Anthropological Approaches

[Final draft of an essay by the same title published by Oxford University Press in The Oxford Handbook of Greek and Roman Art and Architecture, edited by Clemente Marconi, 2015, 621-636.]

§1. Anthropology has had an important role in studies of Greek and Roman art, albeit a role marked by such discontinuity and eclecticism that it does not lend itself to be described as a set of principles. What is possible is a historiographic overview prefaced by some general considerations.

§2. To oversimplify, it may be said that recourse to anthropological models has been a means to address the monuments as documents of the cultural milieu that brought them into being, including not only historical circumstances but also social structures, cultural norms, and their attendant belief systems. The task of charting connections between the monuments and other sources, including texts, in order to expose the network of meaning and practices which the artifacts punctuate, is traditionally carried out in a synchronic perspective. Any anthropological account makes explicit the theoretical underpinnings of its inquiry and acknowledges the distance that separates the society and culture of the observer from those of his object—that is, the measure in which they are alien to one another. [1] In the case of Greek and Roman art, instead, it is continuities and aesthetic affinities that have been stressed, in the reconstruction of a tradition of Western art reaching far into antiquity—the proper subject of art history. [2] In other important respects the methodology dominant in the 20th century marked classical art as different from that of other peoples: taking an atheoretical stance, it addressed the formal characteristics of the monuments in empirical terms and arranged them on the basis of stylistic criteria, typology, and the achievements of great artists. For the modern period, scholars have traced the antithesis of these two approaches to the distinction in Vasari between accounting for a work of art semplicemente, in absolute terms, or secondo che, that is, “with respect for places, times, and other similar circumstances.” [3] The two positions, writes Ginzburg, “are both necessary and mutually incompatible; they cannot be experienced simultaneously.” He adds, however: “But the relationship between the two approaches is asymmetrical. We can articulate the ‘simple,’ direct, absolute approach through the language of history—not the other way around.” Nowadays efforts to bridge the divide seem to move precisely in that direction. But it is a dynamics of exclusion that has characterized the adoption or rejection of anthropological perspectives in classical archaeology. The need to choose between them is apparent in what is arguably the earliest brush of a classical archaeologist with anthropological theory, the case of Carl Robert.

§3. A work of the young Robert, Bild und Lied, published in 1881 presented a comparative analysis of texts and images in terms of narrative structure and relied on a rather bold and clearly stated theoretical premise: the concept of a complex of ideas, customs, and representations that is the common property of the society to which any individual belongs. This Volksbewusstein, folk-consciousness or, as it is commonly known, ‘collective consciousness’, was for Robert the reservoir of Volksvorstellungen, or ‘collective representations’, into which single instances of representation, such as those of poetry and in the visual arts, flow, and from which, in turn, they spring anew. [4] The relationship of images to texts is thus projected as indirect, through the relay of cultural assumptions and modes of representation. Turning to modes of narrative, Robert laid the foundation of a problem that has occupied art historians ever since, that of the structure of narrative in the visual arts and literary genres. In archaic art he isolated several instances of representations in which several stages of a story seem to be compressed into one frame. In a scene of the blinding of Polyphemus on a Laconian cup (Fig. 1), for instance, a viewer intent on following the chronological order of events may find it strange or even comical that the Cyclops cannot accept the wine cup Odysseus is offering him because he holds in either hand the legs of the men on whom he is feeding; and, although he is neither drunk nor asleep, he allows the men to take out his eye. The issue is one of structure: the visual narrative, Robert argued, here emulates the epic, and attempts to give the complete story in a synoptic mode by introducing in the picture elements that evoke an earlier or a later stage of the story. Pictures in which such references are lacking, those depicting a single event, would instead have a structure akin to that of drama. The relationship between the visual and the verbal was thus formulated in terms of a structural analogy.

