Cognition and Representation - Oxford Scholarship Jump to ContentJump to Main Navigation
The Religion of the Mithras Cult in the Roman EmpireMysteries of the Unconquered Sun$

Roger Beck

Print publication date: 2007

Print ISBN-13: 9780199216130

Published to Oxford Scholarship Online: January 2010

DOI: 10.1093/acprof:oso/9780199216130.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE ( (c) Copyright Oxford University Press, 2015. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see Subscriber: null; date: 27 August 2016

Cognition and Representation

Cognition and Representation

(p.88) 6 Cognition and Representation
The Religion of the Mithras Cult in the Roman Empire


Oxford University Press

Abstract and Keywords

This chapter explains how the Cognitive Science of Religion can serve as a powerful new method for exploring the making of representations in a religion and the cognitive processes by which an initiate apprehends a religion's symbol system. Following Dan Sperber's approach (Explaining Culture), all religions may be described in terms of the interplay of representations over time: public representations in the media of sacred spaces, physical images, performed rituals, and words uttered and recorded in text; and private representations in the minds of individual adherents. A fortiori, the negotiation of representations in Mithraism can have been no different. An appendix draws on Lucian's treatment of audience response in his essay On the Dance to show how the negotiation of representation worked in a comparable situation in antiquity.

Keywords:   Cognitive Science of Religion, representation, cognition, Dan Sperber, Lucian On the Dance

1. The Cognitive Approach: Ontogenetic/ Phylogenetic Versus Cultural

It is sometimes diffcult to date the arrival of a new method or approach in scholarship, but at least since the 1990s a new method focused on cognition has been available to scholars of religion—including, since it is not restricted as to date or culture, scholars of the religions of Graeco-Roman antiquity. The method originated in the cognitive sciences, in particular evolutionary psychology, and in anthropology. It is known as ‘the cognitive science of religion’ (Andresen 2001a, b). I shall speak of it here simply as the ‘cognitive method/approach’ and of scholars and scholarship that take it as ‘cognitivist’.

The cognitive method is a powerful explanatory tool. It is not, and does not claim to be, interpretive. We will not understand Mithraism qua Mithraism any the better for using it, but we may understand better how the Mithraic mysteries functioned qua religion.

Cognitivist scholarship tends towards radical reductionism: to explain is to explain away.1 I see no heuristic benefit in making that larger negative claim. Whether the representation-forming minds of Mithraists—or, for that matter, of Christians or Muslims—touched on some otherworldly reality seems to me, in a secular academic context, as idle to deny as to affirm. The cognitive method need not be forced to answer ontological questions at a metaphysical level; sufficient that it addresses, and addresses well, the modal question of how the human mind forms and organizes ‘religious’ ideas in the here and now.

The human mind forms—and, as far back as the record shows, always has formed—representations of supernatural beings. This is not a necessary activity of the mind, for one can get through life without it, but it is certainly a very common activity.

Not all representations of supernatural beings are ‘religious’, in the sense of belonging to that domain of life which we label ‘religion’. Nor do all religious representations necessarily involve supernatural beings. Nevertheless, there is a (p.89) high degree of coincidence: more often than not, the mental representation of supernatural beings is a vital part of practising a religion. Certainly, that was so for the religions current in classical antiquity. Paganism was literally unthinkable without the mental representation of the Olympian and other gods; likewise Judaism without representation of Jahweh, or Christianity without the additional representation of Jesus as the Christ. And so Mithraism: to be a Mithraist, one must have (or feign having) Mithras ‘in mind’.

While there is little religious thought that does not in a primary or secondary way involve supernatural beings, the human mind does construct and entertain representations of innumerable supernatural and paranatural beings entirely outside the religious domain. Folk tale and fantasy literature abound with inventions of this sort, whose connection with ‘religion’ is tenuous or nonexistent. And that is precisely the point. Except in the degree of ontological commitment, there is no essential difference between an ancient Athenian’s representation of Pallas Athena or an ancient Roman’s of Jupiter Optimus Maximus and a twenty-first-century person’s representations of wizards, elves, orcs, and dementors á la J. R. R. Tolkien or J. K. Rowling.

