BMCR 2007.07.33

La cité des mages: Penser la magie en Grèce ancienne

, La cité des mages : penser la magie en Grèce ancienne. Collection Horos. Grenoble: Millon, 2006. 271 pages : illustrations ; 24 cm.. ISBN 2841371905. €25.00.

In this volume Carastro (C.) aims to trace the figure of the Magos introduced into the Greek language in the fifth century B.C.E., arguing in essence that the term μαγεία came to denote a complex of concepts and meanings that were already current in Hellas at the specific time, and that the term evolved in later Greek thought; that “magic”, in short, was not introduced to Hellas from the East. Most of this work, however, is neither about Persian religious practitioners nor Greek itinerant ritual experts, but about Greek mythology. There are 54 pages on the Magoi and the history of the term mageia; almost 100 pages on the “vocabulary of magic”, with long discussions of the Sirens and Circe in the Homeric poems; a mere 44 pages on curse tablets and Plato’s use of magic; and finally 55 pages of bibliography and index. The title and appearance of this book may be slightly misleading: the oinochoe showing three Sirens on the front cover is very appropriate, but despite the background image on the front and back this work neither references the Gortyn Code nor is an epigraphic publication. Aside from the potential misrepresentation, however, this is a very useful and well-executed discussion of the language and mythology surrounding concepts that we might group under the term “magic”.

This volume is based on a doctoral dissertation that C. completed in Paris in 2002. Despite the intervening time and the presentation in monograph form, there seems to have been very little attempt to adapt the work from the form of a student thesis to that of an academic book (although the research has certainly been brought up to date). The thinking is mature but the structure is perhaps unsophisticated, with long stretches of pedantic argument, and not always the most elegant use of citations (in one chapter, for example, seemingly every reference to purification is adorned with a footnote referring Parker’s Miasma). More on the presentation will be discussed at the end of this review.

The current work is part of a fairly extensive, although relatively recent, tradition of scholarly work on ancient magic. Early collections like Faraone & Obbink’s (1991) Magika Hiera and Betz and co.’s (1986) translations of the Greek Magical Papyri, helped to usher in a renaissance of respectable scholarship in the area. Many of the monographs that followed, including volumes by Bernand, Graf, and Dickie, attempted create a synthetic picture of magic in the whole of the ancient world, with mixed success, as did the primarily teaching collections of Luck and Ogden.1 More recently there has been a tendency to build upon these foundations with far more focussed studies of particular aspects of magic, delimited by time and/or place, such as works by Faraone, Janowitz, Trzcionka, and Collins, which have on the whole led to more fruitful scholarship.2 The current volume falls in the latter category of focused works that avoid the pitfalls of over-generalising across the thousands of miles and years of antiquity.

C. begins, sensibly, with a discussion of his terms, his definitions, and the scholarly history of studies of magic. He refers to the important difference between emic and etic approaches to magic (the first of which he characterizes as necessary but not sufficient, and the latter as problematic). He argues that the term μαγεία that was introduced to Hellas from Persia did not reflect the Eastern origin of magic per se, but came to be used to refer to a concept that was useful to name at this time in Greek culture (9). The introduction finishes with a few pages of summary of the anthropological and “evolutionary” models contrasting magic with science and religion, opposing authors who form universal definitions to those whose theories are culture-specific.

This book is divided into three parts, containing two, three, and two chapters respectively. Part I, “L’Avènement des Mágoi“, begins with a relatively cursory chapter entitled “Histoires des Mágoi“. This chapter begins, reasonably enough, with a summary of Herodotus’ mentions of the Magoi, and concludes that, because their rituals are often described as failing, their representation is ultimately negative. The fact that they fail because Xerxes has to fail to invade Hellas is a level of problematisation C. seems to overlook, although it does not entirely invalidate his observations. By the by, C. rejects the reference to Magoi in Heraclitus 14B (DK) as an anachronism, but accepts Aeschylus’ Persae 318 as a significant reference to a Magos. In a discussion of Persian sacrificial ritual, he shows that Herodotus misunderstood their use of theogonic song as a (“magical”?) epaoide. He concludes that in the Greek view the Magoi were “violent”, and that Persian funerary rituals were “outrageux”; that Herodotus represents the Magoi as capable of true divination, but not infallible. The involvement of Magoi in political machinations renders them with ultimately negative connotations, and that their sacrificial practice is described using the verb φαρμακεύειν means that the Magoi are already becoming “magiciens” (36). As an introduction to the issue of Magoi in the fifth century, this chapter feels a little thin, but it is clearly putting the rest of this study into perspective rather than offering a definitive discussion.

