Mythical Structures in Herodotus’ Histories

Madness—The Complexity of Morals in the Light of Myth and Cult

So far, we have seen how mythical parallels function as a kind of commentary, marking sacrilege as such. Now we will have a look at a group of sacrilegious acts that are connected with madness. They, too, have parallels with mythical paradigms, but are also connected to a ritual phenomenon: cultic ecstasy—a dimension which makes the cases in question even more complex than the ones we have already analyzed. I can therefore not promise a simple explanation for the phenomenon of madness in Herodotus. Instead, I will try to demonstrate several mythical and ritual instances related to the kind of madness Herodotus portrays, which will help to illuminate his perspective on madness, especially its paradoxical causality: on the one hand, the connection with sacrilege places madness in the field of divine punishment; on the other hand, the punitive nature of madness in Herodotus’ stories seems to be almost consciously veiled.

1. Double Madness and the Problem of Causality

1.1 Cambyses and Cleomenes:

The Persian king Cambyses and the Spartan ruler Cleomenes are the Histories’ most prominent madmen. The former is constantly violating both a “sociocultural” and a “theological code” (Munson 1991:46 etc.), as Herodotus states e.g. 3.38.1:

Πανταχῇ ὦν μοι δῆλά ἐστι ὅτι ἐμάνη μεγάλως ὁ Καμβύσης· οὐ γὰρ ἂν ἱροῖσί τε καὶ νομαίοισι ἐπεχείρησε καταγελᾶν.

I hold it then in every way proved that Cambyses was quite insane; or he would never have set himself to deride religion and custom.

Among other things, the sacrilegious acts of the mad king consist in the desecration of Egyptian sanctuaries (3.16; 25.3; 37). The utmost profanation is his slaughter of the Apis-bull (3.27–29), whose deity he explicitly denies while stabbing the animal in the thigh:

Ὦ κακαὶ κεφαλαί, τοιοῦτοι θεοὶ γίνονται, ἔναιμοί τε καὶ σαρκώδεες καὶ ἐπαΐοντες σιδηρίων;


Simpletons, are these your gods, creatures of flesh and blood that can feel weapons of iron?

The bull dies, his priests are whipped, the cult forbidden—Cambyses becomes guilty of active theomachy, of the killing of a god.

Of course the deed does not go unpunished. Cambyses will inadvertently hurt himself on his own thigh, in exactly the same place where he had hit the bull-god, and die of the gangrenous wound (3.64.3; 66.1–2). The punitive character of Cambyses’ death seems clear, especially as Herodotus explicitly emphasises the parallel with Apis’ death.

The other great madman of the Histories, Cleomenes of Sparta, is also a notorious blasphemer (6.75–84); however, madness seems to be the punishment for his deeds, not their cause. After having reported Cleomenes’ illegitimate entry into Athena’s sanctuary on the Athenian acropolis as early as 5.72.3–4, Herodotus later tells us about the madness of the king and its genesis (6.75–84). At first, he only relates the fact that Cleomenes becomes mad after having returned from exile (6.75). When put in fetters, the madman manages to get hold of a knife with which he mutilates himself and dies:

Κλεομένης δὲ παραλαβὼν τὸν σίδηρον ἄρχετο ἐκ τῶν κνημέων ἑωυτὸν λωβώμενος· ἐπιτάμνων γὰρ κατά μῆκος τὰς σάρκας προέβαινε ἐκ τῶν κνημέων ἐς τοὺς μηρούς, ἐκ δὲ τῶν μηρῶν ἔς τε τᾶ ἰσχία καὶ τὰς λαπάρας, ἐς ὃ ἐς τὴν γαστέραν ἀπίκετο καὶ ταυτὴν καταχορδεύων ἀπέθανε τρόπῳ τοιούτῳ …


Cleomenes took the weapon and set about slashing himself from his shins upwards; from the shin to the thigh he cut his flesh lengthways, then from the thigh to the hip and the sides, until he reached the belly, and cut it into strips; thus he died …

Herodotus lists three possible reasons as the cause of this madness, all having to do with the desecration of sanctuaries: the bribing of the Delphic Pythia to declare his rival king Demaratus illegitimate (believed in all of Greece corresponding with the Panhellenism of the sanctuary), the destruction of the sacred area in Eleusis, or the burning of the sanctuary in Argos (both credited to local sources). Having already related the story of the Delphic bribe in 6.66.2–3, Herodotus goes on to describe the Argive sacrilege. The passage closes with the repetition of the assumption of the Argives that this must have been the cause of the king’s madness (6.84), and a fourth explanation is mentioned: Cleomenes had gotten used to drinking undiluted wine like the Scythians and therefore had gone mad. Finally, Herodotus ends the discussion with his own verdict: ‘but to my thinking it was for what he did to Demaratus that he was punished thus’, 6.84.3, taking a clear stand on the subject of divine retribution.

At first glance, the punitive character of Cleomenes’ madness seems evident, whereas Cambyses’ punishment does not consist in madness, but in his lethal wounding, rendering the cause of his madness obscure. Herodotus does quote Egyptian sources connecting the beginning of the king’s madness with the killing of Apis (3.30.1), but as we have seen, he himself considers the sacrilege a consequence of Cambyses’ madness. Later, he seems to regard the mental disorder as a consequence of the king’s epilepsy:

Καὶ γάρ τινα καὶ ἐκ γενετῆς νοῦσον μεγάλην λέγεται ἔχειν ὁ Καμβύσης, τὴν ἱρὴν ὀνομάζουσί τινες· οὔ νύν τοι ἀεικὲς οὐδὲν ἦν τοῦ σώματος νοῦσον μεγάλην νοσέοντος μηδὲ τὰς φρένας ὑγιαίνειν.


