Black Bile, Yellow Bile: An Essay on Warrior Dysfunctionality and the Prehistory of Greek Medicine

2020.05.28 | By Roger D. Woodard

Bronze sculpture depicting the horse Pegasus and the hero Bellerophon falling.
Bellerophon falling from Pegaus (anonymous, ca. 1575–1599). Rijksmuseum BK-NM-12026. Image via the Rijksmuseum.


§1. In Problemata physica 30, a work attributed to Aristotle, though typically judged to be a Peripatetic compilation,[1] the composer(s) (to whom I will refer in what follows as “Aristotle” or “the author”) addresses the mental state of dysfunctional warriors. These remarks form a part of the discussion about those who are melankholikoí (μελαγχολικοί),[2] lumping together Heracles, Bellerophon, Ajax, and Lysander of Laconia. Regarding Heracles melankholikós (μελαγχολικός), the author writes (Problemata physica 953a):

Καὶ γὰρ ἐκεῖνος ἔοικε γενέσθαι ταύτης τῆς φύσεως, διὸ καὶ τὰ ἀρρωστήματα τῶν ἐπιληπτικῶν ἀπ’ ἐκείνου προσηγόρευον οἱ ἀρχαῖοι ἱερὰν νόσον. καὶ ἡ περὶ τοὺς παῖδας ἔκστασις καὶ ἡ πρὸ τῆς ἀφανίσεως ἐν Οἴτῃ τῶν ἑλκῶν ἔκφυσις γενομένη τοῦτο δηλοῖ· καὶ γὰρ τοῦτο γίνεται πολλοῖς ἀπὸ μελαίνης χολῆς.

For he also seems to have been of this nature; in consequence the ancients used to call the illnesses of epileptics “sacred sickness” after him.[3] This is clear both from his altered state in regard to the matter of his children, and from the eruption of ulcers prior to his disappearance at Oeta: for this occurs among many [who suffer] from mélaina kholḗ.

Mélaina kholḗ (μέλαινα χολή) ‘black bile’ is the cause of these sufferings, mental and physical. Heracles’ ‘ulcers’ (hélkea [ἕλκεα]) that are here referenced[4] correspond to what is in the mythic tradition of this dysfunctional warrior described as eruptions caused by contact with the poison of the Lernaean Hydra—a poison of toxic bile to which Nagy has recently drawn attention in connection with the khólos (χόλος) ‘bile/anger’ of Hera and of Achilles.[5] Thus, Pseudo-Apollodorus (Bibliotheca 2.157–158) writes that after Heracles had put on the tunic that his wife Deianira had rubbed with a “love potion,” which contained this ‘bile’ (kholḗ [χολή]),[6] and when that bilious solution was then heated by sacrificial flames, ‘the poison of the Hydra putrefied his flesh’ (ὁ τῆς ὕδρας ἰὸς τὸν χρῶτα ἔσηπε). Sophocles (Women of Trachis 573) describes the Hydra’s poison as melánkholos (μελάγχολος) ‘black-bile’, attesting khólos in the sense ‘bile’ (beside common kholḗ ‘bile’), distinct from its sense ‘anger’.

§2. The mentioned ‘altered state’ that Aristotle assigns to Heracles the melankholikós (μελαγχολικός) in the matter of his children is the madness that leads the warrior to slay them. To express the idea of ‘altered state’ Aristotle uses the nominal ékstasis (ἔκστασις), a form derived from eksístēmi (ἐξίστημι) ‘to set out of’, that is, ‘to alter completely’—thus, commonly, ‘to drive to madness’—as in Euripides fr. 265 Nauck (Auge): here Heracles, who had raped Athena’s virgin priestess Auge, when he had been received as a guest by her father, declares: νοῦ δ’ οἶνος ἐξέστησέ μ’· ὁμολογῶ δέ σε ἀδικεῖν, τὸ δ’ ἀδικημ’ ἐγένετ’ οὐχ ἑκούσιον ‘wine drove me out of my mind; I confess that I injured you, but this injury was not volitional’. In the Problemata physica what does ékstasis ‘altered state’ have to do with Ajax and Bellerophon? The author continues:

Ἔτι δὲ τὰ περὶ Αἴαντα καὶ Βελλεροφόντην, ὧν ὁ μὲν ἐκστατικὸς ἐγένετο παντελῶς, ὁ δὲ τὰς ἐρημίας ἐδίωκεν . . .

And the same is so of Ajax and Bellerophon, the former of whom became completely mad [ekstatikós], and the other was pursuing solitary places . . .

Fundamentally these two heroic figures form a contrasting pair with regard to how their warrior madness manifests itself. ‘Seized by madness’ (μανίᾳ . . . ἁλούς; Sophocles Ajax 216), Ajax turned his warrior rage against fellow Greeks, while Bellerophon displaced himself from society—an act that can be also be mechanically described as an ékstasis (ἔκστασις).[7] Concerning Bellerophon Aristotle goes on to rehearse Iliad 6.200–202:[8]

Διὸ οὕτως ἐποίησεν Ὅμηρος
Αὐτὰρ ἐπεὶ καὶ κεῖνος ἀπήχθετο πᾶσι θεοῖσιν           200
ἦ τοι ὁ κὰπ πεδίον τὸ Ἀλήϊον οἶος ἀλᾶτο
ὃν θυμὸν κατέδων, πάτον ἀνθρώπων ἀλεείνων·

Αnd so, Homer wrote thus
Βut when [Bellerophon] too became hated by all gods[9]         200
he wandered alone across the Aleian Plain
devouring his own courage,[10] avoiding the path of humans.

