Why do we continue to find myths so fascinating? The ascendancy of science and the triumphs of modern Western thought leave a large gap, an unsatisfied need, which we attempt to meet with beliefs of very different kinds and very different origins. The Hubble telescope has not killed astrology; antibiotics have to contend with herbal remedies and acupuncture; men and women are still carried off by mysterious creatures from other worlds (though nowadays in interplanetary flying machines rather than in fiery chariots) and return after undergoing experiences which ordinary language is powerless to convey. And the study of myth flourishes as never before. The word “Western,” of course, gives a clue. We enjoy the practical benefits of those triumphs, but we flinch from the triumphalism—rational, patriarchal, colonialist—that often accompanies them, and we hope that comfort and meaning can be found in the very different wisdom of other times and other places.

Wendy Doniger is Mircea Eliade Distinguished Professor of the History of Religions at the University of Chicago and a prolific writer on Indian mythology and on myth in general. She makes no secret of the general direction of her work in both these books, which is political, and on the left. “Winners,” she writes, “have, alas, usually been on the right (though not in the right),” and, of her work, “In pursuing the multivocal, multicultural agenda, we must face the implications of the fact that we use other peoples’ stories for our purposes….” There is a great deal of explicit feminism in her treatment of myths, and patriarchy, sexism, colonialism, and stereotyping all come in for some sharp criticism.

Splitting the Difference is a fascinating book, full of bizarre tales of men and women, gods and goddesses, being duplicated—two for the price of one—or divided into two parts, sometimes the heads of two people being transposed, sometimes a person being transformed from one sex into the other. It is charming to read that “the Tibetan Vinaya casually remarks that anyone who changes sex more than four times a month can’t become a monk.”

The Greek heroine Helen notoriously ran away from her husband with another man to Troy. That story, originally a divine tale of the abduction and recapture of a goddess of fertility, soon became troubling. Why all that fuss, and all those deaths, over a woman no better than she ought to be? And how come she continued to be respected, and indeed worshiped, in the classical period, back home in Sparta? Quite early we find the story that really she did not go at all: it was a double, made by the gods, which went: Helen herself was virtuous. Poets had great fun with that idea, from Euripides to Hofmannsthal.

In India, too, the wife of Rama, Sita, (in the Ramayana) was abducted by a demon and finally brought back, untouched. With time, her virtue, which looked a little fragile, was defended by the invention of a double, Sati, to whom the abduction “really” happened. So far, so good; but poets were not satisfied, and the double began to take on an existence and an erotic career of her own, finally becoming a significant mythological figure, with interesting reincarnations. In another story, Krishna was beloved by all the milkmaids (gopis), and he multiplied himself so as to make love to all of them simultaneously; or he duplicated them, leaving a double in bed with each of their husbands; or, perhaps, with time and moral unease, he made love to the doubles, and so avoided adultery, felt now to be unbecoming to a god.

Females are very liable to doubling in erotic contexts. Doniger sees here the woman’s attempt to distance herself from acts of rape and abuse (“It wasn’t me that it happened to!”), but we see also the male demand that the woman who belongs to me should not have gone astray, even if appearances are against her (as they were against Sita and Helen). Sometimes we come very close to the territory of multiple personality disorder, interestingly discussed here.

Other stories present the woman with an identical pair of male pretenders, often one human and one divine: she must somehow defend her chastity by choosing between them. In Western mythology the chief instance is Alcmena, mother of Hercules, to whom Zeus came in the form of her husband. We have a remarkable Latin play by Plautus on the theme, Amphitryo: a comedy, but with the character of Alcmena dignified and even tragic as she struggles to understand what is happening and incurs the angry jealousy of her husband. The subject appealed to Molière and Kleist and Giraudoux, who play many witty and touching variations on the ancient theme. Especially tantalizing is the question: Did she really know who it was, or was she really deceived?