§4. Were the notions of Volksbewusstein, ‘folk-consciousness’ or ‘collective consciousness’, and Volksdarstellungen, ‘collective representations’, on which Robert relied entirely unrelated to those of conscience collective and représentations collectives that are fundamental to the sociology of Durkheim and his colleagues and successors? These terms have a technical ring and the correspondence is too specific to be a matter of coincidence. [5] Robert credits no one for the idea, but Durkheim did, although much later, reluctantly. In the Elementary Forms of the Religious Life, in his discussion of mythology as a system, he refers to the founders of the Zeitschrift fur Völkerpsychologie und Sprachwissensschaft, Moritz Lazarus and Heymann Steinthal, as the men who had put forward the theory that the rules that govern human societies are dependent upon historical and social factors. [6] It is to these two that one owes the proposition of an over-individual psyche, as it were, the spirit of the nation, which is the depository of cultural norms and systems of representation. [7] For their proposal these scholars drew on the one hand on Hebartian psychology and on the other on Humboldt’s philosophy of language in an ambitious formulation that can be seen as the basis of modern cultural anthropology. [8] In a society, the conceptual activities of individual members form of necessity a system of collective representations that have an existence beyond and outside the individual. The collective consciousness informs and confronts that of the individual. Its “elements” are, first of all, language, Sprache, which embodies the psychological forms of thought, a thesaurus of representations and concepts informing mythology, religion, national poetry, and customs, but also monuments and the visual arts. The following statement uncannily foreshadows some recent proposals for the indexical power of artifacts: [9]

My purpose in focusing at the outset on Carl Robert’s appeal to Völkerpsychologie (albeit without acknowledgement) is twofold: to point out on the one hand its fundamentally anthropological outlook and on the other the far-reaching impact of this concept. “Riddled with conceptual and methodological problems” as it was, [10] Lazarus and Steinthal’s project laid out a new epistemological paradigm. It is against that paradigm that the underlying coherence of theoretical developments, which otherwise may appear unrelated, becomes evident: not only Durkheimian sociology, leading up to the “anthropology” of Louis Gernet, [11] but also the historical psychology of Ignace Meyerson, whose insights Vernant applied to the classical world, [12] as well as Saussure’s structural linguistics, [13] which is the basis for structural anthropology as well as semiotic approaches.

§5. As for Carl Robert, it may be objected that his attempt at a structural interpretation of poetry and image in classical antiquity is little more than a historiographic curiosity: a false start to a dead end, because its connection to the matrix from which it issued was obscured. But the issue is far from dead. Several scholars have turned their attention to the model he set forth for the tantalizing promise of cultural insights it seems to hold. Anthony Snodgrass, for instance, appeals to Robert in the Sather lectures of 1987, where he returns to the issue of the structure of visual narratives, focusing, among others, on the same Laconian cup featuring the blinding of Polyphemus to ask if elements that allude to the cannibalism and drunkenness of the Cyclops have narrative valence or add connotations to the image. [14] But on the promise of the approach taken in Bild und Lied, Robert had firmly shut the door in his Archäologische Hermeneutik of 1919. The epigraph with which that book opens explicitly excludes from the correct method in classical art history any attempt at theoretical formulations, as though empiricism were devoid of theoretical implications:

This adherence to a staunchly positivist approach is reinforced by statements such as: “the first necessary condition for the correct interpretation is to see correctly,” and “meaning becomes self-evident through a careful description.” [15]

§6. By the authority of giants such as Adolf Furtwängler and John Davidson Beazley, from the end of the 19th century onward the formalist approach, relying on taxonomy, stylistic analysis, and attribution to known as well as unknown masters, had become the dominant interpretive mode for both sculpture and the painted vases, although more so in England, Germany, and North America than in France and Italy. That model would not be seriously challenged until after Beazley’s death in 1970. In the intervening period projects relying on anthropological theory developed not in the mainstream of classical archaeology but in the history of religion, most prominently in the work of the so-called Cambridge Ritualists, particularly Jane Harrison, and that of Angelo Brelich. Although it is not directly concerned with art, his Paides e parthenoi reveals the possibilities that a historically grounded anthropological frame of reference opens up for the interpretation of the monuments. Brelich opens his treatise on rituals marking the passage of boys and girls into the adult community in ancient Greece by establishing a morphology of initiations on the basis of comparative ethnology. He then moves on to examine the rites in a historical frame, integrating in his analysis literary sources for myth and ritual and epigraphic and archaeological materials. Once they are situated in a specific social transaction and placed in relationship with other forms of expression, the monuments elicit a different, productive set of questions about their status and function. That is the case with the kouroi, to give one instance. The type is known by hundreds of examples: a youth, not yet a grown man, striding forward, nude (Fig. 2). Outside of the art historical frame of reference in which these statues had been studied, on the basis of ethnographic parallels and numerous references to disrobing in written sources on the occasion of coming of age, Brelich identified nudity as an initiatic “costume,” suggesting for the kouroi, albeit tentatively, a commemorative function in the context of the rituals. [16]