Aristotle here contrasts the altered mental states of Ajax (one of homicidal madness) and Bellerophon (one of self-isolation) but in so doing makes recourse to black bile in both instances. As others have pointed out—see especially the careful discussion of Jouanna (2012:229–258)—Aristotle here departs from normative Hippocratic “doctrine” of black bile: in the lines rehearsed above and in a longish discussion that follows, in which the ingestion of wine (as an intoxicant) is offered as an illustrative parallel, the author of Problemata physica 30 (953a–955a) attributes both violent madness and social separation to the condition he characterizes by the adjective melankholikós (μελαγχολικός), essentially denoting, for this author, ‘black-bilious’.[11] In addition, Jouanna notes, for Aristotle melankholikós can be framed in terms of a psychological condition associated with the humoral temperament. “The very basis of what makes the Aristotelian Problem famous,” concludes Jouanna (2012:240), “does not seem to be of medical, and certainly not of Hippocratic, origin.”

§3. If we acknowledge that the notion “medical” in the archaic Greek system of signification is a shallow descendant of its Indo-European ancestor, then we must allow the prospect that what the author of Problemata physica 30 describes as black-biliousness is “certainly not of Hippocratic,” but can be “of medical,” origin. The antecedent Indo-European notion, preserved lexically in such forms as Latin medior ‘to heal, remedy’ and Avestan vi-mad- ‘to treat a sick person’, as Émile Benveniste made plain to us, is fundamentally one of taking the measure (*med-) of something. The survival of descendent lexemes on the western (Latin) and eastern (Avestan) edges of the ancient Indo-European expansion zone, “in two conservative languages that retained so many vestiges of the common language,”[12] gives us confidence in identifying Indo-European *med- as an element of an ancestral lexicon of healing practice that entailed deliberate (that is, measured) diagnosis and response. Demonstrating the proximity of Greek ideas to the ancestral tradition, Benveniste points to the close parallelism between the Avestan triadic curative practice involving (1) the ‘medicine of the knife’ (karəto.baēšaza); (2) the ‘medicine of plants’ (urvarō.baēšaza); and (3) the ‘medicine of charms’ (mąθrō.baēšaza),[13] on the one hand, and, on the other, Pindar’s enumeration of the curative procedures that Chiron taught to Asclepius in Pythian Odes 3.47–55. The Greek procedures correspondingly involve (1) recovery ‘by lancing’ (tomaîs [τομαῖς]); (2) oral administration of ‘soothing [potions]’ (prosanéa [προσανέα]) or thorough application of ‘remedies’ (phármaka [φάρμακα]) ‘to limbs’ (guíois [γυίοις]); and (3) treatment ‘with gentle charms’ (malakaîs epaoidaîs [μαλακαῖς ἐπαοιδαῖς]). Pindar also enumerates a triad of afflictions that plausibly align respectively with the three elements of curative practice in this way: (1) the ailment ‘of spontaneously occurring ulcers’ (αὐτοφύτων | ἑλκέων); (2) ‘those ravaged in body by either summer scorch or winter’ (ἢ θερινῷ πυρὶ περθόμενοι δέμας ἢ | χειμῶνι); and (3) wounds inflicted on limbs ‘by grey bronze . . . or by stone launched from far’ (πολιῷ χαλκῷ . . . | ἢ χερμάδι τηλεβόλῳ).

§4. Aristotle’s attribution of the altered state of Heracles and Ajax (that of madness), on the one hand, and the altered state of Bellerophon (that of social isolation), on the other, as equally the consequence of black-biliousness does not find expression in Hippocratic doctrine. But Aristotle’s distinction between the two altered states is a meaningful and significant one from the perspective of Indo-European doctrine of warrior dysfunctionality. The potential displacement of the rage of the primitive Indo-European warrior appears to have been a source of crucial, and likely constant, concern for the warrior’s society. On the one hand, that rage could be misdirected at society itself if the warrior were unable to relinquish his maniacal fury following the fight; and, on the other hand, that rage, as a necessary potentiality for society’s protection, could be neutralized by the warrior’s retreat to extra-societal space consequent to the trauma of combat. Both of these dysfunctionalities robbed society of its proper protection and both occasioned a deliberate response by society.[14]

§5. The opposing mental states of warriors of the Heracles-type and of the Bellerophon-type treated in the Problemata physica are taken up in Pseudo-Galen’s Introductio seu medicus in a way that brings Aristotelian analysis into somewhat closer alignment with Hippocratic. For Pseudo-Galen (Introductio seu medicus 14.740–741)[15] there are two prominent forms of ‘altered state’ (ékstasis [ἔκστασις]) of the ‘thinking faculty’ (diánoia [διάνοια]): these are manía (μανία) and melankholía (μελαγχολία). The cause of manía is ksanthḕ kholḗ (ξανθὴ χολή) ‘yellow bile’; in contrast, the cause of melankholía is (self-evidently) identified as mélaina kholḗ (μέλαινα χολή) ‘black bile’, described, in a way consistent with Hippocratic classifications, as a ‘very cold and dark humor’ (ψυχρότερος χυμὸς καὶ ζοφώδης); temperament and altered state are combined:[16]

Διὸ ψοφοδεεῖς τέ εἰσι καὶ δύσθυμοι οἱ τοιοῦτοι. ὕποπτοι δὲ εἰς πάντα καὶ μισάνθρωποί τε καὶ ἐρημίαις χαίροντες, οἷος ὁ Βελλεροφόντης ἱστορεῖται.
Ἤ τοι ὁ καππεδίον τὸ Ἀλήϊον οἶος ἀλᾶτο,
ὃν θυμὸν κατέδων, πάτον ἀνθρώπων ἀλεείνων.

