Introduction: Homeric Anger

{1} Across the landscape of Homeric studies there are monuments that demonstrate how central anger is to whatever the early Greeks—audience and performers alike—imagined the Homeric narratives to be about. By our own various references to the powerful concept of anger, we seem to wave it as a magic wand for conjuring meaning. The titles of some of the most insightful works in Homeric studies at times seem to me part of an inscription recording a magician’s incantation, set up in the agora of Homeric studies:
The Wrath of Athena
The Wrath of Paris
“Der Zorn des Paris”
Götterzorn bei Homer
“The Wrath of the Gods”
“The Anger of Poseidon”
Homers Zorn des Achilleus und den Homeriden Ilias
“The Wrath of Thetis”
This list can easily be expanded to include the analysts’ attempts to find the Ur-Ilias by styling it the mênis- (“anger-”) poem. [1] So too the literary critic will confidently locate the subject of the Homeric Iliad in this central concept; for example, according to C. Whitman, “the Wrath of Achilles had probably been an epic subject for generations when Homer found it.” [2]
Such conversations about anger have continued from antiquity to the present day, with the first scholium telling us that the variety of Homeric terms for anger are just so many synonyms: {1|2}
ὥσπερ ἐπὶ συκῆς πρῶτον μέν ἐστιν ὄλυνθος, εἶτα φήληξ σῦκον ἰσχάς, οὕτω πρῶτον ὀργή, θυμὸς χόλος κότος μῆνις. ὅμως ὁ ποιητὴς ὡς συνωνύμοις ὀνόμασιν ἐπὶ Ἀχιλλέως χρῆται· “ἠὲ χόλον παύσειεν ἐρητύσειέ τε θυμόν” (Α 192)· “οὐδ’ ὄθομαι κοτέοντος” (Α 181)· “αὐτὰρ ὁ μήνιε νηυσὶ παρήμενος” (A 488).
(Erbse 1969-77, vol. 1, 4: Schol. A ad Il. 1.1)
Just as for the fig-tree, there is first the ólunthos , then the phḗlēks , the sûkon , the iskhás , so too there is first orgḗ , [then] thumós, khólos, kótos, mênis . Nonetheless the performer uses these words as synonyms in respect to Achilles: “He would stop his khólos and restrain his thumós ” (A 192). “Nor would I care for you in your kótos ’’ (A 181); “and he raged [ mḗnie ] sitting by the ships” (A 488).
This ancient observation—pointing out that anger has as vigorous a cohort of synonyms as the fig—provides a background question for this study. Do the words for anger in Homeric texts present us with mere synonyms, so that the scholiast’s careful conspectus of terms for Achilles’ anger in Iliad 1 may be accurately called synonymous? Or should we look more carefully? Should we examine anger more closely than we do figs? [3]
Indeed, we have at least thirteen words for anger in Homer, [4] and one of them has special status as the subject of the Iliad, as is made clear by the rigorous poetics of Homeric prooemial style. [5] An epic whose characters cross the Trojan plain to slaughter one another depends upon the very emotion that drives them to pursue the war. We can join critics, ancient and modern, in acknowledging that an understanding of anger is crucial to any reading of the Homeric texts.
One question remains: How surely do we know that to which we have so readily agreed? For all our certainty about Homeric anger, our studies have produced only a few substantial works devoted to analyzing this important concept. Indeed, the semantics of the entire range of anger terms in Homer remains obscure, despite a number of promising excursions into this area of study. In 1879, J. H. H. Schmidt discussed the Greek synonyms for anger in Synonymik der griechischen Sprache. Table 1 groups the Homeric words as they are cited by Schmidt.
Table 1. Schmidt’s list of anger terms
Category 1. μένος, μενεαίνειν; θυμός; χόλος, χολοῦσθαι
Category 2. σκύζεσθαι; χαλεπαίνειν, χαλεπός; ὀδύσσεσθαι; χώεσθαι
Category 3. κότος, κοτεῖν; μῆνις, μηνίειν, μηνιθμός, μήνιμα
From a commonsense point of view, Schmidt’s categories are quite reasonable. The first category shows what we might call natural anger, [6] while the items in the last category are selected on the basis of anger’s temporal duration. In itself, {2|3} longevity indicates a kind of grudge or ill will. [7] The second category is a kind of catchall, where anger can lead to various points on the emotional spectrum, from anxiety to hate. [8]
Schmidt’s categories are to some degree in harmony with mine; after all, I am about to suggest that kótos is characterized by “duration,” and we can all recognize that mênis, Homeric anger par excellence, is extensive in a similar fashion. Nevertheless, Schmidt’s list sometimes separates terms that are, when viewed through Homeric poetic practice, linked by ring-composition. Indeed, no categorization that neglects poetics could tell us that khṓesthai and khólos need to be grouped together, but Homeric poetry does in fact link them as coordinated elements in ring-composition. [9] Clearly, to go beyond lexical completeness, we need to take Homeric poetics into account.
In 1950, J. Irmscher published Götterzorn bei Homer, his 1947 Berlin dissertation directed by W. Schadewaldt. As the title indicates, his treatise is primarily concerned with divine anger. In his first chapter, Irmscher ambitiously sets out to cover thirteen words for anger. [10] He isolates the violation of timḗ as the central motive for anger in the Iliad and goes on to argue that in the Odyssey it is human húbris that motivates Zeus’s wrath; Irmscher, however, falls short of producing a definitive treatise on the semantics of Homeric anger. Indeed, Irmscher exhibits some problems of method, especially in his neglect of the oral-poetic character of Homeric diction. Thus, although he acknowledges the traditional nature of Homeric poetry, we do not see the necessary attention to issues of context and diction in traditional poetics. To proceed in his manner, admittedly, makes the lexicographical work more straightforward, as a variety of Homeric words can be readily cataloged and then included in a single analysis. [11]
Nonetheless, this laudable impulse to lexical inclusiveness itself presents a problem. Do all the words that Irmscher cites really belong together? [12] For example, in the analysis of Zeus Xenios, the terms kótos, mênis, and khólos seem interchangeable. [13] At the very least, we need an analysis of the context and diction in which these terms are found if we are to judge them synonyms. Indeed, we cannot grasp the relationship of these words, one to another, without a close study of Homeric traditional usage, as revealed through a detailed review of Homeric poetics as it relates to each term. To proceed otherwise neglects our most important resource in the understanding of Homeric poetry, the traditional use of context and diction through its customary practice in these poems. [14] Indeed, we will see that khólos and kótos are rigorously kept separate, with very few waverings, throughout Homeric poetry.
