describing how William Blake felt two hundred years ago.

You can imagine my mixed feelings when I read in Ralph Waldo Emer- son's journal the following entry: "If Minerva offered me a gift and an

' David Weiss Halivni, Midrash, Mishnah, and Gemara: The Jewish Predilection forJustified Law (Cambridge, Mass.: Harvard University Press, 1986).

O 1994 by Geoffrey H. Hartman.

option, I would say give me continuity. I am tired of scraps. I do not wish to be a literary or intellectual chiffonnier. Away with this Jew's rag-bag of ends and tufts of brocade, velvet, and cloth-of-gold; let me spin some yards or miles of helpful twine, a clew to lead to one kingly truth, a chord to bind wholesome and belonging fact^."^

On the level of tradition I, too, want continuity, and on the level of intellectual, especially literary analysis, that desired continuity translates into a search for the formal unity of the Hebrew Bible and correlatively of midrashic method. Yet the first lesson Midrash itself teaches the ap- prentice is to be wary of overunifying its words or of unifying them in a totalizing way, It is comforting to anticipate with Emerson "a clew to lead to one kingly truth," and it is intellectually imperative to keep that possibility in mind. The Hebrew Bible, however, is not a classical work of art, nor is Midrash a Patrologia in which a supersessionist revelation draws everything into that regal unity of "wholesome and belonging facts." Emotionally and intellectually I am with Emerson, but empirically and spiritually 1 am closer to the point at which Midrash and Kafka in- tersect.

Let me return to Emerson's image of the Jew's rag-bag. It does not have to be an insult or an indication of abject poverty. Many contempo- rary scholars accept a documentary hypothesis which views the Hebrew Bible as a glorious patchwork, and that the seams show through, as in modern collages, can be an artistic, though not a theological, virtue. Since with Midrash and Talmud there is-perhaps all the more-a marked effort to preserve every significant tradition and law, however elliptical or unclear, many sayings or exegeses are agglutinated: having been stored in a collective memory-the Oral Tradition-and waiting to get into the written text at the appropriate moment, if no such mo- ment comes along, they sneak in nevertheless. The founding rabbis, in short, can be considered as law-rhapsodes, and the halakha as a severe p~etry.~

The literary study of Midrash makes us aware that what is meant by the literary has not been clearly defined. Does it neatly preclude this me- morial and encyclopedic dimension of rabbinic literature? One reason for the modern interest in both Bible and Midrash is their apparent care- lessness about literary effects associated with the concept of genre and formal unity. What we call "rhetoric" and "poeticsn-arts indebted to Greek and Roman thought-did not separate out as technical branches

See Bliss Perry, ed., The Heart of Emerson'sJournals (New York: Dover, 1958), p. 267.

I allude, of course, to Giambattista Vico's famous description, in The New Science (172544) of the Roman Law, chanted by the decemvirs and still memorized by the schoolboy Cicero.

of knowledge during the formative period of talmudic Judaism. It would be exciting to know what the paidea of the sages involved, and while we know something about the rabbinic academies, we do not know enough. But it seems that, despite the impact of Hellenistic learning in the area of dream interpretation or hermene~tics,~

poetics played no role. Even today, the elements of Midrash are taught mainly through immersion. The best an essay can do is to smear a little honey on this or that text.

My essay may not be different in this regard. But I do want to raise the issue of what happens to that cornerstone of poetics, the concept of unity, when Midrash enters the picture. Consider J. B. Soloveitchik's "The Lonely Man of Faith."5 The essay rejects source criticism (the documen- tary hypothesis) as an attaint to the unity and integrity of the divine text. This rejection does not entail, however, a denial of contradiction in Scrip- ture. What ensues-what Soloveitchik liberates himself into-is a strong Midrash on the two stories of Adam's creation. Their incongruity is not a textual accident (as the Higher Criticism proposes) but points to a real contradiction in the nature of man. Soloveitchik develops the two ac- counts into a picture of Adam the first and Adam the second.

These two categories allow him to pursue a sensitive contrast that builds into an entire moral and social philosophy. It sets against the intel- lectual and scientific dignity of man the crisis-feeling of being overpow- ered by God or the world, a feeling which radically deepens the quest for redemption. The first Adam though not alone is lonely because he is committed to an ideal of mastery that results in dominion over the cos- mos yet separates him from it; the second Adam is doomed to explore that loneliness as a loss which can only be illuminated by the dimension of faith and the convenantal idea. The muteness and indifference of an alienated cosmos now sharpen Adam's consciousness of separation; for each "I" is ontologically lonely-that is, incomplete-and so the "natural work community" of Adam the first cannot solace this ordained lack of connection. Hence a turn from cosmos to covenant. "Our sages said that before Abraham appeared mjestas dei was reflected only by the distant heavens and it was a mute nature which 'spoke' of the glory of God. It was Abraham who 'crowned' Him the God of the earth, i.e., the God of men."

By a surprising return to the text Soloveitchik then suggests that Eloh- ist and Yahwist do express very different source experiences. The craving of convenantal man for a personal and intimate relation with God cannot

See Saul Lieberman, Hellenism in Jewish Palestine (New York, 1950).

J. B. Soloveitchik, "The Lonely Man of Faith," Tradition 7 (1965): 5-67.

be realized through the cosmic "E-lohim encounter," so that the locus of transcendental experience must shift to where "the finite 'I' meets the infinite He 'face to fa~e."'~ This more communal relation between man and God is symbolized by the Tetragrammaton in the Biblical account of Adam the second.