The first achievement of the cognitive approach is thus to strip away the special status of ‘religious’ representations of supernatural beings, and consequently to de-mystify and de-problematize them. The ability to form mental representations of supernatural and paranatural beings is simply part of the evolved mental endowment of the species Homo sapiens. As a further consequence, the cognitive approach radically redefines the ‘why’ questions: why religion, why the gods? Granted our natural propensity to entertain representations of the non-natural, it is not the presence of the gods in our minds that requires explanation so much as their expulsion in relatively recent times; not ‘why religion?’ but rather ‘why religion no longer?’ As the cognitivists emphasize, religion is ‘natural’, science is not (McCauley 2000).

The second achievement of the cognitive approach is to divert part of the inquiry into religion from the socio-cultural level both upwards to the phylogenetic and downwards to the ontogenetic. We form representations of supernatural beings not by virtue of membership in societies and cultures but by virtue of membership in the species Homo sapiens. Our particular societies and cultures shape and standardize our representations, conforming them to the various explicit traditions current and licensed in our various times and places. But it is we who construct the gods, not ‘society’, not ‘culture’; and ‘we’ means the human mind functioning in the human brain. Hence the reorientation of the inquiry from society or culture to the individual and the species.2

At first glance this reorientation might seem to doom any project directed towards an ancient, dead religion. How can we possibly access the representations of minds long dead and gone? It is precisely here, however, that evolutionary (p.90) theory comes to the rescue. Societies may come and go, cultures may change with great rapidity, but the adaptive changes which significantly modify a species are measured at the least in tens of millennia, not mere centuries. Evolutionary science postulates no change in the human brain and mind which would have rendered them markedly different now from what they were and how they functioned in classical antiquity. Quite the contrary, we may safely assume that we form our representations of supernatural beings, to all intents and purposes, just as the ancients did. Any adaptive changes, so cognitive theory argues, took place in earlier and far longer epochs as our remote ancestors passed through the hunter-gatherer phase, and they took place in response to the exigencies of the hunter-gatherers’ environment. They occurred because they gave those hunter-gatherers with these adaptations a competitive and reproductive edge over those without. This is not to say that the capacity (and neural circuitry) for imagining supernatural beings itself conferred a competitive edge, merely that it cannot be a recent acquisition, something which radically differentiates us from our conspecifics a score of centuries ago.

Same brain, same mind. Consequently, one may argue with some confidence from the way we form ‘religious’ representations now to the way the ancients formed them then. Given the comparatively rapid and radical shifts of culture, we are actually on much firmer ground with the phylogenetic and the ontogenetic than with the socio-cultural.

While a comprehensive solution to the mind–body problem still eludes both scientists and philosophers (is it even in principle attainable?), much is now known about the neural processes in the brain which accompany various mental states and events. This now opens up, for the first time, the possibility of correlating what happens in the brain with what happens subjectively in the mind of someone undergoing a religious experience. We shall touch on some of this research later, but since it mostly involves unusual states of consciousness (e.g. meditation, ecstasy), we shall pass it by for now and return to the more pedestrian topic of the mind’s representation of supernatural beings. Here one may surely assume that just as there is nothing distinctively ‘religious’ about the mental event of forming representations of beings not normally encountered in the natural world, so the concomitant neural events do not differ, or do not necessarily differ, according to whether or not the representations belong to the subject’s religious world. Different neuronal groups do not fire in different ways whenever the mind is, as it were, ‘doing religion’. The human brain has no dedicated circuits for religion or ‘the sacred’.

The topic of cognition in the religious domain is caught up in a much wider debate taking place in the social sciences and in those disciplines, particularly psychology and anthropology, which straddle the boundary between the life sciences and the social sciences. The debate is about the acquisition and location of ‘culture’. Are cultural systems, of which religions constitute a particular form, downloaded in a process of teaching and learning as content into an originally (p.91) content-free human mind/brain? Or is the human mind/brain already endowed with systems—software running on ‘wetware’, as the saying goes—selected in and by the evolutionary process, which form representations modified, not created, by interaction with conspecifics in a particular society and culture?

