Lenfant ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 Article copyright held by the author(s) and made available under the Creative Commons Attribution License CC-BY https://creativecommons.org/licenses/by/4.0/ Eunuchs as Guardians of Women in Achaemenid Persia: Orientalism and Back Projection in Modern Scholarship Dominique Lenfant N THE MOST RECENT VERSION of the Oxford Classical Dictionary, one can read about eunuchs that “The Persian king employed them prominently as guardians of his harem and loyal protectors of his throne.”1 This is just one example among many, since, as a matter of fact, the function of guardians of the harem has frequently been assigned to ancient eunuchs, including in scholarly literature. It is true that, in the OCD entry, eunuchs are credited with two main functions (they are also presented as “protectors of the throne”), but their role as guardians of the royal “harem” comes first, and this is cer- tainly not incidental. Applying the notion of harem to the ancient Persian context is controversial, and I rank among those who consider the term to be irrelevant, confusing, and conveying anachronistic con- notations. Having already argued against its use,2 I will avoid that topic here, and focus independently and specifically on the question of eunuchs as guardians of women. Since our sources on eunuchs in the Persian Empire are almost exclusively classical,3 the question to which we may 1 E. D. Hunt, “Eunuchs,” OCD3 (1996) 569, unchanged in the 4th on-line- edition of 2012. 2 D. Lenfant, “The Notion of Harem and its Irrelevance to Women of the Persian Court,” AncSoc 50 (2020) 13–27. 3 The interpretation of non-Greek evidence on eunuchs is doubtful. See I DOMINIQUE LENFANT 457 ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 hope to find an answer is not so much whether eunuchs were the guardians of secluded women for the Persians, but whether Greeks saw them as such. In other words, were the eunuchs of the Persian Empire considered and represented by Greeks as guardians of women? As many scholars have thought that the etymology of the word εὐνοῦχος was a reflection of their function as guardians of women, I will first examine this etymology and argue that it is a false proof. Second, I will review Greek allusions to eunuchs in their writings, and show that the function of guardians of women does not form part of the picture. Third, I will argue that this fantasy is due to the back projection onto antiquity of a modern image of the ‘Orient’, and therefore an example of the orientalism which has been denounced by Edward Said. Last, the assignment of this fantasy to the ancient Greeks may for its part be labelled as an example of occidentalism. 1. εὐνοῦχος and the assumed etymological clue The idea that among the Persians eunuchs were guardians of women can be observed in modern dictionaries, which do not hesitate to build on etymology. This is true, first, for some authoritative dictionaries of modern languages. For example, Le Grand Robert de la langue française (2001 ed.) naturally has the French “eunuque” deriving from Greek eunoukhos via Latin eunuchus, but thinks it right to explain that eunoukhos means “properly ‘who guards (ekhein, ‘to have, to hold’) the bed (eunê ) of the women’.”4 More surprisingly, this is also true of diction- aries of ancient Greek like the Greek-French dictionary of Bailly, which reads s.v. εὐνοῦχος “eunuch (litt. guardian of the ___ e.g. R. Pirngruber, “Eunuchen am Königshof: Ktesias und die altorienta- lische Evidenz,” in J. Wiesehöfer et al. (eds.), Ktesias’ Welt – Ctesias’ World (Wiesbaden 2011) 279–312; D. Lenfant, “Ctesias and his Eunuchs: A Challenge for Modern Historians,” Histos 6 (2012) 257–297, esp. 279–281; M. Waters, Ctesias’ Persica and its Near Eastern Context (Madison 2017) 24–29. 4 “proprement ‘qui garde (ekhein, ‘avoir, tenir’) le lit (eunê ) des femmes’.” 458 EUNUCHS AS GUARDIANS OF WOMEN ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 bed, i.e. guardian of the women).”5 The fact is not purely anec- dotal since it is not rare to refer to etymology to explain the authentic meaning of a word—democracy as the sovereign power (kratos) of the people (demos), philosophy as love (philo-) of wisdom (sophia), and so on. Now, what does the etymology of eunuch actually tell us? Some dictionaries are in fact more cautious and factual than those cited above. For example, according to the Dictionnaire de l’Académie française (ed. 9, 1986–), French “eunuque” is “bor- rowed, through the intermediary of Latin eunuchus, from Greek eunoukhos, ‘guard of the bed’, from eunê, ‘bed’, and ekhein, ‘to have’.”6 In other words, women are not mentioned. In the same way, the OED (ed. 2, 1989) concludes from the two com- ponents that “the literal sense is thus a bedchamber guard or attendant.” The same holds for the Dictionnaire étymologique de la langue grecque of Chantraine, who translates “guard of the bed, eunuch.” He refers to an article of Maas on Greek designations of castrated men.7 Maas explains the etymological sense of εὐνοῦχος as “keeper of the bed” of others,8 and he specifies that 5 As will be seen below, LSJ widens the field of action of eunuchs, “em- ployed to take charge of the women and act as chamberlain (whence the name, ὁ τὴν εὐνὴν ἔχων),” but does not illustrate the guarding of women, which is however mentioned in first position. 