Consequently such [as suffer from melankholía] are both alarmed by noises and dispirited. They regard everything with suspicion, both loathing humankind and taking pleasure in solitary places, as is reported of Bellerophon:
He wandered alone across the Aleian Plain
devouring his own courage, avoiding the path of humans.

Compare with this phenomenon those whose altered state is one of manía, the yellow-bile induced ékstasis: these are ‘disordered and carried away and rash and hubristic’ (διὰ τοῦτο ταραχώδεις καὶ ἔκφοροι καὶ πρόχειροι ὑβρισταί τε οἱ τούτῳ ἐχόμενοι τῷ πάθει).

§6. As with Heracles and various other Indo-European warrior figures fallen into dysfunctionality, Bellerophon too displays an ‘altered state’ of the ‘thinking faculty’. His is a warrior-dysfunctionality of melankholía (μελαγχολία)—of ‘black-biliousness’, in the analytic frame of Pseudo-Galen—one of isolationism in the Aelian Plain. The altered state that Bellerophon here displays is that of the dysfunctional Indra in Vedic tradition, the Indic warrior god par excellence who flees from society into remote space subsequent, and consequent, to his combat with the tricephalic monster Vr̥tra.[17]

§7. But Bellerophon also knows the yellow-bile induced dysfunctionality of manía (μανία), of raging madness. The concept of ‘madness’ that is denoted by Greek manía is the proper modality of the Indo-European warrior engaged in combat with the enemy—an idea that finds regular expression in Homeric epic through use of the related verb maínomai (μαίνομαι).[18] But manía is also a source of dysfunctionality—that kind of warrior dysfunctionality exhibited, for example, by the Ulster heroic figure CúChulainn whose warrior rage is turned against his own people, much like that of Ajax.[19] A hint of this dysfunctionality seems to insinuate itself even in the tradition of Bellerophon’s isolation in the Aleian Plain as reported by Joannes Tzetzes—incorporated into an account that attributes the landing of the warrior in that solitary space to a fall from the war-horse Pegasus (Scholia in Lycophronem 17):

Ζεὺς δὲ οἶστρον τῷ ἵππῳ τούτῳ ἐπιβαλὼν ἔπεισε τὸν Βελλεροφόντην ἀποσφαιρίσαι τῶν αὐτοῦ νώτων καὶ καταβαλεῖν *αὐτὸν* εἰς τὴν γῆν. καὶ δὴ περὶ τὸ Ἀλήιον πεδίον κατενεχθέντος Βελλεροφόντου καὶ περιπορευομένου τυφλοῦ ἀπὸ τῆς πτώσεως ὁ Πήγασος ἄνω περιεπολεῖτο καὶ κάτω. ἡ γοῦν Ἡμέρα παρὰ Διὸς τοῦτον αἰτεῖται, ὡς ἂν ἐποχουμένη αὐτῷ τὸν ἡμερήσιον κύκλον βαδίζῃ.

And Zeus let loose a gadfly and set it against this horse to toss Bellerophon from his back and to throw him down to the earth. And when Bellerophon had tumbled down in the vicinity of the Aleian Plain and wandered around, blind from the fall, Pegasus was roaming all about—going up—going down. Then Hemera [‘Day’] asks Zeus for Pegasus, that she might make her daily circuit being carried on him.

The insect that Zeus sends to sting the horse Pegasus, and so by which the god engineers the displacement of the dysfunctional warrior into the remote space of the Cilician plain, is an oîstros (οἶστρος) ‘gadfly’. Intrinsic to this term is a notion of ‘madness, mania’, the response that ensues from the insect’s sting; and oîstros is a lexeme that can itself be used to denote ‘rage, madness’. For example, in Sophocles’ Antigone (1001–1002) the blind seer Tiresias can say of the birds and an unfavorable omen:

ἀγνῶτ’ ἀκούω φθόγγον ὀρνίθων, κακῷ
κλάζοντας οἴστρῳ καὶ βεβαρβαρωμένῳ.

I hear a voice unknown among the birds, screaming
in pernicious madness and barbarous confusion.

But, perhaps most intriguingly in the context of the present essay, oîstros is the term that Euripides uses of the mad sexual desire of Theseus’ wife Phaedra for the young Hippolytus (Hippolytus 1300) in an episode embodying a Greek tragic expression of the Biblical Potiphar’s-wife motif. Such an in-house attempted seduction by the wife of a powerful master finds an exact analogue in the mythic dossier of dysfunctional Bellerophon.[20] Following his murder of a man in Corinth, Bellerophon had fled to Argos to be purified by king Proetus. There Bellerophon would remain until the wife of Proetus, named as Anteia or Stheneboea—when she had failed in her efforts to seduce the young warrior—lied to her husband Proetus, accusing Bellerophon of being a sexual aggressor. Proetus then sent Bellerophon away to Lycia where he would perform heroic deeds, exercising his warrior prowess, acts that included slaying of the Chimaera.

§8. In the fragments of Euripides’ tragedy the Stheneboea, it is the old nurse of Proetus’ wife, Stheneboea, who tries to persuade Bellerophon to sleep with her mistress: in a dramatic twist, Euripides has the nurse confront the young warrior with these words (fr. 661.12b–13a Kannicht):

. . . ὦ κακῶς φρονῶν,
πιθοῦ· τί μαίνῃ; τλῆθι . . .

. . . O such lowly thinking,
be persuaded! Why do you rage? Be daring . . .