For indeed he is said to have been afflicted from his birth with that grievous disease which some call ‘sacred’. It is not unlikely then that when his body was grievously afflicted his mind too should be diseased.

Nevertheless, Herodotus does believe in divine retribution in this case as well, as regards the wounding. And that is not all. If we examine more closely the passage in which the Egyptians identify the killing of Apis as the cause of the madness, we encounter the following phrase:

Καμβύσης δέ, ὡς λέγουσι Αἰγύπτιοι, αὐτίκα διὰ τοῦτο τὸ ἀδίκημα ἐμάνη, ἐὼν οὐδὲ πρότερον φρενήρης.


‘But Cambyses, the Egyptians say, owing to this wrongful act immediately went mad, although even before he had not been sensible.’

This might just mean that the Egyptians have got it wrong. However, a strikingly similar phrase in the case of Cleomenes renders this interpretation questionable (keeping in mind that Herodotus has clearly identified the bribe of the Pythia as the cause of madness, as we have seen):

κατελθόντα δὲ αὐτὸν αὐτίκα ὑπέλαβε μανίη νοῦσος, ἐόντα καὶ πρότερον ὑπομαργότερον.


These similar, ambiguous statements regarding the point, where the madness has started, are confusing, especially in the case of Cleomenes, where Herodotus seems to actually contradict himself. What does this vagueness mean? Is the madness divine punishment or is it not?

Regarding the ambivalent phrases, Rosaria Munson states a change of madness “in the surrounding narrative’s implicit distinction between ‘far’ and ‘near’” (1991:50) with both characters: the madmen first direct their deeds against strangers, then against close relatives (Cambyses kills brother and sister; 3.30; 32) or themselves (Cleomenes kills himself). This is correct, but it does not explain the origin of madness.

Let’s first note the fact that Herodotus does not shy away from the conception of divinely sent illness. The Lydian king Alyattes who has burnt down the sanctuary of Athena in Miletus becomes ill and immediately recovers after having built two temples to compensate for the destruction (1.19–22). The Scythians are plagued by a certain ‘female’ sickness after having plundered Aphrodite’s temple in Ascalon (1.105.4). One desecration is even punished with madness: the Athenians’ attempted theft of the Epidaurian effigies. In thunder and lightning, the thieves go mad and kill each other; the only survivor is stabbed to death by the wives of the victims—a virtual sparagmos (5.87).

Accordingly, Herodotus sees divine punishment as an important possibility—with the one variance of the loose connection between Cambyses’ madness and his epilepsy, and perhaps the Spartan hypothesis that Cleomenes’ madness originates from his drinking undiluted wine—but this explanation, too, could be religiously motivated, as we shall see below. Herodotus definitely does not attempt a detailed medical perspective on madness as pervades the Hippocratean corpus. This fails to explain why the madness of the two kings does not follow the pattern of sacrilege and divine retribution more clearly. However, this traditional pattern is not a simple one, either.

1.2 The Tradition of Madness:

Madness and sacrilege are connected in Greek thought. Yet, the question of the punitive character of madness is always puzzling and complex; there is no clear distinction between ‘going mad’ by oneself and ‘being struck’ by an outside force, as is shown by the change of diathesis in the conjugation of μαίνεσθαι—a passive verb in the present tense and the aorist, but middle in the future tense.

1.2.1 The Doubling of Madness in Myth :

Greek mythical instances of madness most often occur in tragedy, where the protagonists are commonly struck with madness by the gods. Nevertheless, the punitive character of madness is ambivalent here, too: in some way, there is always a double of the godsent madness, a twin situation that has been there from the beginning—whether in the character to be struck with madness, or in another person responsible, or merely as a threat, as shall be shown from several examples.

Although Ajax has sinned against the goddess turning down her help in war (as the seer Calchas reminds us, S. Aj. 758–777) it never becomes clear how much of his madness is punishment and how much is cause for punishment. Sophocles even has Calchas say that Ajax is punished with mania for being ἄνους, ‘mindless’ (763) and acting ὑψικόμπως κἀφρόνως, ‘arrogant and senseless’ (766).

The Euripidean Phaedra is also tormented by erotic mania, as is displayed in the first epeisodion of Euripides’ Hippolytos. In this case, the erotic madness is a form of punishment sent by Aphrodite, but Phaedra is completely innocent of the goddess’ rage. It is Hippolytos who is being disciplined for his irreverence; the maddened Phaedra is merely an instrument causing the youth’s death. However, there is a certain form of madness from the beginning, if we consider Hippolytos’ irreverence a kind of mania. This is not improbable, I think; compare the case of the madly irreverent Pentheus killed by his maddened mother Agaue, which shall be discussed later. Madness, therefore, is present in the Phaedra-story too, before the goddess sends it—but not in the target of the godsent madness: Hippolytos has madly offended the goddess, Phaedra goes mad with love—one kind of mania reacts to another.

Apparently, Orestes has the choice between mania and mania, sent by one or the other side. But do the Erinyes of his father actually drive him to murder his mother? There is no version of the myth in which they actually appear, but Orestes is driven by a mad desire for revenge even before he kills his mother, willing to commit the act with or without divine command. Afterwards, Clytemnestra’s Erinyes replace Agamemnon’s—one madness the other, or at least the threat of the other. The madness is doubled: one form annihilates the other.