6 “Eunuque” est “emprunté, par l’intermédiaire du latin eunuchus, du grec eunoukhos, ‘gardien de la couche’, de eunê, ‘couche’, et ekhein, ‘avoir’.” 7 E. Maas, “Eunuchos und Verwandtes,” RhM 74 (1925) 432–476, re- views the different words designating castrated men in Greek. 8 “εὐνοῦχος bedeutet also ‘die Schlafstätte anderer “behütend” ’ ” (437); on εὐνοῦχος, 437–439. Maas even thinks that in a fragment of Sophocles’ Troilos (fr.789) the word should be understood in its etymological meaning: εὐνούχοισι ὄµµασιν would mean “to the eyes who keep the bedroom.” The idea was perhaps inspired by a Byzantine dictionary: Additamenta in Etymo- logicum Gudianum s.v. εὐνοῦχοι· οἱ τὴν εὐνὴν ἔχοντες, ὡς γεοῦχοι. εὐνούχους δὲ Σοφοκλῆς καὶ τοὺς ὀφθαλµοὺς τοὺς εὐνῆς καὶ ὕπνου µὴ µετέχοντας, τουτέστι τηροῦντας· “οὐχ ὅπου λαµπάδες εὐνούχοισιν ὄµµασιν”). DOMINIQUE LENFANT 459 ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 the word designated castrated men because the keepers of rooms were castrated men and that the use of the word was then expanded to all sorts of castrated men, that is to say to the castrated men with their various functions. As can be seen, women are not mentioned here either. Let us then examine the components of the word. First, -ουχος means “who has/holds” (e.g. λαµπαδοῦχος, “who holds a torch”), but also “who protects” (e.g. δῆµουχος, “protector of the people/land”).9 Although the meaning of “protecting” is not the more common one,10 it is the only one that is possible in combination with εὐνή, the “bed.” The second component, the bed, is not specified to be that of women. In fact, Greek texts, as a general rule, do not give support to the idea that eunuchs above all protected a bed from the incursion of lovers. I know of only one possible exception, in a late partly fictional biography by Philostratus, the Life of Apollonius of Tyana (1.37), from the third century A.D., many centuries after the end of the Persian Empire. Recall that the hero of the story is a Greek charismatic philosopher and wonder-worker who travels across the world with his disciple Damis, and that his behaviour and sayings exemplify how to be wise. When the Greek heroes of the story are visiting the king (of the Parthians) in Babylon, an incident occurs: a eunuch has been caught lying with a concubine of the king and “doing everything seducers do” (ὁπόσα οἱ µοιχοὶ πράττων). Now, the 9 Dozens of words built on -ουχος are listed by P. Kretschmer and E. Locker, Rückläufiges Wörterbuch der griechischen Sprache (Göttingen 1977) 527– 528. 10 This certainly explains why Byzantine etymological dictionaries specify this sense of ἔχειν. For instance, Ps.-Zonaras Lexicon: Εὐνοῦχος· ἀπὸ τοῦ τὴν εὐνὴν ἔχειν ἤγουν φυλάσσειν. ἔχειν γὰρ τὸ ἐπιµελεῖσθαι φασὶ καὶ φυλάσσειν. Etym.Magn. 394: Εὐνοῦχος· Σπάδων, τοµίας· ἀπὸ τοῦ τὴν εὐνὴν ἔχειν καὶ ἐπι- µελεῖσθαι καὶ φυλάσσειν. 460 EUNUCHS AS GUARDIANS OF WOMEN ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 episode is actually forged to demonstrate what Apollonius has maintained just before (against a received idea expressed by his companion Damis), that a eunuch may be in love and feel de- sire. The wise man’s purpose is to demonstrate that the eunuch cannot be the model of self-control (1.33–34). Although many ingredients of the episode are inspired by the conventional image of the Persian court (satraps, magi, eunuchs, concubines, proskynesis, the King’s Eye, the King’s ears, and so on),11 and in spite of references to the Greco-Persian wars (1.25), the setting is not Achaemenid, since the story is set in the Parthian Empire (1.21), under king Vardanes I in the first century A.D. Ad- mittedly, the guilty eunuch has been caught by the “attendants of the women’s quarters” (οἱ ἀµφὶ τὴν γυναικωνῖτιν),12 who are obviously eunuchs themselves (the one who explains the case is “the oldest of the eunuchs,” ὁ πρεσβύτατος τῶν εὐνούχων). This may reflect a real function of eunuchs, although their activity in the women’s quarters is rather depicted as dedicated to the adornment of the concubines. Moreover, this remains an exception in Greek literature, without any equivalent in extant texts about Achaemenid Persians themselves, so that we are not compelled to consider that the bed kept by eunuchs was neces- sarily that of women, especially in Greek eyes. On the other hand, the bed of the King himself undoubtedly needed to be kept. In the Cyropaedia (7.5.59–60) Xenophon explains why Cyrus decided to entrust his guard (φυλακή) to 11 On Greek and Roman pictures of the Parthians see the comprehensive study of C. Lerouge, L’image des Parthes dans le monde gréco-romain (Stuttgart 2007). 