TRAC 1991:
A THEORETICAL FRAMEWORK FOR THE STUDY OF ROMANO-BRITISH VILLA MOSAICS
By Sarah Scott

Studies of Romano-British villa mosaics have tended to be restricted to description, classification and typological survey. Little attempt has been made to go beyond the ordering of data, to consider the relationship between the mosaics, the villa, and the nature of the society itself. The aim of this paper will be to outline a new theOTetical framework fOT the interpretation of villa mosaics. This framework involves a consideration of the nature and ‘meaning’ of the designs; the relationship between the mosaics and their architectural context; the effect of the social position of the interpreter on meaning; and finally, the significance of the mosaics in the broader historical context. It will be suggested that such a framewOTk could provide new and deeper insight into the nature of social relations and material practices within Roman Britain.

INTRODUCTION
The purpose of this paper is to outline an analytical framework for the study of Romano-British villa mosaics. It will hopefully be demonstrated that a new theoretical approach could allow deeper insight into the nature of social relations throughout the Roman period in Britain, and enable an interpretation to be made regarding the significance of the mosaics, both as products of, and ‘participants’ in, these relations.

The discussion will be based on the premise that material culture is a communicative symbolic field, and that it is structured in relation to social strategies and power relations (Hodder 1982). Although this symbolic field is already in place for an individual to use, it is through the manipulation of both material and other practices that the individual is able to establish him/herself within the social order, and is able to make changes to his or her position within this order.

The implication of this premise for the study of Romano-British mosaics is that they can be seen as having the potential to provide us with an insight into the nature of social relations within Roman Britain. These material practices are not simply a source of information about other aspects of society, however; they are a source of information in their own right. The mosaics should not be seen as a passive reflection of Romano-British society; they must instead be seen as integral to, and active within, social relations. Any attempt to interpret the mosaics therefore has to go beyond description and classification, and must endeavour to consider subjects such as: dating; style; the meaning of the art within a given cultural context; and the nature of the society producing them.

Bearing this in mind, the theoretical framework to be outlined involves four basic stages of analysis:

1) the collection and presentation of the data in a manner conducive to further study;
2) a closer look at the nature of representation within the mosaics, and an assessment of the way in which ‘meaning’ is encoded within the system;
3) an examination of the relationship between the mosaics and their architectural context;
4) the interpretation of the mosaics in terms of the historical context of their production and use.

Before discussing these analytical stages in more detail, however, it will be useful to consider briefly the nature of some previous work on RomanoBritish mosaics, and to identify some of the problems which have placed restrictions on interpretation.

The Nature of Previous Work
Work on Romano-British mosaics, such as that carried out by J.M.C. Toynbee (1962; 1964), and D.]. Smith (1969) falls very much within archaeology’s ‘traditional’ culture-historical paradigm. Although extremely valuable in terms of descriptive detail and ordering of data, it has contributed very little to an understanding of the mosaics within the wider social context. In the case of much of the work carried out on villas and mosaics, the manner in which the data has been classified and organised hinders rather than facilitates interpretation. For the most part, studies of the RomanoBritish villa are characterised by a tendency to divide the villa up into several different components, these components then being analysed separately.

A striking example of this type of work is Rivet’s The Roman Villa in Britain (ed. 1969), where various aspects of the villa are compartmentalised and considered in turn by different authors. I suggest, however, that if any one aspect – such as the mosaics – is to be interpreted more fully, it is essential that the villa be considered as an entity. The division of the villa into these various components is artificial in the extreme, presenting a distorted and incomplete view of ‘villa life’. Additionally, if we are to make any attempt to go beyond the descriptive stage, it is necessary not only to consider the villa as an entity, but also to consider the changing historical context of the production of both the villas and the mosaics.

If we are to attempt to interpret the social significance of the mosaics, the ‘traditional’ type of approach must be rejected in favour of an interpretive framework which integrates theory and data more fully. This integration of theory and data is necessary even at the initial stage of the collection and presentation of the archaeological material.

THE PRESENTATION OF THE DATA
It is important not only to locate, identify and describe the mosaics, but also to relate them to the plan of the villa. It is essential that the mosaics can be seen in relation to one another, and in relation to the architectural setting. Some of the eighteenth and nineteenth century plans of villas are useful in this respect as they have the mosaics illustrated in situ (see e.g. Figs 17 and 18 of Newton St Loe and Withington respectively). The drawings of Samuel Lysons (e.g. 1797) are a particularly notable example of this holistic craft.

Before any attempt can be made to relate the mosaics to their historical context of production, it is also necessary to have some idea of their chronological framework. It is suggested that a formal analysis of the mosaics could prove extremely useful for building up a more detailed picture of the chronological and formal aspects of Romano-British mosaics. As Miller (1985) has noted, formal analysis can also provide a description that may be more useful than the mere listing of motifs, and may provide alternative levels at which to examine the articulation between form and society.

