Introduction: Transformation and Contradiction

This is a book about myths, dreams, and illusion. It is about the ways in which they are alike, the ways in which they are different, and what each teaches us about reality. Transformations of one sort or another are at the heart of myths; Ovid called his great compendium of Greek and Roman mythology Metamorphoses. Transformations are particularly characteristic of the great Hindu myths, and here they may appear to take different forms: sometimes they are regarded as actual changes in the physical nature of the world, sometimes as illusions, sometimes as dreams, sometimes as temporary magic changes in the physical nature of the world, sometimes as the unveiling of another level of reality. If the storyteller sets out to tell a tale of illusion, various transformations may seem to take place, in waking life or in dreams, but in the end we cannot tell whether anything has happened or not. If the storyteller sets out to tell a tale of dreams, he may relate events that seem to be physically unreal but turn out, at last, to be real. If he sets out to tell a tale of magic, he may describe some physical transformation that a magician or a god actually caused to take place. And if he sets out to tell a tale of revelation, he may describe events that peel back the physical veil to reveal another, more mystical, reality that was always there but not recognized.

These stories tend to blend into one another; a story that starts out as a tale of magic, or even explicitly announces (as many do) that it is going to be a tale of magic, may be transformed into a tale of illusion. Sometimes it is only the genre of the story, marked by the presence of certain motifs conventionally associated with one sort of transformation or another, that lets us know whether the story is intended to depict a dream or a magic show. These interactions and interchanges are not the result of simple borrowing, back and forth, between related themes. One sort of transformation often becomes transformed, as it were, into another sort of transformation in mid-story because one of the points of the story is to demonstrate how difficult it is to tell one sort of transformation from another.

The tales of dreams suggest, for instance, that dreaming and waking partake of the same reality, which is both spiritual and physical (chapter one). That they do so is the point of the myths of shared dreams (chapter two). The tales of illusion go on to suggest that we cannot tell whether we are awake or asleep during our experience of many sorts of transformation, nor can we judge which half of the dream/waking experience is more real. This is the point of the myths of the mouth of God (chapter three). And the tales of magic may be given a twist at the end to show that the enchanted man, or the dreamer, cannot be certain that he has awakened from the last of a series of dreams within dreams. This is the point of the myths of the receding frame (chapter four). Finally, the dreamer who believes that he is in the last frame of his own dream may still be forced to consider the possibility that he is part of someone else’s dream. This is the point of the myths of the dreamer dreamt (chapter five).

The stories in this book are themselves examples of yet another kind of transformation, the transformation of human experience into words and images through narrative. Both dreams and myths draw their vocabularies from certain intense moments in actual human experience, but it is art that transforms those moments, bringing them from the private realm of the dream into the public realm of the myth. Moreover, our awareness of the experience of art allows us to perceive ourselves as we become aware of the experience of life; art allows us to watch ourselves having the illusion of life (chapter six).

One of the most intense mythic experiences is the experience of events that make us question our certainty about what is real and what is not. From the Indian point of view, the basic condition of human experience is the condition of illusion (chapter three). When we glimpse the power of illusion in waking life, we may resist it or misunderstand it; but when the barriers of rational thought are lowered during sleep, we dream about illusion—that is, we dream about things that turn out to be different from what they seemed to be, about things that seem to be transformed, back and forth, from one shape into another (chapters one and two). When we wake, we may subject these insights to various forms of tests, attempting thus to establish some relationship between the reality of the dream and the reality of waking life (chapter four). But whatever the results of these tests, we find ourselves ultimately facing an ontological cul-de-sac that renders the tests meaningless and requires a leap of faith if we are to guess what is real and what is not (chapter five). This leap is an imaginative, artistic act of creation and discovery (chapter six).

Each chapter turns on its own set of narrative subthemes, though these often spill over into other chapters. These themes include dream transformation—the adventure of the hero who flies or rides to the woman in the other world (chapter two); illusory transformation—change of sex or the creation of a double self or double world (chapter three); social transformation—the dramatic change in status of the king or Brahmin who becomes an Untouchable (chapter four); ontological transformation—rebirth as another person or an animal (chapter five); and artistic transformation—metaphor and simile (chapter six).

All of these themes occur throughout the range of classical Indian texts, and chapters one, two, and three will provide the classical Indian treatment of them, beginning with the earliest recorded text (the Ṛg Veda) and continuing through the medieval Sanskrit Purāṇas into contemporary folktales. But in chapters four and five (and at scattered moments in other chapters as well) I will concentrate on one particular Sanskrit text, the Yogavāsiṣṭha, composed in Kashmir sometime between the ninth and twelfth centuries A.D. (Appendix Three provides a list of the stories from the Yogavāsiṣṭha that are cited in this book.)