12 Translators often give a biased version of the expression as “guards/ guardians of the harem” (so F. C. Conybeare, P. Grimal, C. P. Jones). On the reason why γυναικωνῖτις should not be translated “harem” see D. Len- fant, “Le ‘harem’ du Grand Roi est-il une invention des Grecs? Les enjeux de traductions ‘orientées’,” in D. Agut-Labordère et al. (eds.), Achemenet. Vingt ans après. Études offertes à Pierre Briant (Leuven 2021) 247–256, esp. 250– 252. DOMINIQUE LENFANT 461 ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 eunuchs—his own guard, and not that of women. And just before explaining why guards were needed, the author specifies those occasions on which the King was especially in danger, and which rendered the guards useful: “nowhere are people (ἄνθρωποι) more exposed than in their meals, in their drinking parties, in their bath, in their bed13 and in their sleep” (7.5.59). The bed is one of the places where people are especially weak. In fact, several kings were assassinated in their bedrooms: apart from Candaules, who is murdered in his bed (Hdt. 1.12), the Magi, in Herodotus’ telling (3.77–78), are killed in the “men’s quarters,” whereas in Ctesias (FGrHist 688 F 13.16) the Magus is assassinated in his bedroom where he is with a con- cubine, and his murderers have benefited from the complicity of a eunuch;14 Xerxes is murdered in his bedroom (κοιτών), thanks to the complicity of the eunuch in charge of his bedroom;15 and finally, when Tiribazus and Darius conspire against Artaxerxes II, they plan to enter his bedroom by night and kill him while he is sleeping, but a eunuch reveals the plot (Plut. Artax. 29). Thus, eunuchs are among those who must protect and inform the King, and consequently the ones whose help is necessary for a plot. This role is exemplified when the Seven meet “the eunuchs who carry messages” (τοῖσι τὰς ἀγγε- λίας ἐσφέρουσι εὐνούχοισι): these ask the Seven why they have come, and they both threaten the watchmen for letting the Seven pass and try to prevent the plotters from going on (Hdt. 13 The word used here is κοίτη, not εὐνή, but these are synonyms. See e.g. Hdt 1.9 and 10, where Candaules’ bed is successively designated by these two words. 14 The Seven could enter the palace thanks to the eunuch Bagapates, who held all the keys and could take the arms of the Magus beforehand. 15 Diod. 11.69.1–2, Μιθριδάτην τὸν εὐνοῦχον, ὃς ἦν κατακοιµιστὴς τοῦ βασιλέως καὶ τὴν κυριωτάτην ἔχων πίστιν, “Mithridates the eunuch, who was the king’s chamberlain and enjoyed his supreme confidence” (transl. C. H. Oldfather). 462 EUNUCHS AS GUARDIANS OF WOMEN ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 3.77.2). Of course, there are failures, but eunuchs are especially valued because they are supposed to be faithful (πιστοί),16 and Xenophon compares them in general to faithful dogs. For all these reasons, nothing prevents us from believing that etymology points to the function of a servant expected to be a loyal “protector of the bed” of his master—that is, a protector of his life—far more than to a guardian of women dependent on this master, whom they should watch over from a sexual and moral point of view. Be that as it may, etymology has many limitations, and the meaning of a word is rarely restricted to its original com- ponents—quite apart from the fact that we do not know which was the corresponding word in the Persian world.17 In Greek, then, εὐνοῦχος means a castrated man, and generally a cas- trated servant, who is not necessarily specialised in the keeping of the bedroom.18 Now, we must turn to the texts, in order to see whether the main function ascribed to eunuchs is to guard women. 2. Eunuchs and their functions in Greek texts: guardians of women? Are the functions of eunuchs related to women, and in what way? If we leave the dictionaries behind and turn to specialised historians, the relationship with women seems to be taken as self-evident. Peter Guyot, author of a book on eunuchs in 16 See D. Lenfant, “Le mépris des eunuques dans la Grèce classique: orientalisme ou anachronisme?” in A. Queyrel Bottineau (ed.), La représen- tation négative de l’autre dans l’Antiquité. Hostilité, réprobation, dépréciation (Dijon 2014) 423–442, esp. on Herodotus and Ctesias. 17 Cf. Lenfant, Histos 6 (2012) 257–297, esp. 279–281 (with bibliography). 