Homeric Anger Revisited

1. In the introductory article to a collective volume entitled Ancient Anger: Perspectives from Homer to Galen, D. L. Cairns takes my work on the Homeric word mênis as a case in point for his view that classicists who study emotions “more often that not” shirk their responsibility both to “become familiar with current research on the nature of emotion in other disciplines” and to “situate the study of ancient terminology in a wider context in which full account is taken of aspects of emotion other than the sense and reference of particular linguistic markers.” [1] In particular, he claims that my work is typical of the “cultural determinists who currently dominate both classical studies and the humanities in general.” As his own analysis of a host of Homeric anger terms proceeds, he repeatedly uses my work as a foil for his own, claiming that I have a compelling need to show that the meaning of the Homeric anger word mênis is culturally determined and therefore not universal. According to him, this supposed need on my part has skewed my analysis, an analysis whose results he rejects and, to all appearances, replaces with his own, one based on (though not limited to) a universal definition of anger that is biologically determined. I am interested in engaging with and responding to his criticisms of my own work, but his critique raises larger questions for classicists that deserve response, among them, what constitutes an effective application of the work of other disciplines for classics, and what the nature of scholarly dialogue is and could be among classics scholars.
2. Before I do that, however, I think it appropriate that I provide the reader with a brief statement on the method and results of my research on mênis. In the course of my work, I examined every single attestation of the word mênis along with its verbal and adjectival derivatives in the Homeric corpus, each one of them in its proper context. By context I do not mean the two or three lines that precede or follow the given word, but in depth, each occurrence in its precise place in the unfolding mythical narrative in which it is embedded. It became clear to me, and I sought to demonstrate, that this word is a precisely limited word for anger: it is aroused by consistent, highly classifiable actions, it is expressed in specific, repeated ways, and it is appeased, if at all, also in highly specific ways. In all these contexts (cause, expression, appeasement) the poetic diction surrounding mênis is also highly consistent, to the extent that the word functions in Greek epic as the name of a “theme” in the technical sense defined for that term by Albert Lord. In describing the training and practice of singers who can compose and simultaneously perform the epic songs of the South Slavic tradition, Lord specifies the constituents of their system for generating poetry as consisting of formulae (which are both groups of words and syntactic patterns) assembled into large constellations or themes, which are then combined into still larger narrative patterns so as to become epic songs. [2] I define mênis as “the nomen sacrum for the ultimate sanction that enforces the world-defining prohibitions, the tabus that are basic to the establishment and perpetuation of the world of Zeus and the society of mortals he presides over.” [3] As the first word of the Homeric Iliad, I treat it as both a term for a specific kind of anger and also as a pointer to a whole constellation of mythical ideas and expressions about the world that regularly occur with it and are evoked by its mere mention. That is why, in seeking to define this special term fully, my work examines the system of myths for world maintenance in the Hesiodic Theogony as well as the specific deployments of the mênis theme in the Iliad and the Odyssey, which are conceptually and thematically congruent with the theogonic text.
3. At the beginning of his work, Cairns provides a discussion of essential contributions to the study of emotions from other disciplines, but it is a discussion with an agenda. Since he is out to oppose what he believes is the dominant paradigm in classics and the humanities in general, his survey is intended to show that “cultural determinists” are incorrect. Here is the way he begins his presentation:

Their [=cultural determinists’] approach assumes that all significant features of a culture are products of conditions specific to that culture; and since this holds both of the culture under investigation and of the investigator’s own, it requires the investigator somehow to seek access to phenomena wholly specific to a society that is alien to his/her own experience. Thankfully, however, we do not have to strive for such an impossible goal: all cultures exhibit points of overlap and contact that make them mutually intelligible; and all cultures have been formed and inhabited by a species which evolved its capacity for the creation of culture under a broadly uniform set of environmental pressures. The possibility for dialogue between ancient cultures and our own exists; we must attempt it in a way which avoids both the naïve assumption of shared humanity and unsustainable strategies of alienation. (Cairns 2003:11–12; the highlighting of “all significant features” is mine, not his)
3a. In this statement, Cairns portrays the view that he opposes in extreme terms in order to present his own position as a simple, obvious, and reasonable option. So he places so-called “cultural determinists” in a logical box from which they cannot emerge. The position he imputes to these adversaries of his is that all significant cultural phenomena are wholly culture-specific and that there are therefore no shared features between any cultures. In this scenario, he maintains, cross-cultural study is “an impossible goal.” Having evoked this untenable extreme, he speaks of my work as an instance of it, as follows:

A case in point is Muellner (1996), who claims without argument that “there is no reason to assume that the metaphors, rules, and therefore the emotions that they represent and that we tend to experience as inherent in human nature are actually universal” (p. 1), and argues that Homeric mēnis is “a far cry from any shared, secular notion of anger specific [sic] to contemporary Western culture” (p. 133) because “[f]or us emotions are primarily individualized and internal, and their social dimensions are semantically secondary” (p. 138).

By putting together into one sentence of his fragments of three sentences of mine that occur on pages 1, 133, and 138 of my work, by lifting those fragments out of their contexts, and by inserting a “because” between two fragments that are actually five pages apart, Cairns has cast me as his foil, the “cultural determinist.”