I do not apologize for spending time-little enough-on part of Solo- veitchik's essay. It participates in an intelligent revolt against modernity; it acknowledges yet does not accede to the claim made on us by modern categories. Here is someone acquainted with the language of contempo- rary philosophy, who knows that Jewish thought has often flourished within a foreign environment, and who will not accept the fact that phi- losophy speaks only Greek. His brilliant rejection-conversion of Higher Criticism seems to come right out of the lineaments of midrashic style. It is true that he does not call his procedure "midrashic," yet there is no question that it is so. It explores an incongruity, it puts into play proof- texts and associated authorities in a potentially endless dialogue, one that brings the Bible closer to us even as it brings us closer to it. The essay's fusion, moreover, of homiletic dependence on Scripture with an analysis of contemporary spiritual problems culminates in an appropriation of 1 Kings 19 that makes Elisha into a typical representative of the natural work community, an Adam the first transformed by his sudden encounter with Elijah into an Adam the second. Convenantal mankind replaces ma- jestic mankind.

There is, however, one disconcertingly superficial aspect to this mod- ern orthodox scholar. What he says about the literary is both minimal and wrong. The Higher Criticism, he alleges, based itself on "literary cat- egories invented by modern man." Now contemporary critical theories do use stylistic criteria to distinguish sources, but no literary scholar would stop there. It is precisely the issue of the unity or coherence of art that has exercised poetics since Aristotle. Using the example of Greek tragedy, Aristotle defined a unity that was nonepisodic, that did not de- pend, like epic, exclusively on the presence of a hero but was the outcome of an "action" involving all elements. It seems impossible to apply Aristo- tle's scheme to the Bible, which remains episodic and, though it is said to have been given to Moses, is not unified around Moses as hero, or "one greater man." The unity of Scripture supported explicitly and unre- servedly by Soloveitchik obviously suggests different criteria and forms of analysis. To discover them is also a task of contemporary literary theory.


Recent literary thinkers have challenged the simple location of unity in art.8 They rejoice, like Midrash, in the "interpretive bounty" of a text;g they accept the mediacy of linguistic and interpretive structures. A full-scale rethinking of the dichotomy of creation and commentary is in pro- cess, which has already modified our picture of what constitutes unity. So Claude Lkvi-Strauss engages, like Soloveitchik, disparate creation myths, from Greek literature and South American folklore, because interest has shifted from elegant ideas of artistic coherence to the making and sus- taining of traditions. A consensus is building that cultures stabilize contra- dictions in their belief-system by the interpretive extension of founda- tional texts. On this view, commentary is not, or not only, a by-product of these texts but is itself a functional revelation. The structure of first (Scripture) and second (MidrashIInterpretation)is modified into a hendi- adys or syzygy. Dan Sperber, the French anthropologist, has said suc- cinctly: "Exegesis is not an interpretation but rather an extension of the symbol and must itself be interpreted." 'O

This understanding of the relation between text and commentary is important for Judaism, where commentary becomes the authorized form of creative thought." We have no statement more radical in this respect than that both Oral and Written Law were given on Sinai, for this legiti- mates talmudic commentary as strictly coterminous with Scripture. Therefore the daring intimacy of Midrash with Bible. Remember Moses in Akiba's cheder, marveling at what he hears, and just a little bemused. Instead of a magisterial theology expressed in Aristotelian treatises, our

Northrop Frye indicates the problem while trying to obviate it. He writes: "Unity, a primary principle of works of art since Plato's time, also indicates the finiteness of the hu- man mind, the care that works toward transforming the 'imperfect' or continuous into the 'perfect,' the form achieved once and for all. The Bible, however unified, also displays a carelessness about unity, not because it fails to achieve it, but because it has passed through it to another perspective on the other side of it" (Northrop Frye, The Great Code: The Bible as Literature [New York: Harcourt Brace, 19821, p. 207).

Harold Fisch, "The Hermeneutic Quest in Robinson Crusoe," in Mzdrash and Literature, ed. G. H. Hartman and S. Budick (New Haven, Conn.: Yale University Press, 1986), p. 230.

lo See Dan Sperber, Rethinking Symbolism, trans. Alice L. Morton (New York, 1975). In his recent book, The hice ofJacob (Bloomington: Indiana University Press, 1990), Leslie Bris- man demonstrates how sensitive biblical interpretation can be, which accepts the documen- tary hypothesis (its modification of the Bible's inspirational or authorial unity) yet brings the documents (here J and EIP) into dialogue. To do this requires (1) a definition of the literary as "the primacy of intertextual, as opposed to sociological or political, motives for invention" (p. xiii), and (2) a midrashic sense of how late voices have a standing, how a kind of firstness can be drawn from secondariness.

The best discussion of the innovative and revelatory role of commentary in Judaism is that of Gershom Scholem. See his "Offenbarung und Tradition als religiose Kategorien im Judentum" (Revelation and tradition as religious categories in Judaism), Judaica, vol. 4 (Frankfurt: Suhrkamp, 1984). The English translation appears as The Messianic Idea inJudaism (New York: Schocken, 1971).

main inheritance is the composite, heteroglossic supplement of Midrash: a supplement that converts exegesis into law finding (halakha) or a litera- ture of justification (aggada).