Although all of my main texts are Indian, my arguments rest on a comparison of Indian and non-Indian (primarily ancient Greek and twentieth-century European) approaches to dreams and illusion, and this often involves the citing of non-Indian texts and arguments. Many Western parallels occurred to me as I wrote this book; others will, I am sure, occur to each reader. But I made a strenuous effort to leave most of them out of this final version, in order to highlight the Indian texts. I have had recourse to the Western texts that I have used, not in the hope of saying anything about them that has not been said before by scholars better versed in the Western tradition than I am, but as an aid to understanding the ideas expressed in the Indian texts. It is, in any case, impossible for us to ignore our own assumptions and preconceptions when we read foreign texts, and only by examining those assumptions, to see what it is that makes us find the Indian texts so puzzling, can we hope to understand why some Indians did not find them puzzling, while others did.

The inclusion of these Western texts thus serves a kind of psychological or epistemological purpose: it helps us to understand how we understand the Indian texts. But it is meant to serve an ontological purpose as well: to help us understand the actual problem set by the Indian texts. For our understanding of the Indian ideas is greatly enhanced when we reflect on the insights provided by some of our own sages who have thought long and deep about these same problems.

Rather than muddy the waters by mixing together the Indian and non-Indian theories on every minor point, I have grouped the Western theories together at the end of each of the first five chapters, giving by far the greater weight to the Indian texts, which are lesser known by most people and better known by me. But I would encourage the Western reader to keep both approaches in mind simultaneously as much as possible, one hovering in the wings while the other occupies center stage, and to allow each one to come forth when it has something to say, like the little man and the little woman balanced on an old-fashioned weather-house barometer. It is artificial to go on separating the two approaches forever; I have treated both traditions, Indian and Western, together in the introduction and conclusion, in the final section of chapter six, and from time to time throughout the entire book. Yet, though the two approaches to many of the same problems are, I think, mutually illuminating, they are never truly comparable; for even when the content seems superficially the same, the different cultural contexts often show us that the apparent congruences are the same answers to different questions—and are therefore not the same answers at all.

The reader should also be warned that I have mingled not merely Indian and Western arguments and Indian and Western stories but stories and arguments themselves. I have made a general attempt to separate texts from arguments in chapters one through four, but in chapter five they are formally combined, and in fact there are arguments that link stories throughout the book. In this I am imitating the tradition about which I am writing; that is, the form of my book mirrors its content. Unconsciously, I found myself swinging between text and commentary, as the Yogavāsiṣṭha itself does (see chapter four), and as, in fact, all Purāṇas do. This book may therefore be read as a kind of American academic Purāṇa, leading the reader from one story into another and another and another, with commentary provided as necessary. Moreover, like the Purāṇas, I have felt free to borrow from any tradition that seems relevant to the point, using not only classics from other countries but folk traditions, limericks, and children’s books. That the classics are not only Indian ones does not, I think, nullify the Purāṇic nature of this book, for a Western audience would react to these classics as an Indian audience would react to the Indian classics usually cited in Purāṇas; the effect is thus comparable (though, again, not the same).

But a Western audience may perhaps have more trouble in following the pattern of thought in a Purāṇic text (like this book) than an Indian audience would have. Mine is not a tightly structured argument; it is an argument, I hope, but one that feels free to digress and switch back on itself, to wander into bywaters before flowing on to its next stopping point. Because this is a book that tells stories within stories, the reader may think it is all over when one story ends—but it is not over; each final point turns out to be merely the halfway mark of another final point, which is in itself . . . , and so forth. By way of apology and exhortation, I say, Have courage; all the points are meant to be points, and the book does end.