Chapter Two, “Au Coeur de la Cité”, presents evidence for the reception of terms relating to the Magoi in Athenian literature, principally tragedy, philosophy, medicine, and rhetoric. In a discussion of the passage in Sophocles’ OT where Oedipus calls Teiresias “Magos”, C. points out that this term is used as a general insult with connotations of “beggar” and “trickster”, but also seems to follow Rigsby ( GRBS 1976) in seeing a political reference to the Magoi as king-makers; wrongly, I believe. C. also argues that Iphigenia’s purification ritual in Euripides’ IT is meant to recall those of the Magoi, and that in the same playwright’s Supplices the related noun μαγεύματα connotes attempts to heal that are destined to fail (42). This reference allows C. to segue into a discussion of the Hippocratic writers’ criticism of magical healing, on which he points out, somewhat confusingly, that the targets of the Sacred Disease were not Magoi but “guérisseurs traditionnels grecs”. Here, as elsewhere, it would be useful to have a discussion of what C. believes Magoi in Hellas actually were, and what would constitute evidence for their existence. Further on he discusses the respective characteristics of ἀγύρται, ἀλαζόνες, and καθάρται, almost as if unaware that these terms could at times be largely interchangeable pejoratives. In an argument equally unsatisfactory for its circularity, he insists that the mystical Empedocles was not a Magos, because the term was a negative one. Finally this chapter ends with a section on Gorgias’ Encomium which casts rhetoric as μαγεία (a large part of which is a survey of γοητεία in literature and myth). Again this section lacks overt recognition of the fact that Gorgias’ equation of his rhetoric with magic was itself rhetorical in nature and needs to be read very cautiously through the filter of the writer’s intentions.

The second part of the book, “La Constellation de Thélgein“, which makes up almost half of its length, begins with a chapter on “Des Dieux qui Médusent”. This chapter starts to make clear the true agenda of the book, which is not to discuss the historical or stereotypical Magoi in fifth-century Hellas, but to discuss terminology and thought surrounding the concepts of magic, charming, and bewitching. Most of the discussions in this long section are therefore close readings of particular myths or works of literature (mostly from the Homeric texts); all very interesting, valuable, and convincing in their own right, but with only marginal relation to the argument of the preceding chapters. This chapter focuses on a discussion of the verb θέλγειν and its cognates, on the grounds that these words end up converging on the new concept that will become μαγεία. When the subject of the verb is Zeus or Athena, it seems to inflict a change of emotional and intellectual state upon the object; they mislead; they steal the courage of the object. When Hermes is the subject, he puts to sleep, he affects the eyes. When it is Poseidon, he blinds or steals the strength of the object. In myths involving Narcissus, C. argues, the mollifying effect of the flower recalls the elements of θέλγειν. Apollo uses the verb to inflict fear or forgetfulness upon a victim. Θέλγειν combines with other words to connote persuasion and deception. C. ends this chapter with a discussion of the power of love and seduction in Homer.