I incline to the latter scenario. Fortunately, a concise summary of its underlying model already exists in the introductory essay to a volume of studies by some of its leading proponents (Tooby and Cosmides, in Barkow et al. 1992: 24). These cognitivists call it the ‘Integrated Causal Model’ (ICM). I quote their summary:

  1. a. the human mind consists of a set of evolved information-processing mechanisms instantiated in the human nervous system;3

  2. b. these mechanisms, and the developmental programs that produce them, are adaptations produced by natural selection over evolutionary time in ancestral environments;

  3. c. many of these mechanisms are functionally specialized to produce behavior that solves particular adaptive problems, such as mate selection, language acquisition, family relations, and cooperation;

  4. d. to be functionally specialized, many of these mechanisms must be richly structured in a content-specific way;

  5. e. content-specific information-processing mechanisms generate some of the peculiar content of human culture, including certain behaviors, artifacts, and linguistically transmitted representations;

  6. f. the cultural content generated by these and other mechanisms is then present to be adopted or modified by psychological mechanisms situated in other members of the population;

  7. g. this sets up epidemiological and historical population-level processes; and

  8. h. these processes are located in particular ecological, economic, demographic, and intergroup social contexts or environments.

To understand what is new and diffierent about the ICM, we should also look at the traditional model which the ICM challenges and aims to supplant. This ‘Standard Social Science Model’ (SSSM—again the term is that of the cognitivists), is likewise conveniently summarized by Tooby and Cosmides (1992: 31 f.). Though they are of course opposed to this model, their summary of it is fair and untendentious. I quote the first seven propositions (abbreviating where feasible without loss of substance).4 They convey the implicit assumptions which usually underlie inquiries, such as ours, into cultural phenomena. These are cards which ought to be on the table, but seldom are:

  1. 1. Particular human groups are properly characterized typologically as having ‘a’ culture, which consists of widely distributed … behavioral practices, beliefs, ideation (p.92) systems, systems of significant symbols, or informational substance of some kind. Cultures are more or less bounded entities, although cultural elements may diffuse across boundaries.

  2. 2. These common elements are maintained and transmitted ‘by the group’, an entity that has cross-generational continuity.

  3. 3. The existence of separate streams of … culture … is the explanation for human within-group similarities and between-group differences. In fact, all between-group differences … are referred to as cultural differences and all within-group similarities are regarded as the expressions of a particular culture….

  4. 4. Unless other factors intervene, the culture … is accurately replicated from generation to generation.

  5. 5. This process is maintained through learning, a well-understood and unitary process.

  6. 6. This process of learning can be seen … as a group-organized process called socialization, imposed by the group on the child.

  7. 7. The individual is the more or less passive recipient of her culture and is the product of that culture.

What difference does it make in practice that we adopt the ‘integrated causal model’ (ICM) rather than the ‘standard social science model’ (SSSM) for our inquiry into the Mithraic mysteries? We are already committed, by placing ourselves in the Geertzian symbolist tradition,5 to treating the Mithraic mysteries as a cultural system, which would seem to position us more comfortably in the SSSM camp. Of course, the ICM does not preclude treating religion as a cultural system (propositions ‘f’ through ‘h’); so our question is rather, what advantage does the ICM confer over the SSSM?

At one level the answer is, none at all. When, for descriptive purposes, we uncouple the mysteries from their actual initiates, it makes sense to treat them as an autonomous cultural system, or more precisely as a subsystem of the wider culture of Graeco-Roman paganism. This we did in our ‘template for the redescription of the mysteries’ presented in Chapter 1 (sect. 3), where the propositions were stated in alternative forms: ‘the first line [in each proposition] represents the mysteries as an autonomous system acting on the initiate; the second … from the initiate’s point of view as something apprehended and accepted.’

Clearly the first of the alternative forms coheres better with the ‘standard social scientific model’ and the second with the ‘integrated causal model’. Since, as I have already stated, our ultimate quarry is the initiates’ apprehension of their mysteries (qua symbol system) rather than the mysteries per se, the ICM will be my preferred model both over the longer haul and especially in the present chapter in which I focus on cognition.

(p.93) 2. Gods In Mind: Cognition And The Representation Of Supernatural Beings

Perhaps the best application to date of the cognitivist approach to religion, and certainly one of the most accessible, is Pascal Boyer’s Religion Explained (2001). We shall accordingly hew quite closely to his model of religion in tracing the mental representations of an initiate of the Mithraic mysteries. Remember, as we do so, that the sole and sufficient warrant for this seemingly audacious project is the general evolutionist postulate that the lapsed time between Mithraists and moderns is simply too short for the brains and minds of our species to have undergone adaptations significant enough to render invalid extrapolations from ‘mind doing religion now’to ‘mind doing religion then’. Because of this we are at no insurmountable disadvantage to Boyer and other cognitivists who treat of contemporary religion and have access to contemporary minds by way of psychological, sociological, and ethnographic research.