18 That eunuchs had various functions and that Greeks did not perceive the etymological sense of the word is suggested by the fact that Plutarch calls some of them thalamepoloi, “attendants of the bed-chamber” (Alex. 30.1, about the servants who were captured with the women of the family of Darius III). DOMINIQUE LENFANT 463 ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 classical antiquity,19 more recently has offered this summary in the entry “Eunuchs” in the Neue Pauly: “The origin of the practice of castrating young boys before or during puberty in order to use them as slaves for domestic service is uncertain but it was probably related to the development of polygamy (eunuch as ‘guardian of the harem’).”20 The link between eunuchism and polygamy is here considered probable,21 and the first eunuchs would have been “guardians of the harem.” Guyot certainly deals here with the origin of castration, which means that he does not exclude wider functions after that, but he gives us to understand that this function of “guardians of the harem” persisted afterward. Llewellyn-Jones, in a paper signifi- cantly entitled “Eunuchs and the Royal Harem in Achaemenid Persia,” does not hesitate to speak of “the natural connection and affinity between women and eunuchs found in the ancient sources.”22 In the same way, in a book on ancient Persia in- tended for a general audience, Huyse writes: “il ne fait guère de doute que les femmes à la cour achéménide, concubines comme épouses et princesses, ont disposé d’appartements privés sur- veillés par des eunuques.”23 The reference to “nature” and to self- evident truth are not really arguments at all. What in fact is the connection between eunuchs and women 19 P. Guyot, Eunuchen als Sklaven und Freigelassene in der griechisch-römichen Antike (Stuttgart 1980). 20 P. Guyot, in DNP 4 (1998) 256–258. 21 This can be contested, first because eunuchs are very often in service to men—irrespective of the number of their wives; second, because whereas some kings like Darius were undoubtly polygamous, it is not certain of all of them (kings were first of all polygynous; I have tried to analyse the issue in “Polygamy in Greek Views of Persians,” GRBS 59 [2019] 15–37); third, be- cause there were differences between wives and concubines. 22 L. Llewellyn-Jones, in S. Tougher (ed.), Eunuchs in Antiquity and Beyond (London 2002) 19–49, here 34 (my italics). 23 P. Huyse, La Perse antique (Paris 2005) 240 (my italics). 464 EUNUCHS AS GUARDIANS OF WOMEN ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 in classical sources? Let us review the main allusions to the eunuchs of the Persian Empire and examine their link to men or women, the nature of their relationship with both cat- egories,24 and their functions in different cases. • Athenian tragedy. We know two fragments that explicitly allude to eunuchs.25 One comes from Phrynichus’ Phoenissae (476 B.C.): a eunuch is preparing seats for members of the Per- sian court and he announces Xerxes’ defeat (fr.8). Thus, he appears in connection with men. The other comes from Soph- ocles’ Troilos (admittedly in a Trojan setting, but in tragedy this is usually pictured as a Persian setting):26 a man says that his testicles were cut off by Hecuba (fr.620). Thus, he has been in contact with a woman, but he is far from being her guardian. • Herodotus. In his narrative there are only two cases where eunuchs appear in connection with women, in one case with the King’s wives, in the other with concubines. The first case is well known: when the Greek physician Democedes has man- aged to cure Darius, the King gives him in return a present of two pairs of golden chains, and then sends him off to his wives, in order that they also give him rewards. The physician is escorted by eunuchs, who also inform the women that this is the man who gave life back to the King (3.130.4). Eunuchs are, then, associated with the King, with a foreigner, and with the King’s wives, but this does not make them guardians of 24 Actually, women constitute more than one category, since at the court there were concubines, wives, and other women of the royal family, and among concubines themselves there were probably different statuses. 25 D. Lenfant, “Des eunuques dans la tragédie grecque. L’orientalisme antique à l’épreuve des textes,” Erga-Logoi 1.2 (2013) 7–30. 26 H. H. Bacon, Barbarians in Greek Tragedy (New Haven 1961) 71–72, 101–104, 124, 146–148; E. Hall, Inventing the Barbarian. Greek Self-definition through Tragedy (Oxford 1989) 120–121; D. Lenfant, “L’amalgame entre les Perses et les Troyens chez les Grecs de l’époque classique: usages politiques et discours historiques,” in J. M. Candau Moron et al. (eds.), Historia y Mito. El Pasado Legendario como fuente de Autoridad (Malaga 2004) 77–96. DOMINIQUE LENFANT 465 ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 women.27 The other case is rather different: among the many non-combatants who followed the military forces under Xerxes in his expedition against the Greeks, eunuchs are listed after the cooking-women and the concubines among those who had to be fed, including beasts of burden and dogs (7.187). Eunuchs can be seen here as servants, possibly caring for concubines, but not exclusively. Indeed, it is clear that the majority of the eunuchs mentioned are in the service of men:28 when asked by Astyages, Harpagus claims that he sent some of them to check that Cyrus had been killed (1.117); as we have seen, eunuchs