Some kind of characteristic mania seems to always precede the godsent mania, be it on the part of the person to be punished, as in the cases of Ajax, Heracles and Io, on the part of another person, as it is with Phaedra and Hippolytos, or by the mere alternative of another mania, as happens to Orestes. With this kind of precedent, the punitive character of the madness is always made ambiguous, and the phenomenon of reversal, of the gods transforming human destructiveness into something self-destructive, is very typical for Greek tragedy, always revolving around the problem of human responsibility.

That this general concept of madness influences Herodotus’ ideas is plausible, but if we take another look at the stories of Cambyses and Cleomenes, we find even closer and more concrete mythical models for them.

In Pentheus’ case, too, the punitive character of the madness is ambiguous. He seems mad from the very beginning, rejecting the religious norm. This kind of sacrilegious blindness is madness, as is voiced in the Bacchae by the seer Teiresias: ‘For you are mad in a most grievous way’, he says to Pentheus even before the latter’s hallucinations and ‘actual’ madness start (μαίνῃ γὰρ ὡς ἄλγιστα, 327); and only a little later: ‘You are mad now, and even before you were out of your wits’ (μέμηνας ἤδη, καὶ πρὶν ἐξεστὼς φρενῶν, 359). The latter quote has the same peculiar notion we have encountered in Herodotus: ‘before you were already mad and now you are, too’—meaning that one form of mania has replaced the other.

The story of Cleomenes evokes the myth of another enemy of Dionysus, Lycurgus, at least in its later versions. In Apollodoros (3.5.1 = 3.34–35) and Hyginus (Fab. 132) we find the element of autoaggressive self-mutilation in madness: Lycurgus hacks his foot off just like Cleomenes carves up his leg. The myth then takes another direction; Lycurgus’ death seems not connected with the wounding; Hyginus tells us that he kills himself after having been maddened by Dionysus (Fab. 242). In spite of the syntagmatic shifts, we can recognise a similarity with Herodotus’ Cleomenes, as the punishment is similar and happens as the consequence of a sacrilegious act—even if Lycurgus’ actual theomachy reminds us more of Cambyses’ slaying of the Apis than of Cleomenes’ desecrations of sanctuaries.

Lycurgus, Pentheus, Cambyses and Cleomenes have another feature in common that places them into a Dionysiac context: the motif of wine connected with their madness. We remember the only ‘natural’ explanation of Cleomenes’ madness Herodotus gives: the Scythian drinking of undiluted wine (6.84)—a Dionysiac form of madness. In the case of Cambyses, too, the Persians think he drinks too much (3.34.2–3)—which he himself takes as an allegation of madness (3.34.3); here, too, there may be a Dionysiac subtext.

If the god offended by the mad transgressor is Dionysus—as is the case in our closest parallels, the stories of Pentheus and Lycurgus –, the punishment of the wrongdoers—pathological madness—has special implications, seeing that the cultic reverence the god demands is ritual madness, meaning that the ‘bad’, sacrilegious mania is sanctioned by ‘good’, ritual mania.

That sacrilegious madness is a necessary predisposition to sacrilege is also stated in philosophical texts, namely in Plato’s Laws, explicitly dealing with the desecration of sanctuaries. His ‘Athenian’ describes how to advise a person about to commit robbery in a temple:

My good man, the evil force that now moves you and prompts you to go temple-robbing is neither of human origin nor of divine, but it is some mad impulse bred of old in men from ancient wrongs unexpiated, which courses round wreaking ruin; and it you must guard against with all your strength. How you must thus guard, now learn. When there comes upon you any such intention, betake yourself to the rites of guilt-averting, betake yourself as suppliant to the shrines of the curse-lifting deities, betake yourself to the company of the men who are reputed virtuous; and thus learn, partly from others, partly by self-instruction, that every man is bound to honor what is noble and just; but the company of evil men shun wholly, and turn not back. And if it be so that by thus acting your disease grows less, well; but if not, then deem death the more noble way, and quit yourself of life.

Whoever will desecrate a sanctuary has to be completely mad beforehand.

The phenomenon of reversal, the complexity of madness that generates itself and is cause and punishment at once, cannot be easily characterised; it may result from the nature of the phenomenon as such. Debra Hershkowitz states that even though madness is part of various discursive systems, its own system is by no means consistent. She states a certain constructive aporia:

madness continually eludes all attempts to define or categorize it in a single, understandable way … while madness is inserted in various discourses—for example, the scientific, the philosophic, the legal—it also reinscribes itself outside any ordered, totalizing discourse, any discourse which is comprehensible or which would claim to make madness comprehensible … But while the uncertainty of madness, its lack of comprehensible meaning, threatens to undermine the stability of the meaning of all comprehensible structures around it, while its resistance to ordered discourse calls into question the very validity—one might say the ‘sanity’—of the meaning offered by such discourse, at the same time madness confirms and even defines that meaning, by its antithesis to what is considered comprehensible, rational, sane. This is the paradox of madness. It is outside of the boundaries of comprehension, yet it is also a necessary component of comprehensibility.


1.2.2 The Doubling of Madness in Cult :

The importance of Dionysiac patterns for Herodotus’ cases of madness leads us to the ritual dimension of mania traditionally connected to the cult of the wine god. As we will see, ritual madness, too has a double nature and is therefore closely related to the phenomenon of madness in traditional narrative.

