Artists have a saying: color gets all the credit, but form does all the work. So it is with dreams and sleep. The dream, vivid, remembered, and sometimes interpreted, attracts the attention of poets, psychologists, and philosophers of all kinds; but among humanists, at least, the necessary substrate of dreaming, sleep, is often taken for granted. Stephanie Holton’s book, the revised and improved fruit of her doctoral studies at Royal Holloway, sets out to mitigate this imbalance by exploring accounts of both sleep and dreams in Greek philosophical and medical texts from the sixth and (mostly) fifth centuries BC. Her decision to look at the physiological and psychological dimensions of sleep alongside dreams adds a new dimension to discussions of Greek dreaming. The practical organization of this book and the range of passages that it discusses guarantee that anyone interested in Greek dreams and their interpretations will return to it again and again. Holton’s attempt to trace the origins and development of Greek ideas about dreams before Plato, Aristotle, and Artemidorus has produced a clear, useful map for future explorers of the Greek dream-world. It remains true, I think, that Aristotle is the first Greek author to advance a comprehensive and physiological theory of sleeping and dreaming, but Holton shows that a rich tradition of speculation on these topics existed more than a century earlier.
And not only philosophical and medical—in Chapter 1, “Sleep and Dreams in Context,” Holton begins with a brief analysis of the divine personification Oneiros and the paired Hypnos and Thanatos before offering commentary on a range of passages from Homer, Hesiod, and fifth-century literature. Inevitably, dreams receive more attention than sleep; the bias is not Holton’s but that of her sources. The usual suspects appear—Penelope’s insomnia and her dreams of Odysseus and of geese, Iphigenia’s misinterpreted dream in Iphigenia in Tauris—but there are welcome surprises: Bellerophon’s bridle in Pindar’s Olympian 13 gives an example of the “apport,” the dreamed object that reappears in waking reality, and Hypereides’ For Euxenippus offers an example of the invented dream.
Heraclitus plays a key role in Holton’s study. Her second chapter, “Sleep, Dreams, and Heraclitus,” concentrates on two fragments in which Heraclitus mentions sleep: Diels-Kranz B1, the well-known passage that Sextus Empiricus (Adv. Math. 7.132) assigns to the beginning of Heraclitus’ Peri Phuseos, and DK B26, transmitted to us by Clement of Alexandria (Stromata 4.143). She takes the end of B1, “But other men fail to notice what they do when awake, just as they forget all they do when asleep,” as a statement about the epistemological condition of a person asleep, which, she suggests, entailed for Heraclitus a negation of memory, and as a pointer to Heraclitus’ ideas about sleeping experience (“all they do when asleep”). Because Heraclitus seems to deny the possibility of remembering sleep experiences, Holton downplays the possibility that he has much to say about dreams, but she may be taking the philosopher called “the Obscure” (Cic. Fin. 2.5.15) too much at face value. Surely “all they do when asleep” includes dreaming, as DK B89, discussed on p. 64, may suggest.
Holton’s penchant for literal reading may lead her astray when she takes up the enigmatic DK B26 (in which ἀποθανών is probably a gloss by Clement):
ἄνθρωπος ἐν εὐφρόνῃ φάος ἅπτεται ἑαυτῷ ἀποθανών, ἀποσβεσθεὶς ὄψεις· ζῶν δὲ ἅπτεται τεθνεῶτος εὕδων· ἐγρηγορὼς ἅπτεται εὕδοντος.
Holton translates (p. 66):
A man in the night kindles a light for himself [dead], his eyes having been extinguished; living, he touches that which is dead while sleeping; having woken, he touches that which is sleeping.
As Holton says, this is the one Heraclitan fragment that “plausibly discusses dreaming.” But because she wants Heraclitus to be a theorist of sleep, not of dreams, and because B1 seems to say that people remember nothing done in sleep, she interprets φάος ἅπτεται as the lighting of a literal lamp, rather than as a metaphor for the light of internal consciousness in a dreaming person, who (if B89 gives anything close to Heraclitus’ ideas) leaves the common cosmos of waking life to enter his own private world. Heraclitus may have a clearer notion of dreams than Holton allows, and one closer to her other key text, the Hippocratic Regimen 4.
Two chapters, one gathering physiological texts (Chapter 3, “Sleep, Dreams, and the Body”) and one collecting psychological accounts (Chapter 4, “Sleep, Dreams, and the Soul”), form the center of this book. In them readers will find rich discussions of Presocratic and Hippocratic texts on sleep and dreams, as well as thought-provoking digressions on Herodotus, Sophocles’ Philoctetes, and Pindar. As with the chapter on Heraclitus, I found that Holton’s discussions raised occasional questions even as they provoked me to think about these passages in new ways. Ι wondered, for example, whether it is correct to say that Pindar fr. 131b “neither speaks of the immortality of the soul nor the soul’s involvement in dreams” (p. 113). The fragment is quoted by (pseudo-) Plutarch (consol. ad Apoll. 35, 120C):
σῶμα μὲν πάντων ἕπεται θανάτῳ περισθενεῖ,
ζωὸν δ᾿ ἔτι λείπεται αἰῶνος εἴδωλον·
τὸ γάρ ἐστι μόνον
ἐκ θεῶν· εὕδει δὲ πρασσόντων μελέων, ἀτὰρ εὑδόντεσσιν
ἐν πολλοῖς ὀνείροις
δείκνυσι τερπνῶν ἐφέρποισαν χαλεπῶν τε κρίσιν.
