Pub. Date:
Cambridge University Press
The Language of Images in Roman Art

The Language of Images in Roman Art

Current price is , Original price is $41.99. You

Temporarily Out of Stock Online

Please check back later for updated availability.


Tonio Hölscher develops a new theoretical concept for understanding the Roman art of images by establishing a connection between artistic forms and content and expressions of ideology--such as the glorification of state and ruler, war and triumph. A large role is played here by the reception of earlier images from Greek art. Roman art therefore appears to operate as a semantic system which, from an interdisciplinary perspective, can be compared with the forms of Roman literature as well as the language of images of other cultures.

Product Details

ISBN-13: 9780521665698
Publisher: Cambridge University Press
Publication date: 11/18/2004
Edition description: New Edition
Pages: 188
Sales rank: 364,306
Product dimensions: 6.46(w) x 8.94(h) x 0.51(d)

About the Author

Tonio Hölscher is Professor of Classical Archaeology at the University of Heidelberg. His publications focus on public monuments, political iconography and urbanism in Ancient Greece and Rome and on general art and cultural theory. His is a member of various scientific institutions, including the Academia Europae, London.

Anthony Snodgrass is Emeritus Professor of Classical Archaeology in the University of Cambridge whose books include Homer and the Artists (Cambridge University Press, 1998).

Jas' Elsner is Humfry Payne Senior Research Fellow in Classical Art and Archaeology in the University of Oxford. His books include Art and the Roman Viewer (Cambridge University Press, 1995) and Imperial Rome and Christian Triumph (1998).

Read an Excerpt

The Language of Images in Roman Art
Cambridge University Press
0521662001 - The Language of Images in Roman Art - by Tonio HÖlscher



My aim in this book is to examine the language of images in the art of the Roman Empire as an essential factor in Roman culture. Recent efforts to explore the political and social meanings of Roman figural art have tended to push issues of form and style into the background. By contrast, the problems of iconography and iconology have been emphasised - though the formulation of modern iconology by Erwin Panofsky (1892-1968) did not necessarily entail the one-sided implications of sidelining the significance of form. Certainly, as new art-historical questions are being formulated, we can carry on for a while without recourse to the traditional methods and analyses grounded in formalism. But it is my proposition that the sacrifice of formal analy- sis has much more serious consequences than is often appreciated - especially for social history. For few cultural phenomena have a more pronounced collective and social character than artistic style and the language of artistic imagery.

The exploration of stylistic forms offers a way of measuring cultural identity - not just in the case of individuals but especially of larger groupings, including entire cultures and epochs. It is no accident that history as a whole has borrowed so many of its temporal and cultural subdivisions from the history of art. The units of space and time which may be defined by the style of a group, period or region bear witness to the social interconnections of the original exponents of that style. There is no space here to explore these questions further, but their bearing on history in general has by no means been exhausted.

Moreover, the common visual language of a society - underlying the thematics of its imagery and regardless of minor temporal and local stylistic differences - is a social fact of the greatest interest. Among the fundamental themes for social history raised by the communicative aspects of the history of images are the following: how a society may coin a means of visual communication, how this language then reacts upon the society as it uses and develops it, what the overall visual system is able to achieve as a result, which structures of meaning are implied in its syntax and repertoire of motifs. All these are of real importance for social and cultural history.

The attempt to clarify these issues can only be convincing if it can claim a certain degree of universal validity. Yet - in the case of Roman art - among the obstacles is the fact that so much of the evidence, especially of 'idealising' sculpture, has hardly been examined and set out in a systematic way. Of necessity, in a short account like this, the burden of proof must be borne by only a very few examples. Indeed, despite all efforts to represent a fairly balanced range of works, it remains true that no one can yet claim to have a complete overview of the whole field today. It is the familiar dilemma: if one looks at the whole before its parts, the picture risks being out of focus, even distorted; if one reverses the order, it loses any general structure, becoming a jumble of parts. To my mind, the latter is the greater risk. But at the same time I must admit that this book is only a first, rough sketch.

My project is to explore the language of imagery in Roman art as a semantic system. This 'semantic system' functions according to a sort of grammar, on the basis of certain specific structures (like a language). I mean a system in the broad sense - not a consistent structure shaped by unifying principles but a flexible interplay of elements which together form a co-ordinated whole. This visual language was not consciously devised. Rather it evolved gradually and organically. This means there was a considerable degree of inconsistency in its expression. One must be cautious, therefore, in deploying exceptional cases as arguments against the validity of the system as a whole. Since no culture was ever designed to fit to scientific measurements and methods of research, academic study must develop models and points of view which account for the indeterminacies of cultural life.