To sum up: the structure of a doubled mania in cult is most obvious when one kind of madness is replaced by another kind, that is, if like is cured by like. The doubling of mania in myth, especially in tragedy, works in a rather similar way: mania originating from the madman is turned against him (or another person involved): Ajax, Heracles, Lycurgus and Pentheus demonstrate a kind of—occasionally sacrilegious—dynamic, eventually turning against themselves. With Pentheus, it is not his own sacrilegious mania that kills him but the ritual madness of the thiasos tearing him apart. Some kind of doubling is always there, as in the cases of Orestes, Phaedra and Io. In its structure of replacement, the punitive reversal of mania corresponds exactly with its cultic function of healing.

1.3 The Traditionality of Mania in Herodotus: A Mythical Narrative

1.3.1 The Doubling of Madness in Herodotus :

Apparently, the structure of reversal is connected to the concept of madness in both ancient myth and cult. It is consequently not surprising that the traditional structure is transferred to historical characters as well. If Cambyses and Cleomenes have a sacrilegious madness eventually turning against themselves, the problem of what causes sacrilege and madness, posed earlier, is eliminated; the second, ‘punitive’ part of Herodotus’ stories is not just a consequence of sacrilegious madness, it is its exact reversal, as we have seen above. It therefore functions in the same way as the ritual and mythical patterns of madness. This must be the explanation of Herodotus’ notion of change between a ‘before’ and ‘after’ kind of madness in the case of Cleomenes and Cambyses (the same phenomenon we have seen in Aristophanes and, especially, Euripides). The sacrilegious madness of the evil-doer has been redirected and transformed from destructive to self-destructive. The characters have been mad from the beginning, but the nature of their madness has been changed by the divine.

In contrast to Cleomenes, Cambyses does not die from an act directly resulting from his madness. In a way, however, his case is even closer to the mythical examples of madness, because the pattern of reversal is even more literal: he dies from the very same act by which, at the height of his sacrilegious madness, he has killed the Apis bull—he thrusts the sword into his thigh in the exact same place. This is autoaggression, as in Cleomenes’ case: though inadvertently, the wounding happens only through Cambyses himself and without any external influence. But the reversal happens on the level of the act itself, too: it is not just the mania that is doubled, but the mad act itself is now redirected against the evil-doer himself. The deed can only be undone by the same deed.

Finally, the chaotic chronology in the case of Cleomenes’ madness, as Herodotus depicts it, does not seem coincidental. To summarise briefly: in 6.66 Herodotus reports the bribe of the Pythia that is discovered in chapter 74, after the story of Demaratos and Leotychidas. In chapter 75 we are told of Cleomenes’ madness and suicide, and encounter the statement we have touched on several times now: that he had become mad after returning from exile whilst having been mad before as well. The chapter is closed with a discussion about the possible reasons for Cleomenes’ madness: the bribe of the Pythia, the destruction of the Eleusinian or the Argive sanctuary. The latter is then told in detail from 76–83, after which we are presented with another discussion of the cause of the mania: in chapter 84 Herodotus names the destruction of Argos, the drinking of undiluted wine and, again, the bribe of the Pythia, this time with the author’s explicit support (‘but to my thinking it was for what he did to Demaratus that he was punished thus’, 6.84.3). Crime—madness—discussion of causes—crime—discussion of causes: it seems as if Herodotus consciously blurs all clearness of chronology and thereby imitates the problematic causality of madness on a narrative level. The plurality of causes seems part of the narrative strategy: all acts of aggression result in madness, every new level of madness generates new aggression. The Herodotean kyklos becomes a vicious circle.

As in Croesus’ crossing of the river Halys analyzed in II.2.3, we can see the importance Herodotus places on presenting different models of explanation: the traditional concept of a double mania does not allow for a simple punitive character. It is even possible for Herodotus to have religious and physical explanations coexist side by side—i. e. madness as punishment for religious offences and madness as a consequence of epilepsy in the case of Cambyses. Admitting a medical explanation amongst the traditional concepts or, as James Romm states it, merging “a systematic or even scientific approach, in which natural phenomena are explained by regularly occurring forces and processes, with a more myth-based and religious mode of thought which sees the will of an anthropomorphic divinity behind such phenomena” (2006, 186) obviously does not pose a problem.

1.3.2 More Traditional Elements in Herodotus’ Stories of Madness :

There is more tradition to mania in Herodotus than the mere doubling of it—the stories are part of an intricate net of mythical and ritual associations. It is interesting, for example, that the exact reversal of Cambyses’ deed points to another mythical model that has become eponymous for the structure of reversal: the story of Achilleus’ son Neoptolemos. He kills king Priam at the altar of Zeus Herkeios, the household god, and tries to rob the Delphic sanctuary; consequently, he is slain by the guards or even by Apollo himself. [28] Pausanias tells us of a proverb that has come from this myth in the context of a Spartan anecdote: 200 years after the Lacedaemonians have bought themselves a victory in the Messenian war by bribing the Arcadian general Aristocrates, they become victims of corruption themselves, when the Persians bribe several Greek cities and so drag Sparta into the Corinthian War (Paus. 4.17). This is equated with the mythical model of Neoptolemos:

περιῆλθε μέντοι καὶ αὐτοὺς Λακεδαιμονίους ἀνὰ χρόνον ἡ Νεοπτολέμειος καλουμένη τίσις. Νεοπτολέμῳ γὰρ τῷ Ἀχιλλέως, ἀποκτείναντι Πρίαμον ἐπὶ τῇ ἐσχάρᾳ τοῦ Ἑρκείου, συνέπεσε καὶ αὐτὸν ἐν Δελφοῖς πρὸς τῷ βωμῷ τοῦ Ἀπόλλωνος ἀποσφαγῆναι· καὶ ἀπὸ τούτου τὸ παθεῖν ὁποῖόν τις καὶ ἔδρασε Νεοπτολέμειον τίσιν ὀνομάζουσι.