In India, and occasionally in Greece, we find the motif of the woman choosing the mortal rather than the immortal lover, because of their shared mortality, as Odysseus, in the Odyssey, chose his old wife Penelope rather than immortality on an island with the charming goddess Calypso. Doniger makes the point that the men (like the returning Odysseus) prove their identity by their deeds, the women by their wifely fidelity to their husbands, while in the case of the exchange of heads, women are shown doing it in response to male violence, but men in order to avoid death.

Changes of sex are extremely common in Indian myth, much rarer in Greece; but even in India it appears that one is not reborn to a different sex—it is easier, as Doniger puts it, for a man to be reborn as a male ant than as a woman. Even great gods change sex repeatedly, and some human persons have the privilege, if that is what it is, of experiencing both sexes in series. One element here is curiosity: What is it like for the other sex? Why can’t I ever know? Somebody must have been able to do it—Teiresias was his name; and his story… That train of thought is not so very different from such fantasies as being able to fly (Why not? Somebody did it—Daedalus was his name; and his story…); or being invulnerable in battle, as Siegfried was, except for one place; and he also played the bed trick on Brunhilde, coming to her on the first night in the place of her husband Gunther, a deception which she never forgave.

The fascinating stories Doniger tells are bracing to read, energetically written. She has an enviably wide range of knowledge about India and Greece and modern art forms, and while there are arguments in this book which will not command universal agreement, anyone with any interest in these matters will certainly learn a lot, and derive a lot of amusement, from Splitting the Difference.

The second book by this admirably prolific scholar is more general, more polemical, and more seriously controversial. The implied spider of the title is an extension of an idea in the Upanishads, that God is a spider emitting the world from inside himself. For Doniger, the invisible spider behind the myths, the source of all the stories, is

the shared humanity, the shared life experience, that supplies the web-making material, the raw material of narrative to countless human webmakers, authors, including human anthropologists and human comparatists.

The emphasis on shared life experience is not casual. It is important, because Doniger is anxious to avoid both of a pair of seductive opposites, “the Scylla of universalism and the Charybdis of cultural essentialism.” Both have distasteful political implications: the former suggests that all men, all women, are always the same; it is the parent of sexual and social stereotypes. The latter maintains that different communities, different societies, even different races are immutably fixed in their peculiar natures; it begets Orientalism and racism.

Instead of alleged universals, we are to explain resemblances between stories by invoking shared human experiences: they respond to shared human needs. It is at once clear that this comes dangerously close to universalism. Such experiences as those of childbirth and motherhood, for example, or of male aggression, which all play their part here, can be called “universal” almost as plausibly as “shared.” Doniger herself makes the point, when she insists, rather nervously,

I have argued that the great cross-cultural themes of myth are not given or natural, that they are merely widespread responses to shared experiences that are, in a sense, given and natural.

Merely widespread,” “in a sense natural”; the Scylla of universalism is very close, as we navigate this narrow and rocky passage.

It is part of the method that the comparative scholar is not to be restricted to what we might think of as myths. Modern literature is equally relevant, and there are discussions of such works as Dr. Jekyll and Mr. Hyde, which is declared to be both homosexual—demonstrating “a kind of extreme homoerotic narcissism—“ and also guilty of Orientalism (“When Hyde tramples down a little girl he is twice referred to as a Juggernaut,” a term referring to an incarnation of the god Vishnu who so excited his followers that they threw themselves under the wheels of his cart). A more plausible view might be, I think, that this story is much less concerned with sex, or with the Orient, than with Calvinism, temptation, and original sin. In The Picture of Dorian Gray, too, “Orientalism and colonialism are manifest…in the form of the oppressive presence of luxurious Oriental spoils—fine china, Persian carpets, Indian emeralds and rubies.” It is apparently to be assumed that all these goods—“spoils”—were not paid for. The movies are also effectively drawn on for parallels; in fact “sometimes later texts provide more fruitful material for this kind of study.”