In Holton’s translation:
And while the body of all men is encumbered by overpowering death, a living eidolon of life yet remains, for it alone is from the gods. But it sleeps, while the limbs are active; yet to the sleeping, in many dreams, it shows an interpretation of future delights and pains.
Holton wants Pindar, if he is talking about the psukhe at all here, to be limited to the Homeric soul, which “has little direct involvement in any emotional or intellectual activities” (p. 115). It is true that until later in the fifth century “there are no occasions in which eidolon is specifically and directly used as a synonym for the kind of psukhe being read into the Pindar fragment” (p. 114). But “specifically and directly” carry a lot of weight here, especially when speaking of Heraclitus’ equally obscure near-contemporary, Pindar.
Eidolon in Homer does indeed mean “a statue, an object created in the likeness of something living” (p. 114), like the eidolon of Aeneas made by Apollo in Iliad 5.449–453, and Holton argues that Pindar cannot be speaking of the soul as a psychological faculty capable of generating dreams. Instead, she wants Pindar’s eidolon to be a kind of Democritean emanation from the gods that can appear in other people’s dreams after the individual’s death. She unconvincingly argues (p. 116) that πρασσόντων μελέων “could be a poetic expression of life more broadly: an inversion of the common ‘loosened limbs’ of death.” But as Gregory Nagy and others have shown, Pindar’s Homer is not an eidolon of the epic poet, but rather a living transformation of him. Αἰῶνος εἴδωλον paraphrases Homer’s ψυχή τε καὶ αἰών (Il. 16.453), and Pindar’s soul, which is inactive while we are awake but shows us dreams while we sleep, is closer to the dreaming soul of On Regimen 4, which wakes while the body sleeps, “manages her own household” (διοικεῖ τὸν ἑωυτῆς οἶκονͅ), and performs all the actions of the body (Vict. 4.86), than to the Homeric psykhe.
On Regimen 4, which Holton rightly takes as integral with books 1–3, takes up a substantial section of Chapter 4 (122–35) and makes up the meat of Chapter 5. She has interesting things to say (129–30) about the gendered soul of Vict. 4.86 and its relation to the female household manager of Xenophon’s Oeconomicus and (132–135) about the gaze of the psukhe at Vict. 4.93, which she reads in the light of the gaze in poetry, and especially of Gorgias’ remarks (Encomium of Helen 19) that Helen’s eye transmitted Paris’ beauty to her soul. Chapter 5, “Interpretation of Dreams,” surveys Near Eastern dream books and Greek oneiromancy, including dreams in the cult of Asclepius, before settling into an analytical paraphrase of Vict. 4. Holton follows that book’s classification of dreams: first, “day residue” dreams, then dreams of cosmological bodies, the natural environment, clothing, the dead, and miscellaneous dreams. Her aim throughout is to connect the interpretations in Vict. 4 to “cultural patterns, religious traditions, philosophical ideas, and medical observations” (188) so as to highlight the Hippocratic author’s originality and his claims to authority.
Some features of Sleep and Dreams in Early Greek Thought, for which publisher or editor may bear responsibility, call for praise mixed with blame. On the one hand, notes follow each chapter—not as convenient as having them at the foot of the page, but certainly easier on a reader than finding the note one needs from a multi-chapter compendium in the back of a book. References for quoted passages, on the other hand, like Vict. 4.86 or Ar. Kn. 809, are given only in the notes, rather than in parentheses with the main text. This reviewer, who will own up to being old and cranky, deplores the disappearance of the distinction between “principle” and “principal” (85, and again on 86), the use of “as” for “like” (129), and a few other solecisms. None detracts from the merits of this book.
If someone wanted to pick the two bodies of ancient Greek texts most filled with difficulties, ambiguities, and contested interpretation, one could do worse than name the Hippocratic Corpus and the fragments of the Presocratics. A scholar’s response to the difficulties posed by these works must rest on a philological foundation, but the superstructure rising from it will have room for taste and temperament. If I differ from Holton’s reading of some of her texts, it is not because she is wrong, but because we bring different approaches to the text. I do not share her confidence that a sharp line can be drawn between literature and philosophy in the fifth century, or between medicine and philosophy, and I look at the fragments of Heraclitus with the same lens that I use for his contemporary, Pindar. Our different approaches did not hinder me from learning much from Holton’s book.
List of Abbreviations
1. Sleep and Dreams in Context
2. Sleep, Dreams, and Heraclitus
3. Sleep, Dreams, and the Body
4. Sleep, Dreams, and the Soul
5. Interpretation of Dreams
 I assert the improvement with confidence because I am the “one anonymous reviewer” thanked in the acknowledgements for reading the pre-publication version. My encounter with this earlier version left me, as far as I can tell, with no bias for or against the published work.
 For most scholars, Diels-Kranz remains the edition of reference, but it is puzzling that Holton does not mention the new Loeb Early Greek Philosophers, edited by André Laks and Glenn Most, which appeared in 2016.
 As Holton says, how closely B89 reflects Heraclitus’ words “has been the topic of debate” (64); Laks-Most describe it as “an anonymous paraphrase” and place it among the “Reception” passages (R56).