However, in course of time the punishment of Neoptolemus, as it is called, came upon the Lacedaemonians themselves in their turn. Now it was the fate of Neoptolemus the son of Achilles, after killing Priam on the altar of Zeus Herkeios (Of the Courtyard), himself to be slain by the altar of Apollo in Delphi. Thenceforward to suffer what a man has himself done to another is called the Punishment of Neoptolemus.

It is striking that the mythical model of Neoptolemos—who has murdered and then been murdered in the sanctuary—becomes a very general symbol of the (not necessarily sacrilegious) deed that turns against the evil-doer (in the sense of “harm set, harm get”). In Greek thought, there seems to be an associative link between the pattern of reversal in general and the desecration of sanctuaries in particular.

Miltiades’ sacrilege actually consists in stepping into the temple, consequently, the body part that has been the instrument of his misdeed is injured, turns against him and kills him.

The motifs of the desecration of sanctuaries and of the punishment of like by like seem to be connected in tradition—that being the case with Neoptolemos, Miltiades and Cambyses, the first example having even become proverbial. This seems to explain the connection of mania with the desecration of a sanctuary, because madness, too, is traditionally seen as working in a homoeopathic way.

In this context, the song of the madness-inducing Erinyes in Aeschylus’ Eumenides is interesting, too, here given with Ruth Padel’s translation (1995:241–242):

μάλα γὰρ οὖν ἁλομένα
ἀνάκαθεν βαρυπετῆ
καταφέρω ποδὸς ἀκμάν,
σφαλερὰ ⟨καὶ⟩ τανυδρόμοις
κῶλα, δύσφορον ἄταν.
πίπτων δ’ οὐκ οἶδεν τόδ’ ὑπ’ ἄφρονι λύμᾳ·


With a great leap from above
I bring down my foot’s heavy-falling force.
My limbs make even fast runners fall:
a terrible disaster.
And as he falls, he knows nothing, in mad folly.

The jump does not just cause madness in others, nor does the fall necessarily cause madness in oneself—jumping and falling are interchangeable; the foot is ‘heavy-falling’ in an active ‘leap’. Likewise, Cambyses and Miltiades can be seen as actively jumping into a passively suffered misfortune, and therefore as a metaphor for the structure of mania-reversal in general: active aggression becomes passively suffered autoaggression, jumping turns into falling.

1.4 Madness and Initiation? Cleomenes, Cambyses and Anacharsis and Scyles

1.4.1 Madness and Initiation :

1.4.2 Cambyses and Cleomenes and the Paradigm of Initiation :

The semantic level of initiation connected to mania in mythical and cultic contexts adds further depth to the stories of Cleomenes and Cambyses. If the myth of Pentheus follows the pattern of mystic initiation, we have to consider this dimension, too, if we accept the story of Pentheus as a mythical parallel of the story of Cambyses.

It is not just madness, however, that has to do with mystic initiation: Cleomenes’ self-mutilation points in that direction, too. Dismemberment is a manner of death strongly associated with mystic initiation: Dionysus suffers a Sparagmos as the Orphic god of afterlife and is pieced together again—a symbol of a new life with a new order.

For one, the associative link between the madness of Herodotus’ characters and initiatory rites supports the notion of a change of identity—the characters become mad and enter a new phase of personality. The paradigm of initiation opens another interpretative possibility via the identification of the mystēs with the god. Cleomenes and Cambyses are punished for refusing to worship, a situation that in the Dionysiac resistance myths leads to a forcibly performed initiation. If we read Herodotus’ cases of madness with that in mind, we have a kind of exact reversal again, a Punishment of Neoptolemos: the proud denial of worship (which basically means that the hierarchical differences between god and mortal are blurred) results in the actual identification that the god forces upon the sinner by replacing his personality with the god.

Herodotus’ examples also show the healing of society by the annihilation of the madman. Cambyses and Cleomenes have turned away from the laws of the community; without their being punished, those laws would be seriously called into question.

It is interesting in this context that Cleomenes violates the religious customs of his own culture, whereas Cambyses disrespects those of another people, the Egyptians. The latter is explicitly called madness by Herodotus:

Πανταχῇ ὦν μοι δῆλά ἐστι, ὅτι ἐμάνη μεγάλως ὁ Καμβύσης· οὐ γὰρ ἂν ἱροῖσί τε καὶ νομαίοισι ἐπεχείρησε καταγελᾶν. εἰ γάρ τις προθείη πᾶσι ἀνθρώποισι ἐκλέξασθαι κελεύων νόμους τοὺς καλλίστους ἐκ τῶν πάντων νόμων, διασκεψάμενοι ἂν ἑλοίατο ἕκαστοι τοὺς ἑωυτῶν· οὕτω νομίζουσι πολύ τι καλλίστους τοὺς ἑωυτῶν νόμους ἕκαστοι εἶναι. οὐκ ὦν οἰκός ἐστι ἄλλο γε ἢ μαινόμενον ἄνδρα γέλωτα τὰ τοιαῦτα τίθεσθαι.