To the diachronic postulate of the sameness of the brain/mind now and two millennia ago must be added a synchronic postulate: that there is no essential difference in the way in which believers and non-believers construct mental representations of supernatural beings. The content and associations of a Persian’s, a Roman’s, and a Christian’s representations of Mithras no doubt differed, and all these certainly differed from a modern historian of religion’s representations of Mithras, but the mental and underlying neural processes of representation did not and do not differ.

The cognitive approach, we have already seen, locates a religion in the representation-forming minds of those who adhere to it, whether actively or passively. As humans we all form representations of beings which do not exist in the natural world, at least in a normal, empirically testable way. Such representations are for the most part evanescent. Some representations, however, because they are conformable to the representations of others in the same socio-cultural group, are preserved, fostered, and modified by the interaction of mind with mind; also of mind with the projections of mind in the actual world: text, creed, artistic representations, mimetic ritual, and so on. Very occasionally, a new religion is born, or an old religion substantially modified, when the novel representations of a single mind, then of the minds of a small founding group, successfully commend themselves over wider and wider circles, and the apparatus of the religion, in whatever form is deemed necessary or appropriate, is constructed in the common objective world.6 A religion dies when no one remains to energize its outward forms and, more fundamentally, to make its subjective representations within the context of those outward forms. Because they exist (p.94) in the external, objective, common world, the forms linger on in the fossilized record of text and artefact. From the record we seek to recapture, as far as an external inquirer can, something of the mental representations that were the living religion.

To avoid over-privileging religion or ‘religious’ representations, we should recall that the very same mental processes create and perpetuate representations of make-believe beings, whether natural or supernatural, in the domain of art, both visual and verbal. Sir John Falstaff, inanimate in Shakespeare’s text, lives on in the representation-forming minds of audiences and readers, as do the aforementioned wizards, elves, orcs, and dementors of Tolkien and Rowling. Thus too have the Olympian immortals fled mortality by migrating to the aesthetic domain—where of course they were always at home: a consideration which merely returns us to our postulate that the process of forming representations of the gods did not—and does not—vary; regardless of whether those forming the representations were witnessing a sacrifice in some ancient community, a tragedy in fifth-century BCE Athens, or a pantomime in first-century CE Rome.

Religion, then, exists in, and consists of, a ferment of representations in the minds of its living adherents. From the general to the particular, it follows that the Mithraic mysteries were the mental representations of successive generations of initiates in their far-flung ‘caves’ and autonomous brotherhoods throughout the Roman empire.

3. Negotiating Representations

Adopting a cognitive approach and locating ‘religion’ in the minds of its adherents pays two rich and immediate dividends. First, the cognitive model accommodates well the rather fluid concept of doctrine which I proposed towards the end of Chapter 4: ‘In place of doctrine as a definite body of explicit teaching, we have re-characterized it as a loose web of interpretation, both actual and potential, located in the symbol system of the mysteries’ (sect. 12); alternatively, as ‘an indeterminate set of explanations which senior Mithraists would impart to their juniors or explore among themselves’ (sect. 10). From a cognitive perspective, then, doctrine is that which within the given religious group is negotiated (or negotiable) concerning legitimate representations. As we said above (preceding section) of humanity’s propensity for forming mental representations of supernatural beings, ‘some representations … because they are conformable to the representations of others in the same socio-cultural group, are preserved, fostered, and modified by the interaction of mind with mind; also of mind with the projections of mind in the actual world: text, creed, artistic representations, mimetic ritual, and so on.’

Legitimate representations can—but need not necessarily—be negotiated explicitly and by formal process. They are thereby reified—and sanctified—in (p.95) creeds, catechisms, canons of scripture, and the like. This was the road taken by Christianity. Mithraism followed another and less divisive road: from the overall conformity of its mithraea and figured monuments to certain norms, one may infer that it managed to maintain a coherence of mental representation without resort to explicit doctrinal formulations. Creeds and the like, once formulated, objectify and sanctify doctrine, and so make it something to defend and police, a criterion for inclusion and exclusion. It may be (though I doubt it) for want of extant evidence, but it seems improbable that there were Mithraic heretics. ‘Heresy’ and the odium which attaches to it only become possible when there is an ‘orthodoxy’ to measure it against.