And in fr. 665, we read τοιαῦτ’ ἀλύει ‘in such a way her passion rages’, where Euripides has chosen to express the sexual madness driving Stheneboea with the verb alúō (ἀλύω) ‘to be distraught, rage (with passion)’. Greek alúō can be equally used of the warrior’s raging: in Seven Against Thebes, when an unfavorable oracle hinders Tydeus from crossing the Ismenus to engage in combat, Aeschylus describes him as ‘furious and in a state of longing after battle’ (μαργῶν καὶ μάχης λελιμμένος, line 380)—and he is alúōn (ἀλύων) ‘raging’ and ‘lusting for battle’ (μάχης ἐρῶν). The verb is well attested in the Hippocratic corpus, used to describe frenzied behavior. It also occurs with the sense ‘to wander’: Polybius (26.1.1–7), for example, uses alúō of Antiochus Epiphanes ‘wandering’ about the city of Antioch in a discussion of his behaviors that led some to think him ‘to be mad’ (mainómenos [μαινόμενος]).

§9. This verb alúō (ἀλύω) is generally regarded as derived from the same root that appears in alúskō (ἀλύσκω, both with a -u- extension) ‘to flee from, shun’ and in aléomai (ἀλέομαι) ‘to avoid, shun; to flee’. Added to this set is aláomai (ἀλάομαι; an archaic intensive) ‘to wander’,[21] which Homer deploys in Iliad 6.201 of Bellerophon’s self-isolation from society—coupling it with aleeínō (ἀλεείνω, an epic doublet of aléomai [ἀλέομαι])[22] in line 202:

Ἦ τοι ὁ κὰπ πεδίον τὸ Ἀλήϊον οἶος ἀλᾶτο,
ὃν θυμὸν κατέδων, πάτον ἀνθρώπων ἀλεείνων.

He wandered alone across the Aleian Plain
devouring his own courage, avoiding the path of humans.

These are lines that we have seen both Aristotle and Pseudo-Galen to rehearse in their treatments of black-biliousness. Here the verb aláomai is impressionistically linked with the name of Bellerophon’s plain of wandering (the Aleian Plain) as etymon. For Homer the verb alúō can share conceptual space with aláomai: this seems clearly to be the case at Iliad 24.12, where the participle alúōn (ἀλύων) is used of a distraught Achilles ‘wandering up and down᾽ (here the verb is dineúō [δινεύω]) the shore.[23] For Homer alúōn can equally denote ‘raging’—as at Odyssey 9.398, used of the blinded Cyclops.

§10. These several Greek verbs can be traced to Proto-Indo-European *h2el-, commonly glossed as ‘to wander’.[24] The seemingly wide semantic divide shown by, on the one hand, active alúō (ἀλύω) ‘to be distraught, rage (with passion)’ and, on the other hand, middle aléomai (ἀλέομαι) ‘to avoid, shun; to flee’, aláomai (ἀλάομαι) ‘to wander’, and iterative alúskō (ἀλύσκω) ‘to flee from, shun’ has not gone unnoticed (“les sens de ces divers termes ont franchement divergé” [Chantraine 1968:66]). Remarkably, this semantic diversity finds unity in the scenario of the dysfunctional Indo-European warrior: the etymon * h2el- is pressed into service to describe equally the widely-attested twin dysfunctionalities of rage (directed against the warrior’s own society) and withdrawal (from the warrior’s own society).

§11. Returning to Euripides’ Stheneboea—the unrelenting rage of the warrior Bellerophon will be very much on display as the plot progresses. After slaying the monstrous trisomatic/tricephalic Chimaera in Lycia, Bellerophon wings his way to Tiryns on Pegasus and persuades Stheneboea to fly away with him. As they pass over the sea, Bellerophon drops her from Pegasus and she dies.[25] This is the dysfunctionality of manía (μανία), raging madness not directed against an opponent in combat: Bellerophon exhibits an altered state of mind of Pseudo-Galen’s yellow-bile sort, and, as with Heracles, homicidally so. The dual nature of Bellerophon’s warrior dysfunctionality—the madness of yellow-bilious rage coupled with the self-isolation of black-bilious social separation—is one not unique to this Indo-European warrior. We see an expression of the same fusion of destructive fury against and retreat from society in, for example, the case of Marcus Coriolanus, one of several Roman reflexes of the primitive Indo-European dysfunctional warrior.[26] Aristotle’s (non-Hippocratic) characterization of both the maniacal rage of Heracles and the social withdrawal of Bellerophon as the result of mélaina kholḗ (μέλαινα χολή) ‘black bile’ is consistent with Greek mythic presentations of Bellerophon as displaying both reflexes of ancestral Indo-European warrior dysfunctionality.

§12. Greek khólos (χόλος) ‘bile’, preserved in Sophocles’ compound melánkholos (μελάγχολος) ‘black-bile’, equally, and typically, denotes ‘fury, wrath’. Significant is Cicero’s statement (Tusculan Disputations 3.11) that what Latin speakers call furor ‘rage’ (commonly that of the warrior)[27] the Greeks call melankholía (μελαγχολία): quasi vero atra bili solum mens ac non saepe vel iracundia graviore vel timore vel dolore moveatur, . . . ‘Just as if the mind were in fact stirred by black bile alone and not by severe rage, or fear, or anguish, as is frequent’.[28] The Greek word khólos (χόλος) and its variant kholḗ (χολή) ‘bile’ find their origin in Indo-European *g̑hel- ‘yellow’, color term that describes the appearance of bile and thus, one can infer, appropriated to name the biological material.[29]