try to prevent the Seven from entering the Magi’s apartments (3.77); the eunuchs who escort Democedes to Darius’ wives are first with the King (3.130); and Hermotimus, “who has won a place second to none among the eunuchs serving the King,” is designated by Xerxes to be guardian (φύλακος) of his illegitimate sons during their travel back to Asia after the defeat at Salamis (8.104). Hermotimus, then, is in the service of the King and in contact with his sons, rather than with women. The allusion to Hermo- timus is an occasion for Herodotus to explain the “barbarian” practice of castration: “among the barbarians, eunuchs are more valuable than males with testicles because of their total trustworthiness” (8.105). Nothing leads us to believe that this trustworthiness (πίστις)—a characteristic of eunuchs often men- tioned by Greek authors—especially concerned women. 27 As Keaveney rightly points out in a paper on eunuchs at the Achaemenid court, “the eunuchs simply act as ushers” here: in O. L. Gabelko et al. (eds.), Iranica: Iranian Empires and the Graeco-Roman World from the Sixth Century B.C. to the Sixth Century A.D. (2017, in Russian; an English ver- sion was kindly sent to me by the author in 2013). 28 On their different functions see Lenfant, Histos 6 (2012) 271–272. The fact that eunuchs are more often in service to men than to women also holds for eunuchs in diverse societies at various times: S. Tougher, The Eunuch in Byzantine History and Society (London 2008) 13, highlights how this fact contrasts with popular imagination. 466 EUNUCHS AS GUARDIANS OF WOMEN ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 • Aristophanes. In Acharnians 117 the (alleged) eunuchs who escort the Athenian ambassadors returning from the court with a Persian are shown in the function of servants, in the service of men. • Ctesias. Ctesias is among those who have depicted the in- ternal relations at the court in most detail; he has been strongly criticized by scholars for having promoted “petite histoire” with doubtful historical foundations, and he is without contest the Greek author who mentions the greatest number of eu- nuchs. But even he does not depart from the rule we have seen so far: he did not represent eunuchs as guardians of women (or at least it was not striking enough to leave traces in the extant literature through which we know his work). In fact, eunuchs appear in Ctesias’ fragments as attendants serving members of the royal family, and far more often around men than women.29 The mother of Artaxerxes II admittedly has her own eunuchs, but, far from being guarded by them, she plays dice with her son in order to win another eunuch (one of her son’s), and wreak her vengeance on him.30 Most eunuchs are men- tioned as attendants or advisors to the king, and are assigned such an important place that for nearly every reign Ctesias specifies the names of the most influential of them.31 Their proximity to kings explains why some of them are seen taking part in plots or assassinations,32 or even contributing to install- ing new kings,33 while others are deemed to be most faithful be- yond death: some take the king’s corpse to Persia, and one 29 Lenfant, Histos 6 (2012) 269–270 (the whole paper is more generally on eunuchs in Ctesias’ work). See also Waters, Ctesias’ Persica 29–32, who lists Persica’s eunuchs with their respective actions. 30 Ctes. FGrHist 688 F 26.17. 31 E.g. F 13.9, 24, 51. 32 F 9.6, F 13.16, 33, F 15.48, 54. 33 F 13.15–16, F 15.50. DOMINIQUE LENFANT 467 ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 remains even sitting beside his tomb for years.34 Cyrus the Younger, brother of Artaxerxes II, also has his own eunuchs, who are found mourning near his corpse after his death at Cunaxa.35 All these eunuchs are associated with service to male figures. They do not feature as guardians of women, and the women—at least those of the royal family—do not appear to be especially watched over. • Platonic corpus. Eunuchs are seen educating the sons of the Persian King, alone (Alc. I 121D) or with women (Leg. 694E– 695B). They are not watching the women themselves. • Xenophon. The Cyropaedia mentions eunuchs in three sorts of contexts: (1) Gadatas, an Assyrian noble who has been cas- trated by his king (incidentally the king not of Persia but of Assyria),36 is a special case, since he is not a slave (4.6, 5.2.28, 5.3.8–19). In his desire to take revenge, he joins Cyrus against his own king (5.3.8–17). He is in command of an army, and helps Cyrus in his war against the king of Assyria and Bab- ylonia (6.1, 7.3, 5) before becoming a dignitary at his court (7.3.17–4.2). He has nothing to do with being a guardian of women, but is admittedly a special case. (2) Panthea, an Assyrian woman, has eunuchs as servants, as well as female servants (6.4.11). She orders one of her eunuchs to carry a message to Cyrus (6.1.33–34). In other words, eu- nuchs are here around a woman, but in her service. And her husband Abradatas has his own eunuchs,37 so the possession of eunuchs is not specific to women. (3) In discussing the royal guard established by Cyrus after the capture of Babylon, Xenophon explains the advantage for a 34 F 13.9, 15, 23. 