§7. The model of a historically and contextually grounded analysis that Brelich offered had little following outside the “Rome School” in the history of religion. The emergence of anthropological approaches to classical art (mostly Greek) bears in greater or smaller degree a “structuralist” imprint—although that label is in most cases an approximation. Proposals came in rapid succession from several unrelated directions in the late 1970s. The move was apparently triggered on the one hand by frustration, a sense that the tyranny of formalist methodology had driven the discipline into a blind alley and rendered it increasingly isolated and inaccessible. [17] On the other hand it responded to developments in critical theory that presented new challenges in the allied fields of classics and art history. Herbert Hoffmann’s 1977 presentation to the Royal Anthropological Institute in London provided the opening salvo. [18] The brief essay focused on a particular type of vases, Athenian askoi, small pouring vessels decorated with a variety of scenes: animals pursuing animals, hunts, amorous pursuit and sex, mythical subjects, banquets. The essay confronted important issues that are still unresolved: is there an underlying coherence to this bewildering assortment? And what makes these images appropriate to the particular shape on which they appear, that is, to the function of the vessel? In his attempt to answer these questions Hoffmann turned to semiology and anthropological theory, specifically, the structuralism of Lévi-Strauss, albeit mediated by Edmund Leach. The model on which the interpretation rests is that of structural linguistics and its governing notion that language underlies, and is distinct from, individual acts of speech: it a culture-bound system that encodes fundamental categories of thought, arranged in patterns of binary oppositions. Accordingly, “every part is a variation of every other part and [ … ] reproduces the whole on the level of significance.” [19] In the case of the askoi the varied imagery would consist of “transformations” of the same theme, that of sacrifice, mediating between opposites: life and death, the human and the divine. Such a theme, Hoffmann argued, is appropriate to the function of the vessels in funerary rites. [20]

§8. Two years later an essay by Christiane Sourvinou-Inwood also challenged the prevailing empiricist method in the study of Greek art. [21] Her approach, which was there and throughout her work frankly eclectic, is not easily classifiable as either structuralist or anthropological. It adhered, however, to one of the tenets of cultural anthropology in its insistence on the distance that stands between ancient Greek culture and its modern interpreters and on the need to recover the assumptions and expectation of the ancient viewers, their “perceptual filters.” [22] But it was the work of scholars associated with the Centre Louis Gernet, with its declared anthropological methodology, that had the greatest impact on research on classical art, especially on the area that had long been the preserve of connoisseurship: Archaic and Classical Greek vase-painting.

§9. Through the “historical anthropology” of Gernet, what has become known as the “Paris School” (malgré soi, I think) stands directly in the tradition of the sociology of Durkheim, from whom Gernet drew the fundamental operating concepts of collective consciousness and collective representations. [23] It would be hard to overestimate the importance of the notion of representations in his thought, emphasizing the conceptual import of the concrete aspects, such as gestures, of social activities in relationship to the institutions they sustained. But it was Vernant and the scholars associated with him who brought actual monuments into play—not actions and things described in literary texts but statues, vases, inscriptions, and pictures. Vernant’s own inquiry into the visual universe of the Greeks has addressed both philosophical conceptions of figuration and mimesis and the way in which visual images and artifacts articulate concepts. I think in particular of the essay on the kolossos as a concept and as an artifact, in Myth et pensée, which illuminates the way in which the monument represents the visible double for the invisible dead. [24] Elsewhere he addresses more generally the task of defining the conditions that allow archaeological documents, epigraphical testimonies, and literary texts to throw light upon one another in reciprocal fashion.