§13. That being the case, there are a couple of observations that reasonably follow. First, we can say that in the semantic opposition yellow bile versus black bile, the former (yellow) is the unmarked member and the latter (black) is the marked member of the pair. That this is so is revealed by instances in which unmodified kholḗ (χολή) is set in contrast to mélaina kholḗ (μέλαινα χολή) and the former must be understood in context as referencing ‘yellow bile’. Thus, in the Hippocratic Nature of Man 5, Polybus (the probable author, and Hippocrates’ son-in-law) seeks to demonstrate empirically that there are four separate humors (yellow bile, black bile, phlegm, and blood); he goes about doing so by pointing out that there are four separate and distinctive measures for drawing these from the body: inflicting a wound draws blood; administering a phármakon (φάρμακον) that draws out phlegm causes vomiting of phlegm; administering a phármakon that draws out ‘black bile’ (kholḕ mélaina) causes vomiting of black bile; and administering a phármakon that draws out ‘bile’ (kholḗ) causes vomiting of bile: unmodified kholḗ ‘bile’ specifies ‘yellow bile’.

§14. The second observation follows from the first: the use of the color adjective ksanthḗ (ξανθή) ‘yellow’ in the Greek lexical concatenation ksanthḕ kholḗ (ξανθὴ χολή)[30] is otiose (that is, the phrase encodes notionally ‘yellow [ksanthḗ] material that is yellow’).[31] Ksanthḕ kholḗ is likely a phrase that arose in response to a concatenation mélaina kholḗ (μέλαινα χολή, that is, one that encodes ‘yellow material that is black’). It is worth noting at this point that Jouanna (2012:233) concludes that melancholía (μελαγχολία), nominal derivative of melánkholos (μελάγχολος) ‘black-bile’, “originated in Greek medicine at a stage prior to Hippocrates, and its origin will remain hidden in the shadows of Greek medical prehistory”: this conclusion follows from his observations (2012:229–230, 232–233) that Polybus in effect “invented” the notion of black bile as a humor and that melancholía (μελαγχολία), when the word is first attested (in Airs, Waters, Places, perhaps to be attributed to Hippocrates himself), must have described a familiar pathological phenomenon, as it is presented with no elaboration of its symptoms.[32]

§15. Plausibly we might identify the notion ‘black’ in mélaina kholḗ (μέλαινα χολή) / melánkholos (μελάγχολος) as not, in origin, a descriptor of material color but as, instead, a metaphorical descriptor—and one that cannot be separated from the nature of a wrath that is linked to this biological material—a wrath that is dark and careening toward dysfunctionality. Such a metaphorical use of Greek mélas (μέλας) ‘black’ is one that is well known: Homeric epic, for example, provides phrases such as mélas thánatos (μέλας θάνατος) ‘black death’ and mélaina kḗr (μέλαινα κήρ) ‘black (goddess of) death’. And in the context of warrior dysfunctionality we think readily of the ménos (μένος) ‘rage’ that completely fills Agamemnon’s phrénes . . . mélainai (φρένες . . . μέλαιναι) ‘black mind’ (Iliad 1.103–104) as he declares to Calchas that if he were compelled to release Chryseis to her father, he would require a comparable prize as compensation.

§16. But the semantic shading of the word for ‘bile’, the yellow biological material, into meanings expressing potentiality for destruction—as in Greek khólos (χόλος) ‘fury’—is one that appears to have been antecedent to the evolutionary emergence of the Greek language. Thus, Proto-Indo-European *g̑hel- also eventuates in Latin fel, fellis, a word that denotes ‘bile’ but also ‘poison, venom’, as at Virgil Aeneid 12.857, of a poisoned arrow. The irregular outcome of word-initial * g̑h- as f-, rather than expected Latin h-, indicates the Latin acquisition of fel from a neighboring dialect—likely Sabine: compare, for example, Sabine fēdus beside Latin haedus ‘young goat’ (from Proto-Indo-European *ghaido-) and Sabine fasēna beside Latin harēna ‘sand’. The Latin borrowing of fel ‘bile’ likely took place in the context of the acquisition of some ideological matrix in which fel participates as a signifier. In this instance Latin has no alternative *hel to offer, though a synonym of fel ‘bile’ is provided—that form being bīlis. The Indo-European heritage of Latin bīlis is uncertain, but the form appears to be of at least Italo-Celtic origin, as signaled by Welsh bustl, Cornish bistel, Breton bestl.[33] The synonym bīlis denotes not only ‘bile’ but also ‘anger’, as at Horace Epodes 11.16, of a lībera bīlis‘unrestricted anger’. ‘Rage’ as the sense of bīlis is conspicuous when the noun is accompanied by the descriptor ‘black’—again black bile: ātra bīlis agitat hominem ‘black bile drives the man’, writes Plautus (Captives 596), describing a warrior, within a domestic setting, who is mad with rage. When Pliny (Natural History 11.193) writes of bīlis being used as a term of reproach, he is commenting on his observation that fel nigrum ‘black bile’ is the cause of insānia ‘madness’, using fel and bīlis interchangeably.