I hold it then in every way proved that Cambyses was quite insane; or he would never have set himself to deride religion and custom. For if it were proposed to all nations to choose which seemed best of all customs, each, after examination, would place its own first; so well is each convinced that its own are by far the best. It is not therefore to be supposed that anyone, except a madman, would turn such things to ridicule.

1.4.3 Scyles and Anacharsis (4.76–80) :

Speaking of disrespecting foreign customs, we have to take a look at two characters in the Histories who are also connected with initiation and mania, the ‘martyrs’ Anacharsis and Skyles, in whose stories cultic and mythic structures are combined in a highly complex manner.

Presented in ring composition, they form an excursus intended to demonstrate the radical rejection of foreign customs by the Scythians, as Herodotus emphasises both at the beginning and end of it (4.76 and 80). The Scythian Anacharsis becomes impressed with the rites of Cybele while visiting Cyzicus and vows to the goddess to arrange for a similar celebration for her in Scythia, should he return home safely. When at home, he secretly roams the woods with a tympanon and effigies around his neck. But he is found out and reported to the king, his kinsman or even brother, who immediately shoots the ‘heretic’ with an arrow.

The second story is quite similar. The Scythian king Scyles learns Greek from his Istrian mother and leads a double life as a Greek with the Borythenites, who claim to be of Milesian origin. He even marries one of them and has a house in their town. One day, he pushes the limit too far: he wants to be initiated into the cult of Dionysus. At the moment of his initiation, ‘god’ (ὁ θεός) strikes his house with lightning (4.79.2). Just like Anacharsis, he is denounced, by a Borysthenite who is fed up with the Scythians making fun of the Borysthenite ecstatic cults, telling them that their king is now mad, too (4.79.4).

Scyles is eventually killed by his own brother.

There does not seem to be a concrete narrative mythical parallel for the doublets. On the contrary, we have a strange contrast to the Dionysiac resistance myths: the believer is punished, the resistance is successful; ‘god’ even throws lightning into the house of his follower.

Considering once again the frame of the stories, the ethnographic remarks on Scythian xenophobia, these stories do not seem to deal with cult per se, but with the illustration of cultural borders via cult—a subject that is of great importance to Herodotus (cf. below chapter V.4.2).

The stories of Scyles and Anacharsis, but also the stories of Cleomenes and Cambyses seem generated by ritual patterns or display possibilities of associative connections with them. Of course, this has semantic implications: as we have seen above, the godsent mania of Herodotus’ evil-doers Cambyses and Cleomenes corresponds with the forced initiation of the mythical enemies of Dionysus into cultic mania. It is a different story with the wise Graecophiles Scyles and Anacharsis, who actively undergo initiation rites. The separation from their own community is part of the initiatory vocabulary, too. It ends in death, a part of mystic initiation—as a metaphor, of course. But as mythical, ritual and historical discourses intertwine, one cannot expect a clear separation of metaphorical and ‘real’ elements.

For my purposes, it is negligible whether Herodotus is conscious of the initiatory elements in his stories—their implications are generated by themselves and fit well with both the ‘forcibly initiated’ evil-doers and with the Scythian sages undergoing initiation of their own free will. It is interesting, however—and this is a result of the amalgam of discourses again—that in the case of Cleomenes and Cambyses, the experience of initiation has been reduced to a completely symbolic level which is merely hinted at, whereas the tales of Scyles and Anacharsis are a mixture of ritual and mythical narrative: the cultic event of initiation is named, but the elements of separation and killing belong to the field of mythical narrative and could only figure symbolically in a cultic context.

1.5 Conclusions:

Herodotus’ refusal to clarify the cause of Cleomenes’ and Cambyses’ sacrilege and madness is founded within a mythical, especially Dionysiac narrative tradition, and in ritual patterns of healing and initiation. The madness of the madman turns against him in Herodotus, as it does in myth and ritual: he who commits mad crimes is struck with madness.

It cannot be ruled out that parts of Herodotus’ narrative are historical. But only if we place the episodes into the tradition they stem from are their semantic contents revealed. The reversal-structure of mania and the initiatory association of the forced worshipping attest that the evil-doer’s downfall is his own fault as well, and point to the healing and liberation of society from the evil-doer, or even the liberation of the evil-doer from himself.

That Herodotus does not connect the origin of madness to a single event, as do the Egyptians in the case of Cambyses killing the Apis-bull, does not mean that he is rationalising the old story of godsent madness. On the contrary, he places new cases of mania into a tradition that has always regarded madness as a disposition that can be directed outwards or inwards but does not work as a punishment for a single deed. This traditionality does not exclude rationalism: by acknowledging disposition and multicausality, the traditionalising reflection is more rational than deducing madness to a single deed would be.

Herodotus’ tales of mania are paradigmatic for his dealing with traditional narrative: only by the complexity of tradition, historical discourse becomes ‘three-dimensional’ to an extent that could not be reached by the mere reporting of facts.

2. Dionysus and the Tyrant

In analysing the mania-episodes, this dimension, too, should be considered. Consequently, the stories do not just generally comment on the disposition of the mad evil-doer, they also constitute a historiographical statement: rising too far above the community can result in mania, much as Plato states in the Republic (572c–573c): their cupidities make tyrants mad,

καὶ μὴν ὅ γε μαινόμενος καὶ ὑποκεκινηκὼς οὐ μόνον ἀνθρώπων ἀλλὰ καὶ θεῶν ἐπιχειρεῖ τε καὶ ἐλπίζει δυνατὸς εἶναι ἄρχειν.


and again the madman, the deranged man, attempts and expects to rule over not only men but gods.