I have already admitted one thing that this book does not intend to do: it does not intend to give anything like a complete (or even balanced) view of the Western approaches to the problems raised by dreams and illusion. There are many other things that it does not attempt to do. Historians should not expect a steady chronological development; though a few points are traced through their history where that history seems relevant to their final form, most of the time I have drawn on any text that serves my purpose at a given moment. Anthropologists should not expect a thick description; such data are rarely available for the periods in which most of my texts were composed and are seldom relevant to the arguments I wish to make. This is a book about the history of ideas; it is not a history of social forms or customs. In chapter four (pp. 158–60) I will defend this principle of selection and interpretation at some length. Here I will simply say that my procedure can be defended on the basis of its sympathy with the procedure of the texts with which I am dealing, texts that steadfastly deny us any possible access to the sort of material that anthropologists and historians might hope for, because the Indian authors do not regard such information as relevant to the problem of dreams and illusion. I take quite seriously the conventions of genre and the multiple voices and viewpoints of illusioning illusion, and from this vantage I attempt to ask certain questions of Indian as well as non-Indian stories and theories. The free-wheeling form that results from this process was, I think, well characterized by the reader of an early draft of this book, who described it as “an intricate compilation of texts that sets both Hindu and Buddhist themes into mutual plays of illusioning—kings’ views of monks’ views of Harijans’ views of women’s views of men’s views of hedonist views of asceticism’s view of Vedic views of philosophic views of inverse views of all these views viewed from outside and inside and before and after in foreshadowings and flashbacks of lives now and then and ultimately or cyclically extinguished . . . and all that.”¹ This is a spiraling rather than a linear approach, and it will, I fear, exasperate the sorts of people who like to be able to make outlines of the structure of an argument.

I have used as wide a range of theoretical tools in the interspersed analyses as I have used genres in the texts themselves. Because each chapter takes up a different facet of the problem of reality and illusion, I use different hermeneutical tools, both Indian and Western, at different points. Though each is particularly relevant to a single chapter, they blend into one another, just as the themes themselves do, and they may make brief reappearances in chapters other than those in which they make their principal contributions. In chapter one, I draw on Plato, Freud, Piaget, and several contemporary anthropologists in order to analyze the Indian approach to dreams. In chapter two I have recourse to the Jungian concept of the archetypes. In chapter three I make use of classical Indian philology and Western epistemology to analyze the theory of illusion. Chapter four involves the philosophy of science, particularly the writings of Sir Karl Popper, Thomas Kuhn, and Michael Polanyi. To come to terms with the paradox of the dreamer dreamt, in chapter five, I have found certain Western logical paradoxes useful, particularly as reformulated by Douglas Hofstadter. And finally, in chapter six, I have relied on Sir Ernst Gombrich to illuminate the use of illusion in art.

This is the moment when I ought to produce, like a rabbit out of a hat, a working definition of reality. I cannot do this; the whole book is my rabbit. The man in the street (in India and in the West) has hunches about reality, and he vaguely supports these by resting on various authorities he does not really understand. Because of this rather wobbly ontological base, he is always trying to gather some sort of positive evidence to support his hunches, and, if you push him hard, he will fall back on ad hoc definitions in order to defend his primitive beliefs. The stories in this book press on the raw ontological nerve in that way; they show us what we have believed but not defined. Indeed, some of these beliefs cannot be defined; as Isadora Duncan is said to have remarked, when asked what one of her dances meant, “If I could tell you, I would not have to dance it.”

When we speak of reality, we are usually referring to a cluster of assumptions. We do not have a set of precise definitions of reality in our heads—indeed, we usually do not bother to define reality at all—but we somehow assume (usually with good reason) that we all agree on what we mean. In general, we mean that reality is what we value, what we care about. But often we mean something far more specific, more debatable, less relative than this. Some of us mean that reality is what is solid; others mean that reality is what is not solid. We tend to assume that our way of thinking is simply an expression of common sense, the most obvious way of understanding and manipulating the everyday world. But since common sense carries with it an implicit definition of reality, to define reality in terms of common sense is to commit a tautology. Common sense, as Einstein once remarked, is what we are taught by the age of six; that is, common sense is an attribute of culture, not of nature, a part of myth rather than a part of reality. As scientists and artists become more adventurous in restructuring their reality, common sense is thrust unceremoniously into a dusty old attic, along with the astrolabes and the perpetual-motion machines. As Clifford Geertz has shown, common sense is a highly variable cultural construct, involving massive a priori judgments;² yet each culture thinks that it not only knows what common sense is but knows how common sense differs from other ways of viewing the world—ways that may be, but are not necessarily, mystical or magical.

In the West common sense is often treated as if it were the same as the scientific point of view, but it is not. Reality is not a problem for physicists, except in the most trivial realm of accuracy of measurement; their stable sense of what is real is taken for granted long before they reach for a slide rule.3 Common sense is any world-view that is assumed a priori. Scientists have one—or, rather, several; nonscientists have others; and mystics have theirs. (The Hindus, as we shall see in chapter four, admit from the start that common sense is not an authoritative source of knowledge.) Newtonian (and Freudian) ideas have certainly helped to construct our common sense, but most of us do not usually read or indeed use Newton (or Freud) in our daily, common-sense lives. Our common-sense view of reality is so deeply embedded in us that it is unlikely that any single definition would satisfy us or would meet with general acceptance.