Salman Rushdie is applauded for saying, of myths like those of Niobe, Arachne, Prometheus, that they “seem to say that we must not challenge the gods (or, by implication, those in power), for if we do they will destroy us….” But, we want to protest, the gods are not simply a cover for those in power. Niobe, who insulted the mother of Apollo and Artemis and was punished with the loss of her children, was herself a queen, and it was her royal position that emboldened her to be saucy to the gods. The divine really was different, and a central point of the myths is that it is the mighty who are tempted to forget that terrible distinction—as we might be, were we in their position.

Astonishing on a larger scale is the assertion, taken with approval from Roberto Calasso, that myth includes the opposite of its actions:

The hero kills the monster, but even as he does so we perceive that the opposite is also true: the monster kills the hero.

Now that view may be a stimulating fantasy but it has no place in a serious discussion of the subject. There is no myth I know of in which the monster kills the hero. What kind of story would that be? The dragon cannot kill Saint George; Grendel cannot kill Beowulf, nor can his fearful mother; the Babylonian monster Tiamat must lose to the hero Marduk; the smart money was on Hercules against the Hydra, and on Siegfried against Fafnir. And so on. Nor can the tale of the Quest, that great mythical pattern, end with the hero slinking home, baffled and empty-handed (“the opposite is also true”). The possibility of defeat must exist, but it cannot actually happen. Nor yet can we imagine a myth of the Garden of Eden type, in which the God forbade the people to eat a certain fruit, and they obediently didn’t eat it, and went on not eating it. God himself would share our sense of frustration.

The point is wittily made by Ernest Bramah in “The Story of Hien and the Chief Examiner” (in Kai Lung’s Golden Hours, 1922). The storyteller Kai Lung, whose boast it is that he knows and can recite a story to suit any situation, is challenged: “The test…is that he relates the story of a presuming youth who fixes his covetous hopes upon one so far above his degraded state that she and all who behold his uncouth efforts are consumed by helpless laughter. Ultimately he is to be delivered to a well-earned death by a conscientious official whose leisurely purpose is to possess the maiden for himself.”

The story he tells in response is of course quite different, ending happily for the lover and disastrously for his powerful enemy. There are protests; unperturbed, he replies:

Herein it is one thing to demand and another to comply, for among the Platitudes is the admission made, “No needle has two sharp points.” The conditions which the subtlety of Ming-Shu imposed cease to bind, for their corollary was inexact. In no romance composed by poet or sage are the unassuming hopes of virtuous love brought to a barren end or the one who holds them delivered to an ignominious doom. That which was called for does not therefore exist.

And something not entirely dissimilar goes for myths generally.

It is a leading concern of both books to use the myths in order to recover from them other voices than that of the surface interest of the narration, and especially the voices of women, “to bring women’s voices into texts written by men.” Most of the texts as we have them are probably the work of men, but that can be circumvented by careful reading and attention to other implications which are present to be discovered. And in any case,

Once the story is told, we may see in it simultaneously a woman’s point of view and a man’s, no matter who told it. The author of a myth is a tradition, not just one human male; and traditions have women in them too.

One interesting argument, which makes use of the license to disregard chronology, is that we can not only use earlier texts to illuminate later ones, but also reverse the arrow of time and use (say) Shakespeare to help understand the Bible. A striking instance is the episode in Genesis of the seduction of Judah by his daughter-in-law Tamar, here compared with All’s Well That Ends Well; but this treatment leaves the reader feeling uneasy.

In the biblical narrative Tamar has had the misfortune to have her husband, the eldest son of Judah, struck dead by God. His brother Onan is then shoved forward to fulfill the traditional obligation, take his place, and raise children for his dead brother. Onan dislikes this arrangement and “spills the seed on the ground,” thus giving his name to a solitary sexual practice; an offended deity strikes him dead, too. Asked to produce a third son as husband for this formidable bride, old Judah understandably hesitates. So Tamar takes matters in her own hands, disguises herself as a harlot, and seduces Judah as he passes by, taking the precaution of getting from him some undeniable tokens. Duly found to be pregnant, she is denounced and threatened with burning, but she is able to vindicate herself by showing that her sexual partner was none other than the patriarch. He is forced to admit that she is justified.