Having discovered the nature of the art forms, where they occur, and how and when they were constructed, it is possible to move to the next stage of the analysis. This will involve a closer look at the nature of representation within the mosaics, and an assessment of the way in which meaning is encoded within the system.

Figure 17. Newton St Loe: villa and mosaics.

How Do the Mosaics Encode Meaning?
Regarding Romano-British mosaics, it is useful to make an analytical distinction between representational, or iconic forms, and non-representational, or arbitrary systems of representation.

Iconicity has been defined by Taylor as (1987, 198): ‘the formal resemblance between the signifier, the painted form, and signified, the object or species represented.’ Within an arbitrary, or non-representational system, there is no necessary relationship between the signifier and the signified. It must be emphasised, however, that although there may be no necessary relationship between signifier and signified in arbitrary systems of representation, such a system may be far from arbitrary in that, in order to interpret meaning, it may be necessary to possess an in-depth knowledge of the code employed. Likewise, although the representational forms may be easier to interpret initially, they too may have meanings which require some kind of specialised knowledge.

It is useful to distinguish between the two systems of representation within Romano-British mosaics because, within Roman art, it is necessary to look at many of the representational forms in terms of contemporary mythology and religion. The non-representational forms, on the other hand, need to be understood as part of an historical, aesthetic tradition extending back to classical Greece.

It must be stressed, however, that if the full potential of the mosaics for encoding meaning is to be understood, it will be necessary to look at the relationships within and between the two systems of representation. In order to interpret an element within a mosaic, it is essential to look at its relationship with other elements within the mosaic, and also to look at the relationships between mosaics within a villa. For example, in terms of classical mythology, many figures may be interpreted very differently according to their context: who else, and what else they are represented with. By considering the relationships between various elements, it may be possible to screen out ‘surplus’ meaning. Additionally, various forms may be placed together in such a way as to imply a new meaning, which can only be grasped through an analysis of the relationships between forms. Eriksen’s “re-reading’ of the Hinton St. Mary ‘Christian’ pavement illustrates clearly the importance of this type of approach. He convincingly argues that the design represents a syncretistic allegory: ‘where the Graeco-Roman elements indicate the presence of the old religion, and where scriptural images are so interwoven as to create a striking thematic and symbolic unity’ (1980, 48).

Figure 18. Withington: villa and mosaics.

It is also important here to consider how the form of a mosaic, i.e. the way in which subjects or elements are presented, can affect meaning. I have suggested elsewhere (1991) for example, that the concentric circles of the Romano-British Orpheus design (e.g. Figs 18 and 19 of Withington) lend impact to the idea of Orpheus subduing nature in its strongest and wildest forms. It is argued that this particular feature suggests continuity, and emphasises the fact that the animals are unable to escape the power of Orpheus.

All of these factors have to be considered in terms of who would have seen the mosaics. The mosaics would obviously have been open to different levels of interpretation according to the social position of the interpreter. If we are to consider this question further, however, it will be necessary to place the mosaics within the architectural context.

THE ARCHITECTURAL CONTEXT

It is important to remember that the mosaics themselves are an integral part of the villa architecture, being of a much more permanent nature than furniture, for example. The following quote from Vitruvius emphasises clearly the significance of the architectural context (Vitruvius in Preston-Blier 1987, 1):

“In all matters, but particularly in architecture, there are these two points: the thing signified, and that which gives it significance …. It appears then, that one who professes himself an architect should be well-versed in both directions. Let him be educated, skilful with the pencil, instructed in geometry, know much history, have followed the philosophers with attention, understand music, have some knowledge of the jurists, and be aquainted with astronomy and the theory of the heavens.”

What this highlights is that the significance of architecture is grounded in the experiences and intellectual background of its makers and users. As Preston-Blier (ibid.) has pointed out, architecture is invariably anthropocentric, being bound up with human activity, experience, and expression. Architecture provides an objectification of pre-existent patterns and perspectives, ordering space, and therefore human action. Preston-Blier (ibid., 2) notes that when architecture borrows its imagery from human experience, it encourages those who move within it to reaffirm essential features of human identity and activity. However, at the same time, architectural meaning can also have a basis in metaphor; a structure may provide the means for seeing one thing in terms of something else. As such, it may bring a number of diverse ideas and activities together into a coherent whole.

Figure 19. Withington: Orpheus and other mosaics.