There are two basic lines of common sense in the West, one derived from Plato and the Christian tradition, the other derived from Hume and Locke and scientific empiricism. These two world-views are so dramatically opposed that it is remarkable that our culture has been able to hold them in suspension for as long as it has. It makes no sense, therefore, to speak of a “Western” view without specifying whether we have in mind Plato or Hume. Yet we do speak of “the West,” just as we speak of an “Indian” view, and some limited purpose is served by such generalizations. In selecting “Western” texts to contrast with my (equally selected) “Indian” texts, I have chosen those that provide the most dramatic (and often the most extreme) antithesis within each culture. This may give a misleading impression of the schizophrenia of Western ontology; it may also provide a more striking contrast with the Indian point of view than might have appeared had I chosen texts from other, perhaps more recondite, Western traditions. But one of my central concerns is to show the contrast between what most people think and what philosophers think, not only in the West but in India, and another is to show the contrast between what most people think in India and what most people think in the West. To make such contrasts, one is necessarily led to concentrate on the famous Western credos. To have dealt more fairly with the Western tradition in all of its complexity would have weighed this book far too heavily away from the Indian materials that are its primary raison d’être.

It is useful to distinguish, however tentatively, between the two major Western approaches to the problem of reality. Most people think that reality is physical, public, external, and somehow “hard,” and they think that what is not real is mental, private, internal, and somehow “soft” (chapter three).⁴ The terms hard and soft are also used to distinguish people who accept this distinction (hard thinkers) from those who feel that all phenomena (physical or mental, public or private, external or internal) are equally hard and soft (soft thinkers). Hard thinkers think that you should always define, sharply, at the start, what you think and that you should always continue to think it; soft thinkers feel that you can play it by ear and shift your definitions as your understanding grows. Hard thinkers think that you cannot believe two contradictory things at once; soft thinkers think that you can. These definitions are self-referential: soft thinkers do not think that they, or hard thinkers, exist as a separate category; hard thinkers think that they do. In the light of my attitude to the terms “hard” and “soft,” I would be called a soft thinker by people who think that there is such a thing as a soft thinker as opposed to a hard thinker—a distinction that I would challenge, as a soft thinker should, though I used it myself only a few paragraphs back. The metaphysical implications of such a circular definition, spiraling in upon itself, will be discussed in chapter five, in the analysis of the dreamer dreamt. But within both the Indian and the Western traditions we will encounter thinkers that hard thinkers would define as hard and thinkers that hard thinkers would define as soft.

“Hard” and “soft” are lamentably gross terms for dealing with the complexities of ontology. The assumption implicit in their use—the fond hope that reality can be pinned down so simply—is so full of holes that one could drive a carriage and pair through it. In the course of this book, the inadequacy of hard and soft as basic terms will become increasingly apparent, and once we have used them to scramble up, untidily, to a spot from which there is a good metaphysical view, we will kick them out from under us, ungratefully. Still, it is not easy to replace them with other, more satisfactory, terms. This is so because the conceptual dichotomy that they represent—and that will be represented, in turn, by any similar pair of contrasting terms—is itself at fault.

How is it that the myths in which these categories are blurred seem to violate common sense, and why is it that people have attempted to apply hard scientific criteria to phenomena that they themselves have defined as soft? This is the question that will be asked in chapter four. Here the problem of common sense—that is, the conflict between common sense and other primitive beliefs—comes into sharper focus as it conflicts with both soft phenomena (the experience of dreams) and hard criteria (the standards of scientific falsifiability). At this point, the argument runs aground on the problem of contradiction. I have just now maintained that hard thinkers think that you cannot believe two contradictory things at once, while soft thinkers think that you can. We owe to Plato our belief that it is impossible at one time to hold contradictory opinions about the same thing; many Indian texts, by contrast, would argue that, if two ideas clash, both may be true.5

Sir Ernst Gombrich has formulated several questions that are highly relevant to our present inquiries:

Do all cultures make the same radical distinction between “appearance” and “reality” which ours has inherited from Plato? Are their hierarchies the same? In other words, do they necessarily accept the demand that contradictions must be ironed out and that all perceptions that clash with beliefs must force us either to change our views of the “objective world” or declare the perception to have been a subjective experience—an illusion? Even in our rationalist culture we don’t often live up to this logical precept. We try to evade it, especially when our emotions are involved.⁶

Many Indian texts are troubled by contradiction; their attitude in this may seem to us Platonic. And all of them distinguish, at least nominally, between appearance and reality. But they do not ultimately iron out the contradictions; they alter their definitions of reality in order to let the contradictions survive intact.