§17. Pliny has returned our attention both to fel and to black bile and its affiliations with rage. In light of the likely Sabine origin of fel, the appearance of ātrum fel ‘black bile’ in the tradition of Hercules and his raging encounter with the tricephalic monster Cacus, and the aftermath of that encounter, is conspicuous. As I argued in Woodard 2006,[34] Roman “Hercules” in the Cacus-tradition, as preserved by Latin poets, stands in for the Sabine warrior-god Semo Sancus, and the myth of Semo Sancus’ fight with the cattle-thieving Palatine tricephal (Cacus) provides an Italic homologue to Vedic Indra’s combat with the dragon Vr̥tra—the two showing striking agreement. If the Latin account is coated, unsurprisingly, with a thin varnish of Greek influence,[35] it is most probably a Sabine import to Rome. Following his slaying of Vr̥tra Indra flees into remote space in a traumatized state, abandoning society: this is Indra’s warrior dysfunctionality. Hercules’/Semo Sancus’ flight to the “remote” (trans-pomerial) space of the Aventine—as he is possessed by ira ‘rage’[36]—is a homologous expression of the Italic warrior’s retreat. Virgil (Aeneid 184–305), whose presentation of the myth is the most expansive, describes in the following way Hercules’ response upon hearing the bellows of one of his stolen cows (lines 219–220a): hic vero Alcidae furiis exarserat atro | felle dolor ‘at this with madness indeed Alcides’ wrath blazed by black bile’—and we are put in mind of Cicero’s comments in Tusculan Disputations regarding the semantic equation of melankholía (μελαγχολία) and furor. The myth of the Sabine warrior deity Semo Sancus slaying a tricephalic monster, a myth of primitive Indo-European origin, provides precisely the sort of ideological matrix by which Sabine fel ‘bile’ would be transmitted to the Latin lexicon.

§18. Indo-European *g̑hel- ‘yellow (stuff)’ develops directly into reflexes that denote ‘bile’—of the yellow variety—and ‘rage’. What was the semantic intent of adjoining to that term a descriptor signifying ‘black’? What was the ‘black yellow stuff’ conceived to be? If yellow bile (the unmarked member of the opposition) is the cause of manía (μανία), as Pseudo-Galen reports, then black bile, I would posit, must be in origin the manía that the warrior turns against the “innocents”—the members of his own society. As there are two canonical forms of primitive Indo-European dysfunctionality, black bile must be equally the cause of the warrior’s abandonment of his own society—of his withdrawing his protective power from those it is his function to protect. Aristotle tells us as much in the Problemata physica in his assigning both the altered state of Heracles and Ajax (that of madness) and the altered state of Bellerophon (that of social isolation) equally to black-biliousness. This Greek notion of the bile that is black, melánkholos (μελάγχολος), as operative in the dual expressions of warrior dysfunctionality, is properly assigned to the prehistory of Greek medicine—as is the Greek act of taking the measure of melancholía (μελαγχολία). The problem that is being measured is one endemic to primitive Indo-European society.



Baltussen, H. 2015. “Understanding Odours in Problemata 12–13: Peripatetic Problems Concerning the Elusive Sense of Smell.” In Mayhew 2015, pp. 172–189.

Benveniste, E. 1945. “La doctrine médicale des Indo-Européens.” Revue de l’histoire des religions 130:5–12.

Benveniste, E. 2001. “The Medical Tradition of the Indo-Europeans.” Trans. Arthur Goldhammer. In Loraux, Nagy, and Slatkin 2001, pp. 422–427.

Bodnár, I. 2015. “The Problemata Physica: An introduction.” In Mayhew 2015, pp. 1–9.

Calame, C., and P. Ellinger, eds. 2017. Du récit au rituel par le forme esthétique: Poèmes, images et pragmatique cultuelle en Grèce ancienne. Paris.

Chantraine, P. 1968. Dictionnaire étymologique de la langue grecque. Paris.

Chantraine, P. 1973. Grammaire homérique. Volume 1. Paris.

Driessen, C. M. 2003. “Evidence for *g̑helh2, a New Indo-European Root.” Journal of Indo-European Studies 31:279–305.

Eijk, P. J. van der. 2005. Medicine and Philosophy in Classical Antiquity: Doctors and Philosophers on Nature, Soul, Health, and Disease. Cambridge.

Ernout, A., and A. Meillet. 1959. Dictionnaire étymologique de la langue latine. 4th edition. Paris.

Gamkrelidze, T. V., and V. V. Ivanov. 1995. Indo-European and the Indo-Europeans. 2 vols. Trans. J. Nichols. Berlin.

Hansen, B. S. S., B. N. Whitehead, T. Olander, and B. A. Olsen, eds. 2017. Etymology and the European Lexicon: Proceedings of the 14th Fachtagung der Indogermanischen Gesellschaft, 17–22 September 2012, Copenhagen. Akten der Fachtagung der Indogermanischen Gesellschaft 14. Wiesbaden.

Jouan, F. 1989–1990. “Femmes ardentes et chastes héros chez Euripide.” Sacris Erudiri 31:187–208.

Jouanna, J. 2012. Greek Medicine from Hippocrates to Galen: Selected Papers. Trans. N. Allies. Studies in Ancient Medicine 40. Leiden.

Kannicht, R., ed. 2004. Tragicorum graecorum fragmenta. Vol. 5, Euripides. Göttingen.

Loraux, N., G. Nagy, and L. Slatkin, eds. 2001. Antiquities. Vol. 3 of Postwar French Thought, ed. R. Naddaff. New York.

Mallory, J. P. and D. Q. Adams. 1997. Encyclopedia of Indo-European Culture. London.

Mayhew, R., ed. 2016. The Aristotelian Problemata Physica: Philosophical and Scientific Investigations. Philosophia antiqua 139. Leiden.

Melchert, H. C. 2016. “‘Western Affinities’ of Anatolian.” In Hansen et al. 2016, pp. 297–305.

Meurant, A., ed. 2011. Routes et parcours mythiques: Des textes à l’archéologie; Actes du septième Colloque international d’anthropologie du monde indo-européen et de mythologie comparée (Louvain-la-Neuve, 19–21 mars 2009). Langues et cultures anciennes 17. Brussels.