35 F 20.12.1. 36 The king was jealous because Gadatas had appealed to his concubine. 37 When he dies, his eunuchs and servants (τοὺς µὲν εὐνούχους καὶ τοὺς θεράποντας αὐτοῦ) dig his grave. 468 EUNUCHS AS GUARDIANS OF WOMEN ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 king in having eunuchs in his service: they are the most trust- worthy men (7.5.59–65). No woman is mentioned here either. • Alexander histories. Here we note three sorts of data. (1) When describing the non-combatants who accompany Darius III’s army, Curtius mentions an important group of eunuchs, but they are associated with the children (and not with royal women or with concubines),38 just like the women “who edu- cated them.”39 (2) Later, Curtius associates eunuchs and concubines, but without implying that the first are the guardians of the second. Instead he describes eunuchs as sexual slaves, saying that Alex- ander, like Darius, had 365 concubines, who were followed (sequebantur) by herds of eunuchs “who were also used to serve as women” (et ipsi muliebria pati adsueti, 6.6.8). (3) The last noteworthy case is that of a eunuch who looks after the King’s wife. According to Arrian (Anab. 4.19–20),40 after the battle of Issos, where the daughters, wife, and mother of the King were captured by Alexander, Darius was joined by the eunuch whom he had made the guardian (φύλαξ) of his wife (4.20.1, τὸν εὐνοῦχον τὸν φύλακα αὐτῷ τῆς γυναικός). The eunuch has fled, and Darius asks him first whether his daughters, wife, 38 Keaveney (n.27 above) has pointed out that “When the royal army was on the march, the eunuchs travelled separately from the royal women and the concubines. In no sense are they there to guard them and are explicitly attested as waiting upon them.” He cites Curt. 3.3.22–25, 4.10.18, 6.6.8, and Plut. Alex. 30. 39 Curt. 3.3.22–25: after the chariot carrying the King’s mother comes the chariot with his wife (coniunx), escorted by many women on horseback, then fifteen harmamaxai in which are the children (liberi) and the women who educate them (quae educabant eos) as well as a “herd of eunuchs” (spadonumque grex), then the 360 royal concubines (regiae pelices), mules and camels carrying the King’s money, then the wives of the relatives and friends of the King, then the many servants, and finally light-armed troops who bring up the rear. 40 See also Curt. 4.10; Plut. Alex. 30. DOMINIQUE LENFANT 469 ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 and mother are still alive, second whether his wife remained faithful to him, and third whether she had to suffer violence from Alexander. The eunuch clearly has the function of a φύλαξ, a protector, for the wife. Yet it is a function which does not exclusively concern women, but also—as we have seen with Hermotimus—the royal children and even the King himself (as seen above, in the Cyropaedia, Cyrus surrounds himself with eunuchs because he needs a φυλακή around his person, 7.5.58). Darius’ question about his wife’s faithfulness is not enough to make us confine the eunuch’s role to scrutinizing her virtue. After all, after being away, a Greek could put the same ques- tion to the female slave of his wife, just because of her closeness to her.41 Eunuchs had to take care of all, women or not, and to protect them against possible threats. To summarize: in Greek literature on the Persians, eunuchs have a range of functions (from the simple servant or preceptor to the court dignitary), most frequently in the service of men and in their company, and when they are with women it is more as servants at their command (as with men) than as mon- itors of their lifestyle.42 That is especially true of royal women (the mother of Artaxerxes II, Parysatis, has “her” eunuchs, just as the King has his own),43 but even when eunuchs are with concubines, and even if in practice they probably had to supervise them, they are explicitly mentioned as equivalent to servants, among the non-fighting people following the army of 41 See Lys. 1.18. 42 Cf. Lenfant, Histos 6 (2012), esp. 269–270 on Ctesias’ eunuchs. Keaveney (n.27 above) rightly points out that in most sources “eunuchs appear as the servants of noble women, not their guardians or keepers,” and that Ctesias certainly “has a eunuch in charge of female singers but it should be noted that this is at the Median court”—even if Ctesias “could be back- projecting practices of his age to a previous age,” it should be noted that the singers are not to be equated with wives and prominent concubines. 43 Ctes. F 26.17 = Plut. Art. 17. 