It is difficult to assess the extent of the validity of the semantic system I propose in the following pages. That the phenomena exist can hardly be doubted in the light of Roman documentary sources (some of which I shall examine below). But it is equally clear that there are exceptions to how far we can apply the structures of the system I shall discuss. The limits of my suggestions about the language of Roman art and the extent of their applicability are questions which future research must clarify. My sketch needs to be tested not only by a review of additional monuments; it must also be enlarged beyond its present margins. For a 'grammatical' system of visual language can never encompass all manifestations of art in a culture. It can only be set alongside other phenomena - for instance, temporal changes in style and various formal constraints, such as concepts of space and body. The semantic system presented here was inevitably mediated by the development of style and by such pre-linguistic structures of form as space and body, and my theoretical presentation of it needs to be filtered through a theorisation of these issues.

It is illuminating to compare the visual arts with other aspects of cultural production, above all with Roman literature and literary theory. Modern theory (I have linguistics and semiotics particularly in mind) is also helpful. I have profited from the awareness of these possibilities but confess that I am far from having exhausted them.

This essay is, in fact, a very preliminary and open-ended sketch. My aspiration is that it will be assessed, clarified and (let us hope) enriched by discussion and criticism, rather than that it be set in stone. Similar reflections have, on occasion, been offered before. Instead of overburdening the text with numerous scattered notes, I shall mention a few authors who have been particularly important for me. Of prime significance are the projects of Otto Brendel (1901-73) and Peter H. von Blanckenhagen (1901-90), who between them first pointed the way towards a pluralistic understanding of Roman art.1 More recently, Paul Zanker has identified some characteristic features of the Roman language of images in his general sketch of Classicism in Roman Imperial culture. Likewise, Adolf H. Borbein has pursued similar questions in an unpublished lecture to the Mommsen-Gesellschaft, delivered at Würzburg in 1980.2 Otherwise, my approach derives - more perhaps than I am consciously aware - from day-to-day life experience: the modern restriction of personal initiative in daily life by an overriding tendency towards cohesion and social collectivity seems to me to be becoming ever more oppressive.


Style of references: as in Archäologischer Anzeiger 1985, 757ff. Additionally, Classicisme = Fondation Hardt, Entretiens sur l'antiquité classique 25: Le Classicisme à Rome aux 1ers siècles avant et après J.-C. (1979).
By contrast with the form and style of artworks, iconography is the study of their subjects and themes. Iconology is a broader analysis using the results of both iconographic and stylistic study to place a work of art in its full cultural context and hence to delineate its complex of meanings. [JE]
1. Brendel (1936) ; (1953) ; (1979). See also Settis (1982) ; Blanckenhagen (1942).
2. Zanker in Classicisme, 283-306. I myself, after prolonged preparation, first put forward these thoughts in a lecture at Münster in 1980. They represent an extension, to cover the language of imagery, of the phenomena which made up the Roman view of history (which in turn stand as proxies for a wide range of themes in Roman art), which are set out in Hölscher (1980a). For further references, see chapter 3, n.22 (p. 22).


The Greek paradigm: a model for life-style, a case of academic classicism, or a building block of Imperial culture?

Roman works of art, with rare exceptions, correspond so little to modern ideas of 'creative art' that scholars have had to resort to an unusual array of theories in the effort to make sense of them. In one respect, at least, this is an advantage. Greek art, by contrast, readily arouses feelings of unmediated familiarity which seem universally valid and are as a result obstacles to understanding. Roman art, on the other hand, has always been handled at a distance, has only been made accessible by intellectual bridge-building. As a consequence, Classical art history has over the past century developed its most important tenets and theories in relation to Roman rather than Greek art.

These theoretical concepts have especially examined general issues of form, conceptions of space and the body, variations of style across time and artistic genre, as well as the problematics of function and the sociological conditions under which art has been created. However, surveying this set of conceptual investigations,1 one is struck forcefully by the fact that one basic and fundamental element of Roman art - namely, the indelible stamp of Greece - has been largely ignored as a subject for theoretical reflection in its own right (although it has never been explicitly denied). This omission entails both premisses and consequences. At the heart of the matter lies the often unspoken demand for originality, which is defined as an absolute quality, confined to artistic forms and thus denied any concrete historical context. Closely linked with this is an equally absolute conception of individuality, applied not just to persons but to whole peoples. Originality is what guarantees the nature of 'Romanness' in its own right. In this context, 'Greek' and 'Roman' come to be terms of polar opposition.2

In so far as Roman art was indebted to Greek models, it could never satisfy this (modern) call for originality. Thus the question of its Greek foundations was taken as a given and lost its interest both for the positive and the negative commentators on Roman images. Johann Joachim Winckelmann (1717-68), for example, saw the dependence of Roman art on Greek models as the basis for making his dismissive judgment of a 'style of imitators'. Conversely, the positive re-evaluation of Roman art since the beginning of the twentieth century emphasised the independent and fruitful nature of Roman forms. Behind this opposing banner, however, art historians remained committed to the old call for originality. The effect was that the Greek elements within Roman image-making had to be declared a foreign disruption to the indigenous and independent genius of Roman art. The Greekness of Roman art was pushed aside as a secondary and marginal issue. Where Roman art had at first been of secondary rank because of its imitation of Greek styles, it was now significant in spite of its temporary domination by Greek styles. In effect, both these perspectives - their polar opposition notwithstanding - largely lost sight of what remains a crucial aspect of visual production in Rome: that is, the Greek roots of Roman art.