It is hermeneutically liberating to shake off the task of reconstructing an explicit but no longer extant Mithraic doctrine; likewise the task of rebutting the opposite but more insidious charge that Mithraism, lacking formulated doctrine, was for that reason a second-class or inferior religion. If anything, its coherent yet unprescribed way of normalizing the mental representations of its initiates seems, if not more sophisticated, at least more admirable. It is certainly more irenic.

The cognitive approach compels us to interpret and explain the mysteries always in terms of actual on-the-ground Mithraism. We focus not on an abstracted system (although for descriptive purposes one must sometimes treat it as such), but on those interactive processes of mental representation by which successive cohorts of initiates in their autonomous ‘caves’ apprehended and communicated their mysteries. These surely are the hyponoiai, the ‘under-thoughts’ which, as we saw in Chapter 4 (sect. 10), Plutarch so perceptively listed together with the ‘images’ (eikones) and ‘imitations’ (mimēmata) as the gifts of another mystery-cult deity to her initiates.

As a benign consequence, much that was in dispute can now be deproblema-tized. For instance, the issue of ‘generalizing about Mithraic doctrine from unusual monuments’ which we addressed in Chapter 4(sect. 9) becomes less urgent. We still want to explore whether or not an unusual monumental feature points to a more widespread element in Mithraic thought, but whether it is orthodox or heterodox is not at issue. First and foremost, it is the product of negotiation, explicit or tacit, between the representation-forming minds of those particular initiates in that particular group. It is what they thought consonant with their mysteries then and there. We return to our point that Mithraic doctrine is whatever accords with what a Mithraic Father thinks is Mithraic doctrine. Since there was no overarching objectified code to which appeal could be made, merely norms perpetuated for the most part in iconography and the design of sacred space, there was little constraint on innovation. The surprise is not the occasional variant, but the unpoliced integrity, amounting almost to unanimity, of the mysteries over such an expanse of time and territory.

The cognitive approach also lets us deproblematize what appear to be doctrinal contradictions. For example, I have referred to the paradox that Mithras both (p.96) is and is not the Sun: is, because the Unconquered Sun is his cult title; is not, because Sol is a separate person in certain episodes (notably, the banquet following the bull-killing) in which the two deities participate. Cognitive theory predicts, and can demonstrate experimentally, the generation of such contradictions in the formation and transmission of religious concepts (Boyer 2001: 78–89).7 Although their contents may defy logic, their genesis in the mental representation of supernatural beings is a normal enough process psychologically. Consequently, while they may be described and explored, there is nothing to explain or resolve—unless the religious group itself chooses to problematize the paradox, as Christianity notoriously has done with, for example, the dual nature of Christ.

4. Reintegrating The Wise And The Vulgar

Now to the second dividend paid by the cognitive approach. Here I shall risk a value-judgement. To some extent I have already done so, in that I have suggested that Mithraism, in eschewing explicit doctrine and yet successfully transmitting doctrinal norms by instantiation in the symbol complexes of its monuments, was in this regard superior to Christianity. That judgement entailed of course a more general judgement about the baleful effect of formulated doctrine as a cause of discord—and worse. If you consider explicit doctrine benign or at least harmless, my claim for Mithraism’s superiority will not have been persuasive.

Now, however, I make a larger claim, not for the mysteries of Mithras but for the paradigm of religion which the cognitive approach implies. In locating ‘religion’ in the representation-forming minds of its adherents, the cognitive approach emphasizes its creative and egalitarian strain.

That assertion might surprise both the cognitivists and their opponents. The cognitive approach is necessarily associated with evolutionary psychology. We form our representations as we do because the minds/brains of our hunter-gatherer ancestors evolved as they did. So we are prisoners of the mental/neural adaptations they underwent, doomed to repeat their thinking and behaviour until our post-Neolithic age has clocked sufficient tens of millennia to generate and select for further mind/brain adaptations more suited to an environment which will then itself be slipping inexorably into the irrelevance of the past.

(p.97) This gloomy scenario may be countered in one of two ways. Opponents of evolutionary psychology reply that because we have reached a level at which we can communicate and store information transgenerationally, societal evolution has more or less replaced biological evolution and so emancipated our species from the snail’s pace of the latter. A better response, so it seems to me at least, is to retain evolutionary psychology’s model, since it is supported by a growing body of hard neurological evidence, but to set aside the entailment of determinism with which it is invested. That our brain/mind adaptations were selected for in our hunter-gatherer ancestors manifestly does not condemn our inference systems, thus evolved, to think only hunter-gatherer thoughts and to solve only hunter-gatherer problems until at long last natural selection rolls out a phenotype better adapted to its environment.