4. In actuality, it would be difficult to find anyone who believes that all the significant phenomena in every culture are specific to that culture and therefore shared with no other culture, nor was the statement that he quotes from my work uttered in support of any such notion. In fact, the first sentence of mine that Cairns quotes actually begins as follows: “One society may share some of its elaborate metaphors and moral rules with other societies, but there is no reason to assume, etc.” [4] So the sharing of cultural phenomena about emotion across cultures was explicitly evoked in my presentation, and an absolute view of the differences between cultures was neither taken nor implied. Cairns simply omitted the beginning of my sentence from his excerpt.
5. Furthermore, there is a problem with the logical box that Cairns tries to construct around what he claims is the cultural determinists’ view of culture-specific phenomena (in his words, cited above: “it requires the investigator somehow to seek access to phenomena wholly specific to a society that is alien to his/her own experience…an impossible goal”). It is not necessary that the phenomena of another culture be the same or overlap with those of one’s own culture in order for any of them to be comprehensible. If it were, an observer’s understanding of other societies would be limited to those phenomena which resemble her/his own, which is patently false. Human languages are open-ended systems. Although they arise to function in the context of a specific culture, languages are capable of articulating and making it possible to conceive of new phenomena, such as the worldviews or specific phenomena of other cultures however different they may be from the observer’s. [5] Certainly there is a difference between being born into a society and being introduced to it from outside, and there are also societies that either happen to be closed systems through isolation or that try to enforce their impenetrability to others. But language can break such barriers—ever more so when we are dealing with cultures and languages like those of Ancient Greece and Rome, civilizations that were imperialistic and sought to “universalize” (in their terms) and export their worldview. So we can try to understand ancient cultures that have left us verbal as well as physical monuments along with visual representations, although we have no native informants to interrogate by means of any language. We can observe cultural phenomena that may not be our own and at least try to rebuild their systems as much as possible from within their own system, especially by way of our ever-evolving understanding of their languages, to include an articulation of their way of thinking into our own language and thought. Unless we make the effort to rebuild other cultural systems, we easily and unconsciously impose our own systems on them. Consequently, in studying other cultures, ancient or otherwise, it certainly makes sense to begin by assuming nothing at all about their similarity to ours, in order to enhance our ability to see as clearly as possible both what is similar as well as what is different. [6]
6. There is even a further point to make about understanding universal features across the boundaries between cultures: I used the word “system” to describe the way cultural phenomena are organized, which means that even though there may be common aspects to vastly different cultures, there is no guarantee that a feature that appears to be universal when looked at outside of the context of its culture actually functions in the same way within different cultural systems. The question is best exemplified by the way that Ferdinand de Saussure first defined sign systems, with a clarity and simplicity that made the modern discipline of linguistics possible: signs consists of signifiers and what they signify. [7] For instance, physically speaking there are only so many sounds that the human vocal apparatus can make and combine into a language. But even though two languages may have similar phonological systems, there are no guarantees that the same sequence of sounds in those languages (related or unrelated languages) will have the same meaning within the system that a given language constitutes, because sounds are signifiers, not signs in themselves. The key question to ask about the cultural “universals” that Cairns posits is at what level within the cultural system do they exist? If they function within a system of signs, are they signifiers, or are they signifieds, and if they are signifiers, how do they function within the system of signs? For within a given system of signs, there again lies the possibility of variation and differentiation between cultures that can be articulated and understood across cultural boundaries.
7. To return to Cairns’ criticism of my work, the argument that he claims is missing from my statements (see above, his words: “without argument”) is simply one that he has unaccountably not noticed. Its starting point on the first page of my book, in a chapter entitled “Approaching Anger,” was a reference to a study done in 1987 by the linguists George Lakoff and Zoltán Kövecses called “The Cognitive Model of Anger Inherent in English.” [8] As its title implies, this work displays in precise and telling detail the elaborate metaphors, metonymies, and implicit social and moral rules about anger that are a persistent feature of the English language and that native speakers acquire with the language whether they want to or not; it goes further than that, though. It organizes the metaphors and metonymies of English into a “cognitive model,” in other words, what I would call a system, and it suggests that this cognitive model may well be universal, pending further research. Using the work of Lakoff and Kövecses, I concluded from their demonstration and from my (limited, to be sure) knowledge of the conventional metaphors and metonymies and moral rules for emotions in several languages, that it is decidedly unreasonable to assume that every language in every culture has and constitutes in the same way the same elaborate and specific set of metaphors, metonymies, and social/moral rules about emotions as every other. [9] I did not rule out the possibility that emotions might be the same, only that “there was no reason to assume” that they were the same. This is not a quibble. My purpose at that point in my work was not to disprove the universality of any and all emotions, as Cairns claims. It was to formulate and justify a methodological starting point and a discovery procedure for a research project focused on a single term for anger, the first word of the Homeric Iliad. So what Cairns has done in his portrayal of my work is to take my words out of context and then mistake a methodological discussion for an ideological one. To be sure, ideologies and theories of culture are not unrelated to methodologies and vice versa, but in this case, the only thing resembling an ideology on my part was skepticism about taking putatively shared features of human emotions as a point of departure, as a starting assumption in approaching the study of a loaded emotional term from another culture. [10]
8. In his study, by contrast, Cairns has an agenda, which is to promote the opposite strategy, to take as his starting point for a discussion of the Homeric terms for anger what he claims is not only the universal definition of anger but the one which is primary in evolutionary biological terms, the prototypical, built-in notion of anger. How has he arrived at this prototypical definition of anger? Not, it turns out, on the basis of comparative historical linguistics, or of a collection of anthropological data about anger across a wide variety of cultures, or of ethological study of primates evolutionarily close to humans, but on the basis of “the emerging discipline of evolutionary psychology.” [11]
9a. Before he takes that step, however, he starts with the premises that he attributes to this emerging field. For Cairns, emotions are both cognitive (in other words, they include cognition, however that is defined) and, more importantly for him, physical, because they include “neurophysiological and visceral changes, and these, since they depend on the evolved organic natures of the species, cannot be entirely given by culture.” [12] He cites Darwin’s last work (1872), The Expression of Emotions in Man and Animals, on the universality of facial expressions across cultural boundaries, and links it to the modern research of Paul Ekman, who produced an edition of that work (London, 1998). Ekman seeks to show how Darwin’s ideas can be supported with clinical studies based on protocols that he (Ekman) developed for specifying and identifying emotions and facial expressions on the basis of Darwin’s own intuitions and experiments. Ekman (after Darwin) reduced the field of study to seven emotions and the facial expressions that accompany them: anger (of two types, open- and close-mouthed), [13] fear, disgust, surprise, happiness, sadness, and contempt. For these seven emotions, the hypothesis is that, as Cairns puts it, “Individuals in one society generally have no difficulty in correctly interpreting the facial expressions of members of another.” [14]
9b. As Cairns concedes, however, the fact that the facial expression of these seven emotions is mutually understood leaves a lot of latitude for cultural differences:

[T]his leaves all the many aspects of emotion or emotional scenarios which can vary from culture to culture: the eliciting conditions (if disgust and its facial expression are universal, what is occasion for disgust in one society is clearly not in all); the valorization of different emotions (cultural differences in the valorization of anger are a pertinent consideration here); the forms of expression which are and are not culturally approved for different categories of individual…; the imagery of emotion; and, of course, emotion-words. These aspects are not negligible. But this enumeration of universal and variable factors would be misleading if it suggested a sharp distinction between the two categories: for (on the one hand) all aspects of emotion are subject to construction through their incorporation in the complex syndromes of factors that constitute emotions. The cognitive-evaluative aspect of emotions as responses to states of affairs in the world will always presuppose an element of constructionism—the typical scenarios in which emotions are deployed and to which emotion-words are attached must be understood within the practices of a particular culture, and thus all aspects of the total emotional experience will be subject to construction and interpretation in terms which may be specific to a specific culture. On the other hand, there are aspects of the cultural construction of emotion that will be firmly rooted in our nature as humans: as George Lakoff and his collaborators have shown, the cultural models and folk theories that structure a culture’s emotional life and the metaphors and metonymies that structure our emotion-concepts derive to a very large extent from our experiences as physically embodied human beings in the world. Thus there will remain fundamental and substantial aspects of any genuine emotional scenario…which are not determined by culture but rather depend on biological factors; and “biology vs. culture” is a false antithesis when the biological must be experienced and constructed in a cultural context and when shared cultural categories draw on our nature as a physically embodied, social species. [15]

By the words “constructed / construction / constructionism,” Cairns is designating phenomena that are subject to local determination by a specific culture. After saying that all aspects of “an emotional scenario” are “subject to construction and interpretation in terms which may be specific to a specific culture,” he then states that “fundamental and substantial” aspects of emotions are “not determined by culture but rather depend on biological factors,” whereupon he states that the antithesis (which he has just made) between biological and cultural aspects is false because each depends on the other. I appreciate that this formulation represents an effort to describe a complex situation in complex terms and to avoid oversimplification, but it also makes any arguing against his view difficult at best, since he at least appears to have taken every possible position and its opposite. At a minimum, Cairns’ argument against a hard and fast opposition between biological and cultural determining factors in human emotions should have mitigated the opposition that he exhibits to those whom he considers “cultural determinists” in the study of emotion, since he himself validates the possibility that “all aspects” of emotion can be culture-specific. A fortiori, even if his notion of cultural determinism is not the same as the notion of it held by those whom he argues against, if all aspects of an emotion can be culture-specific, it would be completely valid as a research methodology to assume nothing about the biological factors that (as he believes) constrain a culture’s representation of an emotion. [16] Those factors become part of a complex cultural system and must be interpreted as cultural in any case (compare my remarks above about phonological systems).