Chapter Four is a study of the Sirens, and this topic seems to be at the heart of C.’s interest in the subject matter. (One wonders why this book was not more obviously centred on the magic of the Sirens rather than bringing in the Magoi in the first place.) In this section also the structure of the work as a dissertation is most obvious and most intrusive upon the scholarly writing. C. begins the chapter with a reinterpretation of the Homeric episode involving the Sirens in the light of the semantic range of θέλγειν discussed above, both positive and negative. Siren-song, he concludes, is essentially unlike the epic poems themselves because it is deceptive; because it is deadly (as opposed to epic which brings about immortality); because it is funerary. Then follows a discussion of the Sirens in art: C. summarizes four vase paintings, with proficient use of iconographic methodology and terminology, but very little discussion of the paintings or how they contribute to the argument as a whole, except for the question of why the Sirens bear instruments in the images when they do not in Homer—and this despite the obligatory proem stating that vase paintings are not mere illustrations of texts (108). C. then spends some time on the onomastic coincidence between some of the individual Sirens and Muses, showing how this casts light on the domain of θέλγειν. He focuses much of this argument on the adjective λιγύς or λιγυρός, and on musical vocabulary and birdsong (115-21). In a further digression on auloí, C. over-emphasizes the funerary and mournful context of the Sirens’ song in order to draw attention to parallels between it and the mind- and mood-affecting powers of lamentation (121-7). A further detour into discussion of the chorus of Deliades in the Hymn to Apollo centres around the philological argument as to whether the instruments described in that passage are castanets or percussion (the image of a Siren with a tambourine is also recalled in this discussion, in what may be the only sophisticated iconographical argument in evidence.) Finally, C. rehearses two telling parallels: that between Sirens’ song and ritual funerary lamentation, and that between the Sirens themselves and rhapsodes or storytellers.

Chapter Five, “La Pharmacie de Circé”, which together with the preceding dominates the volume, moves away from music as the instrument of charming to the use of drugs. C. begins this chapter by disagreeing with Scalera McClintock ( PP 1999), whom he characterises as arguing that Circe is not a magician (using the “emic” approach), but that what she does is, somehow, magic, according to the rational/irrational dichotomy. C. proposes to argue that Circe is indeed a “magicienne”, if only retrospectively, in terms of Greek concepts that in the fifth century were labelled μαγεία (141). From this promising start, C. then goes on to use genealogy and geography to further tie Circe to the Magoi. This is less convincing: Circe is related to Medea who is the mythical ancestor of the Medes and therefore the Magoi. For a few further pages, then, the discussion covers the episode of metamorphosis from the Odyssey, where the magical terms φάρμακα, ῥάβδος, and μῶλυ occur. C. argues that Odysseus’ sailors are not physically transformed into swine, by translating ὥς τε σύες in Od. 10.283 as “comme des porcs”. He then claims that the ῥάβδος is a herding stick, not a magic tool of any kind (following Stanford 1947 on 10.238-9, whom he does not cite), but this argument needs to take account of the fact that a similar ῥάβδος is used by both Athena and Poseidon in the Iliad and Hermes in the Odyssey in the performance of mind-altering activities (passages cited by C. in a previous chapter). He then goes on to point out, rightly, that the herb μῶλυ cannot be identified as any real-world plant (145).3

A discussion of the verb κηλέειν is less well-structured and so less convincing; C. uses passages from Hesiod’s Opera 464 and the Hymn to Demeter 227-30 to link κηλέειν and its cognates with φάρμακα. The section begins with the suggestion that θέλγειν operates through the medium of song and κηλέειν through drugs, but ends up having to admit that both verbs include both domains (153). C. discusses the ambiguity of φάρμακα in a few pages, calling upon Plato here as he will again in a later chapter. This single Greek word may represent a cure or a poison; good or bad connotation; herbal or chemical drugs; and may be taken in a literal or metaphoric sense. C. passes rather briefly over the question of women and φάρμακα, only mentioning that all users of these drugs are females such as Circe, Medea, Deianeira, and Helen (omitting to mention the facts that Deianeira was given her φάρμακα by a male, that the Asklepiadae in Homer are also πολυφαρμακός, as is Chiron the centaur, and the god Paieon). He explains this gender imbalance by referring to mythic geography, the focus on Egypt and places far from Hellas such as Colchis and Aeaea.4 Finally he argues that the ambivalence of φάρμακα throughout Greek thought helps to set the scene for the ambivalence of the Magoi, concluding—not entirely convincingly—that the Magoi are somehow equivalent to φάρμακα in Herodotus’ account of them (159).