For we find, often side by side in the same Indian text (such as many of the myths in this book), two basically contradictory views of the world. These views interact in ways that cannot be entirely contained within the categories of hard and soft. The first is the world-view preserved in the Ṛg Veda (c. 1200 B.C.) as well as in many Purāṇas and vernacular devotional texts from the medieval period and even in present-day India. This world-view holds that worldly life (saṃsāra) is real and good; to be born, to eat, to make love, and to work are real experiences that one values and hopes to go on experiencing as long as possible, even in as many rebirths as possible. The Kauṣītaki Upaniṣad⁷ suggests that a man who has died travels to the moon as a way-station in the round of rebirth; there he may ask to be engendered by a man in a woman, or else he may proceed to a world from which there is no rebirth. Though this passage is apparently unique in the Upaniṣads, the idea that, given a choice, one might choose rebirth rather than mokṣa (release) is characteristic of much of later Hinduism. Purāṇic Hinduism rejects mokṣa implicitly, for Hindus in their private pūjās and in their temple worship ask the gods not for mokṣa but for health, children, and, sometimes, rebirth as another, better, human being. The Hinduism of bhakti (devotion) rejects mokṣa explicitly, often mocking the ascetics, who have become stuck in this inferior religious goal instead of progressing to the higher goal of eternal life in the heaven of the loving god. The commitment to the goal of saṃsāra is thus both ancient and still widespread in India. In this view, death and dreams are also real, an essential part of living and waking experience. To a limited extent, this might be seen as a hard view of reality. Against this we might set the world-view that first appears in the Up-aniṣads (c. 700 B.C.) and continues throughout Indian tradition to this day; it, too, is found in the Purāṇas and the devotional texts, as well as in the Yogavāsiṣṭha. This view holds that birth, sex, and marriage are not good and that the wise man seeks release (mokṣa) from them and from the whole vortex of worldly life (saṃsāra). Both views believe in both mokṣa and saṃsāra as true facts of human life, but they value them differently; mokṣa is what is real for one, while saṃsāra is what is real for the other—real in the sense of valued, and hence true.

But the mokṣa school is Janus-faced in its turn, for it exists in two significantly different forms. One, which we might characterize as soft, is the school of extreme idealism; it maintains that everything that we think of as real (our lives) or unreal (our dreams) is in fact illusion—equally real and unreal. Yet something else is real, and that something is Godhead (brahman).8 (The very softest view of all, denying the reality even of Godhead, is limited to a few esoteric Buddhist and Hindu schools, to which I will refer only in passing.) The second significant form of the mokṣa view, a kind of modified idealism (or modified realism) or hard-soft school, argues that there is something real in the world but that we constantly mistake it for something that is unreal. Moreover, since it is often impossible to know when we are making this mistake, it is impossible to know precisely what is real. This is, I think, the dominant Indian view, but it often appears interwoven with more realistic (saṃsāra-slanted) or more idealistic (mokṣa-slanted or Buddhist-influenced) views.

We have seen that there are at least two major kinds of common sense in the West—Platonic and empiricist—and that we usually feel constrained to choose one or the other. Both are indeed still strong in our culture, and because of their strength we continue to see them engage in serious combat. We see it when our legal system (materialist) steps in between Christian Scientist parents and their child, whose life may be threatened by the parents’ confidence that God (spirit) will cure the body as well as the soul. We see it in the passionate, though still confused, arguments about creationism versus Darwin. In India, too, there are two modes of common sense: common sense A (saṃsāra-linked, and materialistic) and common sense B (mokṣa-linked, and idealistic). But Indians are often able to hold both kinds of common sense in their heads at once and to reconcile both kinds with their perceptions of their lives. In the West only a gaggle of professional metaphysicians or fanatics will have enough genuine familiarity with such ideas to be able to internalize them and act on them instinctively, but Indian children learn about metaphysics as Western children learn about sex, on the street.

In order to illustrate these ideas, I have selected the myths in this book out of thousands of stories in Indian texts. I chose them because I think they are deeply important to Indian culture but also because they are full of meaning for us. True, one could have made a different selection to illustrate other facets of Indian thought; it is notoriously simple to prove anything with citations from the great compendia of Indian texts, so rich is their pluralism. But I would argue that the themes in this book are truly central, and in support of this contention I would point to the great number and variety of the tellings and retellings of the myths about them. I have collected here only a fraction of these subspecies; most of the myths in this book are paradigms for many others.

The various Indian approaches to the problem of reality are woven in and out of a tradition that spans three thousand years and reached its climax in the masterpiece of Indian philosophical narrative, the Yogavāsiṣṭha. My goal is to set that text in its cultural context and to listen to what it tells us about illusion, dreams, and myths.