9c. What remains unclear at this point from Cairns’ generalized and abstract treatment of the relationship between biological and cultural factors (given that a complete antithesis between them is false in the first place) is what precisely are the cross-culturally consistent, “substantial,” fundamental biological features that constrain the variety of possibilities for cultural determinism of emotion. Cairns cites the works of Lakoff and Kövecses (separately and together) to assert that because we are all “physically embodied human beings,” there is consistency in our emotions, and our emotions must thereby be constrained and to some extent universal. Lakoff and Kövecses speak of general categories of imagery (such as anger metaphors that feature bodily “containment” or “hot fluids”) for the expression of emotions and do not propose them to be determinative of emotions themselves. In fact, they understand their image categories as common conceptual constituents of social, cultural, and psychological constructs about emotions, not as tokens of universality in emotions themselves. [17] In any case, to say that being a physical human is a defining or restrictive or constraining factor for emotions is a truism. Is there any way to measure, count, or identify the number or characteristics of emotions that humans can have as opposed to those that they can’t? It looks as though Cairns has confused the signifiers of emotions, like metaphors and metonymies, which is what Lakoff and Kövecses are focused upon, with the things that they signify, the emotions themselves in their social and cultural contexts, despite Cairns’ obvious awareness of the “construction” that builds emotions in specific cultures in specific ways. Those signifiers may be relatively restricted by the biological consistency of our species (Cairns also speaks of environmental consistency, though I am not sure what he means, given the diversity of the biosphere), but what they signify is what we are after in trying to understand emotions across cultural boundaries. In other words, Cairns’ mitigated antithesis between biological and cultural factors in human emotions may also be the relationship between a signifier and what it signifies. From within a given system, as Saussure said, that relationship is totally determined and inevitable, but outside it, that same relationship is actually arbitrary. So there are contradictory things to be confused about, but there is also a way out of that confusion.
9d. Despite the claim made by Cairns following Ekman that there are only seven “basic” emotions—anger, fear, disgust, contempt, surprise, happiness, and sadness—such an inventory by no means represents the gamut of emotions that humans do have. [18] Furthermore, such a reductive, even impoverished view of human emotions takes no account of the variable intensity that they may have—for instance, just in the category of sadness, we can speak of a broad range, from disappointment to grief at the loss of a loved one, that may all be expressed by more or less the same facial expression or by no facial expression at all—or of the fact that humans often have complex, conflicting emotions at the same time, while our faces may be expressing only one of them, [19] or of the fact that there are emotions that have no consistently identifiable accompanying facial expressions at all, such as guilt or hope. And lastly and perhaps most importantly, just how many different emotions can lurk behind a human’s smile? The assumption of the experimental psychologist is that a smile signifies “joy” or “happiness,” but in actual human interactions as opposed to experimental contexts, there is no guarantee that a smile or even any of the other “basic” facial expressions really signals the supposedly unique and universal emotion that such research blithely assumes. What Darwin had in mind was a simplified starting point that demonstrates our animal ancestry (thus his title: The Expression of Emotion in Man and Animals), to point out what we share with primates and other mammals. Chimpanzees, for instance, and dogs, can express these seven emotions, even including disgust, with sounds and body language. [20]
9e. So does the putative universality of seven facial expressions prove that they signify only the emotions that they do in dogs and chimpanzees? I say “putative universality” because one recent test of Ekman’s work has cast doubt on the validity of its conclusions. In 2009, a peer-reviewed journal in this field, Current Biology, published a paper entitled “Cutural Confusions Show that Facial Expressions Are Not Universal.” [21] A team from the Universities of Glasgow and Montreal using advanced techniques of measuring and analyzing eye movements on two groups of experimental subjects, East Asian and “Western,” showed that “Eastern observers use a culture-specific decoding strategy that is inadequate to reliably distinguish universal facial expressions [“universal” in terms of Ekman’s protocol] of ‘fear’ and ‘disgust.’” [22] This different decoding strategy features fixating upon the eyes rather than “distributing their fixations evenly across the face as Westerners do.” The report concludes “Our results question the universality of human facial expressions of emotion, highlighting their true complexity, with critical consequences for cross-cultural communication and globalization.”
9f. Even without the doubt this last work casts on Ekman’s conclusions, an outside observer may be entitled to have serious concerns about the applicability of Darwin’s research perspectives to a cultural representation as sophisticated and complex as that of Homeric poetry and its wide range of terms for anger. It seems clear that facial expressions are signifiers in a semiotic subsystem for the expression of emotions that are admittedly less arbitrary (though how much less is unclear) than the sounds that constitute the signifiers in human languages, but they are also much less well-defined and subject to even more slippage than the words that normally accompany facial expressions. Taking facial expressions out of the verbal and gestural contexts as well as the social contexts in which they actually function (if that is possible) and proving that they can be decoded as expressing universal emotions may simplify the task of experimenters, but it certainly looks like an experiment that is set up to discover only what it wishes to find in the first place (the most basic of emotions signified by the most basic of signs), and it may not give a true account of the functions that facial expressions can have in social interactions or of the more complex, higher grammar of sign combinations to which they actually belong. Such research inspires one to ponder Roman Jakobson’s example of the forty different ways of saying the two words segodnja večerom (“this evening”) that was required of an actor auditioning for Stanislavskij at the Moscow Art Theater. The auditioner’s audience had to be able to recognize the forty distinct emotional settings of each one. One particular actor, in redoing the experiment, actually extended the number to fifty, and most of them “were correctly and circumstantially decoded by Muscovite listeners.” [23]
10a. Cairns speaks of the work of Darwin, Ekman, Lakoff, and Kövecses to justify the turn that his argument takes towards evolutionary psychology. But before taking that turn, he weighs in on another methodological topic, the limits of classical philology as a discipline for understanding topics like emotion. “[S]tudies of emotional terminology will always be necessary…but they will never be sufficient,” and he continues as follows:

We do need to start with the terms that the Greeks themselves used to describe their emotional experience, and we should certainly not foist our own preconceptions about particular emotions on to our data; but philological study must not stop with the investigation of the sense and reference of particular linguistic labels, and the resources of the philologist, even of the classicist in general, will not suffice for the study of ancient emotion: for the right methodological principles, for the proper interpretative categories, we also need to look to the insights of other disciplines; and we need to supplement our studies of the semantics of ancient terms with such data as we can recover about other aspects (both linguistic and non-linguistic) of emotional experience. [24]
10b. Just like his earlier remarks about the broad areas for cultural construction of emotion and the need to be aware that all aspects of an emotion may be culturally constructed, there is much here that is sensible. Although anyone would agree that classicists should “not foist our preconceptions” on data about the ancient world, and that we should “supplement our study of the semantics of ancient terms” with data about non-linguistic aspects of emotion, the idea that we need to look to other disciplines for “the right” methodology or “the proper” categories for interpretation is less straightforwardly sensible. Classics is an interdisciplinary field that consists of the study of all aspects of the ancient Greek and Roman worlds, and classicists are trained in a range of disciplines, from history and literary studies to philology, from archaeology to linguistics to philosophy. However, it is not unfair to say that classics as a field has historically been conservative and resistant to the inclusion of insights from newer fields that have much to contribute to the intellectual goals and the vitality of research in classics. On the other hand, saying that “the right” methodology or that “the proper” categories for interpretation are to be gleaned from newer disciplines is of a piece with the rigid and conservative thinking that excludes the insights of such disciplines in the first place. The practice of scholarship, and especially of classical scholarship given its long history, is heuristic, and we are standing on the shoulders of generations before us as others will, hopefully, stand upon ours. So, on the one hand, scholars in classics or any other field should be judicious, open-minded, and objective in the way that they study, present, and, incorporate new disciplines into their work, and, on the other hand, no scholars should feel obliged to destroy the work of their predecessors in order to build a new understanding upon the ruins of that work on the assumption that these predecessors were using “wrong” methodologies or “improper” categories for interpretation. In the case of Cairns, his expression of such an assumption is not just a matter of nuance. His statements here are sending mixed signals in a long-standing, destructive conflict whose benign resolution would be universally beneficial.
11a. After all his broad, critical, and generally welcome reflections on theory and method, Cairns begins his case in point: a presentation of what he calls the “emerging discipline of evolutionary psychology” and what it contributes to the study of Homeric anger. First there is the neurobiological justification for an evolutionary approach to emotion, based on the demonstrable fact that humans have complex brains that are physical organs wired directly to their autonomous nervous systems, so that no cognitive process, emotional or otherwise, can be separated from physiological processes and physical symptoms. [25] I would add that there is as yet no experimental evidence to link specific, recurrent neurobiological phenomena to specific emotions across cultures. It might be misleading to omit that point from a summary of what neurobiology has shown about the physiological links between mind and body. From there he proceeds to a definition of anger as a socially effective adaptation (in Darwinian terms) in a “tit-for-tat” model for interaction (in other words, in a society based on reciprocity in a relatively simple form) in which mutually beneficial as well as genuinely altruistic behavior occurs. Basing his views on the work of Robert Axelrod on modern corporate strategies for conflict resolution and of Robert H. Frank on the strategic role of emotions, here is the way that Cairns defines prototypical human anger:

In this scheme, anger is a response to a breach of co-operation which, by its character as an affective reaction liable to evade rational control, guarantees and makes credible the threat inherent in the offended party’s reaction, even where to carry out such a threat would be costly for its agent. The adaptiveness of anger lies in the ability of those who possess a recognized disposition to be angry to deter others’ attempts to exploit them. [26]

This definition is tailored to the notion that anger as an evolved emotion must have a socially beneficial “design,” must function as a biologically appropriate adaptation for the perpetuation of the social group, but Cairns is using semantically loaded technical terms here in ways that may not be transparent. First there is the description of anger as “an affective reaction liable to evade rational control,” which uses terminology from the neurobiological and psychological description of some emotions as reflex-like reactions that are not subject to the higher and evolutionarily newer cognitive powers of the brain—I do not think that Cairns literally wants to retroject the concept of rationality into the pleistocene epoch, which is the time frame for the evolutionary processes that evolutionary psychology strives to uncover. Second is the phrase “those who possess a recognized disposition to be angry,” in which the word “disposition” designates a genetic phenomenon that is not shared by all members of the species because it is an inborn potential that is only expressed in some members of the biological class due to individual circumstances—this would be a qualifiedly universal trait. Then there is the “adaptiveness” clause. This provides the justification for the development of the trait in question in evolutionary terms. In this case, the explanation is that the person with the disposition to anger enforces cooperation, a process which is, presumably, an ESS, an “evolutionarily stable strategy” that promotes the well-being of the species. In sum, what Cairns is reporting as the form of anger that evolved in the pleistocene is a reflex-like response to breaches of social cooperation that is not a universally-inherited genetic trait but that some people have a disposition to, and this form of anger evolved genetically because it worked in a positive biological sense, as an enforcer of social cooperation for the well-being of the species.

11b. I used the word “reporting” to describe the way that Cairns treats this idea as a fact, but it is a highly theoretical hypothesis in a field, whether one calls it sociobiology or evolutionary psychology, whose scientific basis has been radically questioned by eminent biologists, geneticists, and philosophers for the past forty years, ever since it was first proposed by the entomologist/ethologist E. O. Wilson. One problem is this: as a hypothesis about the psychological and genetic makeup of humans in the pleistocene epoch, how can it be tested? We are not or at least not yet in a position to know how genes were programmed then (between 2.588 million and 12,000 years BP) with respect to such a disposition. Not only that, but I do not believe that geneticists and neurobiologists are as yet able to determine if any of us nowadays are the bearers of such a supposed genetic trait. Cairns repeatedly invokes what he calls the “significant” and “fundamental” biological aspects of emotions that require an evolutionary explanation—in evolutionary psychology, as he claims, “the answer is being sought in the right place”—and he also claims that this account of emotions is a “relatively close fit” to “our ordinary intuitions as to what they actually do.” [27] But in actuality, there is no way to prove convincingly that the detailed scenario about anger that he describes must be genetically instead of culturally determined. Here is what Stephen Jay Gould wrote about the assumptions and methods of sociobiology in 1978, and they seem to apply perfectly to this dilemma about Cairns’ definition of the prototypical human form of anger:

Much of human behaviour is clearly adaptive, but the problem for sociobiology is that humans have developed an alternative, non-genetic system to support and transmit adaptive behaviour—cultural evolution. An adaptive behaviour does not require genetic input and Darwinian selection for its origin and maintenance in humans; it may arise by trial and error in a few individuals who do not differ genetically from their groupmates in any way relevant to this behaviour spread by learning and imitation, and stabilised across generations by value, custom and tradition. Moreover, cultural transmission is far more powerful in potential speed and spread than natural selection—for cultural evolution operates in the “Lamarckian” mode by inheritance though custom, writing, and technology of characteristics acquired by human activity in each generation…Thus the existence of adaptive behaviour in humans says nothing about the probability of a genetic basis for it or about the operation of natural selection. [28]

The fact that the expression of many emotions has an integral physiological component does not of itself prove that this or any other particular form of anger must be genetic.

11c. Furthermore, and lastly, there are serious questions about the adaptiveness or natural selection arguments that are central to the practice of evolutionary biology and evolutionary psychology. These are the arguments that Gould calls “just-so stories:”

When evolutionists study individual adaptations, when they try to explain form and behaviour by reconstructing history and assessing current utility, they also [like Rudyard Kipling] tell just-so stories—and the agent is natural selection. Virtuosity in invention replaces testability as the criterion for acceptance…But as Darwin insisted vociferously, and contrary to the mythology about him, there is much more to evolution than natural selection. [29]

Here is what Noam Chomsky, who has called evolutionary psychology “a philosophy of mind with a little bit of science thrown in,” has said about adaptiveness arguments:

“You find that people cooperate, you say, “Yeah, that contributes to their genes’ perpetuation.” You find that they fight, you say, “Sure, that’s obvious, because it means that their genes perpetuate and not somebody else’s. In fact, just about anything you find, you can make up some story about it.” [30]

A discipline that is never at a loss for an explanation does not carry conviction, whether those explanations are mutually contradictory or not; [31] such explanations are “stories” in several senses of the term, in that they are easy to produce and multiply. P. E. Griffiths, who has provided a detailed, sympathetic, but still substantively critical treatment of evolutionary psychology, has this to say about the way that adaptiveness stories mislead about the scientific value of the conclusions of evolutionary psychology with respect to human emotion:

These putative evolutionary explanations have the “how possibly” form typical of adaptationist reasoning. They sketch plausible scenarios for the evolution of various features discovered by psychologists. But…it is a mistake to put too much weight on such storytelling. The most basic trap to avoid is that of taking the empirical findings about the emotion system to be confirmed by their evolutionary plausibility. Past experience suggests that whatever was discovered would have been made to look evolutionarily plausible. Only actual empirical psychology can provide good grounds for theories about the structure of the emotion system. The second trap to avoid is taking the mere plausibility of an evolutionary explanation as adequate confirmation of that explanation. [32]

An important current debate on the evolution of language takes on these disciplinary problems of evolutionary psychology, namely the debate between one set of thinkers like Hauser, Chomsky, and Fitch, on the one hand, who hypothesize that human language evolved as an exaptation, a kind of secondary or indirect consequence of other adaptations, [33] as against another set of thinkers like Pinker and Jackendoff, who defend an evolutionary-psychological view that it is a “complex adaptation for communication which evolved piecemeal.” [34] To work towards a resolution of the question, Hauser, Chomsky, and Fitch propose a multi-disciplinary research program that features “the comparative method, which uses empirical data from living species to draw detailed inferences about extinct ancestors” and involves “testable hypotheses” and “much less of the adaptive storytelling and phylogenetic speculation that has traditionally characterized the field.” [35]

12a. Cairns shows his audience of classicists none of the intense debate about the methods and conclusions of evolutionary psychology. [36] Instead, he uses these adaptiveness arguments to attack his straw man:

The legitimate emphasis which such approaches place on the selective pressure exerted by the environment locates emotion firmly in the development of human sociality; they also therefore stress the communicative function of emotion in social interaction; and in emphasizing the fact that emotion is an aspect of group, not just of individual life, they break away importantly from the Western tradition from Descartes to Freud…the evolutionary perspective suggests that what some feel they have discovered as a specific feature of Greek culture is in fact of much wider, indeed universal application. Muellner’s remarks on the primacy of the social dimension of emotion in Homer versus “our” supposedly individualized and internalized conception construct an “us” versus “them” antithesis that errs, as such constructs often do, not in its understanding of ancient sources, but in its unreflective definition of a modernity that has to be the polar opposite of antiquity if cultural determinist assumptions are to be maintained. If nothing else, the crucial importance of emotional signals, especially facial expressions, in “our” culture (as in all cultures) should have inhibited the formulation of this antithesis, for if visible expression is important in the emotional scenarios of all cultures, then sociality is a universal attribute of emotion. [37]

I find the remarks impugning my motives and inventing pretensions for me uncollegial and unjustified, but, more importantly, I believe strongly that Cairns misrepresents what I wrote.