Of course, literary commentaries and Midrashic ones have their differ- ences. But these are not easy to define except by pseudo criteria. The very advance of contemporary theory toward Midrash makes Jewish scholars more zealous to avoid contamination. There is fear that the mo- tive for Midrash will be mistakenly reduced from Everything is in the text, and what the text szgnijies is its relevance to the actions or thoughts of the interpretive community to Everything is text, and the text is a structure of imagznary relations, a tissue without issue. I acknowledge the danger, but why be frightened by those who insist on being superficial?

The literary establishment has its anxieties, too. Is it not foolhardy to extend from sacred to secular texts the principle that an apparent inco- herence signals an overarching integrity? Yet every good literary inter- pretation takes precisely that risk. It comes upon an obvious or less obvi- ous difficulty: a crux, a deviation from the norm, a contradiction. On that evidence it either impugns the authority of the work, or it values the anomaly and so augments the authority of the work.

The secular critic, of course, has a choice between negative and re- demptive approaches, while the religious interpreter does not. The dar- shan cannot stay in the negative. His inventiveness is spent on ways to redeem the text's negative features (incoherence, ellipses, the disparity of historical fact and religious expectation, gaps between tenor and text). His basic choice may be characterized as follows. Does the harmony he evokes in order to repair the text point to a higher unity? Or is the truth plainer, a rectijication, even polemically directed against a mystijication, one that comes from our need to think of the sacred in sublime rather than down-to-earth terms?

In brief, though the words of the Torah can be made to fly up, more often Midrash infers from ellipses or condensations a very human story and introduces dialogues that draw God deeper into the affairs of man- kind. Let me bring an example from Midrash Tanhuma, which has paral- lels in Genesis Rabbah and the Babylonian Talmud (Berakhot 60a).

Genesis 30:21-22 reads: "Afterwards she [Leah] bore him [Jacob] a daughter and named her Dinah. And God remembered Rachel and listened to her and opened her womb." Here there are four possible ellipses: "Afterwards" could raise the question, Precisely after what? "Di- nah" is not followed by an explanation of the name, as is the norm in Genesis (the name Joseph is explained twice in lines 23-24 of the same

chapter).12 The transition from Leah to Rachel is abrupt. And Leah as well as Rachel could be referred to by "God . .. listened to her." Here is how Tanhuma interprets the verses:

After Leah had given birth to six sons, she saw by way of prophecy that there would be twelve tribes in the future emerging out of Jacob. She had already given birth to six sons and was pregnant with her seventh. The two maid servants [Bil- hah and Zilpah] had given birth to two sons each, hence a total of ten. Leah arose and called out to the Holy One, blessed be He: "Master of the World, twelve tribes in the future will arise out of Jacob; I already have six sons and am pregnant with my seventh and the maid servants have two each. If my next child is a male, my sister Rachel will be even less than the maid servants!" Immediately, the Holy One, blessed be He, listened to her prayer and transformed the fetus within her into a female, thus it says: "Afterwards, she bore him a daughter and named her Dinah . . ." Why did she name her Dinah? Because Leah the righteous stood before the Holy One, blessed be He, seeking a judgment (din)and the Holy One, blessed be He, said to her, "You are rahmnit [kind person, merciful person] and so I will be merciful to her, hence: "Now, God remembered Ra~hel."'~

The interpolated story, typical of a patriarchal society, makes the ellip- ses fertile, and suggests that Leah, after giving birth to six sons and car- rying a seventh, remembers the plight of her sister and calls on God to grant Rachel the next male child. God converts the fetus in Leah's womb to a female, while mercifully (a pun intervenes here linking the word for mercy with that for womb) granting Joseph to Rachel as her first male offspring. Whereas the Bible story mentions nothing about Leah's frame of mind and, indeed, has God abruptly "remember" Rachel after Leah has given birth, the Midrashic writer ascribes that "remembering" first to Leah and then uses it as a charming psychological touch to explain the meaning of the name Dinah. In brief: he assumes that Leah has an open line to God and can influence Him. He also shows God making a decision not autocratically, or in a great consult with his angels, but after a plea from a mortal. Leah's rachmones seems to activate his own.

Is there a literature from that time which is so down-to-earth? It may be strange to call Midrash literature since it remains a mode of commentary explicitly linked to the very words of Scripture. Yet we recognize the cre- ative and parafictional result of its interpretive elaborations. Moreover, at a certain level Midrash is not satisfied with the text as it stands, and while it refuses to produce a new or transformed writing it looks for more of the original in the original, for more story, more words within the

l2 It is also rare, moreover, that a female child is given a formal naming phrase. I am grateful to Barry Holtz for drawing my attention to this passage in an interesting talk held at a Midrash colloquium at Yale.

words.14 But this potentiality in the sacred text is rarely treated as a mys- terious void. Leah, as she stands before God, exerts a theurgic force, how- ever weak,I5 and such force, such pressure, is also what the darshan ex- erts on and through the Bible. The rabbinic interpreter is characterized by participation exkgitique.

That Midrash is not satisfied with the text-in the sense that it wishes for something more, not something different-means that its labor of the negative can be very daring. Gaps or obscurities, everything that could be characterized as indeterminate, are emphasized before being resolved by one interpretive or interpolative davar after another. And though what is potential is usually understood in accord with the dictum that the lan- guage of the Bible addresses the normal person's capacity for under- standing, in the tricky area of theurgic interpretation text moves close to theory in one respect: it must be "falsifiable" (in the Popperian sense of that word).

