Bryn Mawr Classical Review

[Full disclosure: William Harris was one of my teachers at Columbia 45 years ago. Readers must decide for themselves whether that long-ago experience has, or could have, affected this review.]

Early in this magisterial, entirely wide-awake study of dreaming, William Harris asks a question (p.21): "Can a cold-blooded rationalist write about ancient religion?" This is evidently a question intended to make readers aware of his stance and approach. I am not sure what "cold-blooded" means here, but Harris is certainly rational, as well as lucid. He intends to discover what can be known in the bright world of day and at the beginning of the twenty-first century about the experience of dreaming in the Greco-Roman world: how people described their dreams, what they believed about them, and how they explained the phenomenon of dreaming. As he did in Ancient Literacy (1989), Harris subjects received opinions to rigorous, diachronic testing and finds them wanting. This important book makes it impossible any longer to generalize carelessly about what Greeks and Romans believed about their dreams.

Harris may also succeed in persuading those he describes as "humanists who still sometimes refer to Freud, and even Jung, as the embodiments of modern research about dreaming" (p.7) to abandon their belief that psychoanalytic theory can tell us very much that is meaningful about dreams. (He does, however, acknowledge the importance and accuracy of parts of Freud's descriptions of dreaming.) Instead, he hopes, they will begin to take account of the large body of modern scientific research into the physiology of dreaming and anthropological research into dreaming in non-Western cultures.

These issues appear in an introduction, which is followed by four chapters each focused on an important question about dreaming in antiquity. Each of these questions has a complex answer. I will not summarize Harris's carefully qualified arguments here; they resist condensation. It may, however, be possible to sketch his answers. Did Greeks and Romans describe their dreams in the same way that we do (Chapter I)? For the most part, yes; but one fairly frequent kind of ancient dream, the epiphany, seems to have nearly disappeared from the night-world of modern, secular dreamers in the West. Can we say that any of the dreams described in our sources were actually dreamt (Chapter II)? We cannot, or at least we can only give estimates of probability; but Aelius Aristides "emerges as a rare and relatively convincing informant" (p. 280). The third chapter, "Greek and Roman Opinions about the Truthfulness of Dreams," is nearly a monograph in itself, at 105 pages with 612 footnotes (to the credit of Harvard University Press, they are at the bottom of the page where they belong). It asks, did Greeks and Romans believe that their dreams sometimes came true? The answer depends on which Greeks, which Romans, and when -- fifth century Athens is not Ptolemaic Egypt, and neither is Republican Rome; but skepticism was commoner than often believed and not confined to philosophers. The social class of the dreamer also mattered. Chapter IV surveys non-supernatural explanations that Greeks and Romans (mostly philosophers) offered for the existence of dreams and the mechanism of dreaming. Empedocles receives welcome emphasis. The analytic table of contents, which I reproduce at the end of this review, will give some idea of the range and depth of argument in this book. Yet Harris hardly wastes a word, and anyone even slightly interested in dreams in the ancient world will have to read the entire book carefully.

Let me focus instead on methodology. Harris's approach raises at least two important questions about our relationship to the ancient world. First, what do we believe about human nature? Second, what do we believe about the Greek and Roman self? To attempt to know the interior experiences of individual Greeks or Romans -- their thoughts, their emotions, their dreams -- we must have recourse to our own sense of what it means to think, feel, or dream. It is easy to move from this touchstone to belief in a permanent, unchanging human nature that we share with the Greeks and Romans. When Harris provides a list of thirteen universal or nearly universal characteristics of dreaming (pp. 14-17), he seems to endorse this simplistic view. His position, in fact, is carefully balanced. He recognizes that even as fundamental an experience as dreaming changes as cultures evolve, so that "Greeks and Romans from time to time dreamed a kind of dream that we do not dream and have not for several hundred years" (p. 14). On the other hand, evolution has not had time to create radical discontinuities in the physiology of dreaming between, say, the fifth century BCE and the present. If we had no experiences at all in common with Greeks and Romans, we would not be able to say anything meaningful about them.

As its title suggests, Harris's book is concerned to give an account of dreaming as a reported experience in the Greco-Roman world. This approach needs to be distinguished from one that might consider dreaming as an aspect of something called "the self" or "personality" that can be studied as an independent phenomenon. It remains likely, in my view, that people in antiquity did not have anything like the post-Cartesian, post-Kantian conception of the self as an autonomous, self-aware locus of thought and will capable of disinterested moral rationality. Instead, what Christopher Gill has called the "objective-participant" self, continually negotiated and re-negotiated through language, dialogue, argumentation, and other social activities, seems to offer a better account of the human beings that we find in the surviving evidence from the Greek and Roman world.