470 EUNUCHS AS GUARDIANS OF WOMEN ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 the King, Xerxes or Darius III, with a possible sexual function. The only story in which eunuchs might appear at first glance to be guardians of women is the passage of Herodotus in which eunuchs lead Democedes to the King’s wives and inform them that the physician gave life back to the King (3.130.4). They act as servants who do what their master ordered and may go to his wives. As already stated, however, this does not imply that they are the guardians of these women, but at most that they are servants who may come into contact with them. Were they the only ones? After saying that each of the King’s wives dipped a bowl into a chest of gold, Herodotus points out that a servant (οἰκέτης), “whose name was Skiton, collected an enor- mous sum of gold just by picking up the staters that fell from the bowls.” Was this servant who followed (ἑπόµενος) Demo- cedes a eunuch? We do not know and that question does not seem important to Herodotus. All in all, according to Greek representation, there is re- markably little sign of any concern to watch over women, and this is especially striking for women of the royal family such as Atossa, Roxane, Amestris, Amytis,44 or Parysatis. 3. The back projection of modern orientalist prejudices In these circumstances, one can be surprised by the recurring modern definition of eunuchs as guardians of women. The prejudice is so strong that dictionaries quote passages without seeing that these do not exemplify their definition. For in- stance, the Dictionnaire of Bailly refers to Herodotus 1.117 and Xenophon Cyr. 7.5.60. As seen above, in the first passage the eunuchs have been sent by Harpagus to check that Cyrus had been killed and the second text concerns the eunuchs as con- stituting the personal guard of Cyrus. In both cases, this has nothing to do with women. 44 According to Ctesias, Amestris and Amytis had many lovers (F 13.32, 34, F 14.44). DOMINIQUE LENFANT 471 ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 A second example: LSJ admittedly mentions two fields of action for eunuchs, saying that they were “employed to take charge of the women and act as chamberlain (whence the name, ὁ τὴν εὐνὴν ἔχων).” But the three following references do not really illustrate that they “take charge of the women”: neither Cyr. 7.5.60 (already seen) nor the Acharnians of Aristophanes (117) includes any women. The only one where women are present is the Democedes episode (Hdt. 3.130), where the role of eunuchs is not explicitly confined to having charge of women. Why, then, have eunuchs been imagined as guardians of women? In my view, it seems clearly due to the back projection onto antiquity of a modern vision of the eunuchs of the ‘harem’, especially that of Ottoman sultans or Safavid Persia. It is significant that the authoritative Grand Robert de la langue fran- çaise, already mentioned for its etymological aspect, gives as a first definition: “In the Orient, Castrated man who guarded women in harems.”45 The first of the quotations that follow refers then to the eunuchs of the ancient Persians, and it is a passage of the Essays of Montaigne (1.25), which reads: “Après sa naissance, on le donnait [le fils aîné du roi de Perse] non à des femmes, mais à des eunuques de la première autorité autour des rois, à cause de leur vertu” (“After his birth, he [the eldest son of the king of Persia] was delivered, not to women, but to eunuchs of the greatest authority about their kings for their virtue”).46 Curiously, the author of the dictionary’s entry did not notice that this quotation contradicts his definition, since eunuchs do not appear here as “guardians of women in harems,” but as men of quality with influence on the King and entrusted with the education of his eldest son. But this quotation dating from the sixteenth century which 45 “En Orient, Homme châtré qui gardait les femmes dans les harems.” 46 Transl. Charles Cotton (New York 1910). 472 EUNUCHS AS GUARDIANS OF WOMEN ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 drew on a Greek source47 is followed, in the dictionary, by two quotations from the eighteenth century, which betray clearly what is in the lexicographer’s mind. The first one is extracted from Montesquieu’s Persian Letters, a novel in which eunuchs do in fact play the role of guardians of the seraglio, in which the wives of their master are secluded—in the (imagined) Persia of the early eighteenth century. The illustration is all the more perfect as it is the probable source of the definition! The second eighteenth-century quotation speaks volumes about the orien- talist component of the confusion between ancient and modern ‘Orient’. It is from the Essai sur les mœurs of Voltaire (197): Une autre différence (entre nous et les Orientaux), qui naît de nos usages avec les femmes, c’est cette coutume de mettre auprès d’elles des hommes dépouillés de leur virilité ; usage immémorial de l’Asie et de l’Afrique, quelquefois introduit en Europe chez les empereurs romains. Nous n’avons pas aujourd’hui dans notre Europe chrétienne trois cents eunuques