Part Three of this work is titled “Penser la Mageía“, and begins with the sixth chapter, “Au Pays des Liens”, which focuses on curse tablets. This chapter begins with a few pages summarising the state of the field, which is useful, and is riddled with footnotes referencing the magisterial works of David Jordan and Roger Tomlin, as one would expect. C. does make some precarious claims, however, including the idea that the practice of creating defixiones on lead was a Middle Eastern invention, perhaps even brought to Hellas by the Magoi (63). He argues that the intention of this chapter is to put curse tablets into their full context in the Greek world, both as texts, as material objects, and as objects of the Greek “imaginaire”. The argument about the status of these curses as written or as oral texts is rehearsed, with C. remarking that curse tablets were not meant to be read, and later concluding that they were not merely (pace Faraone and Graf) records of a spoken ritual. He argues that one cannot, after all, comment on the “missing” oral text that these written words would be a record of, and also makes the point that such a reading privileges the more elaborate texts at the expense of the majority of texts which are little more than names and sometimes a significant verb and a chthonian divinity. C.’s position comes close to relying on an ex silentio argument of sorts; recognising that we can not comment with any confidence on the spoken text of which no record survives is not the same as concluding that there was no such oral dimension behind the often brief written remains.

This chapter then proceeds to focus upon two example texts, both unusual in their own way, and neither especially relevant to the question of imports from Asia Minor. The first of these is a curse tablet of the early fifth century from Selinous in Sicily; C. gives a sketch, the text, and a translation, then mentions a couple of textual difficulties. This text, however, seems principally to offer C. the opportunity to discuss the word καταγράφειν, which is common in such texts. That this is a term of official terminology, usually meaning to list in an account or a legal accusation, is adduced as further evidence that these texts are not oral. C. also rightly observes that curse texts such as this one draw legitimacy through their borrowing of vocabulary from official and legal documents. The second text discussed is a very usual tablet from Carystos, which is a hybrid of the defixio and the “Voodoo Doll”. This text, again sketched, quoted, and then translated, allows C. to kick off a discussion of the cognate words κάτοχος and κατέχειν (although the verb does not appear in this tablet). Some detailed discussion of the role of binding and body parts also follows from this (179-80), along with some general remarks about the function and effectiveness of cursing and magic. C. concludes that given this background of Greek ritual and belief, there is no need to assume that the Magoi introduced “oriental” magic to Hellas (183). This chapter then ends with some brief discussion of Plato’s references to and condemnation of cursing, putting these in the context of his philosophical and political thought. Plato is not condemning cursing per se, according to C., but the concept of absolution and purification that leads only to further crime. C. concludes again at the end of this chapter that the practice of cursing was a product of Greek culture, and not introduced by the Magoi, but that it became associated with the concept that the Greeks called μαγεία (188).

Staying with Plato, the seventh and final chapter, “La Magie de Platon”, deconstructs the dichotomy between “elite” philosophy and “vulgar” magic. This chapter traces the use of words related to Magos and other language that C. has highlighted in this volume, through the works of Plato. In the Republic, a Magos is a sophist; in references to Homer, Plato associates the charming of the Sirens to bad, corrupting persuasion. Meno compares Socrates to a γοής while simultaneously using language that recalls the vocabulary of curse tablets. C. takes this one step further and argues that Plato is being doubly mischievous, playing with words in the same way as Gorgias who links rhetoric with magic; Plato is therefore not only allowing his Socrates to be called a trickster, but also a sophist. The argument goes into more detail with an analysis of the Phaedo, which also uses the language of magic to describe philosophy: Socrates’ teachings are purification, are an ἐπωϊδή against the fear of death. C. discusses this “metaphorical” sense in which Socrates is a γοής, and talks about the power of magic and shamanism (an unhelpful term which detracts from rather than adding to the argument).5 Other references in Plato (especially the Alcibiades) make Socrates a Magos in the sense of an inheritor of oriental wisdom, a religious sage who incarnates the concept γνῶθι σεαυτόν, the same sense of Magos that Apuleius draws on several centuries later to defend himself from the criminal charge of using magic.