§10. A landmark in the activity of the scholars at the Centre Louis Gernet, [25] in collaboration with colleagues at the Institut d’archéologie et d’histoire ancienne of the University of Lausanne, was La cité des images, a traveling exhibition of Greek vases accompanied by a book of essays. [26] The project represented a clear démarche with respect to current trends. Instead of the objects themselves, bearing the imprint of the master’s hand, photographs of the vases were displayed. The setting was not exclusively museums or private collections, but rather more public venues that extended to metro stations. And the commentaries that accompanied the pictures and the essays in the book made no mention of the painters, and showed little concern for either artists or style. Unlike earlier attempts at structural analysis, [27] the book presented the reader—without apologies and without much warning—not with an argument but with a fait accompli: a new frame of inquiry and a different vocabulary in which to talk about the vases. The outlook of the project was expressed by a metaphor emphasizing the distance that separates modern interpreters from the world of ancient Greece, turning them into anthropologists. The imagery in its entirety is cast as a “city” populated by the collective representations of its inhabitants, and to us an “exotic and distant world,” which we enter as strangers. [28] This was joined by the concept of visual representation as language, and so a code unintelligible to someone lacking the local knowledge that the members of the city shared. The model of structural linguistics was the interpretive paradigm explicitly or implicitly acknowledged by all the participants, although in different measures. In Saussure’s formulation, language is a system of signs that express concepts in an entirely conventional manner and, in its entirety, belongs to the given social community of speakers. Language is actualized in acts of speech, in which a sign is joined to other signs (the syntagmatic chain) or may be substituted by another, with which it stands in a paradigmatic relationship. In this regard, there are differences among the members of the team. The semiotic approach, in the formulation given by Claude Bérard, may be briefly summarized as follows: the artist manipulates a set of icons (“minimal formal units”) that are many but finite in number. These “refer” to objects one may think of (parts of the body, features such as dress or hairstyle, attributes) but have no meaning in and of themselves. They are simply available elements—comparable to the phonemes and morphemes of language—in the minds of the painters and the viewers. [29] They acquire meaning only in combination, or syntagmatic association. Accordingly, interpretation should proceed by observing the combinations and substitutions of such units against the background of the entire repertoire of icons. In contrast, François Lissarrague adopts the model of structural linguistics as a useful heuristic device but one that cannot be applied strictly in the interpretation of pictures. [30] His sustained study of the theme of war on Athenian vases proceeds by identifying anthropologically significant categories—such as the hoplite, the archer, the woman, the satyr—to analyze their occurrences in thematically related series, with the aim to establish their semantic range as well as binary oppositions.

§11. The insight gained from the analysis of the pictures is then related to representations in other forms of cultural expression. I give an example not from La cité des images but from a seminal article by Lissarrague and Schnapp, which analyzes the representations of the death of the hero, the moment in which the body is carried away from the battlefield and saved from outrage. [31] Often the body is marked by its large size and in particular by the long, flowing hair. The typical iconographic scheme for the transport may be seen on the François vase (fig. 3), where inscriptions identify the characters as Ajax with the body of Achilles. Here as in many other instances of this theme, particular emphasis is placed on the hero’s hair. This, Lissarrague and Schnapp argue, is one of the ways in which the concept of the heroic death, such as the “Spartan beautiful death” is given visual expression: the long, flowing hair is a signifier of the beauty of heroic death. One finally understands the sense of an episode reported by Herodotus (7.208): before the Thermopylae, a Persian scout was startled by the sight of Spartan warriors combing their hair before going into battle. By this gesture they ready themselves to become beautiful corpses, thus advertising their determination to fight to their death. In this manner, the visual representations, poetic representations of the ideal death on the battlefield, and, if Herodotus is to be believed, actual customs converge. One is not the illustration of the other, but complementary and inseparable in the mind of the viewer, calling up series of other related images, gestures, and customs.

§12. The possibility that visual imagery functions in a manner analogous to language—as had been postulated in Carl Robert’s analysis of the structure of narratives—is illustrated by Françoise Frontisi-Ducroux’s comparative analysis of verbal and visual tropes in Du masque au visage, particularly her interpretation of the frontal face in Archaic and Early Classical painting. [32] The frontality that characterizes the mask and particularly the Gorgon is the exception in representations of human figures. The frontal face confronts the viewer in a way that represents an infraction of the norm, and its pointed connotations are stressed by the fact that often it is the face of characters who are dead or about to die, such as the Amazon squire in an Early Classical painting of the duel of Achilles and Penthesilea (Fig. 4). But it may also be the face of someone asleep or a symposiast in his cups, or a whirling dancer, or a man in a funeral procession, turned around to face the viewer. The figure thus turns away from the figural context into which it is embedded, in a way that is structurally analogous to the rhetorical figure of the apostrophe: the point at which, for instance, an orator would turn from addressing the jury to confront his adversary directly. Moving to an instance of this trope at Iliad 16.787, where the poet turns to his character with a direct address (“then for you, Patroklos, the end of life appeared”) Frontisi-Ducroux observes (92): “the analogy between the rhetorical figure of the apostrophe and the graphic procedure of frontality is made evident by their application to the same motif, the death of the warrior, sung in the epic on the one hand, depicted on the vases on the other.” If frontality is the means to represent the evasion, or falling out of the figure from its surroundings, one may also understand in the same sense other cases in which the figure is used for other extreme states, such as madness, drunkenness, and ecstatic dances.