Let us then postulate a measure of individual human autonomy within the broad scenario modelled by evolutionary psychology. There is no contradiction, unless one is committed a priori to determinism as a universal principle.

Evolutionary psychology’s theory of mind frees us from another and more insidious tyranny, societal determinism. This you may readily appreciate by reviewing the two models, ‘Integrated Causal’ and ‘Standard Social Science’, presented earlier in this chapter. It is of course true that we are largely conditioned by and into our social systems—what else, after all, is education? But that we are entirely the product of these systems, that our minds are virtually empty slates on which our cultures inscribe themselves, this we need not—indeed should not—accept; for psychological research shows otherwise.

A religion, then, while it lives, is a dialogue of minds: admittedly, a lopsided dialogue in which the voice of authority preponderates, whether the actual voices of leaders and exegetes or the recorded and materialized voices of sacred text, creed, liturgy, artefact, and so on; but a dialogue nonetheless, because the representation-forming minds of the led are as necessary to it as the representation-forming minds of the leaders.

Our cognitive approach focuses, in an egalitarian way, on individual minds calibrating their representations in negotiation, mostly below the level of conscious intent, with other minds in the culture and environment of Mithraism—which itself is the product or instantiation of those negotiated representations. In so doing, we return some measure of agency to the rank and file, breaking the mortmain on the mysteries assigned to the ‘wise’ in antiquity and in modern times to the system itself as socio-cultural construct. We return ownership, not of course to the led alone, but to the members, the ‘hand-claspers’ (syndexioi), each in his proper status, gathered in their ‘caves’.

I take this position not just to indulge a sentimental egalitarianism, but to correct a ‘folk theory of mind’8 which lurks unexamined among the learned, (p.98) though fortunately less so now than in earlier generations. This folk theory of mind assumes that the less learned are substantially less sophisticated, less discriminating in their mental representations, than the more learned. As a purely aesthetic judgement about informed and uninformed taste, this is perhaps so; likewise as an intellectual judgement about understanding an argument and failing to understand an argument. But as a cognitive judgement about processes of representation and inference it is nonsense—and pernicious nonsense at that. In religion and in the study of religion, as we saw in Chapter 4, it licenses the spurious separation of the ‘wise’ and the ‘vulgar’ on which the ancients harped and a no less spurious class dichotomy which modern scholars of antiquity occasionally still apply.

In contrast, the cognitive approach adopted here treats religion in general, and of course ‘mysteries’ in particular, as works continuously in progress, re-created across the generations as their members, leaders and led alike, fashion their mental representations by interaction both with each other and with the religion’s external memory archived in text, ritual, symbolic iconography, and so on. In this way the cognitive approach, as I claimed above, emphasizes that which in religion is creative and egalitarian.

As it happens, Mithraism towards the end of its historic lifespan furnishes an excellent example of a religion in which ownership, through an implicit claim to monopolize the agenda of representation, gravitated to the leadership. The lost set of inscriptions V400–5 from the S. Silvestro in Capite mithraeum in Rome records a series of grade initiations undertaken by two members of the pagan senatorial aristocracy, Nonius Victor Olympius and Aurelius Victor Augentius, between the years 357 and 376. The former was the latter’s (biological) father; in the mysteries he held the rank of ‘Father of Fathers’ (pater patrum) and his son that of Father. By 376 Aurelius Victor Augentius had ascended to his father’s rank, probably on the latter’s death, since in the inscription of that year (V403) his father is not mentioned. What is so interesting about these inscriptions is what they do not say. There is no mention of Mithras, and there is no mention of those inducted into the various grades—with one exception: in 376 (V403) Aurelius Victor Augentius initiates his 13-year-old son Aemilianus Corfinius Olympius into the initial grade of Raven. Clearly, what the mysteries were largely ‘about’ for this noble family was the noble family itself, not Mithras, not the cult brothers, but themselves. What it was ‘about’ for the rank and file of this cult group, presumably composed largely of the family’s clients and household, we can never know. But one may reasonably conjecture that appropriate representations of the patronal hierarchy were encouraged. This is not to say that the mysteries practised under this noble family’s aegis were less genuine, less alive, than those of earlier times and other places. Rather, it is to infer from the evidence of surface symptoms a change in how the initiates represented the mysteries to themselves and each other.