In Shakespeare’s play the young nobleman Bertram thinks himself too grand for the bride, Helena, daughter of a medical man, whom the king forces him to marry. He goes off to the wars without touching her. She follows in disguise and succeeds in sleeping with him without his knowledge, by pretending to be a local girl whom he is planning to seduce. Doniger has the interesting idea of using Shakespeare’s play to fill in aspects of the Genesis narrative which did not interest the rabbis and do not appear in the biblical text, most notably Tamar’s feelings about what she does. By reading Shakespeare,

we can imagine what a woman in Tamar’s situation might have felt, and imagine a woman’s voice in the Bible asking that, and being suppressed…. I would hope that scholars of the Bible might come to read All’s Well for new insights into the story of Tamar….

The motif of the bed trick is common to both stories; but that does not really license Doniger to say, “In both stories a woman whose husband refuses to sleep with her disguises herself as someone else and tricks him into bed.” The stories are vitally different. Shakespeare’s Helena performs her trick in order to sleep with the husband whom she loves, and she wins him for keeps. Tamar’s story is much more complicated and more local: she tricks not her husband but her old father-in-law, in order to get a child for her dead husband. All’s Well may offer “insights into the psychology of sexual rejection” (though in fact Helena is not very explicit on the subject), but they are inevitably of a very different kind from those suggested by the Tamar story. Shakespeare’s people do not live in the world of the “levirate,” the traditional obligation to take the place of a dead brother; while as for Judah, we read, “And he knew her again no more.” Tamar is not out to secure a husband, and she does not get one. Nor does love enter into it.

A better parallel to All’s Well would be the story of Jacob and his two wives: he wanted Rachel, the younger sister, and worked for her, and after seven years of toil was put to bed with his bride, and in the morning discovered that it was the sister whom he had refused, the elder, Leah. That really is a story of sexual rejection in the more central sense, as poor Leah continues to be “hated” by her husband, though she bears him a series of sturdy sons, and we see her and Rachel competing for the patriarch’s affection (no avoidance of the female viewpoint there). And we might bring in another Shakespearean play, too: Measure for Measure, in which the rejected and lovelorn Mariana sleeps with Angelo in place of the woman he really wants, the icy Isabella, and wins him as her husband. It is true that the biblical narrative does not fill in the emotions of Tamar, but it does not follow that we can supply them with confidence from another tale in a very different setting.

What this illustrates is the extreme slipperiness of this sort of comparison. The question suggests itself: When is a story the same story? Or indeed: When, and in what sense, are two narratives comparable? Doniger faces it, but in the end she is forced to admit that there is a large element of subjectivity in our choice of stories and approach. Here is another example, developed at considerable length: the story of Adam and Eve and the Fall. We are offered what is called “an alternative reading” to the “dominant reading” in our text:

Other tellings of the myth cast it differently (a benevolent Goddess in her form of life-giving serpent and tree, giving the blessing of the fruit of useful knowledge that makes human life possible). This alternative reading of Genesis implies not a Fall but a progression from the Garden of Eden (a place of ignorance and constriction) to the wider world….

This “reading,” it has to be admitted, is never actually found in any ancient text, although it “is often said (usually by feminists) to have existed in the ancient Near East alongside the biblical text.” Also the word “useful” seems to have wormed its way in uninvited: Genesis speaks only of the knowledge of good and evil, a very different thing.

What is going on here? The unasked question is, surely: What is this story about? Doniger knows. It can be reduced, we read, to something like this: “A woman and a serpent gave a man a fruit to eat, created by a deity, which brought him knowledge.” That is “an imaginary text, a scholarly construct that contains the basic elements from which all possible variants could be constructed”; it is “a neutral structure,” “a myth with no point of view at all.” And yet:

This already tells us a great deal. It tells us that women rather than men are responsible for change…, that women rather than men supply the original food….