Comparable in scope to Irmscher’s work, P. Considine (1969) produced a study of “divine wrath” in the Near East and Greece, assembling many important parallels that abound in Homeric, Mesopotamian, Canaanite, and Hittite texts. As with many other studies on anger, Considine’s method here does not focus on the poetic usage of particular words for anger. Thus, we are left without a clear notion of the boundaries that mark the territory covered by different lexical {3|4} items. That Homeric narrative is intimately bound up with a complex poetics must simply never be ignored. For example, in discussing Hera’s “disarming address” to Aphrodite, where Hera alludes to Aphrodite’s kótos against the Trojans as a possible impediment to her helping Hera in the Diòs Apatḗ, Considine seriously underestimates what is being said when he translates Il. 14.191-92: “Will you be cross [kotessaménē] and refuse me, because I am on the Danaans’ side and you are on the side of the Trojans?” [15] Now the divinity’s reference to kótos here is far from an attempt to minimize their conflict. Rather, it is an attempt to remove the most powerful hindrance to Hera’s request for Aphrodite’s help, namely, the kótos of Zeus. In his work on the lexicon for anger in Greek (1966, 1985, 1986), Considine carefully considers the lexical variety of Homeric diction, especially in opening and exploring the semantic field, all the while capably demonstrating that adequate analysis requires a detailed study of context. One needs to add, however, that in Homeric poetry, context is more than a paraphrase of the immediate narrative moment in which the term occurs. The richness of the language requires us to study each passage’s form and place it in the wider diction of the poems in addition to the narrative context in which the lexical item occurs. [16]
One of the most frequently cited articles on the terminology for aggressive impulses in Homer is A. W. H. Adkins’s “Threatening, Abusing and Feeling Angry in the Homeric Poems” (1969). The author is concerned with five words: neikeíein “to quarrel,” apeileîn “to threaten,” okhtheîn “to be troubled,” khṓesthai “to be angry,” menaeínein “to rage.” [17] Adkins discusses two distinct words for anger (khṓesthai, menaeínein), even while attending to the relationship between khṓesthai and okhtheîn, where he concludes that the subject has a more positive attitude toward action if he is in a situation where khṓesthai is appropriate than he has when okhtheîn is appropriate. As was the problem with the studies by Irmscher and Considine, Adkins makes no attempt in his brief article to set out systematically the meaning of khṓesthai within the formal poetic context of Homeric poetry. Instead, one passage is juxtaposed with another in a manner that depends on the associative skills of the researcher. Adkins’s distinction that okhtheîn signifies “a psychological response of frustration, distress and anger” [18] and that khṓesthai means “confusedly grief and anger at once” is neither systematic nor satisfying. [19]
What remains to be done is to make clear the syntagmatic relationship between anger terms and their immediate context and the paradigmatic relationships they have with one another; without such study, the relationship that obtains between, for example, khólos and khṓomai remains invisible; without a commitment to the system in which these terms have their function, it is difficult to arrive at the hypothesis that drives Feuding Words and Fighting Words, that khólos and kótos are distinct from one another.
Before addressing my own method, I want to suggest that as difficult as the problem of studying a large lexical field is for the Homerist, it is even more so {4|5} for the cultural historian who tries to sort out from the galaxy of Homeric vocabulary the notions that best characterize Homeric anger in order to establish large-scale cultural trends of great importance. A cultural historian may, with understandable dismay, balk at the suggestion that a full-scale formulaic analysis is in order before anyone may venture into the terrain of Homeric narrative.
Yet it is precisely this problem that will be encountered by those who, trying to put the Greeks in context, take the words for granted in the interest of a seemingly more important enterprise. For example, E. Voegelin makes a problematic claim about Homeric khólos:
It will have become clear … that in Homeric society a lordly wrath is not a private state of emotions. A cholos, a wrath, is a legal institution comparable to a Roman inimicitia or a medieval feud … Hence, in the compact Homeric cholos one must distinguish between the emotional, wrathful reaction against damage inflicted on a man’s status and the customs that regulate the course of the emotion. [20]
Here Voegelin is justified to say that there is an anger that as a legal institution parallels cultural structures such as can be observed in the Roman institution of inimicitia / amicitia. But that anger is not khólos—it is kótos. For kótos is directly related to the incipient vocabulary of Greek legal institutions (including díkē “judgment,” “justice”; húbris “outrage”; and the like), especially as the terms relate to the traditional institutions of reciprocity. And, as I will demonstrate, khólos bears little resemblance to a “legal institution.” The only way to sort out khólos from kótos, however, is to study systematically the semantics of anger that makes clear such a distinction. Indeed, without such work, it is hard to imagine how a cultural historian can distinguish which word describes “a lordly wrath” and which “a private state of emotions.”