Let me clarify this scandalous idea. The midrashic skill, for instance, that divides up words or verses, together with an ingenious repunctua- tion (revoweling) of phrases, is a combinatory art that questions the can- onized letters before us. While these letters have a received meaning on the grammatical level of word and sentence, they are taken to be, at the same time, anagrammatic, in the sense of constituting the elements of a divine name coterminous with the Torah and guaranteeing every mark in it. This perspective is made explicit by the Kabbalah. "The whole Torah is the Name of the Holy One," we read in the Zohar (Yithro 87a); the letters of the Torah, the Ramban comments, were written continuously, without break of words, which makes it possible to separate them into Divine Names when read by that "path" (Nachmanides, Commentary on the Torah, Introduction to the Book of Genesis). Scholem remarks that these names of God constitute a language without a grammar.I6

The established link between signifier and signified can therefore be heuristically modified by viewing the signifier anagrammatically as a com-

l4 God Himself, in Menahot 29b, the famous story about Moses already cited above, is shown ornamenting the letters of Scripture (making crowns or wreaths for them). Scholem in "Revelation and Tradition" interprets Moses's query to God, "Who is restraining you?" to mean, "Why are you not satisfied with the letters as you have constituted them, so that you add to them crowns, that is, the hooks which are found on certain letters in the Torah scrolls?"

l5 If "theurgic" is objected to, a different word should be found; I am not interested here in the differentiation of theurgy and magic or whether "theurgy" comes in only with a Neoplatonic influence.

l6 Scholem, "Revelation and Tradition." Scholem's further remark, that Revelation is es- sentially that of the name or names of God (linked to the signatures or rishumim found in created things), suggests how intrinsic the mystical mode of commentary is that reached exoteric status with the Kabbalah.

bination of letters that yield, by permutation if necessary, another signi- fier. So "Israel" Y = yod, S, R, 'L) is reinscribed as "Y-SAR-EL" to reveal the mystical number 10 (Y = 10) which in the series of Sephirot or sepa- rate intelligences is the last emanation (Prince of the Presence, or SAR- EL).17 The extreme of this combinatory art is found in Abulafia and the prophetic Kabbalah, but normative Midrash never loses sight of that po- tential.18

Isaac Heinemann, it is well known, sees Midrash as "creative philol- ogy." The difference, even known to Milton, between kri (the voiced text) and ktiv (the written, consonantal text) permits pun and paranomasia, such as substituting "builders" (bonecha) where the received text has "sons" (banecha), and far more startling shifts. Such revoweling contains an entire allegory, extracted by readers alert to the inner flow, the lava as it were, of the Sinaitic text. Even consonants can be changed, or trans- posed, at the level of kri. So in Jacob's struggle with the angel, wayyasar; he "fought" (and prevailed), is repunctuated and read as wayyashar; he "sang" (and prevailed)-suggesting a significant fusion of the roles of pa- triarch and angel.

There is usually, of course, a homiletic or moralistic aspect to this play with words. Indeed, the segmentation of the Bible for the sake of com- mentary may have produced fragments memorable enough to circulate as proverbs or to pass more fully into the common language as well as the common understanding.Ig The creation, by an art of division and combination, of quotable fragments ("citemes," as Arnold Goldberg calls them)" allows their recitation in the most varied circumstances: released from their immediate context, the interpreter or preacher can find new

The fullest account of such practice in the Kabbalah may be found in Moshe Idel's Kabbalah: New Perspectives (New Haven, Conn.: Yale University Press, 1988). For gematria in Midrash, see, e.g., Rabbi H. Freedman and Maurice Simon, eds., Midrash Rabbah: Numbers (London: Socino, 1939), 2:734-39, especially on the yod. For information on Abulafia and on the Kabbalah I am indebted to conversations with Moshe Idel as well as to Gershom Scholem's chapter in his Major Trends in Jewish Mysticism, 3d revised ed. (New York: Schocken, 1939). Scholem first pointed out the crucial importance to the Kabbalah of mitz- vot and halakha as more than "allegories," indeed, as often resembling mystery rites of cosmic importance. Idel, in his revision and extension of Scholem, distinguishes between ecstatic Kabbalah (ascendental and unitive) and theurgical Kabbalah (descendental, work- ing more through the mitzvot to harness divine powers).

The hermeneutic devices of Midrash, as they border on permutation, are described by Saul Lieberman in his chapter "Rabbinic Interpretation of Scripture," in Hellenism in Jewish Palestine (n. 4 above).

l9 I owe this observation to Moshe Greenberg. Compare Kenneth Burke on proverbs, The Philosophy of Literary Form, 3d ed. (Berkeley: University of California Press, 1973), pp. 2, 294 ff.: "Could the most complex and sophisticated works of art legitimately be considered somewhat as 'proverbs writ large'?"

20 Arnold Goldberg, "Der verschriftete Sprechakt als rabbinische Literatur," in Schrz? und Gedachtnis, ed. A. Assman,J. Assman, and Chr. Hardmeir (Munich: Wilhelm Fink, 1983).

frame narratives for them in any Bible episode or invent conversations between Israel and God. This has the effect of renewing Scripture words in startling ways. They come at us now from another context as if it were their original site or as if a later text (their source) had lodged in an earlier one, emanating from it as Midrash itself does at a still later point.