§13. The impact of this approach, relying on structural linguistics in an anthropological perspective, has been considerable, so much so that it was recently characterized as a “time honored tradition of dealing with Greek art as if it were a language, a codified system of signs available to our reading skills.” [33] Even as its premises gained acceptance, however, there emerged problematic aspects of a more or less strictly structural reading. The focus on the conceptual structure informing a number of different and varied representations—whether myths or pictures—is ultimately reductive, it is argued, because it leads to neglect of the object itself: its details, concrete aspects, the circumstances of its production and reception. This charge is in line with the critique of structuralism that began in France fifty years ago, exemplified —to remain within structural linguistics—by Émile Benveniste’s influential essay of 1964. [34] Examining the levels at which linguistic analysis may be carried out, Benveniste pointed out that one moves from that of phonemes to that of morphemes, or words, then reaches the level of the phrase. But, while the signs that are its components are part of the language system, the phrase is the site of creativity, “language in action:”

§14. As a segment of discourse the phrase has a particular content, deployed in a specific frame of reference—a given situation—and it implies an audience. Discourse, in other words, takes place in the social domain and it involves much that is outside of language. Let us take this insight to the analysis, cited above, of a specific visual “phrase,” the figure of the warrior carrying the body of his dead comrade on his shoulder. The authors recognize in this group Ajax and Achilles if their names are inscribed, as they are on the François vase (Fig. 3). If there are no inscriptions or other explicit elements that mark the picture as epic, one is free to interpret the figures as either heroes or historical hoplites, since the concept of the hero applies equally to either. But the specific “content’’—whether the image refers to the age of the heroes or the present of the intended viewer—indeed matters if one is to account for its discoursive context, that is, the frame of reference within which it is issued. Arguably, at least some things that may be imagined in a distant place or time are unimaginable in the here and now. [35]

§15. More broadly, criticism has been leveled at insufficient attention paid to the particular historical and social circumstances of cultural production. For instance, Tanner voices this concern after considering Alain Schnapp’s analysis of representations of the hunt, one that he finds persuasive. He faults it, nevertheless, for not doing enough, that is, for not accounting for the fact that the subject occurs in visual images on Attic black-figure vases of the sixth century BCE while literary accounts date to the fourth century: [36]

The call for more “contextual” and historically grounded approaches to the monuments has been accompanied by attempts to reclaim in a postmodern vein what had become largely discredited fundamental principles of traditional art history, namely: stylistic analysis and the definition of “art” as the proper object of classical art history. As regards the first, Hölscher made a strong case for its importance already in the wake of the “structuralist” turn of the 1970s and 1980s: [37]

§16. The issue of what constitutes “art” as a privileged category of cultural production that cannot be subsumed under rubrics such as “material culture” or “visual culture” has been the subject of much discussion in recent years and with it the question of the value of aesthetic appreciation, and the very possibility of a history of art. [38] Tanner, for instance, put forward a most sustained effort to arrive at a working definition of what is considered art: the creation that serves no function outside of itself but has a purely expressive-aesthetic valence. [39] In this respect such an object differs from monuments whose production is tied to religious or other social institutions but it should nevertheless be approached in a sociological and anthropological frame of inquiry. The question of what “art” is and how it operates in society has also recently come to the fore in the field of anthropology. [40] Most notable, and controversial, here has been Gell’s proposal of art as artifact endowed with agency and operating in the social realm. [41]

§17. My aim in introducing at the end these authoritative examples of the revaluation of the essential underpinnings of traditional classical art history is not to engage with them, nor do I pretend to do them justice by such brief remarks. I wish simply to point out that what they advocate is not a return to the past. Rather, each conceives of the production of art as cultural expression and social practice. Such a premise also underlies several other approaches to classical art that are the subject of this section of this Handbook—socio-historical, semiotic, gender studies, reception theory, contextual—that emerged over a generation ago, appealing to different sources. That they share a common ground, I would argue, eventually depends on a common, however distant, matrix in Lazarus’ and Steinthal’s sociological paradigm, with which we began.