My emphasis on Homeric matters steers the discussion away from later Greek material. Indeed, the reader will notice that missing from this book is an extended analysis of the literature on anger in the ancient world after the early Greek period. Fortunately, as this book went to press an important book on ancient anger appeared, one that provides just such an extended analysis. W. V. Harris’s Restraining Rage (2001), focusing on the restraint of anger, exhaustively covers a vast panorama of Greek and Roman thinking about anger, with a central aim of examining Greco-Roman antiquity’s yearnings for the repression of anger. The importance of this project goes without saying, especially given the philosophical tradition’s tenacious speculation on this topic (Harris 2001, 84-128).
In contrast, I focus my discussion of anger on Homer in order to fill what seems to me a lacuna. As can be surmised from my comments above, the study of anger in antiquity has often taken Homeric anger for granted or at least as worthy only of being confined to a review-chapter serving as background for the rest of antiquity. Despite this drawback, the studies of anger in the later literature {5|6} have been admirably vigorous (as now with Harris and with even more recent work [Braund and Most 2003]); we Homerists should emulate that vigor.
Yet, to study anger in Homer is not the same as to account for anger in the whole of Greek and Roman antiquity. It is, I must say, quite different. For the Homeric texts present problems of poetics and of form that simply cannot be sped over in the interest of getting to later philosophical texts and historical data. If philology is the art of slow reading, the study of Homer is the art of slow and steady reading: for to deal with poetics is not merely to correlate lexical glosses with immediate context. One must view the lexical data within the traditional matrix that is its life’s blood. The task of the Homeric scholar, now, is to treat Homer not merely as an antecedent to “the polis” (Harris 2001, 27), but as a poetic intervention into profound cultural issues, an intervention that deserves a full and robust intellectual engagement. Such a vigorous Homeric poetics is a humane way to make more meaningful the dialogue between Homerists and other readers of the ancient world. [21]
At the heart of a systematic poetics of Homeric narrative are the traditional meanings together with the traditional forms that produce those meanings within the practices of oral poetry. In the words of L. Muellner: “Obviously semantic analysis cannot be effective without an awareness of such considerations of technique.” [22] This kind of complexity of Homeric poetic semantics can be enriched by looking at how the highly developed anger systems of other cultures such as the Ifaluk are deployed within their traditional forms of discourse. By turning to the anthropological literature on emotions, where anger is conceived of as a social phenomenon, I hope to give the Homeric material a comparative context. The following examples can begin to show how an anthropologically informed approach might begin to work. [23]
For example, C. Lutz, in Unnatural Emotions, her ethnography of the Ifaluk in the Southwest Pacific, gives an account of their emotional life as it relates to the way the Ifaluk organize their cultural perception of anger:
The Ifaluk speak about many types of anger. There is the irritability that often accompanies sickness (tipmochmoch), the anger that builds up slowly in the face of a succession of minor but unwanted events (lingeringer), the annoyance that occurs when relatives have failed to live up to their obligations (nguch), and, finally, there is the frustrated anger that occurs in the face of personal misfortunes and slights which one is helpless to redress (tang). But each of these emotions is sharply distinguished from the anger which is a righteous indignation, or justifiable anger (song), and it is only this anger which is morally approved. [24]
Lutz establishes the way words for emotions are charged with meanings that are conceptual, political, moral, and, more generally, cultural. In short, such a study implies that through the consideration of emotional vocabulary, it is possible to come to terms with critically important conceptual features of a given culture. {6|7}
Similarly, M. Rosaldo in her Knowledge and Passion (1980) describes the dynamics of emotion as manifested among the Ilongot in Northern Luzon in the Philippines. [25] Rosaldo’s discussion is worth quoting at length for her account of how the terms that define the Ilongot cultural understanding of anger are to be taken:
My grasp of Ilongot ideas of the dynamics of emotion grew through a series of interviews in which I addressed myself to questions about the use of liget, bea, and related words. I knew that the importance of heart and feelings in Ilongot accounts of action had more to do with general notions of vitality, conflict, and cooperation than with a “psychologistic” sense of persons, but had little understanding of how Ilongots thought emotions worked. The net of senses associated with emotion words was my introduction. By asking informants … I found, first, that of words characterizing the heart, liget was by far the richest in its associations; and in exploring ways that words like “difference” (‘apir), “strength” (‘eret), and “quick” (awet) resemble liget, I began to see in a term that I had understood initially to mean no more than “anger” a set of principles and connections with elaborate ramifications for Ilongot social life … And so, building from families of terms and contexts, I developed a grasp of certain Ilongot metaphors, and of the loose conceptual scheme that underlies them and describes the principles working “energy, anger, passion” in their world. [26]
The Homerist should emerge from an anthropologist’s analysis of anger with renewed expectations that semantic work can move beyond the well-tended groves of lexical classification and into the wilder forests of words and things. If, as M. Rosaldo describes them, metaphors for anger might be based on a “loose conceptual scheme that underlies them,” it may be that philological rigor will not have the flexibility to unearth loose schemes as well as it does the tightly constructed ones. [27]
What will be the link between an analysis of Homeric anger words and an analysis that draws from work outside classical studies? The links will be various, but a secure way for the reader of Homeric poetry to begin is, at the outset, to lay the foundation for the work by close attention to and analysis of early Greek diction. R. Sacks suggests that this approach to “the problem of recovering tradition” involves two steps. [28] The first step is to collect all the attestations of a traditional phrase; and the second step is to search out “those patterns which, with all their complexities, are actually displayed by Homeric phrases.” [29] Sacks reminds us how far we are from an adequate description of Homeric practice.