Consider how, in the Mekhilta de' R. Ishmael, the verse from Song of Songs, "0my dove that art in the clefts of the rock, let me hear thy voice," is applied to Israel caught between Pharaoh and the Deep Red Sea. "What were the Israelites at that moment like? Like a dove fleeing from a hawk, and about to enter a cleft in the rock where there is a hissing serpent. If she enters, there is the serpent! If she stays out, there is the hawk! In such a plight were the Israelites at that moment, the sea forming a bar and the enemy pursuing. Immediately they set their mind upon prayer. Of them it is stated in the sacred writings [Kabbalah]: 0 my dove that art in the clefts of the rock," and so omz1

Through this re-citing (resiting) of citations we seem to hear something speaking from the "cleft" of the Torah's words. In this uncanny sense also "there is no earlier or later." Contextual unities of time, place, narrative, or grammatical meaning are overruled by a dynamic that points to an intertextual coherence. When the midrashic author associates a passage from the Writings (ketuvim or kabbalah) with a passage from the Penta- teuch, this synoptic method works because the special authority of the Pentateuch actually promotes a sense that the earlier contains the later- a sense that extends itself (for this reader at least) to Midrash, as if mikrah (the canon) were always already Midrash or as if, were we to lose mikrah, it could be largely reconstructed through the "distributive justice" of in- terpreters whose prooftexting skills have opened the canon to a text- faithful imagination.

Let me return to the relation between the Bible as, on the one hand, a language of names and, on the other, a language of common words adapted to human capacities. The interpreter's emphasis on names is based, formally, on etymological traditions that explain place names like Penuel or personal names like Israel and Joseph. The name is converted into a paraphrase and comes with a story. It functions like a hook for

21 Jacob Z. Lauterbach, ed., Mekilta de-Rabbi Ishmael (Philadelphia, 1942), 1:211. My atten- tion was drawn to this passage by Daniel Boyarin's suggestive essay on intertextuality, alle- gory, and the meaning of Midrash by way of "handles" provided by the Song of Songs. See his "Re-citing Scripture," Ohm:A Jewish Journal at Yale, vol. 3, no. 2 (1988); cf. his Intertextuality and the Reading of Midrash (Bloomington: University of Indiana Press, 1990), chap. 7. Does this Midrash recall a speculation that the Song of Songs was revealed to Israel at the Red Sea?

memory, whether or not that was the formal intention. What is important from a hermeneutic perspective is that the name-paraphrase-story pat- tern can suggest a transformation from story and paraphrase back to name, from signified back to a material yet numinous signifier. In that way the common words, under hermeneutic pressure, may disclose an uncommon language, the presence of sacred name or alphabet. And when that occurs, exegesis moves closer to prayer and becomes participation exkgitique.

Midrash, in brief, can motivate and potentiate words as if they were names. By "motivate" I mean that an opaque piece of language, like the proper name "Dinah" in the passage from Tanhuma, is explained by an etymological byword or narrative, and by "potentiate" I mean that this motivation sensitizes the reader to a numinous element in all Scripture words, as if they originated in theophoric voicings. What is evoked is a reversibility that recovers the divine names within Scripture's accommo- dated language.

Consider once more the passage from Tanhuma. It comments on the Bible via a mishnaic rule about a certain kind of prayer. The Mishnah discourages the wish for a male child-so strong in patrilinear societies- once conception has occurred. A prayer expressing that wish is consid- ered tejilat shav: pointless or worthless. Yet both in the Talmud and in Tanhuma the mishnaic stricture is not accepted without qualification. The Gemora asks, "Is there no room for mercy [rahmani]?"that is, Surely God's decision is modifiable? The question seems humorous to the con- temporary mind, and a pun linking "mercy" and "womb" may be active. Yet the theological coherence of the mishnaic ruling has been challenged, and the Talmud meets the challenge.22

Rabbi Yosef, for instance, supports the worth of the disputed prayer, arguing that it was after (v'ahar) Leah's appeal to God, not immediately after her son Zevulun's birth, that God "remembered" Rachel. This argu- ment for the worth of the prayer is rejected, however, on the ground that 'en rnaskirin m'aseh nisim, the precedent of miracles is not accepted for the making of halakha. But the Talmud does proceed to discuss in detail what prayers are permitted and when: in the first three days one may wish that conception take place, from the third to the fortieth day that the child be a male, and so forth. (Rabbi Yosi goes even further in the Tanhuma, in- sisting that such prayers are not worthless until the moment the woman is actually on the birthing stool.) After rejecting, that is, an imputed mira- cle as the basis of law, the rabbis revise an apodictic Mishnah in the light

22 See the Babylonian Talmud, Mishnah 3, and Tractate Berachot 60a.

of their own human and empirical knowledge. A timetable is suggested as in the recent abortion debate.23

Seen in context, then, Tanhuma's Midrash still develops in the shadow of a legal restraint that reaches deeply into personal, albeit socially condi- tioned, attitudes. Whether that conception of law appeals or appalls is not at issue here. Nor whether it leads to a mingling of sublime and triv- ial.24 The field of halakha is obviously very broad, and the field of so- called aggadic or nonauthoritative Midrash equally so. Like secular types of imaginative prose, aggadic midrash is overdetermined. Its relation to Mishnah, folklore, and the Bible itself exceeds a unity imposed by talmu- dic law finding. We can rarely claim that a passage from aggada has been introduced for only one reason.