Meurant, A., and M. V. García Quintela, eds. 2017. Traditions indo-européennes et patrimoines folkloriques: Mélanges offerts à Bernard Sergent. Paris.

Nagy, G. 2020.04.03. “Comments on Comparative Mythology 7, Finding a Cure for the Anger of Hērā.” Classical Inquiries.

Nauck, A. 1889. Tragicorum Graecorum fragmenta. 2nd ed. Leipzig. Reprinted 1964, ed. B. Snell, Hildesheim.

Papamichael, E. M. 1983a. “Bellerophon and Stheneboea (or Anteia).” Dodone 12:45–74.

Papamichael, E. M. 1983b. “Peleus and Hippolyte Cretheis (or Astydameia).” Dodone 12:139–152.

Petit, C. 2009. Galien. Vol. 3, Le Médecin; Introduction. Collection des universités de France, série grecque 471. Paris.

Positano, L. M., W. J. W. Koster, and D. Holwerda. 1960–1964. Jo. Tzetzae commentarii in Aristophanem. 4 vols. Groningen.

Poultney, J. W. 1959. The Bronze Tables of Iguvium. Philological Monographs 18. Baltimore.

Rix, H., M. Kümmel, T. Zehnder, R. Lipp, and B. Schirmer. 2001. Lexikon der indogermanischen Verben: Die Wurzeln und ihre Primärstammbildungen. 2nd ed. Wiesbaden.

Simpson, M. 1976. Gods and Heroes of the Greeks: The Library of Apollodorus. Amherst, MA.

Untermann, J. 2000. Wörterbuch des Oskisch-Umbrischen. Heidelberg.

Vendryes, J. 1918. “Les correspondances de vocabulaire entre l’indo-iranien et l’italo-celtique.” Mémoires de la Société de Linguistique de Paris 20:265–285.

Walde, A., and J. Pokorny. 1927. Vergleichendes Wörterbuch der indogermanischen Sprachen. Volume 2. Berlin.

Walde, A., and J. Pokorny. 1930. Vergleichendes Wörterbuch der indogermanischen Sprachen. Volume 1. Berlin.

Walsh, T. R. 2005. Fighting Words and Feuding Words: Anger and the Homeric Poems. Lanham, MD.

Woodard, R. D. 2006. Indo-European Sacred Space: Vedic and Roman Cult. Chicago and Urbana, IL.

Woodard, R. D. 2011. “The Roman Regifugium: Myth and Ritual of the King’s Journey Beyond the Boundary.” In Meurant 2011, pp. 304–332.

Woodard, R. D. 2013. Myth, Ritual, and the Warrior in Roman and Indo-European Antiquity. Cambridge.

Woodard, R. D. 2017a. “Bellérophon et l’agressivité féminine: diachronie et synchronie dans les mythes et la pratique rituelle.” In Calame and Ellinger 2017, pp. 305–336.

Woodard, R. D. 2017b. “Hated by All Gods: Lycurgus, Bellerophon, and the Twin Maladies of the Indo-European Warrior in Homer’s Iliad.” In Meurant and García Quintela 2017, pp. 843–866.

Woodard, R. D. 2020. “Coriolanus and Fortuna Muliebris.” Japan Studies in Classical Antiquity 4:3–32.



[1] On the work as Peripatetic compilation, see, among more recent treatments, the various contributions to Mayhew 2015, especially Bodnár’s and Baltussen’s chapters.

[2] On the relationship of the ideas on the melankholikoí (μελαγχολικοί) that are expressed in the Problemata and Aristotle’s views as otherwise evidenced (and so on the attribution of the work to Aristotle), and on the complexity of the Aristotelian notion of melankholikós, see van der Eijk 2005:139–168.

[3] On the Hippocratic link between epilepsy and the melancholic condition see the discussions of Jouanna 2012:235–236, 238n26, 242–243, with bibliography.

[4] On Lysander and black biliousness (melankholía [μελαγχολία]) see also Plutarch Life of Lysander 2.3 and 28.1.

[5] See Nagy 2020.04.03, especially §§11–15.

[6] For the constituents of the potion, in addition to bile, see Diodorus Siculus 4.36.5.

[7] On Bellerophon as paradigm of the Indo-European dysfunctional warrior, see Woodard2017a.

[8] With line 200 showing a slight textual variation; compare below.

[9] The epic poet judges both Lycurgus and Bellerophon to have become “hated by all gods.” On the relevance of the phrase to the phenomenon of warrior dysfunctionality, see Woodard 2017b.

[10] Or ‘spirit’—though Bellerophon’s flesh is not devoured.

[11] On Aristotle’s broad use of the adjective melankholikós (μελαγχολικός) to denote fundamentally ‘relating to black bile’, rather than continuing the Hippocratic sense ‘relating to melancholy’, see Jouanna 2012:238.

[12] Benveniste 2001 [= English translation of Benveniste 1945]:423. Regarding Benveniste’s characterization of the cultural conservatism reflected in the lexica of Italic (as well as Celtic) and Iranian (as well as Indic) see earliest Vendryes 1918.

[13] Benveniste 2001:424. For the Avestan practice see Vendidad 7.44; compare Yašt 3.6 (cited by Benveniste in his note 2).

[14] For extensive discussion of both of these warrior dysfunctionalities and of societal measures designed to address them see Woodard 2013.

[15] The relevance of the passage to the analysis of Problemata physica 953a was also noticed by Jouanna, who (2012:246) describes it as “a parallel little known by commentators on Aristotle’s Problem.”

[16] Here I follow the text of Petit 2009:57 (13.24.14); see also her page 152.