One consequence has been a unilateral and self-imposed constraint on scholarly discussion. Research has been devoted to those works of art which - individually or collectively - could be seen as particularly 'Roman'. Above all, this has meant a focus on portraiture - especially portraits from the Republic and from late antiquity; on the so-called 'historical reliefs' - especially those from Flavian and Trajanic times, or of popular art; and on certain fields within architecture. Other substantial classes of Roman figural images - for instance, idealising sculpture or mythological reliefs - have long been neglected, not only as subjects for theoretical study, but even as objects of systematic and definitive publication. Explicitly or implicitly, the 'classicising' factors in Roman art have been effectively declared obsolete. The measure of success in an artistic movement was the degree to which it distanced itself from the 'classical' forms of the Greek paradigm.3

This model of interpretation has, for some time, been losing its appeal.4 It is beyond argument that Roman art, in each of its fields and in as many different ways, rests on Greek foundations. In particular, it is clear that by the late Republic Rome was in many ways a Graeco-Hellenistic metropolis,5 and that many traditions of later Roman art go back to this period. It may be an oversimplification to see Roman image-making as no more than a continuation of Greek art,6 but it is certainly wrong to regard its fundamental 'Romanness' as an autonomous structure of basic forms that are polar opposites to the forms of Greek art. The supposed antithesis of 'Greek' and 'Roman' art conceals a decisive and continuous historical process. At the same time, more nuanced understandings have opened us to the fact that some Roman works of art show particularly strong Greek influence.7 Beyond this, the issue of Classicism itself has become an area of increasing discussion.8

A precondition for this new understanding of Roman art (to which my own work is intended as a contribution) is the rejection of certain premisses that have been taken as axiomatic. Above all, we must break free of the expectation of ceaseless innovation, often in the form of progressive 'development', and of the assumption that a unified, free-standing, style is somehow expressive of historical individuality.9 But a further requirement is also necessary. We can no longer approach works of art exclusively from the standpoint of production, as the expressions of artists or patrons, but we must also examine them as forms of communication - that is, as a factor in the collective life of a society.10 It is from this point that the present enquiry into the visual language of Roman art and its function in Roman culture will start. A culture - and even one of its branches, such as figural art - can only be meaningfully described as a system of all the elements belonging to it. Only when we grasp the historical function of the system as a whole can we begin to asess the importance of its component factors.

The problematic use of the term 'Classicism' itself exemplifies the difficulty of defining the relationship between Roman art and its Greek models.11 In a narrow sense, we can use the term to indicate reference to paradigms from the 'Classical' period of Greek art (that is from the fifth and fourth centuries BC), giving special attention to the 'High Classical' decades of Pheidias and Polykleitos. On the other hand, 'Classicism' in a wider sense denotes the reception of Greek art as a whole, from the Late Archaic period down to the end of the Hellenistic. This is no mere terminological problem.

First, we must start from the fact that both these forms of reception actually exist. Next, we need to acknowledge that the conscious reference to the 'Classical period' in the narrower sense is a very different phenomenon from the all-embracing usage of different styles within the repertoire of Greek art. In the first case, we are offered a consistent intellectual attitude, a striving for scale and order, a programme - in short - that evokes 'timeless' values, establishes norms, eliminates conflicting attitudes and makes pedagogic demands.12 In the second case, we find a response to the past that registers and refers eclectically to a variety of historical traditions, a response directed at a rather more diffuse educational ideal.

One might, at this stage, simply conclude that both traditions existed - appealing diversely to the moral ideologues, aiming at a specific goal, and to the educated public, with its range of broad-minded interests. Yet, between them, the definitions of these two forms of 'Classicism' address only part of the question. Neither of them succeeds in explaining why works of art in the Greek traditions achieved so vast a distribution, over the entire extent of the Roman Empire. Equally, the great bulk of Roman work recalls its Greek prototypes only in a very general (one might even say tired and artistically weak) form - thus responding neither to the demands of a higher enlightenment nor to those of a programmatic intellectual regime.13 The broad population of the Empire consisted neither of backward-facing ideologues nor of trained art-historians - and yet this art won widespread approval. One might wonder therefore whether current explanations are not pitched at too high a level. Perhaps the visual language of Roman art had a more rudimentary set of functions for the general public as a whole.14 This brings us to this question: what exactly was the achievement of the Greek tradition for Roman Imperial culture?