12b. Let the reader judge. At a crucial moment in the story of Achilles (XVI.282), the text of the Iliad explicitly contrasts in a single line the Homeric word for anger whose analysis is the subject of my book, mênis, with the word philótēs, which means ‘friendship.’ In the passage that Cairns refers to above, I was trying to explain this puzzling contrast, to make it understandable that two such terms could be opposed, and to explicate what their opposition tells us about the meaning of mênis:

A distinction between inner and outer self does not exist for [Achilles], so that mênis is in the first instance the opposite of philótēs because it is the absence of philótēs even for one’s self. It is also the inverse of philótēs. Just as Achilles’ extreme gesture of friendship is literally to identify himself with a third person, his friend Patroklos, so the boldest mark of his alienation is actually to perceive himself as a third person, as an other. The opposition between mênis and philótēs has significant consequences from the standpoint of lexicography or translation. The conventional translation of mênis in English by the word “wrath,” an epic term for a violent emotional response by a powerful personage, divine or human, does not suit a word whose opposite term is “friendship.” In English, the opposite of friendship is enmity, and of wrath, delight. The essential problem is the distinction that we draw between emotional and social terms. For us emotions are primarily individualized and internal, and their social dimensions are semantically secondary. With mênis, however, its social dimension is neither secondary to its emotional one nor divisible into inner and outer aspects. What English word is the opposite of both friendship and delight yet also includes alienation? [38]

To characterize the contrast between modern English and ancient Greek terms for emotion that I was drawing in this discussion as an “antithesis” (a word that Cairns uses twice in the space of two sentences) between modern and ancient conceptions of emotions and as “an unreflective definition of a modernity that has to be the polar opposite of antiquity if cultural determinist assumptions are to be maintained” is not an accurate description of the complex distinction I was trying to articulate. The contrast I drew is on two levels, 1) between a worldview that distinguishes inner and outer selves (our own) and one that does not make that distinction (the Homeric one), and 2) between a notion of emotions in which the social aspect is secondary while the individual aspect is primary. [39] Our antonyms for anger [delight, happiness] as against the Greek one [friendship] are the example that I was trying to explain, as against one in which, since there is no outer life as opposed to an inner one in Homeric consciousness, the social aspect of emotion is its only aspect. No part of the contrast I was making can validly be described as a “polar” opposition or an antithesis. A person reading Cairns’ critique of my work without looking at my words would certainly believe that I had completely denied the social aspect of emotions in modern life, whereas I did no such thing. As for his intimation that I claimed to “discover” something unique about Ancient Greek, my goal was only to make sense of what I was reading. I made no such pretentious claim.

13. Cairns’ mistaken criticism of my work also has its paradoxical aspects. At the beginning of the passage of his article cited just above, he himself draws a relevant distinction. Here are Cairns’ words in praise of evolutionary psychologists: “in emphasizing the fact that emotion is an aspect of group, not just of individual life [my emphasis], they break away importantly from the Western tradition from Descartes to Freud.” I infer that Cairns ascribes the polar opposite notion, that emotions are individualized and not social, to the whole Western tradition from Descartes to Freud. In other words, the “unreflective view of modernity” that he mistakenly ascribes to me is one that he himself ascribes to the Western tradition since Descartes. In actuality, the position that he advocates, that emotions are “an aspect of group, not just of individual life,” is compatible with my view, that for English-speakers the individual aspects of emotions like anger are primary and their social aspects secondary.
14. But even that is not the biggest paradox. The biggest of all the paradoxes that I find in the work of Cairns is that the definition that I arrived at for the Greek word that was the subject of my research, mênis, is totally compatible with the hypothetically genetic, prototypical notion of anger that Cairns presents. By means of an in-depth analysis of all the occurrences of the word and its derivatives in their thematic contexts, taking into account the poetics of an oral tradition and the systematic, formalized, deeply resonant way in which Greek epic diction functions, I concluded that this word for anger was a term that functioned as the ultimate sanction for behavior that breaks the social fabric of the epic world, divine and human alike. In the case of the hero Achilles, whose mênis is the principal theme of the Iliad, it also has the added aspect of being an example of anger that causes suffering first of all for its subject, as the definition says, “even where to carry out such a threat would be costly for its agent” (Cairns 2003:16). That was part of my point in the passage I quoted above: when Achilles has mênis, he does not even have philótēs for himself, let alone the other members of the community (see Iliad 1:444–492). There is a difference, and an important one, though: in the Homeric Iliad and Odyssey, this special form of anger is, I believe, a cosmic term with devastating consequences on a huge scale. If Achilles has mênis, that is partly because Zeus sanctions his having it, and it is common for divinities to possess and exercise mênis: one of my points was that the Iliad begins with just such an instance, a case of mênis incurred by Agamemnon and expressed by the god Apollo in the form of a devastating plague that threatens to destroy the whole of the Achaeans’ army. But that transformation of the essential idea is surely a predictable one in the poetic and mythological context of Greek epic, and it is also appropriate to the religious norms of this culture as well as the style of Homeric poetry.
15. So Cairns’ agenda and rhetoric made it necessary for his straw man to say things that he did not actually say. In addition, what his straw man actually did say converges with the major idea that Cairns presents, but he does not see that convergence. This is surprising, especially in view of his assertion that “it would…be a mistake to expect that rationale [for the prototypical case of anger as defined by evolutionary psychology] to function as a rigid template for every form of the emotion,” [40] and “we need to be attentive to the possibility that the terminology of other cultures may map the emotional landscape in ways that are different from our own.” [41] He claims that the agenda of his article is to test “whether the evolutionary rationale is too monolithic for an adequate appreciation of the specificity of [Homeric] terminology [for anger], whether, indeed, there exists any substantial degree of universality behind the specificity of the terminology.” [42] And, in fact, the specific words for anger in Homeric poetry that Cairns actually does discuss fit the evolutionary prototype less well than my definition of Homeric mênis; in fact, Cairns more or less abandons it in favor of Aristotle’s definition of orgḗ ‘anger.’ (I note that Aristotle was both a philosopher and a biologist, so his definition of anger in general will of course have neither the sacral nor the mythical aspects that I have found in the Homeric usages of mênis. Cairns’ move to Aristotle is de parti pris.) Many of the instances that Cairns considers are incurred by slights to an individual hero’s timḗ ‘prestige,’ [43] a word that Cairns does not translate, and they have a less powerful, less clearly positive social function (are they in favor of cooperation? or of conflict?) and none of the potential cost to the agent that mênis has and that the evolutionary psychological prototype specifies.
16. In reality, these temperate, objective remarks of Cairns about his own procedure and the way to apply the results of evolutionary psychology are not his actual practice when it comes to his straw man. Cairns’ clear agenda is to assert the universality of emotions in general and of anger in particular, even when it is not being contested. Consider what he says when he does discuss my definition of mênis:

That mênis has this particular character which associates it with the prodigious and awesome wrath of gods, heroes, and the dead does not warrant Muellner’s insistence that it is “a sanction meant to guarantee the integrity of the world-order” (1996:26; cf. 32 “an irrevocable cosmic sanction”), upholding “cosmic prohibitions” (27–8) and “cosmic justice” (35), in fact “nothing less than the nomen sacrum for the ultimate sanction that enforces the world-defining prohibitions…that are basic to the establishment…of the world of Zeus and the society of mortals he presides over” (129). Muellner’s approach has the merit of recognizing the continuity of divine and human values (34, 50–1) in Homer, but his emphasis is too “top-down,”—the “cosmic” is prior in definition to the social in his approach, whereas, both logically and temporally, the opposite is the case: the mirroring of human ethical and social structures at the supernatural level is the result of the modeling of the latter on the former. [44] The effect (?goal) of Muellner’s approach is to err on the side of the specific and to drive a wedge between Greek mênis and anything that we might recognize as anger. [45]