pour les chapelles et pour les théâtres ; les sérails des Orientaux en sont remplis. Another difference (between us and the Orientals), which is born of our usages with women, is that custom of putting near them men stripped of their manhood; immemorial usage in Asia and Africa, at some point introduced in Europe with the Roman Emperors. We do not have today in our Christian Europe three hundred eunuchs for chapels and theaters; the seraglios of the Orientals are full of them. The idea of an eternal Orient (with its despicable features) is here absolutely explicit: what Easterners do in Voltaire’s time is what Easterners have done at all times (and vice versa). Such a postulate is far from being exceptional, including in scholarly literature, where the analogy is more implicit and obviously considered self-evident. In his exposition on eunuchs in RE, Hug systematically links eunuchs with the harem: he does not hesitate to translate εὐνοῦχοι with “Haremshüter” (449), i.e. 47 Montaigne draws on Plato Alc. I 121D. DOMINIQUE LENFANT 473 ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 “guardians of the harem,” and explains that this expresses their main function among Easterners (obviously conceived to be a uniform and timeless category).48 He argues that the castration of men was used “in the Orient” and was the sad consequence of slavery, polygamy, and a suspicious despotism from most ancient times. This preconception of a timeless Orient still has consequences in the mind of the more recent scholars quoted above,49 and it lies behind the implicit assimilation of the eunuchs of the Achaemenid Empire and those of Montesquieu. This is an illustration of the orientalism analysed by Said in his famous book.50 But that is not all, and it may be that all the possible traps have not yet been escaped: in the last decades, scholars who following Said have been critical of the notion of the Orient have attributed to ancient Greeks such a vision of eunuchs as linked with the harem.51 Now, to ascribe to Greeks a vision of 48 A. Hug, “Eunuchen,” RE Suppl. 3 (1918) 449–455, esp. 449. 49 Hunt, Guyot, Llewellyn-Jones, Huyse. 50 E. Said, Orientalism (London 1978). Historians may themselves be in- clined to reason on the basis of prejudices and analogy. The Encyclopaedia Iranica entry on eunuchs, which successively considers the Achaemenid, Sasanian, Early Islamic, Safavid, and Qajar periods, begins with a general statement about eunuchs as “castrated males who were in charge of the concubines of royal harems,” although the many functions then detailed for the Achaemenid period are nearly never linked with the ‘harem’ (the only exception is the eunuch who escapes “after the capture of Darius III’s harem by the soldiers of Alexander the Great”—in Plutarch’s Alexander 30, he is actually qualified as θαλαµηπόλος, “attendant,” the word for waiting- maids in the Odyssey (7.8, 23.293). 51 See H. Sancisi-Weerdenburg, “Decadence in the Empire or Deca- dence in the Sources? From Source to Synthesis,” in Achaemenid History I Sources, Structures and Synthesis (Leiden 1987) 33–45. More recently, as already seen, Llewellyn-Jones, in Eunuchs in Antiquity 34, although not especially critical of classical sources (and thus writing from a different perspective than did Sancisi-Weerdenburg), invoked the connection between women and eunuchs “found in the ancient sources.” Other examples in Lenfant, 474 EUNUCHS AS GUARDIANS OF WOMEN ————— Greek, Roman, and Byzantine Studies 61 (2021) 456–474 ‘the Orient’ which is the same as that of modern Westerners, despite the available sources, and to postulate that ancient Persia and the modern ‘Orient’ have been seen in just the same way and given rise each in their time to the same stereotypes in the mind of ‘Westerners’—is that not the same approach as what is denounced as orientalist? What becomes confused and timeless here is the ‘Occident’, which is assumed to have always had the same (regrettable) fantasy, and this can be seen as a form of occidentalism. The examination of Greek texts has shown that this was wrong on the specific issue of eunuchs and women, and that moderns could sometimes unconsciously rely on ‘facts’ that are in reality false, the anachronistic projection of later representations onto antiquity. The timeless Occident is no less disputable a notion. This shows how much, beyond the common opinion, doubtful analogies may pervade scholarship and enter into a circle where they are both the consequences and the causes of orientalism and occidentalism.52 July, 2021 Faculty of Historical Sciences University of Strasbourg CNRS-UMR 7044 Archimède dlenfant@unistra.fr ___ Histos 6 (2012) 277 n.105. 52 I would like to thank Tom Harrison for inviting me to deliver a first draft of this paper at the University of St. Andrews, and for having then polished my English. I am also grateful to the editorial board of GRBS and the outside referee whose comments helped me improve the paper.