The approach to “searching out” needs to be determined from the outset, since the meanings we find, as hermeneutics has taught us, are intimately bound up with the meanings that we search for. Do we have any control over our first steps, or are we bound to choose our objects of study at random? I return to Sacks’s observations for direction: “Studying Homer is not like studying an existing language. If we want to focus on the theoretical in our analyses of the {7|8} English language, we may take some comfort in being native speakers and thus experts in the language’s synchronic ‘traditions.’ No such expert can ever again exist for Homer …” [30] The problem of the nonliving tradition confronts us immediately. On whom could we rely to be as fluent in Homeric diction as the singer and audience once were? However, certain revelations in the narrative can bring us closer to that expert understanding. These revelations can occur in the form of the folk definition. I will return to this term later in this Introduction, but for now I note that within the text there survive statements about terms such as mênis, khólos, and kótos that go beyond exemplifying the meaning of the term. As we will see, once a native explanation of a term is found, the meaning of that term can be tested alongside that definition. We will also see that this process can guide and ground our study more firmly than even the best of our own intuitions as a nonnative audience.
In the give-and-take of an anthropological field situation, researcher and informant, ideally, find ways to communicate the principles that underlie their statements. I have, of course, no informants to interview. Nonetheless, though we can interact with neither poet nor audience, we may take heart that the Homeric poems show human actors both involved in and commenting upon their social structure, in what will be a variety of ways. [31] Indeed, that these texts are epics in form makes possible moments such as that of Calchas’s definition (described below). One aspect of epic, its expansiveness, makes it a good place to look for an informed definition of a term denoting anger. [32] Once that definition is given, if it is confronted by an inquiry that both illuminates rigorous connections and exploits loose conceptual schemes, it may yield insight, to use M. Rosaldo’s words, into Homeric culture’s “set of principles and connections [that have] elaborate ramifications for [its] social life.” [33]
The student of Homeric poetry is left with a struggle if not a dilemma: to use the methods of linguistics, anthropology, and the other intellectual disciplines for studying human culture and at the same time maintain sharp focus on the poetry, a task the social scientist sees as secondary. Thus, the student of Homeric poetry must keep in mind the poetic object in question, all the while avoiding the siren- song of the applied methodology. This act balances the real and the fictional, and it is indeed difficult to draw documentary evidence about a culture from its creative narrative art.
Because I have suggested that this study will turn to anthropological and linguistic cultural inquiries in order to understand the poetry of the Iliad and the Odyssey, and because it will be evident that this process will be engaged after we have addressed formal issues, it is wise to ask whether social or literary perspectives have priority in my work. The methodological distinction between a social study and a textual one demands more theoretical rigor than I intend to apply here. But it will become evident that I am not trying primarily to reconstruct archaic or bronze-age social structures by combing through the evidence of the Homeric texts. Since I begin my analysis with close readings of {8|9} words, passages, and texts, my focus is primarily on those texts, the aim being to draw out of the cultural categories examined, such as the notion of “feud” or “aggression,” a conceptual field that will be of value to students of the Iliad and the Odyssey. Thus, even if an archaic Greek cultural practice is found out by sheer hard work, it is placed immediately into the service of the text. In sum, this is a work of poetic analysis using the tools at our disposal—which include social theory—to organize evidence and give shape to conclusions. But those conclusions remain, unflinchingly, about the text.
In pursuing the usefulness of bringing to bear anthropological insights onto the study of literature, in trying to understand the way that structural analysis of texts illuminates those texts, I have found compelling the insight of Marshall Sahlins as he, from the anthropologist’s perspective, confronts the way structural analysis interacts with history. For Sahlins, “a set of historical relationships … at once reproduce the traditional cultural categories and give them new values out of the pragmatic context,” so that to analyze the “structure” of a cultural system is not to deny its historicity, nor can a study of history be innocent of the structural system that, for a given culture, provides meaning for the events undergone by the human actors within that culture. [34] Sahlins thus sees structure and history as inextricably related. My use of this observation is to apply it to the experience of a text, so that the structure that we labor to bring forth is set beside the text itself. Although this is not the place to establish securely the analogy on which my use of this idea depends (text : poetics :: history : social structure), let me again assert that the observations on Homeric poetics in this book are made not only in order to understand the system that lies behind the texts but also in order to deal directly with and illuminate the Iliad and the Odyssey. [35]
I return to M. Rosaldo for a clue to our opening enigma: Why does anger, important as it is for Homer, still need to be explored at the most basic levels? Rosaldo’s research provides us with a theoretical basis from which to understand why the central concept of the Iliad is not yet clearly grasped. She suggests that the distinction between symbolic and ordinary language has led students of culture to underrate the complexity, and the cultural specificity, of ordinary words. It might be thought, she suggests, that
in contrast to the “symbol” … are the more ordinary, strictly referential terms of daily language—words perhaps like “anger,” “red,” or “blossom,” the understanding of which requires little more than an ability to recognize the discrete emotional states, plants, or colors that their names conventionally denote. [36]
But then:
Having distinguished what appear to be symbolic terms from the transparent common sense on which to found translation, such analysts prove incapable of appreciating the ways in which apparently foreign and peculiar deeds may by {9|10} themselves have common sense interpretations. And at the same time, their approach ignores the fact that common sense in other cultures is ultimately as demanding of interpretation as is apparently obscure “symbolic” form. [37]
Anger is so much a common part of the way of life for a Homeric warrior that we have come to set aside the terms for the concept of anger as an item of inquiry. This mistake is even harder for us to observe since in our own lives the term “anger” is also part of day-to-day existence; in other words, we think we already know what it is. In fact, our referral of mênis to English wrath is itself an attempt to claim a transparent meaning for this term. Surely, we think, this anger is just like the wrath of the Old Testament Yahweh, an exalted anger. [38] I mean to say that anger, for Homer, has been taken by readers as a “strictly referential term,” the meaning of which we recognize immediately. Further, the lexica do not indicate the systematic nature of the distinction between khólos and kótos to be elucidated in this study. [39] The dictionaries acknowledge a difference in intensity between the two, but crucial contextual features have yet to be examined for their rich elaboration of the semantic force of these terms. Such terms are “demanding of interpretation” in M. Rosaldo’s words, just as much as terms of great social significance—such as, for example, “justice.”