As before, the criticism is not based on an interpretation of the word’s actual attestations, on the analysis of the data, but on theoretical grounds only, and it concerns the result of the analysis, irrespective of how it was arrived at. My definition was based on the thematic and dictional contexts in which mênis is threatened and the consequences when it is incurred, exercised, and finally stilled, not a theoretical agenda. To say that it is “not warrant[ed]” without pointing at attestations in which it does not apply is to argue by authority, not reason. Furthermore, there is a fundamental error in Cairns’ theoretical critique. He has confused an external historical approach to the evolution of a culture’s concepts with a systematic approach to the way those concepts function within a given culture. No character in the Homeric poems could or would or does express the notion that the behavior of the gods is modeled on the behavior of the heroic humans. To state the obvious, the opposite is the case: the cosmic is prior to the social, and the gods are models, though in no simple-minded way, because irrespective of the history of the culture, within the cultural system the gods are on the top of the cosmic hierarchy and were generated before humans. What is at stake is a worldview represented and embedded in a poetic tradition, not in “reality.” If we should wish to extend the same question to the “real” world in which Epic thrived, the Homeric poetic tradition evolved over generations in the context of a similarly evolving social and cultural system, but Xenophanes’ (deviant) consciousness of the principle of anthropomorphism did not exactly “catch on” in Ancient Greece. [46] Furthermore, we are not dealing with one religion among others that is the object of belief by some and not others. What is at stake here is a worldview shared by a group of diverse but interconnected and historically related city-states. A culture’s worldview is a paper bag that is hard to fight one’s way out of in the absence of competing options. From within either the Homeric poetic system or the moving target of its exterior “reality,” the notion that the divine ethical and social structures are based on those of humans makes no sense—if it did, those systems would be destroyed by such an idea. My point is that the semantic and mythological analysis in my work was from within the Homeric as well as the Hesiodic tradition and their poetic systems. It makes no sense to object, as Cairns does, that it does not take account of the fact that the pantheon was constructed as a projection of human attributes on the divine. That process has to be a prehistoric one whose mechanism was only discernible from outside the system, since the pantheon of gods is extant in Proto-Indo-European, before the invention of the wheel, millennia before the text of Homer. On the contrary, it is worthwhile and productive to consider how Indo-European traditions about epic heroes actually evolved from myths about divinities, in the opposite direction of the one that Cairns argues for, as in the three-volume work of Georges Dumézil entitled Mythe et épopée, because there is a sense in which the internal sequence is true even from an external point of view. [47]

17. This is not to say that an external diachronic perspective is irrelevant to the study of concepts like Epic mênis, only that it is a different undertaking and that the two should not be confused. In an appendix to my work, [48] I tried to show that two cognate terms, Vedic manyu– and Avestan mainiiu-, which are not cognates to Greek mênis but both derive from the same root as mênis , had similar cosmic connotations and thematic associations. Avestan mainiiu– is a primordial cosmogonic force, and Vedic manyu-, in the words of the indologist Charles Malamoud, is a “faculty placed on the same plane as the essential forces of cosmic life.” [49] Although Cairns claims that my notion of its meaning “err[s] on the side of the specific and…drive[s] a wedge between Greek mênis and anything that we might recognize as anger,” [50] its hypostasis as a cosmic term is not only parallel to many others in archaic cultures (including that of Ancient Greece) but the result of the same process that creates anthropomorphic divinities, in this case taking emotions as cosmic forces in the same way that the complex categories of social behavior like sexuality or sovereignty are embodied as gods in a polytheistic system. So much for Cairns’ tendentious claim about the intended unfamiliarity of my definition. And the notion that he thinks undermines my work in fact reinforces its plausibility when applied at the appropriate level of analysis.
18. In responding to Cairns’ criticisms of my work, my primary goal has been to defend myself against unfounded assertions about my intentions, mistakes about what I actually wrote, and a view of scholarship that requires that one scholar’s ideas be built on the wreckage of another’s. In our field as in others, such critiques as Cairns’, if there is no response, are transmitted as correct to the next generation of scholars. In this case, that process has already begun. In a recent article entitled “Homer and Herodotus,” Christopher Pelling gives Cairns credit for “improving on Watkins (1977)…and Muellner (1996).” Here is what Pelling cites as the substance of the improvement:

Cairns emphasizes that mortals too feel μῆνις, but still ‘[t]he preponderance of “supernatural” applications, in Homer and after, lends some weight to the view that there is something about menis that makes it particularly appropriate [his italics] as a term for divine anger…,’ and suggests that ‘it is the gravity and intensity of menis that makes it suitable as a term both for divine wrath and for human anger which exceeds the norm in those two respects.’ [51]

There is no demonstration in Cairns’ article of any these statements, only assertions of them and the citation (five times in the paragraph whose sentences Pelling cites) of Hjalmar Frisk’s 1946 article on the concept of μῆνις. [52] At one point, Cairns credits me, as cited above, for “recognizing the continuity of divine and human values” even though in his own discussion of this issue (31–32) he cites Beck’s article in the Lexikon des frühgriechischen Epos s.v. μηνίω and does not mention my work examining the attestations of the derivatives of mênis and showing the intimate relationship between those derivatives and the noun itself. So the statements of Cairns that Pelling quotes are not in fact Cairns’ own contribution to the study of the word, since they are not fundamentally different from my own as well as others’ conclusions. What then, motivates his statement that they constitute an improvement on my work? [53] What is missing from Pelling’s two quotations of Cairns are the words that immediately precede the second one, which are “I hazard a guess that it is the gravity and intensity…,” as though Cairns himself was actually the first to believe that these qualities distinguish mênis from other words for anger. To my mind, they only account for part of the difference, since Cairns does not explain in what the gravity and intensity consist. That is what my work does in concrete detail, and that is the basis on which I attributed cosmic dimensions to the word. So the only sense in which Cairns might be construed as improving on my work is in rejecting, on what I believe are mistaken theoretical grounds, what I demonstrated by working with the word’s attestations in context. I believe that what motivates Pelling’s view is not any significant improvement over Watkins’ or my own work, but the consistently negative stand that Cairns takes towards my work and by extension Watkins’. Such rhetoric is often more readily taken as true than constructive and creative engagement with one’s predecessors.

19. I have spent more than enough time discussing the details of Cairns’ critique of my work, and I now wish to address what is, I believe, the larger issue in his paper, the way in which it advocates for, presents, and makes into its overall argument the value for classical scholarship of work from other fields. Here is the concluding statement of his paper on Homeric anger:

Thus in order to fulfill its potential, the dialogue between ourselves as classicists and the societies we study must be supplemented by dialogue between classics and other disciplines—not only historical, philological, and literary, but also philosophical, sociological, and scientific.

In this connection, the title of his paper is relevant: “Ethics, ethology, terminology: Iliadic anger and the cross-cultural study of emotion.” Aside from its unexpected use of the word “ethology” instead of sociobiology or evolutionary psychology, Cairns also avoids the word and the discipline that one would suppose to be most obviously concerned with “the cross-cultural study of emotion,” namely cultural anthropology. Only once does he cite the work of anthropologists on emotion, in a footnote on the “antithesis” he attributes to me between modern and ancient concepts of emotion, in which he cites works of Michelle Rosaldo and Catherine Lutz as the “pedigree in contemporary ethnology” for the (false) antithesis. [54] In studying anger in cultures other than one’s own, why not look to cultural anthropology or ethnology for helpful methodologies, theories, cross-cultural parallels, universals — in short, a collection of data and ideas with which to understand the problem? In effect, Cairns has made me a straw man for cultural determinism on emotions, and it appears to him as though I am therefore also a front man for anthropological perspectives on the subject, which are assumed to be the same. My work makes no mention of cultural determinism, and his ellipsis of the whole field of anthropology because of its false attribution to my views is a serious flaw in Cairns’ approach. The focus of Lutz’s work is “cultural discourse on emotion,” [55] and the work of Rosaldo does not take an extreme view of the social or the cultural aspect of human emotions—she insists, in fact, on their bodily aspects. [56] If Clifford Geertz believed that “passions are as cultural as the devices [of politics],” Edmund Leach considered that idea “complete rubbish,” and Lévi-Strauss said that if emotions do not, on a conscious level, either block or accelerate understanding, they belong to biology, not les sciences humaines (which includes anthropology). [57] Classicists should know that there is a quantity of captivating, pertinent, and enlightening debate in anthropology on all the open questions in the study of emotions that are mentioned in Cairns’ paper and much more, along with enlightening comparative evidence of a great variety of ways to conceive of, describe, experience, and translate them. All these pursuits are central to the work of classics scholars.