What I want to emphasize, however, is the wordplay that animates mid- rashic discussion. Such wordplay enters not as a disposable device but as a crucial aspect of the traditionary materials. The continuity of the Tanhuma Midrash with its talmudic and halakhic counterpart revolves less on a dissident opinion than on the retelling of a heartwarming story and the reinforcement of linguistic devices that allow the story in. What is said about Leah is grounded verbally in the Bible: close reading, similar to that of modern language-obsessed exegetes, discovers a textual opacity and correlates it with a hidden-though not esoteric-meaning. What the darshan does with the language of Scripture (with 'aharl 'aheret, Dinah and rahmah) provides a glimpse into a level of action that is more personal and comforting, more "anthropopathic" than what transpires at the ex- plicit narrative level. The midrashic habit of segmentation, moreover, works here so brilliantly that it seems to be a technique inspired directly by the text rather than ingeniously imposed on it. We are encouraged to read wajishma 'eleyha ("and He listened to her") as if the verse had Leah and not Rachel as the object of the verb, a reading motivated by hearing in 'eleyha the name "Leah." The process that enriches meaning (the signifier-signified axis) also enriches the word as a phonemic rather than semantic entity (the signifier-signifier axis), for since there is no obvious

23 Yet what is science here, and what theology, is hard to say: for Thomas Aquinas, too, a male child becomes a soul by the fortieth day. (For a girl it takes longer!)

24 The tendency to interpret down as well as up, and perhaps the redactors' unwillingness to let go of any interpretation that illustrates human behavior, can lead Midrash aggada to include silly or dubiously humorous exegeses. For example, see the following, in the Buber edition of the Tanhuma: "(Genesis 30:23) Then she said: God has taken away my shame. What is the meaning of Has taken away? Simply that before a wife gives birth there is shame found within her house. How? When she breaks a vessel in her house, whom does she have to blame? When she gives birth, she blames her child. She therefore said: God has taken away my shame" (Salomon Buber, ed., Midrash Tanhuma [Vilnius: Wittwe & Gebriider Romm, 18851).

agrammaticality the verbal thickness is rather an anagrammatic discovery of the interpreter. Sensitized by this kind of reading, every word in Scrip- ture could become a name or a crux open to elaboration.

Thus Tanhuma's commentary centers on an adverb ('ahar); a noun (rahmah) not in the Bible episode but introduced through talmudic dis- cussion; also perhaps the root zakhor; male, with its link to jidhor (He remembered, i.e., "maled" Rachel); the name "Dinah" construed as a verb ("she strove") and the name "Leah" heard in an inflected pronoun ('eleyha). No one can read such interplay of Bible, Talmud, and Midrash without slowing up at every word and marveling how this augmenting of Leah's name contributes to God's name, which the Bible writes out, as it were, in every particle of its language.

There is, then, a double movement, which maintains at one and the same time the familiar, grammatical sense of the Bible words and suggests that any or all of them are open to an anagrammatic, divinatory read- ing.25 The secular reader also knows about this doubleness, for it is differ- ent only in degree and not in kind from his own interpretive ventures. Ferdinand de Saussure's experiment with "hypograms" and such coin- ages as Walter Benjamin's "Agesilaus Santandern-perhaps inspired by Jewish name mysticism via Gershom Scholem-are dramatic instances that point in the same direction. Midrash invalidates the distinction be- tween the muses as Daughters of Memory-here textual memory-and Daughters of Inspiration.

Can I gather what I have said into a single hypothesis? The attempt to put imagination in the service of memory reverses an emphasis in Helle- nistic culture, if we recall the opening invocation of the Iliad and charac- terize fiction as putting memory in the service of imagination. More tech- nically stated: the art of memory that Midrash may reflect depends on textual instances (topoi or citemes) rather than images (imagines) or places (loci). This is no more than a guess on my part; yet the mnemo- technics operative here do not seem to rely on "visual impressions of al-

25 Another way of putting this is that in Midrash (but not, perhaps, in the extreme permu- tations of prophetic Kabbalah) the idea of a divine (pure, absolute) language did not lead to a rejection of the accommodated or current language. A strictly analogous problem in contemporary thought is the relation of scientific language (in its as yet unrealized perfec- tion) and ordinary discourse. The perfect language or metalanguage of science, writes Henri Lefebvre, "would be a 'pure' construction, closer to an elaboration in logic pushed to its final term than the natural and spontaneous 'expression' of feelings, emotions, passions, images. It could turn out . . . that this perfectly rational language is characterized by the displacement or elimination of stops, blanks, cuts, pauses, which abound in our spoken or written language. The latter segments and punctuates our 'expressiveness' in language; it introduces articulations but also haltings, uncertainties, choices that are doubtless arbitrary (between words, turns of phrases, ways to compose a discourse). . . . A mathematical dem- onstration is obviously not spaced and articulated like a discourse. Its continuity follows through without gaps." See his Le hngage et la societe' (Paris: Gallimard, 1966), chap. 1.

most incredible in tens it^."^^ The visuality directs itself to, and perhaps breaks itself on, scriptural words and letters, even their shape. I describe a tendency, of course, not something invariable. What is important here is that imagination, whether it sponsors or represses images, is disciplined by and yoked to the sacred in theform of a text, at least during the talmudic era. So the sharpness and imaginative acuity shown by Midrash, its fertil- ity and wit, remain commentary; they find their way back to the words of the Bible, and however many meanings are discovered, these never re- place Scripture. While some interpreters play with fire, they continue to recognize the Torah as God's Promethean gift to man. Moreover, though passages may be defamiliarized (which can actually aid memorization), by being taken out of their immediate context and paired intertextually through the devices ofgematria or pun and paranomasia, the "infinity" of interpretation suggested does not-in rabbinic Midrash-favor mystical over everyday kinds of meaning. (The balance shifts with Kabbalah, but that is another story.) Like a good psychoanalyst, the redactor allows his attention to float or distribute equally: the impression, in fact, is that sto- ries and similes from daily life predominate. As in Soloveitchik's reading of the two accounts of creation, Midrash speaks with Bible or God in intelligible, even intimate terms and, so, produces either a halakha or an aggadic flourish that puts the strangest experience within reach.