The importance of anger has been acknowledged in Homeric studies primarily for mênis but not for the dozen or so other Homeric terms for anger. A word like khólos seems less likely to draw our attention precisely because it is the mere unmarked term for anger. [40] M. Rosaldo’s observation that studies of other cultures tend to gravitate to what is considered a priori the symbolically laden material might account for our neglect of the other words for anger, including khólos. [41]
Thus, Homerists still need to address systematically, even if belatedly, the central concept of the field. Perhaps, we can show, as M. Rosaldo shows for the Ilongot, that the emotion of anger is part of a system, a net of senses intricately related to Homeric social life. Such a result should not surprise us, since the most recent anthropological and linguistic studies of the emotions have indicated the central importance of anger to cultural studies. [42]
Before we proceed, I note that Homerists are not alone in coming late to these issues. Awareness of the importance of anger is a recent development even in the social sciences. Sociological and psychological studies of aggression, some researchers complain, pay too little attention to anger. Favoring behavioral, and hence measurable and quantifiable models, our own culture’s tendency has been to exclude the emotions from analyses of aggression. Consider the summary given by R. Novaco: [43]
Until recently, the concept of anger has taken a backseat to that of aggression. The extensive research conducted on human aggression in psychological laboratories in the 1960s and 1970s virtually excluded anger except as a {10|11} precondition induced to facilitate aggressive behavior. Behavioristic traditions favored the study of observable action over subjective experience. Consequently, anger received little attention. However, this neglect of anger as subject matter also occurred in the psychoanalytic literature … In fact, one rarely, if ever, finds the word “anger” in the subject index of psychoanalytic books prior to the last decade. [44]
We are thus presented with a remarkable series of facts. First, the proem to the Iliad signals to us that anger is central to the epic’s analysis of the forces that lead human beings to destroy each other. Second, we fall short of coming to terms with the complex Homeric diction relating to anger. Third, and finally, the social sciences, even as they pursue the ethnography and psychology of primary cultures, have also until recently neglected to observe that human aggression may be rooted psycho-emotionally and culturally in anger. [45]
In terms of method, then, we need three approaches. First, we must focus on anger itself as a semantic field in Homeric poetry. Second, we need to study closely the specific diction for anger in the medium of Homeric verse. And third, we need to set this research within the context of anthropological and social research into the formation and expression of emotions in human groups. The main methodological focus of this book depends upon the first two approaches, especially as they relate to the system of anger displayed in the two words kótos and khólos, in the course of which the third approach is drawn into service for a thorough study of the linguistic and poetic evidence. But how then should we approach that poetic evidence?
A good overview of the approach to diction taken here is presented by R. Sacks in the introduction to The Traditional Phrase in Homer (1987). Sacks’s emphasis, along with that of those he cites in the field (including Frame, Lowenstam, Muellner, Nagy, Shannon, Sinos, Slatkin, Watkins, and others), is to discover the “depth” of traditional phraseology through “the method of collecting all attestations and looking for a pattern or patterns, contextual and otherwise.” [46] I find it comfortable to link the present endeavor to Sacks’s declaration in his work that “the central assumption … is … that we should still be searching out those patterns that, with all their complexities, are actually displayed by Homeric phrases.” [47] I add only that by using poetic analysis coupled with a look at the comparative material displayed through anthropology, historical linguistics, and other approaches beyond the purview of ordinary Greek language study, we have acquired a powerful tool for the analysis and interpretation of the Greek text.
Where has that kind of work been done with respect to anger in Homer? It should not surprise us that the most rigorous examination of anger has been produced in the study of mênis, especially since that word initiates Homeric discourse. I introduce the material on mênis here because in the study of this important word, method has made all the difference for our coming to terms with anger in Homer. {11|12}
That such approaches can have powerful results is clear from the work of Leonard Muellner and Calvert Watkins, [48] who have both shown that mênis can be illuminated by the rigorous study of Homeric diction. Indeed these studies demonstrate that mênis is a taboo word, never used by speakers in reference to their own anger. [49] Mênis is part of a systematic cultural taboo whose roots can be found in the violations that lead to it. [50] That mênis has as its source the productive root men- is certain, although the process of formation has posed problems, reviewed in detail by Muellner (1992 and 1996, with Watkins 1977a). I am encouraged by Muellner’s success in establishing the taboo meaning of mênis through a rigorous approach to Homeric diction. Among Muellner’s findings (Muellner 1996), I single out as relevant to my work in this book the following conclusions about mênis as a word for anger in Homeric narrative: first, mênis is avoided in contexts when one speaks of one’s own anger; second, in the Iliad the noun mênis refers exclusively among mortals to Achilles; and third, there is a complex sociolinguistic structure that houses this term. [51]
Given the success of this method as applied to mênis, the other words in the lexicon of Homeric anger should receive the same rigorous attention that has been given to the first word of the Iliad. In Feuding Words and Fighting Words, then, I focus on two of those words for the following two reasons: first, in order to expand our understanding of the range of anger terminology in early Greek epic; and second, because a specific passage in Book 1 of the Iliad singles these two words out for special consideration.
In this book, my approach, then, is to gather and discuss all the examples of two terms (khólos and kótos), in order to examine them with reference to Homeric diction. In the process of doing this, the anthropological, linguistic, and social topics that arise are to be treated insofar as they are helpful toward understanding Homeric narrative. Note that even as I adhere to the principle of collecting all the examples, the collection is ultimately grounded in an observation from within the culture. Rather than producing an abundance of linguistic data that would be true to lexical completeness while capsizing the central goal of coming to terms with the productive use of anger in this poetry, I ground my arguments in the words of the prophet Calchas in Book 1 of the Iliad. I emphasize that the group of words selected for study is not arbitrary but can be said to stem as a group from an interpretation in the text.