Harris rightly doubts (p.3, n.9) that concepts like "selfhood" or "personality" have much to offer historians studying dreams in antiquity. It is hard enough to know our own dreams or to say how it is that we know them. For the ancient world, all we have are reports of certain kinds of experience, from which it may be reasonable to draw some conclusions about beliefs. Like all description and narrative, ancient accounts of dreaming come to us shaped by cultural expectations, generic demands, and the imperatives of performance and publication. Some combination of these conventional influences, Harris argues ("From Epiphany to Episode: A Revolution in the Description of Dreams," pp. 23-90), accounts for the relative frequency in antiquity of the epiphany dream, in which an authoritative figure visits the dreamer and makes a significant statement, and for its rarity in the post-Enlightenment West.

No doubt some reader of this review is now saying, "But I had an epiphany dream just the other night!" That is the problem with studying dreams: one must work hard to free oneself from dependence on anecdote and from the powerful attraction that dreams have for those who dream them. Appealing to concepts of "selfhood" or "personality" will only reinforce these tendencies by compelling the question, "What does this dream tell us about you?" Harris chooses instead to concentrate on ancient descriptions of dreams and reports of actions based on them. This is a book about dreaming, not about dreams; that is, about behavior and experience in antiquity, not about the ancient self.

Harris is fond of the comprehensive first person plural and of comparing or contrasting ancient dreaming with what "we" experience. (In five pages chosen at random, 154-158, "we" or "us" turned up eight times.) Because to some readers the answer may suggest a limitation of his study, it is worth asking who Harris means by "we." Modern beliefs about predictive dreams, he suggests (pp. 124-125), can be classified by believer. Scientists will believe one thing, and those who "consult books of innocent superstition" like T. Cheung's The Element Encyclopedia of 20,000 Dreams (New York 2006) another. Freudians, post-Freudians, and other psychoanalytically inclined persons will take yet a third view. Finally, there are "normal people who, while they may not be above recounting dreams at the breakfast-table, assume that what they saw in their sleep has no large significance whatsoever." Harris's sympathies are clearly with the scientists and "normal people," but much of the evidence in his study necessarily comes from representatives of the second category: those who were, if not entirely guileless in their recording of dreams, at least superstitious enough to think them worth recording. Sometimes Harris's lack of sympathy for such persons leads him to understate their belief.

Take, for example, Xenophon's report (Anabasis 3.1, 11-12) of the dream that came to him after the battle of Cunaxa, when the Greeks who had followed Cyrus were demoralized, leaderless, and far from home. Xenophon dreamed that a thunderbolt struck his father's house, and that the entire house lit up λάμπεσθαι. Harris wants to argue that belief in the truth of dreams was far less common among "ordinary" Athenians (that is, those not disposed to philosophy) in the fourth century than scholars have supposed; in particular, he asserts that "no Greek politician or general is known to have invoked a dream or a prophet's dream-interpretation as a reason for action or inaction--not even Xenophon" (p. 155).

It is true, as Harris points out, that in the speeches in which he convinces the Greeks to appoint him one of their generals, Xenophon does not appeal to this dream. He does, however, make the dream his motivation for action. As he tells his readers, "What sort of thing, indeed, it is to have such a dream can be seen from what happened after the dream" (Anabasis 3.1, 13). Immediately (εὐθὺς) upon waking he began to deliberate with himself (Anabasis 3.1, 13-14--this remarkable internal monologue deserves more attention than it usually gets in studies of the psychology of decision-making in Greek literature). This deliberation leads directly (ἐκ τούτου) to his decision to address the surviving junior officers in Proxenus's regiment. Although Xenophon does not use the dream in the speeches he gives to his audience within the narrative, he does make it part of the explanation for action that he offers his external audience, readers of the Anabasis.

This book constitutes no exception to the rule that great scholars bring themselves to the ancient world that they study. Less cold-bloodedly rational students of ancient dream-experience may find somewhat fewer Greek and Roman skeptics or less skepticism in specific texts than Harris does, but they will find it difficult to overthrow his basic conclusions. Like Ancient Literacy, this wide-ranging, now indispensable book will become the new point of departure for those navigating the depths that it describes.