[17] On Indra as dysfunctional warrior, see Woodard 2006:206–209 and Woodard 2013, especially pages 120–121, 139–148, 202–203, 216–217, 229–230, as well as the introductory discussion of pages 90–99.

[18] See Woodard 2011, 2013, and 2017b.

[19] On CúChulainn as dysfunctional warrior see Woodard 2013, especially pages 91–92, 121–124, 130–133, 206–207, 218–219, 230–231, as well, again, as the introductory remarks on pages 90–99.

[20]On Anteia/Stheneboea and Bellerophon see Woodard 2017a:309–312. Still other Greek examples can be identified: see Simpson 1976:81–82; Jouan 1989–1990; Papamichael 1983a and 1983b. A notable instantiation of the motif, vis-à-vis Bellerophon’s experience, is that one involving Peleus, who was desired by Astydamia, the wife of Acastus, king of Iolcus; Peleus refused her advances, whereupon Astydamia told Acastus that Peleus had tried to have intercourse with her, and, in response, Acastus attempted to destroy Peleus by deserting him in the wilderness space of Mt. Pelion (see Pseudo-Apollodorus Bibliotheca 3.13.3).

[21] On these verbs see Chantraine 1968:58, 66.

[22] See Chantraine 1973:345, as well as 1968:58.

[23] Compare Iliad 5.352, where alúōn (ἀλύων) is used participially to describe a distraught Aphrodite, who, wounded by Diomedes, ‘was suffering terribly’ (τείρετο δ’ αἰνῶς) and so ‘went off’ (apobaínō [ἀποβαίνω]), that is, ‘retreated’, to Mt. Olympus.

[24] Proto-Indo-European *h2el- is source also of Latin amb-ulō ‘to go on foot, to go abroad’ and Umbrian amb-oltu ‘walk about’; Latvian aluôt ‘to go astray’. See Walde and Pokorny 1930:87–88; Poultney 1959:295; Mallory and Adams 1997:629; Untermann 2000:84–85. See also Rix et al. 2001:264.

[25] Tzetzes Commentarium in ranas [= Positano, Koster, and Holwerda 1960–1964] 1051.

[26] On Coriolanus as dysfunctional warrior see Woodard 2020. On Hercules/Semo Sancus see Woodard 2006:211–219 and Woodard 2013:124–128, 133–138, 207–208, 219, and 231–233. On Camillus see Woodard 2013:105–117 and Woodard 2020:14–15.

[27] On the significance of furor in the description of warrior dysfunctionality see Woodard 2013:92, 133, 171–172, 177, 193, 199, 205, 218, 229, 234, 241, 245, 256.

[28] Cicero’s is an intriguing statement about which more undoubtedly needs to be said but which will have to wait for another occasion.

[29] For a careful discussion of the diachronic semantics of Indo-European *g̑hel- ‘yellow’ and Greek khólos (χόλος) and kholḗ (χολή), framed by other Indo-European outcomes of the etymon, see Walsh 2005:219–223. The identification of *g̑hel- ‘yellow’ as etymon of Greek khólos and kholḗ is reasonable and nearly universally accepted: see, for example, Walde and Pokorny 1930:624–625; Chantraine 1968:1268; Gamkrelidze and Ivanov 1995:618, 715; Mallory and Adams 1997:217, 654. Driessen (2013) would divorce khólos from kholḗ etymologically, arguing for a “new” (we might say “reformatted”) Proto-Indo-European root *g̑helh2– ‘to be in a fury’ as etymon of khólos. Positing the occurrence of some such root—one producing reflexes in Celtic (a root previously reconstructed as *gal-/*ghal-, as in Walde and Pokorny 1930:539) and in Anatolian (see Melchert 2016:299)—appears noncontroversial. However, it hardly seems plausible to separate khólos ‘bile’ and ‘rage’ from its first-declension counterpart kholḗ and the yellow/bile complex of which they are a part, especially when the cultural context of khólos is given due consideration. We might instead want to consider the prospect that Indo-European *g̑hel- ‘yellow’ may have had synchronic lexical/semantic affiliations with the ancestor of Old Irish gal ‘warlike ardor’ and Hittite kallar ‘baneful’; it is an interesting idea (and one that Walsh [2005:221, with note 52] essentially evokes) to which I would like to return in a future essay.

[30] The phrase is first attested in the Hippocratic corpus.

[31] Clinically this is so. Linguistically one might contend that it is only so along the diachronic dimension, but this is not a necessary limitation to the extent that linguistic transparency at certain synchronic moments may be entailed. Even in the fifth century BC, for instance, the relatedness of the terms kholḗ (χολή) ‘bile’ and khlóē (χλόη), describing greenish-yellow plant materials, may have been relatively transparent. In the second century CE khloḗbaphos (χλοήβαφος) can be used to denote ‘person with jaundiced complexion’ (Aretaeus De causis et signis diuturnorum morborum libri duo 1.13.2).

[32] Jouanna (2012:236) further offers:

The illness called melancholy pre-dates Hippocratic medicine. Black bile was born as a humour in its own right in Hippocratic medicine, which supposes that melancholy was attributed firstly to a pathogenic variety of bile. The melancholic temperament is rarely attested. There is no characterisation of it and it is not brought within the context of the theory of the four humours.

[33] See Walde and Pokorny 1927:111; Ernout and Meillet 1959:71.

[34] In Woodard 2006 see especially pages 184–224; see also Woodard 2013, pages 125–127, 133–138, 160–163, 207–209, 231–232.

[35] See Woodard 2006:193, 211, 216.

[36] See Propertius 4.9.62. On the significance of this feature of the Italic tradition, see Woodard 2013:133, 173, 255–258.