Now it is time to introduce the passage that will be often in the foreground and always in the background of this book. In this Introduction, the passage highlights the intricacy of Homeric diction for anger. The final four verses of this passage provide the original stimulus for this book.
Near the opening of the Iliad, the Achaean prophet, Calchas, requests that Achilles protect him from Agamemnon’s wrath once the source of the plague is revealed:
ὦ Ἀχιλεῦ, κέλεαί με, Διὶ φίλε, μυθήσασθαι
μῆνιν Ἀπόλλωνος ἑκατηβελέταο ἄνακτος. {12|13}
τοιγὰρ ἐγὼν ἐρέω· σὺ δὲ σύνθεο καί μοι ὄμοσσον
ἦ μέν μοι πρόφρων ἔπεσιν καὶ χερσίν ἀρήξειν·
ἦ γὰρ ὀίομαι ἄνδρα χολωσέμεν, ὃς μέγα πάντων
Ἀργείων κρατέει καί οἱ πείθονται Ἀχαιοί.
κρείσσων γὰρ βασιλεὺς, ὅτε χώσεται ἀνδρὶ χέρηι·
εἴ περ γάρ τε χόλον γε καὶ αὐτῆμαρ καταπέψη,
ἀλλά τε καὶ μετόπισθεν ἔχει κότον, ὄφρα τελέσση,
ἐν στήθεσσιν ἑοῖσι· σὺ δὲ φράσαι εἴ με σαώσεις.
(Il. 1.74-83)
O Achilles, beloved of Zeus, you order me to speak the anger of Apollo, king, far-shooter. So, I will tell you, but promise me and swear that, yes, you will truly help me in word and deed. Now, I think I will enrage a man who has mighty rule over all the Argives, and the Achaeans obey him. For a king is stronger when he comes to be angry with a lesser man; for, if in fact he can swallow his khólos down on the very same day, he can later keep kótos in his heart until he brings it to its télos . You must take thought of protecting me.
In line 75 the text shows that mênis signifies the anger of Apollo represented by the plague that ravages the Achaean camp. Quite rightly, Calchas worries that his revelation of the source of Apollo’s anger will lead to a mortal’s anger (ándra kholōsémen, Il. 1.78). This observation may be true of anyone, but Calchas is concerned about the political power this man wields: hòs méga pántōn / Argeíōn kratéei kaí hoi peíthontai Akhaioí “Who has mighty rule over all the Argives, and the Achaeans obey him,” Il. 1.78-79. Next, Calchas introduces yet another word for anger (khṓsetai).
Finally, Calchas opposes one kind of anger to another in a systematic fashion, that is, in a way that implies some regular difference between the two: khólos is immediate (autêmar “on this very day,” Il. 1.81) and is related metaphorically with the physiological process of digestion (katapépsēi “swallows down, digests,” Il. 1.81). In addition, khólos is surpassed in its consequences by the last kind of anger here cataloged, kótos; for kótos has consequences that reach out far beyond the present (metópisthen “in the future,” Il. 1.82), and kótos may, in the course of things, reach some end to which it by nature tends (óphra teléssēi “until he brings it to pass,” Il. 1.82).
Calchas’s own words indicate that there is an opposition between khólos and kótos. Grammatically, K. Ameis and C. Hentze point to the chiastic pairing of khólos and kótos, helped by the placement of the particle ge after khólon in Il. 1.81. [52] The arrangement formed by Calchas clearly plays on a set of oppositions between khólos and kótos in a way that needs clarification. First, how can we say that khólos may last only the day, while kótos has a teleology of its own? How active is the metaphor in katapépsēi? How consistent is any of these propositions with the rest of Homeric language?
Moreover, what about their formal features? Of all the words for anger these are the two that are closest in form to one another. They both begin with a velar {13|14} and belong to the same noun-paradigm; each has two syllables and could take part in similar alliteration and assonance patterns. [53] Thus, on purely formal grounds, these two words for anger may be studied together as part of the sensitive workings of Homeric formular diction. Yet only three times in the manuscript tradition do these two words serve as variants for each other (Il. 14.111, Il. 24.584, and Od. 13.342).
I will be looking closely at the issues that arise from my selecting these two words for anger as exemplary of Homeric culture’s construction of anger. To close this Introduction, however, let us consider a feature of Calchas’s speech that might independently indicate to us the speaker’s value for the study of Homeric language: Calchas presents us with a folk definition of anger terms.
This latter point is important enough to elaborate further, at least briefly. In reference to the theoretical value of such kinds of folk speech, consider the following two statements by a folklorist as he considers the force of folk definitions:
Naive native speakers’ definitions represent a basic source of such [empirical] data. Their value lies in the fact that folk definitions express not merely the meanings of individual lexical items but also underlying semantic relationships between the defined term and the defining term(s). [54]
And again:
To understand the significance of folk definitions for the study of semantics it is necessary to recognize two fundamental facts about defining. First, definitions are not in any way an artifact of linguistic analysis or of formal lexicography. It seems likely … that “defining is a cultural universal.” … Second, all definitions reflect one basic law of natural language: a natural language serves as its own metalanguage … Folk definitions, then, offer more than just recurrent patterns in the definitions of individual terms which may point to possible semantic components. They are expressions of significant relationships among lexical items which form the basis for native speakers’ understandings of the meanings of those items. [55]
Seen in the light of Manes’s observations, Calchas’s statements constitute a folk definition of two Homeric terms for anger. [56] Thus, let this study in the semantics of anger begin with the definition of khólos and kótos provided us by Calchas, the prophet of the Achaeans. Indeed, we can prepare for our encounter with the basic Homeric concept by quoting Odysseus’s fillip to his friends:
τλῆτε, φίλοι, καὶ μείνατ’ ἐπὶ χρόνον, ὄφρα δαῶμεν
ἢ ἐτεὸν Κάλχας μαντεύεται ἦε καὶ οὐκί.