(p.99) Appendix: Comprehending the Pantomime: Lucian, on the Dance

The capacity to form representations of non- or supernatural beings, we have accepted, is a constant of the human mind, and it does not operate exclusively in some special religious domain. As an example from antiquity I gave the audience’s representation of supernatural beings at the pantomime in first-century CE Rome. The example of Roman pantomime was not chosen at random, for it also helps to give the lie to the assumption challenged in the preceding section: namely that cognition in the learned differs radically from, and is superior to, cognition in the unlearned. (To avoid confusion latent in the term ‘representation’, note that I am not of course talking about the artist’s public and performative ‘representation’ of the supernatural being, but of the audience’s inner mental representations prompted by the artist’s representation in performance.)

The pantomime, as an art form in ancient Rome, was both hugely popular and very demanding on the audience because of the extreme artificiality of its conventions. To comprehend the pantomime, to ‘get it’, the audience had to run, simultaneously and subconsciously, an array of mental programs (to use the computing metaphor) to coordinate and translate into a unified and meaningful experience an audio-visual input quite remote from the input that they would receive if viewing/hearing the corresponding events in real time and real life. To the artificialities of other forms of ancient drama (principally tragedy and comedy) the pantomime added the convention of divorcing the spoken or sung word from the physical action, with the primary focus on the latter. The pantomime—the term properly belongs to the actor rather than the genre—conveyed the action by dance and gesture alone, accompanied by music. Words sung by a choir or spoken by a narrator were optional and strictly unnecessary in that to rely on them for communication of meaning would be considered an artistic failure. The pantomime was of course masked, and so could convey nothing by facial expression. Moreover, the acme of his art was the ability to carry all the roles sequentially, even of characters who in real time would be interacting concurrently.

The pantomime is irretrievably lost, since, unlike tragedy and comedy, such fragments of texts as have survived are obviously of slight importance to the art. We do, however, possess some testimony to the pantomime of much greater value to our present purpose: a comprehensive and intelligent critique of the pantomime as an art form, written while it was still performed. This work, composed as a dialogue, is entitled On the dance (De saltatione); it was written by Lucian, a satirical essayist of the second century CE. The work is particularly relevant to our present topic, for like much ancient criticism of drama it is more concerned with how the audience comprehends and responds to the performance than with a formal definition and description of the art.

Lucian makes it clear that the mass audience of the pantomime did indeed ‘get it’: that is, they reacted in a way that, to a cognitivist, shows that the inference and data-processing systems of their minds, operating for the most part below the conscious level, took the pantomime’s highly complicated and unnaturalistic conventions in their stride. They translated effortlessly the stylized movements and gestures of a single silent performer into a sequence of meaningfully related interactions and verbal exchanges between a multiplicity of persons. Lucian tells the story of an unnamed pantomime who in Nero’s reign (p.100) undertook to refute the charge that his art owed its success solely to the accessories (‘the silk vestments, the beautiful mask, the flute and its quavers, and the sweet voices of the singers’) by performing without music and chorus:

Enjoining silence upon the stampers and flute-players and upon the chorus itself, quite unsupported, he danced the amours of Aphrodite and Ares, Helius tattling, Hephaestus laying his plot and trapping both of them with his entangling bonds, the gods who came in on them, portrayed individually, Aphrodite ashamed, Ares seeking cover and begging for mercy… (De salt. 63, Loeb trans.)

The challenger, a Cynic philosopher named Demetrius, withdrew the charge, shouting to the pantomime still on stage: ‘I hear the story that you are acting, man, I do not just see it; you seem to me to be talking with your hands!’ The point, for us, is that Demetrius now ‘gets’ and verbalizes what the groundlings have ‘got’ or apprehended all along.

Lucian has much to say about the way in which ordinary people respond to the pantomime, particularly about their emotional responses, of which he—or perhaps we should say, the speaker whom he privileges in the dialogue—generally approves, thus accepting the moral and educative value of the performance (De salt. 72, 79, 81). People weep when they rightly should, and they feel and express indignation when they rightly should. This, of course, they can only do if they have comprehended the intent of the highly abstracted and artificial show which their eyes and ears have taken in. Successful cognition is the necessary precondition for appropriate affective response.