Allen, Nicholas J., “The Category of the Person: A Reading of Mauss’s Last Essay,” in The Category of the person: anthropology, philosophy, history, ed. Michael Carrithers, Steven Collins, and Steven Lukes (Cambridge University Press, 1985), 26-45.
Susanna Morton Braund and Glenn W. Most, eds., Ancient anger : perspectives from Homer to Galen, Yale Classical Studies 32 (Cambridge University Press, 2003).
D. L Cairns, “Anger and the veil in ancient Greek culture,” Greece & Rome (Second Series) 48, no. 1 (2001): 18–32.
D. L. Cairns, “Ethics, ethology, terminology: Iliadic anger and the cross-cultural study of emotion,” in Braund and Most 2003.
M. J. Clarke, B. G. F. Currie, and R. O. A. M. Lyne, eds., Epic interactions: perspectives on Homer, Virgil, and the epic tradition: presented to Jasper Griffin by former pupils (Oxford: Oxford University Press, 2006).
Antonio R Damasio, Descartes’ Error: Emotion, Reason, and the Human Brain (New York: G.P. Putnam, 1994).
Charles Darwin, Paul Ekman, and Phillip Prodger, Expression of the emotions in man and animals, 3rd ed. (London: HarperCollins Publishers, 1998).
Georges Dumézil, Mythe et épopée, 3v. (Paris: Gallimard, 1968–1973)
W. Tecumseh Fitch, Marc D. Hauser, and Noam Chomsky, “The evolution of the language faculty: Clarifications and implications,” Cognition 97 (2005): 179–210
Robert H. Frank, Passions within reason : the strategic role of the emotions, 1st ed. (New York: Norton, 1988).
Hjalmar Frisk, “ΜΗΝΙΣ· Zur Geschichte eines Begriffes,” Eranos 44 (1946): 28–40.
Clifford Geertz, Negara : the theatre state in nineteenth-century Bali (Princeton, N.J.: Princeton University Press, 1980).
S. J. Gould, “Sociobiology: The art of storytelling,” New Scientist 80, no. 1129 (1978): 530–533.
S. J. Gould and R. C. Lewontin, “The Spandrels of San Marco and the Panglossian Paradigm: A Critique of the Adaptationist Programme,” Proceedings of the Royal Society of London, Series B: Biological Sciences 205.161 (1979): 581–598.
P. E Griffiths, What emotions really are: the problem of psychological categories (University of Chicago Press, 1997).
Marc Hauser, Noam Chomsky, and W. Tecumseh Fitch, “The Faculty of Language: What Is It, Who Has It, and How Did It Evolve?,” Science 298: 22 November 2002: 1569–1579
Karl G. Heider, Landscapes of Emotion: Mapping Three Cultures of Emotion in Indonesia (Cambridge: Cambridge University Press, 1991).
Dorothy C. Holland and Naomi Quinn, eds., Cultural models in language and thought (Cambridge: Cambridge University Press, 1987).
John Horgan, The Undiscovered Mind: How the Human Brain Defies Replication, Medication, and Explanation (New York: Free Press, 1999).
Alexandra Horowitz, Inside of a dog : what dogs see, smell, and know (New York, NY: Scribner, 2009).
Rachel Jack et al., “Cultural Confusions Show that Facial Expressions Are Not Universal,” Current Biology 19 (September 29, 2009): 1543–1548.
R. Jakobson, “Linguistics and poetics,” T. A. Sebeok (cd.) Style in language. (NY: Wiley, 1960).
Z. Kövecses, Metaphor and emotion: Language, culture, and body in human feeling (Cambridge [UK]: Cambridge University Press, 2003).
George Lakoff, Metaphors We Live By (Chicago: University of Chicago Press, 1980).
George Lakoff and Z. Kövecses, “The cognitive model of anger inherent in American English,” in D. Holland and N. Quinn, Cultural Models in Language and Thought, Cambridge, 1987: 195–221.
E. Leach, “A poetics of power,” New Republic 4 (1981): 30–3.
J. Leavitt, “Meaning and feeling in the anthropology of emotions,” American Ethnologist 23, no. 3 (1996): 514–539.
C. Lévi-Strauss, L’homme nu (Paris, Plon, 1971).
John A. Lucy, “Whorf’s Theory of the Linguistic Mediation of Thought,” in E. Mertz and R. J Parmentier 1985.
Catherine Lutz, Unnatural emotions : everyday sentiments on a Micronesian atoll & their challenge to western theory (Chicago: University of Chicago Press, 1988).
Charles Malamoud, “Manyúh Svayambhúh,” in Mélanges d’Indianisme à la mémoire de Louis Renou, 1968, 493–507.
E. Mertz and R. J. Parmentier, eds., Semiotic mediation: Sociocultural and psychological perspectives, (Orlando, FL: Academic Press, 1985).
Leonard Muellner, The anger of Achilles : mēnis in Greek epic, Myth and poetics (Ithaca, N.Y.: Cornell University Press, 1996).
Leonard Muellner, “Discovery Procedures and Principles for Homeric Philological Research,” Classics@ Issue 3 (2007) https://chs.harvard.edu/wa/pageR?tn=ArticleWrapper&bdc=12&mn=1321
Christopher Pelling, “Homer and Herodotus,” in Clarke, Curry, and Lyne 2006, 75–104.
Steven Pinker and Ray Jackendoff, “The faculty of language: what’s special about it?” Cognition 95 (2005): 201–236.
M. Z. Rosaldo, “Toward an Anthropology of Self and Feeling,” in R. A. Shweder and R. A. LeVine 1984.
Ferdinand de Saussure et al., Cours de linguistique generale, 1st ed. (Paris: Payot, 1916).
S. Schachter and J. Singer, “Cognitive, social, and physiological determinants of emotional state,” Psychological Review, Vol 69 (5), Sept. 1962, 379–399.
R.A. Shweder and R.A. LeVine, eds., Culture Theory: Essays on Mind, Self, and Emotion (Cambridge: Cambridge University Press, 1984).
Bruno Snell and Hans Joachim Mette, Lexikon des frühgriechischen Epos (Göttingen: Vandenhoek & Ruprecht, 1955).