(Il. 2.299-300)
Take heart, comrades, and wait awhile, until we determine whether Calchas prophesies the truth, or not. {14|15}


[ back ] 1. The titles in the list refer respectively to: Clay 1983; Collins 1987; Heitsch 1967; Irmscher 1950; Woodhouse 1930, Chapter 4; Lidov 1977; Petersen 1920; and Slatkin 1986. And see now The Anger of Achilles (Muellner 1996). For a typical analytic statement, see Leaf 1886, I.xxiii: “The original poem, the work of ‘Homer’ himself was the Μῆνις Ἀχιλλέως …”
[ back ] 2. Whitman 1958, 182.
[ back ] 3. On synonyms in Homer see Paraskevaides 1984 and Martin 1989, 14 (“When we do pay attention to context, synonymity recedes”). For the question of functional synonymity and the formula, see Bakker and van den Houten 1992.
[ back ] 4. Using Irmscher’s list, see note 11.
[ back ] 5. For the significance of the opening word in prooemial style, see van Groningen 1946, 284-87 and Kahane 1992 and 1994 (for mênis, see 52-58).
[ back ] 6. “Zorn als Ausfluss des natürlichen Wesens des Menschen bezeichnen, als eine starke Regung oder Offenbarung des Gemütes oder der Seele” (Schmidt 1879, 551).
[ back ] 7. “Den Kern der dritten Gruppe machen zwei Substantive und zwei Verben aus, welche den dauernd in der Seele wonenden Groll bezeichnen” (Schmidt 1879, 551).
[ back ] 8. Schmidt 1879, 551.
[ back ] 9. See, for example, Od. 8.227 with Od. 8.238. So, too, Calchas refers to khólos just after using the phrase basileùs . . . khṓsetai (Il. 1.80-81) “A king … will be angry.”
[ back ] 10. mênis, khólos, khōomai, kótos, odússomai, mênos, ékhthos, miséō, alastéō, skúzomai, ágamai, némesis, and khalepaínō.
[ back ] 11. For further problems with Irmscher’s work (especially its lack of statistical information), see Considine 1966, 15-16. Considine acknowledges that both mênis and khólos receive “rather unsatisfactory treatment” in Irmscher.
[ back ] 12. Most disturbing is the absence of analysis of poetic context. Without contextual analysis firmly rooted in the study of traditional diction, it would be nearly impossible to uncover the relationship of, say, khólos to khṓomai.
[ back ] 13. Irmscher 1950, 81-86. On the relation of Zeus Xenios to anger see Watkins 1987b, 298-99, Muellner 1992, 130-31 and 1996, 37.
[ back ] 14. Again, see Martin 1989, 12-14, for articulation of this method.
[ back ] 15. Considine 1969, 139-40.
[ back ] 16. Considine’s recent work on mênis has helped advance a more rigorous study of this term (Considine 1985, 1986). Cf. Muellner 1992 and now Muellner 1996, 177-89.
[ back ] 17. Adkins 1969, 7-20.
[ back ] 18. Adkins 1969, 15.
[ back ] 19. Adkins 1969, 17.
[ back ] 20. Voegelin 1956, 89.
[ back ] 21. In addition to Harris 2001, see also an excellent brief review of the state of this research in Laurenti and Indelli 1988, 1-8; for the relationship of the philosophical to the literary implications of anger see Anderson 1964, especially 149-73, with Galinsky 1988 for the later literature. As this book went to press, appeared Braund and Most 2003.
[ back ] 22. Muellner 1976, 15.
[ back ] 23. For a brief introduction to the anthropology of anger see Tavris 1989, 48-69; see also, especially for a review of the debate between the physicalists and social constructivists, Myers 1988; Hochschild (1983, 201-22) succinctly reviews the study of emotion “from Darwin to Goffman”; a collection of classic essays on the topic of the emotions can be found in Calhoun and Solomon 1984. In the social sciences, a valuable reference for research on the emotions is the collection of Lewis and Haviland (1993; on anger, especially 537-46). The classic study in the social sciences is that of Averill (1982, 1983). Lutz (1988) reviews the anthropology of emotions in general.
[ back ] 24. Lutz 1988, 157.
[ back ] 25. For the Ilongot, see also M. Rosaldo 1984, and her discussion of the relationship between feuding, raiding, and beheading.
[ back ] 26. M. Rosaldo 1980, 45.
[ back ] 27. As informed by anthropology, classical studies have blossomed of late, but the field of anthropology is too vast for me merely to cite it passing. Yet some works provide methodological perspectives that have informed this book: besides the important works of Detienne (see especially 1981, Chapter 6, for an analysis of the interaction of 20th century anthropology and the Greeks), Vernant, and Burkert, let me cite works that consciously take on the relationship between the literary text and anthropology: Kurke 1991, 7-12 and 88-96; Nagy 1979, 1990a and b, 1992; and Martin 1989. As to social life and anthropological modeling for the ancient world see D. Cohen 1991, where he applies a model derived from Mediterranean anthropology to the moral life of Athens (see Chapters 1 and 2 for the theory and method); see too Humphreys 1978 and Zeitlin’s overview of Vernant’s work (in Vernant 1991b, 3-24).
[ back ] 28. Sacks 1987, Introduction, 1-30.
[ back ] 29. Sacks 1987, 17.
[ back ] 30. Sacks 1987, 23.