Another indication of successful cognition among the groundlings can be seen in their banter with the performers. Again, Lucian (or his dialogue persona) treats this not censoriously but positively as a form of popular quality control by which ‘entire peoples’ (holoi dēmoi)—he is speaking here specifically of the Antiochenes—‘regulate (rhythmizein) its [i.e. the dance’s] good and bad points’. The examples he gives seem quite trivial:

When a diminutive dancer made his entrance and began to play Hector, they all cried out in a single voice, ‘Ho there, Astyanax! where’s Hector?’ On another occasion, when a man who was extremely tall undertook to dance Capaneus and assault the walls of Thebes, ‘Step over the wall’, they said, ‘you have no need of a ladder!’ (De salt. 76, Loeb trans.)

However, for all their simplicity, these and other examples indicate a high level of cognitive sophistication. Together, the anecdotes demonstrate the audience’s ability not only to comprehend the performance but also to discriminate consciously between that which is represented in performance and the performative representation. They heckle because they can detect, and consciously represent to themselves, an amusing dissonance between character and actor. Metatheatricality, it seems, was well understood by the dregs of the Orontes, and presumably of the Tiber too. To reach that level of conscious appreciation, the mind—of the learned and unlearned alike—must first have developed a massive capacity for complicated feats of cognition below the threshold of reflective consciousness.

I offer one final example of audience response, because Lucian uses it to contrast the reaction of the vulgar with the reaction of the refined. It is the story of a pantomime who got carried away by his own performance of the mad Ajax:

He tore the clothes offone of the men who beat time with the iron shoe, and snatching a flute from one of the accompanists, with a vigorous blow he cracked the crown of Odysseus, who was standing (p.101) near exulting in his victory….9 Coming down among the public, he seated himself among the senators, between two ex-consuls, who were very much afraid that he would seize one of them and drub him, taking him for a wether. (De salt. 83, Loeb trans.)

The crowd, as the saying goes, went wild, ‘leaping, and shouting and flinging up their garments’. Lucian (or his dialogue persona) treats it as a case of triple madness: the actor, miming the madness of Ajax, goes mad himself, and the ‘riff-raff and absolutely unenlightened (surphetōdeis kai… idiōtai) went mad along with him (synememēnei)’. However, Lucian has already said far too much to make this diagnosis credible, for all that it is rhetorically effective and conforms to class stereotypes. More likely, the groundlings comprehended the actor’s transgression of performative boundaries just as well as did the ‘politer sort’ (asteioteroi) who Lucian says ‘understood and were ashamed’, covering their embarrassment with lukewarm applause rather than stony silence. The vulgar, I suggest, understood the transgression perfectly well; they thoroughly enjoyed it and entered into its spirit. If ‘Ajax’ could break convention and export his madness into the real world, then the audience could reciprocate from the real world by ‘madly’ entering the performance. The audience didn’t go mad with Ajax, they played mad with ‘Ajax’.


(1) Notably S. E. Guthrie (1993, 1997).

(2) For an overview of this new approach, see Tooby and Cosmides 1992.

(3) In this sort of context, ‘nervous system’ includes, but is not limited to, the brain (my footnote).

(4) The remaining four propositions (8–11) in effect elaborate the first seven. Though useful, they can be omitted here without prejudice to the SSSM or to its alternative, the ICM.

(5) Above, Ch. 1, sects. 1–2.

(6) Boyer 2001: 46–7. Dan Sperber (1996) treats cultures, and a fortiori religions, as epidemics of mental representations. He intends ‘epidemics’ literally, not metaphorically.

(7) Note the experiment (p. 88) in which a group of subjects was asked first to read stories about an omnipotent and omnipresent god who (e.g.) saves a man’s life and simultaneously helps a woman find her lost purse, and then to retell the stories. Without giving up the divine attributes of omnipotence and omnipresence, ‘many subjects said that God had helped one person out and then [italics in original] turned his attention to the other’s plight’. The observed effect was the same with believers and non-believers and in India and the USA. (Not without interest is the gender stereotyping both of God and of his imagined beneficiaries, so easy is it to represent a male god and a female human losing her purse.)

(8) I use the term in its technical sense of a non-specialist commonsensical set of working assumptions about the way other minds—and by inference one’s own—operate.

(9) This performance clearly involved more than a single dancer.