[ back ] 1. Cairns 2003: 11. My work is Muellner 1996.
[ back ] 2. Lord 2000, Chapters 3, 4, and 5 (pp. 30–123).
[ back ] 3. Muellner 1996: 129.
[ back ] 4. Muellner 1996: 1.
[ back ] 5. Cairns does not invoke the Whorf hypothesis about the relationship between language and thought, but it does not apply here in any case; see Lucy 1985 for the way in which language can mediate between languages and cultures.
[ back ] 6. This is a general principal of cross-cultural study in anthropology, as articulated for example, in Allen 1985.
[ back ] 7. Saussure 1916: 97–103.
[ back ] 8. Cairns’ remarks (given just above) citing my work come in a footnote at the end of a sentence inveighing against classicists for shirking their responsibility to become familiar with research on emotions in other disciplines, but, strangely enough, he himself cites this same work and other work by these authors in the previous footnote as an example of a study that does not confine itself to “the labels that a culture attaches to the emotions it recognizes,” that broadens the study of emotion in a way of which he approves.
[ back ] 9. Kövecses in 2003 assembled empirical evidence from several other languages in an attempt to discover how universal the linguistic attributes of anger in English described in Lakoff and Kövecses 1987 were.
[ back ] 10. For a more detailed account of my discovery procedure, see Muellner 2007.
[ back ] 11. Cairns 2003:15.
[ back ] 12. Cairns 2003:12.
[ back ] 13. Cairns never discusses these two types of anger as exhibited, for example, in chimpanzees. The open-mouthed form of anger is a threatening display of aggression and dominance; the close-mouthed one is not. More on the stance towards empiricism of evolutionary psychology and of Cairns below.
[ back ] 14. Cairns 2003:12.
[ back ] 15. Cairns 2003:12–13.
[ back ] 16. Cairns 2003:14.
[ back ] 17. Kövecses and Palmer 1999:250 put it most clearly: “there is still no physiological substrate on which each emotion can be precisely located, there is no universal cluster of features for any emotion, and there is no invariant conceptual content for any emotion.” The study cited by Lakoff and Köveceses 1987:219, namely Ekman et al. 1983, tested heart rate and skin temperature of either actors mimicking facial expressions linked by Ekman to specific emotions or scientists “reliving” emotions. It resulted in relative measures that distinguished ‘positive’ from ‘negative’ emotions and only distinguished among ‘negative’ emotions, such as higher heart rates associated with anger or fear than with happiness and higher skin temperature associated with anger rather than fear. The evidence was significantly more distinct in actors pretending to express emotions with their faces, and half of the data of those “reliving” emotions was thrown out because the persons reliving the emotions reported not feeling them strongly enough. The classic study, Schachter and Singer 1962, showing that people do not themselves reliably associate specific physiological changes with specific feelings, does not appear to have been shaken by Ekman et al. 1983, which itself raises the larger questions about biofeedback and the still unknown nature of the nexus between feelings and the autonomous nervous system.
[ back ] 18. Heider 1991, for example, speaks of forty distinct “clusters” of emotions among the Minangkabau in Indonesia, with little or no overlap between them. I have this reference from Kövesces and Palmer 1999.
[ back ] 19. Though there is the amazing example in Homer of Hera expressing two emotions at once…
[ back ] 20. This is not to minimize the complexity of the canine emotional repertoire. There is a YouTube video of a dog with a distinctive facial expression, perhaps of guilt: http://www.youtube.com/watch?v=B8ISzf2pryI, and for an accessible account of new research on the psychology and evolution of dogs, see Horowitz 2009.
[ back ] 21. Jack, Blais, et al. 2009: 1543-1548
[ back ] 22. Jack, Blais, et al. 2009: 1543
[ back ] 23. Jakobson 1960:23
[ back ] 24. Cairns 2003:14.
[ back ] 25. The popularizer of this hypothesis about brain function in its technical form is Anthony Damasio (Damasio 1994), but Homeric scholars are all too familiar with a way of describing thoughts and feelings that does not distinguish between the two or between mind and body, since the Epic language ascribes all mental activity to physical organs like phrénes and thumós.
[ back ] 26. Cairns 2003:16
[ back ] 27. Cairns 2003:16.
[ back ] 28. Gould 1978:532.
[ back ] 29. Gould 1978: 530.
[ back ] 30. Horgan 1999:179.
[ back ] 31. Cairns 2003:20n33 is aware of the problem of alternative explanations for the capacity to anger in mentioning that there are several which he claims are not mutually exclusive.
[ back ] 32. Griffiths 1997:91.
[ back ] 33. Hauser, Chomsky, and Fitch 2002 and, in response to Pinker and Jackendorff 2005, Fitch, Hauser, and Chomsky 2005; my thanks to S. Jay Keyser for pointing out this exchange to me. For the notion of exaptation and the architectural term “spandrel,” see Gould and Lewontin 1979.
[ back ] 34. Pinker and Jackendoff 2005:202.
[ back ] 35. Hauser, Chomsky, and Fitch 2002:1572 and Fitch, Hauser, and Chomsky 2005:180.
[ back ] 36. For another classics scholar’s comments on sociobiology, see Burkert 1996:8–12. I cite one comment: “Behavior is hopelessly complex already at the level of primates.”
[ back ] 37. Cairns 2003:16–17. See also Cairns 2001:31–32n42 for the same point and the attribution to me of “underlying” cultural-determinist motivations.
[ back ] 38. Muellner 1997:138.
[ back ] 39. I note that in Cairns’ version of my statements, he attributes to me remarks on “the primacy of the social dimension in Homer.” Such “primacy” is something that is not possible, since there is no contrasting secondary aspect, no way of conceiving or describing interior feelings as opposed to exterior ones in Homeric poetry. I also note that Cairns omits the specific example I was trying to explain, preferring to criticize theoretical statements without showing the data that they are framed to account for.
[ back ] 40. Cairns 2003:20.
[ back ] 41. Ibid.
[ back ] 42. Ibid.
[ back ] 43. I do not agree with his contention that all of the instances are of this type, as my book shows.
[ back ] 44. Cairns here cites Burkert 1996:80–102, a learned description of the way that Classical and Near Eastern religions represent the dominance of divinity, the subservience to them of mortals, and the way intermediaries function between them. It does not explicitly mention the notion that divine behavior is modeled upon human behavior since it is an analysis like my own, of the way in which a system functions internally, not of its history. The fact that Cairns has referred to Burkert’s chapter evinces the same basic confusion on his (Cairns’, not Burkert’s) part as is evident from the objection to my work that he raises: see below.
[ back ] 45. Cairns 2003:33.
[ back ] 46. Xenophanes Frs. B 15 and 16 (D-K): just as the gods of the Aithiopes and the Thracians resemble themselves (Fr. 16), so cows and horses (Fr. 15), if they could paint like humans, would paint gods who look and act like cows and horses. Significantly, the understanding is comparative and anthropological—in order to extricate Greeks from their worldview, Xenophanes had to get them to think like Thracians or cows. My assumption is that Cairns knows nothing of Bourdieu’s concept of miscrecognition.
[ back ] 47. Dumézil 1868–1973.
[ back ] 48. Muellner 1996:177–186.
[ back ] 49. Malamoud 1968:499.
[ back ] 50. Cairns 2003:33.
[ back ] 51. Pelling 2006:95–96n57.
[ back ] 52. Frisk 1946.
[ back ] 53. As for putative improvements on Watkins’ work, Cairns argues against one aspect of it in a footnote (32n93) that uses (without attribution) the detailed information that I provided (Muellner 1996:189–194) about the usage of derivatives of μῆνις and its occurrence in Euripides’ Hippolytus in trying to defend Watkins’ idea that the form of mênis was the result of a tabu deformation, but he has misunderstood the basic problem. Cairns apparently understood that Watkins believes that the word μηνιθμός is the result of deformation, whereas what he is explaining is the form of mênis itself. In any case, at the end of his note 93, he dismisses my defense of Watkins’ view (“Muellner’s attempt…fails to convince me”, which is a way to appear to engage with someone’s ideas without actually doing so). I not only supplied but did my best to answer the objections that he raises.
[ back ] 54. Cairns 2003:17n21, citing Rosaldo 1980 and Lutz 1988. In an earlier footnote, he criticizes the classical philosopher D. P. Fowler’s citation of the absence of anger among the Utku, an Inuit people (Cairns 2003:12n2).
[ back ] 55. Lutz 1988:5.
[ back ] 56. Rosaldo 1983:136n4, where she calls emotions “a mix of intimate, even physical experience and a more or less conscious apprehension of, or ‘judgment’ concerning, self-and-situation, ” or Rosaldo 1984:143, “Emotions are thoughts somehow ‘felt’ in flushes, pluses, ‘movements’ of our livers, minds, hearts, stomachs, skin. They are embodied thoughts…”
[ back ] 57. Geertz 1980:124; Leach 1981:32; Lévi-Strauss 1971:597; these and the preceding two references are from Leavitt 1996, an excellent introduction to research on emotions in anthropology over the thirty years that preceded it. Among other things, Leavitt’s paper highlights extraordinarily intuitive perspectives on emotions by Spinoza. I owe thanks for all sorts of help with thiis essay to Gregory Nagy, Samuel Jay Keyser, Douglas Frame, and Richard Parmentier.