2.1 Arion (1.23–24):

The antityrannical element of the Dionysiac is especially clear in the story of Arion. Placed almost at the very beginning of the Histories, it seems to be of programmatic content. It goes as follows: The singer Arion, friend to the Corinthian tyrant Periander, is on his way back to Corinth from Italy, when the sailors decide to rob and kill him. Arion begs them to be allowed one last song and attracts a dolphin performing it. He jumps into the water and the animal carries him to the shore. When he tells Periander of his miraculous salvation, he does not believe a word and imprisons him, until the sailors have returned and unwittingly confirm Arion’s story by claiming to have left him safe in Italy, therefore being exposed as liars. Since then, Herodotus tells us, there has been a bronze memorial of the miracle.

It has caused much surprise that Herodotus reports the story so early on in the Histories, not in Book 3, where he comments in detail on the other main character of the story, Periander. Plausible explanations have been given by Rosaria Munson, who sees the story as structurally analogous to the main subject of the Histories, “in that it focusses upon an overwhelming aggression against a numerically weaker victim,” and as a demonstration of the general ethical explanation of “divine punishment and reward” (1986:98). Vivienne Gray, too, finds a display of divine justice in the tale of Arion, and draws parallels to the stories of Alyattes and Croesus (2001). Periander, who by his research reveals this divine justice, may be a model of Herodotus himself (Packman 1991; Gray 2001).

This programmatic illustration of divine justice is without doubt one important aspect of the tale of Arion and its position at the beginning of the Histories. However, by analysing the rather obvious mythical parallels of the story, we can find even more impact in it, namely Herodotus’ critical view of monarchy.

Consider the episode’s main plot elements:

  1. The protagonist is threatened by sailors.
  2. An animal is attracted by music.
  3. A dolphin appears.
  4. The threat (1) motivates the protagonist to ride through the sea on an animal.
  5. The protagonist is imprisoned as a fraud.
  6. The protagonist is freed and rehabilitated.
  7. There is a monument to commemorate the story.

This first mythical parallel shows that Arion has near godlike status in Herodotus’ story. This is confirmed by further mythical parallels with reference also to other gods, i. e. the dolphin-story of the Homeric hymn to Apollo (h.Ap. 3.388–546): looking for priests to serve in his new sanctuary in Delphi, the god notices a ship with brave men from Crete going to Pylos. In the shape of a dolphin, he jumps on the ship, so that the sailors do not dare to steer it anymore—whoever tries to touch the dolphin is hurled aside. The ship then sails to Tainaron on its own, where the crew wants to land; but the oars do not follow them and the dolphin pushes the ship into the harbour of Delphi with his breath. Here, Apollo appears in his true shape and tells the sailors to serve him from now on.

The story is related both to the myth of Dionysus and the Tyrrhenic sailors and to Herodotus’ tale of Arion. It is the functions 1 and 3, in a broader sense 4 and 7, too, which structure the story. The element of threat (1) is practically turned around, but still recognisable: this time, it is the sailors who are threatened and the god who threatens; the element of the dolphin (3), however, connects the situations, as in both stories it is the transformed image of the threatening party. Being driven to Delphi by the dolphin is not the same as riding on a dolphin (4), still, the animal acts as transport for the human characters of the story. Finally, the institution of the sanctuary is a kind of commemoration (7), if not for the story itself.

In any case, both the myths of Apollo and Dionysos deal with an antagonism between man and god. Even if Apollo’s relationship with the sailors is rather friendly at the end, the acts of the god against those who want to hinder his plans are aggressive and threatening (the hurling aside of the people who touch him, the blocking of the oars). In both stories, the mortals resist the god without recognising him.

Furthermore, what about function 2, the attraction of animals with music? It obviously reminds us of Orpheus, another godlike figure serving as a parallel for Arion. There is apparently no fixation on one divine role-model (which might even have been theologically questionable, if we keep in mind the consequences of Xerxes’ imitatio Iovis; cf. above, chapter II.1).

This places unbelieving Periander—who imprisons, but never actively frees Arion—in the position of Pentheus, the sacrilegious enemy of the god. Also, if Arion is the first writer of dithyrambs, Periander is at least the son of the first Greek tyrant Cypselus (cf. below chapter V.2). What Herodotus has to tell of him in the following books is not flattering: ‘to begin with milder than his father’, but later ‘much more bloodthirsty than Cypselus’ (5.92.ζ.1), he kills all prominent citizens (5.92.ζ–η), murders his wife (3.50), rapes her corpse and so forth; the latter story is told by Sosicles in a passionate pro-democratic speech against tyrant terror (5.92.ζ–η). Periander is, most of the time, portrayed as an intelligent but wicked typical tyrant.

Besides doubtlessly representing divine justice, the tale of Arion is also a richly suggestive characterisation of Periander, who—playing the part of Pentheus—appears as a potentially negative figure, which paves the way for his drastic portrayal of the later books and for Herodotus’ negative view on monarchy in general. Maybe the story’s position at the beginning of the Histories is programmatic in more than one way.

Munson’s and Gray’s assumption of a more general programmatic statement in the Arion-story is plausible. However, it does not exclude another programmatic level: the demonstration of Herodotus’ scepticism against the ‘barbaric’ form of government that is tyranny.