1. For critical analysis see Brendel (1953) 9ff. and (1979) 3ff., 139ff.; Bianchi Bandinelli (1960) and (1979) 108ff.; Settis (1982) 161ff.
2. On this see Brendel (1979) 47ff., 68ff., and Settis (1982) 178.
3. On this see Sichtermann (1974).
4. Brendel (1979) 84ff. and Bianchi Bandinelli (1960) and (1979).
5. Thanks above all to the researches of F. Coarelli; for one representative example out of many, see Coarelli (1970-1). Important too is Zanker (1976).
6. Exemplified by Toynbee (1934), on which see the fundamental comments of Brendel (1979) 72ff.
7. Above all, in recent years, in the work of Zanker (1974).
8. e.g. Jucker (1950), Wünsche (1972) 45-80, Sichtermann (1974). Preisshofen and Zanker in Classicisme, 263-77 and 283-306 respectively.
9. For Roman art, this begins with Brendel (1936) 122ff. and (1979) 122ff. and Blanckenhagen (1942) 310-41. In practice, however, the phenomena of temporal styles have continued to attract much greater attention than the various more static factors in Roman art.
10. For an example of the aesthetic dominance of the categories of production and expression, see the (nonetheless stimulating) essay by Wünsche (1972). Compare n.14 below.
11. For an attempt to define this expression, without the distinction made here, see Gelzer in Classicisme, 1-41, esp. 10ff. More closely related to the phenomena under discussion, Zanker, Classicisme, 45, 299ff. In this work (to have done with the burden of terminology) I make very restricted use of the term 'Classicism': not for the aspects of typology, but only for stylistic attitudes (see pp. 11, 21 below), and then only for the programmatic concentration on fifth- and fourth-century BC models. But I am here concerned only with the phenomena: one might perfectly well give them different labels.
12. See most recently Gelzer in Classicisme, 1-41.
13. These features, among other questions, are given special emphasis by Zanker in Classicisme, 283ff.
14. This forms a fundamental ground for objection to the all-too-general use of the high-cultural, production-centred aesthetic categories of interpretatio, imitatio and aemulatio for Roman public sculpture by Wünsche (1972). They can have been valid for only a narrow range of thinking artists and viewers.


The monuments: how the language works

At first sight, one might adopt a simple and consistent model for the reception of Greek prototypes in Roman art. Each succeeding epoch of Roman history might be presented as embracing a different phase of Greek art, in which a sympathetic stylistic ideal was identified and seized upon. Thus, in the Augustan period, the dominant type of imperial portrait witnesses a return to the orderly 'Classical' (in the narrow sense) language of the fifth century BC - especially that of Polykleitos.1 In the Flavian period, by contrast, the portraits of Vespasian follow the colourful, 'baroque', formal language of Hellenistic art.2 According to this model, these choices of specific prototypes are seen as representative of their respective epochs in Roman culture. A similar case can be made for the Classicism of the Hadrianic and the 'baroque' of the Antonine periods in the second century AD.

This model of Classicism will not do. It is based on an idea of the development of Roman art as a pendulum, swinging between classicising and baroque phases.3 Classical and baroque are here understood, in the sense first proposed by Heinrich Wölfflin (1864-1945), to be polar opposites and - at the same time - temporal styles. These period styles of Roman art are defined by the conscious selection of specific period styles in Greek art. The very identity of the relevant Roman epoch finds its expression through the stylistic choices made in this way. Obviously, this way of seeing 'Classicism' is just an extension of the history constructed, though partly by different means, for Greek art.4 That is, it is a story of a developing progression through fundamentally unitary temporal styles - though now no longer a process of independent innovation, but rather a series of selections from specimens of older exemplary styles.

© Cambridge University Press

Table of Contents

Foreword; Preface; 1. Introduction; 2. The Greek paradigm: example for lifestyle, academic subject, or building block of imperial culture?; 3. The monuments: questions, categories, theses; 4. Battle-scenes: the tradition of Hellenistic pathos; 5. Battle-scenes: their reception in Rome; 6. State ceremonial: the tradition of Classical dignity; 7. The semantic system: the elements and their use; 8. The semantic system: premises and structure; 9. The origins of the system: dynamics and statics; 10. Language of imagery and style; 11. Formal system and style in the theory of rhetoric and of imagery; 12. Conclusion: language of imagery and culture of empire; Bibliography, supplementary bibliography.

Customer Reviews

Most Helpful Customer Reviews

See All Customer Reviews