Already, then, we have succeeded in drastically changing the sense of the text which we actually have, and reading it in a more acceptably feminist sense. But even more is at stake than it seems, for Doniger’s point is that “any cross-cultural analysis will have to be sufficiently multivalent at least to acknowledge the validity of all these sets of variants,” including, in the case of Adam and Eve, not only versions which make Eve a beneficent goddess, but also Mark Twain’s wisecrack that Adam only wanted the apple because it was forbidden—“the mistake was in not forbidding the serpent; then he would have eaten the serpent”—and a cheeky poem “composed by Royall Tyler in 1793,” in which the fruit is the male organ, and Eve identifies both Death and the Fall with her partner’s detumescence.

But objections crowd in. Why should we suppose that the original form of the myth was “neutral” and had no moral point? Is it not at least equally plausible that the story came into existence in order to answer very important questions: Why must we work? Why do we feel shame? Why are women subordinate to men? And, above all, the most central and terrible question of all: Why do we die? Myth expresses such concerns by saying that, of course, originally we did not have to, but then there was a change, and fundamental to that change is the idea of disobedience, of sin. To excise the notion of fault is not merely to remove an optional extra, but to lose the point of the whole.

“It is not the task of myth to moralize explicitly,” says Doniger mysteriously; but it is hard to see why such an assertion should be true. Central, surely, to this one is the assertion that our first ancestors forfeited, by their fault, the happy and deathless life which was—must have been—within their, and our, grasp. The idea that “women rather than men are responsible for change” seems, by contrast with these large concerns, both eccentric (do we in fact see women bringing about most changes?) and peculiarly of our own time. Can we imagine that anyone really attached importance to saying that, three thousand years ago?

And here we bump into another difficulty. What is a myth? Doniger starts out by declaring, plausibly enough, that “all myths are stories, but not all stories are myths.” She goes on with the thoughtful proposition that

In its positive and enduring sense, what a myth is is a story that is sacred to and shared by a group of people who find their most important meanings in it, a story… about…an event that continues to have meaning in the present because it is remembered….

It is not clear how that definition of myth, which is a good one, is compatible with the practice of ascribing to a myth, with equal validity, all the meanings that have been built onto it by parodists, jokers, and others who were far from taking the old story seriously. For the aim is now to maintain “each of several conflicting views in a balanced tension, like chemical elements resolved into a suspension rather than a solution.” Perhaps what we should rather say is that the old myths can indeed be treated in this way, but only at the cost of ceasing to be, in any significant sense, myths. So treated, they will not be sacred and important to any group of people. They will just be a set of more or less interesting, suggestive, and amusing tales.

It might have been helpful to distinguish more sharply myths on the one hand and motifs on the other. The story of what Tamar does with Judah is not really the same as the story of Helena: it is a different myth, which makes use of a similar motif (the bed trick: he doesn’t know whom he’s doing it with). Such a motif can be comic or serious in its application. And another distinction which Doniger does not quite consistently keep in mind is that between a myth and a work of literature which tells it or uses it. In particular tellings of stories we can indeed find different viewpoints: that is in fact one of the marks of a serious and admirable work of art.

In the Iliad of Homer we find sympathy for the defeated Trojans, and sympathy for the unhappy wives and mothers of the warriors who fight and die to win glory. That is part of the reason why we continue to find the poem powerful and moving. But all that is true of Homer’s poem, not of the myths themselves that it draws on. Hecuba, Queen of Troy, curses the hero Achilles, who has killed her son; Penelope, in the Odyssey, curses the war which has kept her husband away for twenty years. In these great works of literature we are in no doubt that the voices of the oppressed and suffering are present, and not merely because while “Homer tells the Iliad from the vantage point of a dead white male, [he] allows Thersites to show us the underside of the Trojan War.”

We can also detect them in less great compositions. Doniger skillfully shows how we can find something of the sort in such writings as the medieval French prose satire Les Quinze Joies de Mariage, which is on the surface a pretty misogynistic work, where utterances put in the mouths of women reveal their grievances against unsatisfactory husbands. So does the Kamasutra, low as that work’s general view of women is. But that has nothing specifically to do with myth: neither of those is a mythical work. Doniger observes,

It is interesting to note that in both cases—the French and the ancient Hindu—we find meanings that have value for us only by transcending, if not totally disregarding, the original context and, in fact, the fairly patent intentions of the author.