[ back ] 31. Sacks 1987, 19. For problems related specifically to the relationship of field worker to informant see Clifford 1988, 48-50 and 72-75. Though we should not romanticize the “informant” as a source, to have access to no living informed member of a culture threatens the success of any anthropological inquiry. Just such a situation imperils a project on Homeric discourse, since we deal without the perspective provided by dialogue with a competent member of the culture. Yet the limitations thus imposed on work in early Greece parallel the problems newly discovered with respect to the dialogue between an outsider anthropologist and an insider informant. Anthropology’s maturation—its realization that its very hallmark, the relationship between outsider and insider, is loaded with difficulties—emboldens those of us who work in pre-ethnographic cultures to attempt to reconstruct, even if imperfectly, the relationship that mirrors that of the dialogue between an informant and a researcher. For anthropology’s cognizance of the pitfalls faced by the researcher in approaching any culture, the “crisis in ethnographic authority,” see Clifford 1988 and the collection of essays in Clifford and Marcus 1986. Cf. Finley 1986, 118.
[ back ] 32. For the importance of the expansion aesthetic see Martin 1989, 206-30.
[ back ] 33. See again M. Rosaldo 1980, 45.
[ back ] 34. M. Sahlins 1985, 125. See Dougherty and Kurke 1993, 5-6, for an application of Sahlins’ work to cultural poetics.
[ back ] 35. Besides Sahlins 1985, see also Sahlins 1976, 1981.
[ back ] 36. M. Rosaldo 1980, 21.
[ back ] 37. M. Rosaldo 1980, 22.
[ back ] 38. On this kind of anger, see Considine 1969.
[ back ] 39. E.g., Cunliffe 1963, LSJ, LfrGE, s.vv.
[ back ] 40. There have been exceptions to this general neglect, as for example in the study of éris (“strife”) by Hogan 1981, and Nagler 1988 and 1992; see also Dumézil 1980 on ménos (“anger”) and lússa (“rage”).
[ back ] 41. That khólos should be neglected because it is common fits in nicely with M. Rosaldo’s remarks. For khólos as the unmarked term for anger, see Considine 1985, 162 n. 2, “In Homer the unmarked word for anger is khólos, in Classical Greek orgḗ and in later and Modern Greek thumós”; cf. Considine 1986, 53. See also Turpin 1988, 258 (“dénomination non-marquée de la colère”). Curiously Considine 1986 omits kótos in his review of Homeric anger terms.
[ back ] 42. See, for example, Lakoff and Kövecses 1987 and Lutz (in Holland and Quinn 1987), who covers the ethnographic importance of anger and its role in a community’s organization of its social life. See also M. Rosaldo 1980, and R. Rosaldo 1984; on the meaning of emotions as part of a culture’s semiotic practice, see Myers 1988 and 1991. A variety of critical work continues to be produced in this area (see, e.g., Heider 1991). Anthropological work in the area of emotion and aggression needs to be attended to by the Homerist, who can also make a contribution to ethnographic projects from the evidence provided by our archaic Greek texts.
[ back ] 43. In Novaco 1986, 3.
[ back ] 44. Add here Tavris’s complaints (1989, 39). For more on the slighting of emotion in the sociological literature, see Hochschild 1983, 201-4. The literature on aggression has begun to include extensive study on anger’s role as a social problem; see Geen 1990, 115-25. In the context of world mythology, note that Aho 1981 does not single out anger as a specific cause of violence in the sacral mythology of war. See also Lemerise and Dodge 1993, 537.
[ back ] 45. Two psychoanalytically informed studies of Homer consider the psychological questions relating to Achilles, while leaving aside the question of anger (MacCary 1982 and Holway 1989). For psychoanalysis and anger, see Novaco 1986, 2-6.
[ back ] 46. Sacks 1987, 17. See also 1-3.
[ back ] 47. Sacks 1987, 13.
[ back ] 48. Muellner 1992, 1996; Watkins 1977a.
[ back ] 49. Muellner’s book on mênis (1996) is a watershed in the study of anger in Homer, both for providing rigorous method with sophisticated readings of crucial passages, and for engaging the necessary comparative material. A major thrust of this work is to apply a social and linguistic category (taboo) in the course of analyzing the traditional diction that is Homeric usage. Important questions regarding the nature of such a taboo are raised in Turpin 1988; see also Muellner 1992; Muellner’s work details the nature of this constraint (1996, 191-94). See also Nagy 1979, 74 and Nagy 1990a, 227 and 238.
[ back ] 50. Watkins 1977a; Muellner 1992. For more on mênis and khólos see chapters 2 and 3. Also see Considine 1966, 16-22; as well as 1985 and 1986; Nagy 1979, 15-18; 1990, 227, 238, and 263; Turpin 1988. Janko 1992, 106 questions the hypothesis that mênis refers to divine wrath; see also Kahane 1994, 50-58. While providing the definitive study of mênis, Muellner (1996) analyzes the way the taboo mechanism functions in the Homeric poems.
[ back ] 51. Schwyzer 1931, rejected in Schwyzer 1939, 260; see now Muellner 1996, 18689, where the appendix (“The Etymology of Mênis” 177-94) reviews the comparative evidence and comes to the conclusion that “Watkins’s revival of Schwyzer’s etymology appears to be the most consistent with the word’s meaning and usage” (194).
[ back ] 52. Ameis and Hentze 1894 at Il. 1.81 for a succinct and telling note on the style of this passage.
[ back ] 53. On sound-patterns in Homer see Shewan 1925, and on word-play in Homer see Louden 1995; note that Louden (30) catches a pun involving khólos (with khōlós “lame”) at Od. 8.304-8.
[ back ] 54. Manes 1980, 122.
[ back ] 55. Manes 1980, 123.
[ back ] 56. Compare Nestor’s definition of mêtis “cunning intelligence” (Il. 23.313-18), as discussed by Detienne and Vernant 1974, 11-23; also Frame 1978, 85; and note the contrast with nóos at line 590. See too Menelaus’s observations on kóros “satiety” in Il. 13.636-39. Such passages can profitably be approached with Manes’s notion of the folk definition.