[ back ] 1. Godley softens the first expression to ‘not entirely in his right mind’, but there is no indication of this relativisation in the Greek word ὑπομαργότερον.

[ back ] 2. Procl. arg. Il. parv. Davies EGF p. 52, 3–5 = Bernabé PEG I, p. 74, 3–5.

[ back ] 3. S. Aj. 44–45; 50–54.

[ back ] 4. Cf. Simon 1981:135–136 and Schlesier 1985:10–11, 33–34 on HF 840–842.

[ back ] 5. E.g. Aesch. Pr. 673–677; also, the gad-fly, οἶστρος, is a Greek metaphor for madness; cf. Padel 1995:13–17.

[ back ] 6. E.g. Plat. Phdr. 241a; cf. also Mattes 1970:69–70; Hershkowitz 1998:29–34.

[ back ] 7. Cf. e.g. Aesch. Cho. 1061; Eur. IT 281–294; Or. 407–408.

[ back ] 8. E.g. A. Cho. 269–296; 558–559; 583–584, 1021–1042; S. El. 37–41.

[ back ] 9. Fr. 183 Radt TrGF 3, p. 298–299, cf. Bierl 1991:11.

[ back ] 10. Fr. 301–311 Bernabé PEG II, p. 246–255; cf. West 1983:140–175; Iles Johnston 2007a.

[ back ] 11. Fr. 59; 322; 326–328 Bernabé PEG II, p. 66–68; 265–266; 267–271.

[ back ] 12. Cf. Padel 1995, 174–187.

[ back ] 13. Zum Begriff οἶστρος vgl. oben Anm. 204.

[ back ] 14. Cf. Padel 1995:82.

[ back ] 15. Cf. Ba. 337–341; 367–369.

[ back ] 16. Ov. Met. 4.1–415; Plu. Quaest. Gr. 299e–300a; Ant. lib. 10; Ael. VH 3.42.

[ back ] 17. Cf. Sidwell 1990; Reckford 1977.

[ back ] 18. Ar. V. 71; 76; 87; 114; 651.

[ back ] 19. Ar. V. 1485; 1496.

[ back ] 20. Cf. Linforth 1946:169: “We must acknowledge that we do not know what legend Socrates had in mind, and I repeat that the vagueness of the relative clause, with πόθεν and τισί, and τισί in the plural, may really mean that he had no legend in mind.”

[ back ] 21. Cf. Linforth 1946; Velardi 1989:79–81; Scullion 1998:108–110.

[ back ] 22. Fr. 350.5 Bernabé PEG II, p. 288.

[ back ] 23. Cf. also Scullion 1998 for a collection of loci that point to Dionysiac therapeutic ecstasy.

[ back ] 24. Henrichs 1994:35 and n. 13.

[ back ] 25. Cf. Ustinova 1992–1998, esp. 511–515; Jeanmaire 1951:131–138; Dowden 1989:85; Seaford 2006:105–108.

[ back ] 26. Cf. Pigeaud 1981:500–503.

[ back ] 27. Cf. e.g. Grof 1983.

[ back ] 28. E.g. Procl. arg. Iliup. Davies EGF p. 62, 19–20 = Bernabé PEG I, p. 88, 13–14; Pi. Pae. 6.113–115 fr. 52–53 Maehler; E. Hec. 23–24; Tr. 15–17; Apollod. Ep. 5.21.203–206 = Ep. 5.21; Ep. 6.108–114 = Ep. 6.14; Hyg. Fab. 113; Paus. 10.7.1; Strab. 9.3.9.

[ back ] 29. Cf. Seaford 1981:254–255 and n. 42; 1994b:284–285 and n. 66.

[ back ] 30. Cf. Padel 1995:241–242.

[ back ] 31. Schol. D ad Il. 1.59; Lyc. Alex. 206–215 and Tz. ad Lyc. 211; Lyc. Alex. 1246–1247; Apollod. Ep. 3.17.126–133 = Ep. 3.17; Hyg. Fab. 101.

[ back ] 32. Cf. Seaford 2006:44.

[ back ] 33. Cf. e.g. Burkert 1997 [1972]:57; Seaford 1981:261–263; Seaford 1994a:318–326.

[ back ] 34. Fr. 309 Bernabé PEG II, p. 252–253.

[ back ] 35. Cf. Seaford 2006:117–119.

[ back ] 36. Cf. Dowden 1989:87–90.

[ back ] 37. Cf. West 1983:140–175, esp. 143–150 and 160–161.

[ back ] 38. Fr. 463 Bernabé PEG II, p. 390–391; cf. West 1982:19; Iles Johnston 2007b:125–127.

[ back ] 39. Cf. e.g. Vernant 1985 [1965]:192–193; Graf 1998:1002.

[ back ] 40. Cf. Munson 2001:118–123.

[ back ] 41. Cf. Hartog 1980:93–100, Bierl 1991:13–20, Henrichs 1994:47–51.

[ back ] 42. Cf. Munson 1991:55–56.

[ back ] 43. Cf. i. e. Griffiths 1989:esp. 71–72.

[ back ] 44. Cf. Bierl 1991:esp. 45–110.

[ back ] 45. E.g. Usener 1899:160; Burkert 1997 [1972]:218–226.

[ back ] 46. In an as yet unpublished paper on the Arion episode.

[ back ] 47. Ibid.

[ back ] 48. H.Bacch. 7.13–14; E. Ba. 642–649.

[ back ] 49. Cf. e.g. Hdt. 5.78.