From a simple starting point, the literary study of Midrash, we have arrived at a complex result. Let me summarize and conclude. Literary study that has Midrash as its subject faces a number of difficulties. The first of these is that contemporary thought is moving away from the liter- ary-from essentialist definitions that are said to be ideologically moti- vated and favor a fixed, eurocentric canon. The very existence, at the same time, of a writing like Midrash, that cannot be said to have a literary purpose yet is intensely textual and intertextual, helps to expand the contested notion of the literary. What interests literary scholars in Mid- rash is, therefore, that it cannot be constituted as literature, that it chal- lenges narrow aesthetic or genre-conscious theories.27

26 See Frances A. Yates on "The Classical Art of Memory," in her The Art of Memory (Har- mondsworth: Penguin, 1966), p. 19. David Stern suggests, however, that some elements in the important type of parable called the king-mashal "represent the literary equivalent in Rabbinic Judaism to the pictorial art of early Christianity." See his Parables in Midrash: Narra- tive and Exegesis in Rabbinic Literature (Cambridge, Mass.: Harvard University Press, 1991), pp. 94-97.

27 This does not mean, ofcourse, that important types of discourse like thepetich (proem) should not be isolated and defined; "literary" indicates a development of form-criticism liberated from source-criticism, though not from a speculation about the genesis of struc- tures that could provide insights into their historicity.

(Another way of stating the originality of Midrash is to point out its exceptional position as a form of commentary with strong residues of oral transmission which elude a concept of the literary based on manuscript and especially print culture. Midrash, even in the age of manuscript cul- ture, is allowed a freedom characteristic of oral tradition, which does not congeal the transmitted communication but leaves room for diversity of performance. In Midrash, as in oral forms, "the integrity of the past [is] subordinate to the integrity of the present." With one, essential excep- tion: the integrity of the letter of Scripture must be preserved. The text commented on is accorded sacred status and, though itself reflecting ear- lier forms of oral transmission, is not allowed further change. Thus Mid- rash, which tends toward a retelling of the Bible that might eventually have replaced it, becomes instead a spiritual means to preserve a letter- perfect Hebrew Scripture.)

A further difficulty springs from pious fears about secular misuse. Yet the scholarly study of Midrash was among the first fruits of the nineteenth-century "Wissenschaft des Judentums." Zunz's book on the rabbinic homilies remains a classic.29 What is strange and exciting is that in our time a form of close reading has developed which displays some features of Midrash. So far, however, the similarities between rabbinic Midrash and this new Midrash, inquisitive, open, highly text-dependent, have elicited more antagonism than sympathy.30 There is already an at- tempt to disqualify the secular mode by imputing to it mystification or privatism: ingenious academics and punsters should not be compared, we are told, with rabbis whose utterances are historically situated and make a truth claim.31

These turf-protecting charges assume that the historical situation of

Walter Ong, Orality and Literature: The Technologizing ofthe Word (London: Routledge, 1982), p. 48. 2g Leopold Zunz, Gottesdienstliche l'ortrage der Juden (Frankfurt am Main: J. Kauffmann, 1892).

30 Exceptions are Susan Handelman's The Slayers ofMoses: The Emergence of Rabbinic Inter- pretation in Modern Literary Theory (Albany, N.Y., 1982); and Boyarin's Intertextuality and the Reading of Mzdrash.

31 Howard Eilberg-Schwartz, in an important article, "Who's Kidding Whom? A Serious Reading of Rabbinic Word Plays," Journal of the American Academy of Religion, vol. 55 (1988), reads rabbinic paranomasia from the perspective of "linguistic assumptions (that) would have made a dichotomy between activities called philology and word-plays inconceivable. Any linguistic transformation that we might designate as a pun or word-play would consti- tute, from the rabbis' linguistic understanding, a reasonable interpretation of a word." It is "reasonable" because they believed in the pre-existence of Hebrew and Torah and the interdependence of both. This belief produced what he usefully calls a "molecular" analysis of biblical Hebrew in terms of consonantal recombination and revoweling: "There is a seemingly unproblematic level that tells a story. But the words that comprise this narrative are linked in various ways to one another and thus force the reader onto the level of com- mentary." I can follow Eilberg-Schwartz so far. But the rabbinical assumptions he offers as

those who are evolving the new Midrash is not worth examining. Yet why do paramidrashic readers come on the scene? And why, long before Derrida, is there a revival of interest in Patristic typological commentary as a complex literary form and, beyond that, in a critique of any method of reading that insists on "unity" as the literary work's highest value de- spite ambiguity, polysemy, and complexity of reference?