In a very short, two page conclusion tacked on to the end of the book, C. attempts to draw together some of the disparate topics covered in his dissertation. The Magoi, he begins, “entered the Greek city” in the fifth century B.C.E., both as a term in language and as a concept for understanding certain practices and beliefs. Gorgias associated μαγεία with Homeric concepts, thus entrenching it in Greek culture. Plato turned this around and made magic equivalent to sophistry and therefore to corruption and moral degradation. In post-classical times and up to the present magic has come to be equated with bad religion, but also—and also thanks to Plato—with mystic wisdom.

It would have been helpful if C. had somewhere in this book explained his own beliefs about who the Magoi that “entered” Hellas actually were, and summarised the evidence for their presence and existence. Were the Magoi to whom he refers lone Persians, coming to Greece in the wake of Darius’ and Xerxes’ defeats? Were they in fact members of the caste of Magoi described by Herodotus? Or were they perhaps Greeks come from Ionia (or even natives to the mainland), itinerant ritual practitioners and mendicants but posing as wise foreign religious experts? Did they exist at all outside of the popular imagination and a new vocabulary for describing native figures who always had existed, or was Magos merely another word for the agurtai and goetes ? This reviewer finds it difficult to assess C.’s arguments in places without understanding in precisely what sense he means the reader to take this term.

At times the content of this book is disjointed and uneven. The striking amount of space spent discussing the Sirens and Circe in Homer in great detail, and lengthy digressions on philological details, for example, while not uninteresting or worthless in themselves, do seem disproportionate to their function in the discussion of Magoi and the language of magic. The discussion of the Magoi in Herodotus and other fifth century sources would make a good article or monograph; the close reading of Odyssey 10 is also good and should stand alone; the discussion of curse tablets is good if thin. All of these lead to conclusions that are important to C.’s argument, but these conclusions would stand up with less background and more discussion of their significance. These passages are clearly necessary to the dissertation, where the candidate is proving his worth and demonstrating his research skills and argumentative method, but they feel excessive in a monograph. This may seem like an invidious criticism, but it should not detract from the value of the research in this volume.

There are few errors and inconsistencies in this text. C. adopts an unambiguous transcription from the Greek that combines accents with macrons for eta and omega; sometimes, however, especially in footnotes and extended quotes, he uses Greek script. χθόνιος is rendered variably as Chtonios or Chthónios; Χ may become either ch or kh, etc. On p. 11 n. 14, S. J. Tambiah’s initials are given as “S. H.” (but it is correct in the bibliography). Most unfortunate is the (admittedly minor) typo resulting in the absence of an iota in line 1 of the already much corrected Greek text of the Carystos tablet (174).


1. André Bernand, Sorciers grecs, Paris, 1991; Fritz Graf, La magie dans l’antiquité gréco-romaine: idéologie et pratique, Paris, 1994; Matthew Dickie, Magic and Magicians in the Greco-Roman World, London 2001; Georg Luck, Arcana Mundi: Magic and the occult in the Greek and Roman worlds, Baltimore, 1985; Daniel Ogden, Magic, Witchcraft, and Ghosts in the Greek and Roman Worlds: a sourcebook, Oxford, 2002.

2. Christopher Faraone, Ancient Greek Love Magic, Harvard, 1999; Naomi Janowitz, Magic in the Roman World: Pagans, Jews, and Christians, London, 2001; Silke Trzcionka, Magic and the Supernatural in Fourth Century Syria, London, 2007; Derek Collins, Magic in the Ancient Greek World, Oxford, 2007.

3. But he needs a reference to J. Stannard, ‘The Plant Called Moly’, Osiris 14 (1962), 254, or, more recently, J. Scarborough, ‘The Pharmacology of Sacred Plants, Herbs, and Roots’, in Magika Hiera, 1991, 165 n. 24.

4. But needs to refer to, e.g., J. Winkler, The Constraints of Desire, 1990 (esp. ch. 3, “The Constraints of Desire”) and other recent work on social and cultural explanations for the dominance of women in literary magic.

5. He cites Meuli, Dodds, and Burkert on “Greek shamanism”, but not J. N. Bremmer’s 1983 The Early Greek Concept of the Soul and 2002 The Rise and Fall of the Afterlife, which between them strongly challenge the view that there is anything convincing to tie Central Asian shamanism to Hellas in any way.