That is doubtless true of myth, too; but there she prefers to say not that we are transforming the Adam and Eve story by disregarding the obvious intentions of its author, but that the correct comparative procedure is to treat all versions, however late or frivolous, as equally valid.

We should, on this view, emulate in our criticism the Hindu myths themselves:

By refusing to modify its component elements in order to force them into a synthesis, Indian mythology celebrates the idea that the universe is boundlessly various, that everything occurs simultaneously, that all possibilities may exist without excluding each other… [that] untrammeled variety and contradiction are ethically and metaphysically necessary. And what is true of the myth is also true of the world.

That is powerfully said. But two limitations strike us. One is that Doniger, suddenly alarmed (it appears) by this infinite vista, hedges surprisingly:

But there are limits to the pluralism of interpretation…. It is possible both to misinterpret and to misuse myths. We misinterpret them whenever we ignore context and difference, and we misuse them in the ways to which post-colonial discourse has alerted us. We can therefore eliminate some interpretations, but we should still be left with more than one.

This sounds cryptic, and it is not followed up. After all, is there not to be infinite freedom of interpretation? Are only interpretations that are in some way colonialist forbidden? And why?

The second limitation is more serious. It is not easy to bring the welcoming of complete, world-encompassing variety into harmony with another strongly expressed wish: to use knowledge of the myths for specific political ends. There is an agenda in Doniger’s work, and it is to be multicultural, feminist, anticolonialist. “The cross-cultural comparison of myths,” she writes, “is pragmatically possible, intellectually plausible, and politically productive.” Structuralism has indeed been criticized for not being political. But Doniger thinks that structuralism needs to be brought back to deal with “domination and subversion,” and to advance into “political, theological, ethical territories.” “Comparison of myths from other cultures opens up our political vision,” and “Hindu texts …challenge our ideas about gender,” and so on. But something is surely loose here. Whose political vision is it that is to be opened? Not that of enlightened academics in Anglophone academia. Whose ideas about gender is it that are to be challenged? Not Professor Doniger’s, but those of the rest of us, sexists and essentialists that we are (or are assumed to be).

We may well wonder what could be the effect, in political terms, of taking seriously myths which present opposite visions as equally and simultaneously true. If we did take that idea seriously, it surely would not lead to any kind of involvement in, or even desire for, political change. If absolute rule and popular will are equally valid, if parity of the sexes is no more valid as an ideal than the misogyny which undeniably pullulates on the surface of so many of these myths, if life itself is simultaneously real and illusory—these myths “challenge us simultaneously to see that our lives are real, and to see that they are unreal”—then what can be the meaning of progress, or even of change?

That may seem a rather captious criticism. We know, after all, that when we say that something is “liberating,” we don’t mean that it will liberate us: we are liberated already (or we shouldn’t be so keen on liberation). “Shocking,” “challenging,” “subversive”: when we use this familiar vocabulary it does not mean that we are to be disturbed.

But seriously: it really does seem strange, in two books devoted to Indian myth, and to the salubrious power which it should have to liberate people from stereotypes and from bad old accepted ways, that we find not a word to suggest that these stories (or anything else) might be so used as to subvert, or to challenge, or even to question, what is perhaps the greatest single evil in the modern world: the Hindu caste system. Now, there really is a set of ideas that stereotypes the Other. And yet, Doniger writes, Indian myths “teach us”—us, this time, but not them: Anglophone liberals, not the people whose myths these are—

that the real problem is racial indiscrimination—the unwillingness to discriminate between two different members of another race, the tendency to regard them all as doubles of one another.

And yet the caste system asserts precisely that all members of a given caste really are indistinguishable: an untouchable simply is an untouchable, to be treated as such. The system will back that view with force, and all too often with death. Can it be that the “colonialists” of the West still have something to teach the East; that enlightenment does not run all one way? Or is that one of the interpretations that must be eliminated?

This Issue

November 4, 1999