If we are troubled today by unifying modes of commentary, it is be- cause, in the past, they have been triumphalist. Midrash, however, is sub- tle in this regard. It encourages the intellect to find room in the strictest laws or else to support by proof-texts the most imaginative turn on Scrip- ture. It is as if the original text had become unreadable except through an extreme fragmentation that paradoxically confirms its unity. The frag- mentation deconstellates phrases and words, producing what Maurice Blanchot has called an "icriture du disastre" yet installing all the more firmly the law of citation. "If citation," Blanchot writes, "in its fragment- ing force, destroys in advance the text from which it is not only torn but which it exalts to the point of being nothing but this having-been-torn- away, then the fragment without text or context is radically ~ncitable."~~

It is crucial to separate the concept of unity from that of triumphalism by a hermeneutics of the faithful fragment, and the disaster of the hurban may have contributed to just this separation in Jewish commentary. The multivocal or aspectival qualities of Midrash do not seem to be the result only of historical distance-of our ignorance of the mode, its forms or genres. Goldberg can talk of the "aporia" of midrashic texts and their special linguistic and literary characteristic^.^^ How to interpret that "apo- ria" without oversimplifying it by referential history or a purely contem- porary appropriation, is the critical task.

superintending the interpretive process do not determine it. His anxiety about the notion of play leads him into a restrictive intentionalism that deprives the interpreter of a freedom within those assumptions (not just against them), a freedom without which the entire rab- binic enterprise of interpretation would lose much of its interest and integrity-for if the divine words were totally presignificant, they could not be used to argue with or even "de- feat" God in the drama that plays itself out between God and man but also between conflict- ing interests in the community (e.g., the Schools of Hillel and Shammai), each of which claims authority.

32 Maurice Blanchot, The Writing of the Dirasteq trans. Anne Smock (Lincoln: University of Nebraska Press, 1986).

33 Goldberg (n. 23 above), p. 139. In his careful "Introduction: Law as Literature," to Semeia, vol. 27 (1983). William Scott Green points out that halakha too should be studied as a cultural construction, as texts that are "sources" rather than simply data. He emphasizes the "literary strategy of juxtaposing opposites without resolution" in halakhic law finding, which makes it difficult to treat it as mimetic, as an unmediated form of actual rabbinic practice. "Indeed, the structure and form of rabbinic writings suggest that our first reading of these documents must see them not as records of collective behavior or institutional legislation, but as works of intellect and imagination."

Yet we need not give up all speculation about the mimetic character of midrashic style, as long as it is taken to be heuristic rather than dogmatic. The impact of the hurban has to be considered, but also the shift in con- temporary intellectual life from identity-philosophies to theories of dif- ference based on an appreciation of the intertextual character of writing, and the light this throws on psyche and structure of belief. If quality of belief matters in midrashic texts as well as belief itself, if interpreters test themselves and their listeners over and again, this would recall a modern revolt against the automatism of creeds. "No forcing of image, plan or thought," writes A. R. Ammons, admiring how nature undoes art. At the same time he senses, as we do in Midrash, a darker side to this apparent freedom:

.. .all possibilities of escape open: no route shut, except in the sudden loss of all routes .. . 34

It is true that there is also a deeply unmodern aspect to Midrash. For the self-consciousness we characterize as "modern" often seeks to es- trange itself through the deceptive "anti-self-consciousness" of fiction. Ev- ery trace of secondariness is then displaced or denied. "To forget the text which has engendered the text. We write starting from that forgetful- ne~s."~~

There is no such forgetting in either the old or the new Midrash, which are citational to an extreme degree. They reject amnesia and cryptomnesia. Even when, as in Jabks, the absolute authority of Scripture is lost, words of the other are recalled, or simulacra that continue to haunt us.

Once the religious thinker understands that the literary study of Mid- rash does not turn everything into aggada and then into a web of language-inspired forms (that linguistic aspect is fascinating, of course, and poses an obvious ~hallenge),~~

then it is possible for Midrash to be studied comparatively, as an exceptional form of commentary. It is excep- tional because of its close yet supple relation to a canonical text, because of the way exegesis turns into exegesis plus, or "literaturen-in short, because of matricial qualities that allow us to see the creative yet text-

34A.R. Ammons, Corsons Inlet: A Book of Poems (Ithaca, N.Y.: Cornell University Press,


35 "Oublier le texte qui a enfant6 le text. Nous kcrivons ipartir de cet oubli" (E. Jab&, Le lzvre du dialogue [Paris, 19631).

36 If the anagrammatical reading of Scripture points to a "revelation" of the divine names or to a mystical alphabet, a logos that created the world and still has creative (or destructive) force, then there is a strange convergence with a radical kind of deconstruction that sees literature as "revealing" nothing more and nothing less than the remarkableness of the alphabet we already have, and by whose combinations everything that we do or think is ar- ticulated.

permeated mind at work. The dichotomy between (primary) text and (secondary) interpretation is alleviated: exegesis augments instead of dis- placing the received symbols and words. There is a fictional momentum, but it does not become freestanding even in the mashal or Midrashic para- ble, a form that goes farther in devious naivete than the Homeric simile. Just as the mashal can be accused of stylistic insubordination, so Midrash is an insubordinate type of exegesis,37 a wonderfully inventive yoking of one text to another by a virtuosity that saves the whole-in Soloveitchik's words, the "unity and integrity" of Scripture.

As for the future, and the field that may eventually be created by the awareness that Midrash and literary study take of each other, I can say only one thing with confidence. A knowledge of Midrash will prove more interesting for the literary critic than a knowledge of literary criticism for the scholar of Jewish texts. Ask not what deconstruction may do for Mid- rash; ask what Midrash may do for deconstruction.

37 David Stern's Parables in Mzdrmh: Narrative and Exegesis in Rabbinic Literature (Cambridge, Mass.: Harvard University Press, 1991),in a section called "Theorizing Midrash" (pp. 6367), shows finely how "in midrash, exegesis may be the mashal's occasion, but its exegetical occasion does not exhaust the mashal's meaning, which goes far beyond both exegesis and narrative alone, lying instead in their intersection."

  • Recommend Us