In order to analyse the complex symbolism of architecture, however, it is obviously necessary to consider the occupants and their use of the architectural space and symbols. In particular, it is essential to account for the social position of the interpreter when attributing ‘meaning’ to the architecture. Interpretation is always bound up with social inequality and with power relations. It is particularly important to bear this in mind in the interpretation of mosaics. The question of physical access to the mosaics has to be considered in relation to the nature of representation. Access to the meaning of the mosaics may have been constrained both physically and through varying degrees of education and understanding. It is important to consider, for example, how the mosaics and architecture might have been viewed by the people who worked in the villa, or how they might have been viewed by women. This may be critical to understanding the ways in which they functioned within the social and political context.

In terms of Romano-British mosaics, therefore, it is necessary to look very closely at the various trends in design construction, and at the changing popularity of the subjects depicted. These patterns must then be related to developments in domestic architecture more generally. At this stage it will also be necessary to look at the ways in which Roman beliefs and ideals imposed on domestic architecture and interior decoration, and at the various functions of rooms within the villa. On the basis of this, the question of who may have had access to those rooms which possessed mosaics can be considered, and the relationship between these people and the owner of the villa assessed. It may then be possible to suggest reasons why a certain design may have been chosen, and to address the question of what the villa owner may have been trying to say about himself and his relations with the rest of society.

Integral to this stage of the analysis must be a consideration of the relationship between the architectural forms and their immediate environment. For example, how did the siting of a villa contribute to its overall aesthetic effect, and how did the villas relate to other buldings in the landscape?

THE SOCIAL, POLITICAL AND ECONOMIC CONTEXT
In order to explain the relationship between the mosaics, the villa, and society, it will be necessary to place them within the broader social, political and economic context.

If we are to understand why certain practices were employed in certain places at certain times, it is important to consider the social, political and economic context of their production and use. Material culture is always integral to power strategies and ideological practices. As Shanks and Tilley (1987, 72) point out, any analysis of power should be concerned with the social roots of power, attempts to achieve and maintain power, and counter attempts to subvert power strategies.

Regarding Romano-British mosaics, the data must be considered in all of its specificity, and must be related to the changing historical context. It will be necessary to consider who the elite were at anyone time, and on what their power and wealth was based. The relationship between the elite and the other elements of Romano-British society also has to be taken into account. All of these factors then have to be considered in the light of Britain’s changing relationship with the core provinces, and with other areas of the Empire. Having placed the mosaics and architecture within this wider context, it should then be possible to address the question of why certain material practices were employed at certain times, and to suggest how these practices may have played a role in maintaining or changing the very structure of society.

CONCLUSION
To summarise, it has been suggested that much of the pessimism inherent within studies of Romano-British villas and mosaics is due not to a lack of evidence, but to the nature of the approach employed. In order to move beyond the stage of description and classification, a new approach has been proposed, in which the mosaics are considered within their architectural context, and as integral to social strategies and power relations. It must be emphasised that the analytical stages outlined above are not mutually exclusive. In practice, there will have to be a constant movement back and forth between the various levels. However, perhaps the most important feature of such an approach is that it provides a single framework within which to analyse the different levels of discourse inscribed in the material text. Within such a framework, it is no longer possible to consider various components of the villa in isolation, or to discuss the villa without placing it within its specific cultural context. An attempt must be made to develop theories regarding the possible function and significance of these material practices.

Although the integration of theory and data has only begun relatively recently in Romano-British archaeology, it has hopefully been demonstrated in this paper that the potential of such an integration is considerable. It is possible for an interpretation to be made regarding the significance of the villa mosaics and architecture, allowing new and deeper insight into the nature of social relations and material practices throughout the Roman period in Britain.

Bibliography

Coward, R. and J. Ellis 1977 Language and Materialism. R. K. P.

Eriksen, R. 1980 ‘Syncretistic symbolism and the Christian Roman Mosaic at Hinton St Mary: a closer reading’ Proceedings of the Dorset Natural History and Archaeological Society 102,43-48.

Hodder, I. 1982 The Present Past. Batsford.

Hodder, I. 1986 Reading the Past. Cambridge University Press.

Lysons, S. 1797 An Account of the Roman Antiquities discovered at Woodchester. London.

Lysons, S. 1817 Reliquiae Britannico-Romanae volume 2.

Miller, D. 1985 Artefacts as Categories. Cambridge University Press.

Moore, H. 1990 ‘Paul Ricoeur: action, meaning and text’, in Tilley (ed.), 85-120.

Nichols, W. L. 1838 Description of the Villa at Newton St Loe, Bath.

Preston-Blier, S. 1987 The Anatomy of Architecture. Cambridge; Cambridge University Press.

Rivet, A.L.F. (ed.) 1969 The Roman Villa in Britain. London.

Scott, S. 1989 Symbols of Power and Nature: A Contextual Approach to the Orpheus Mosaics of Fourth Century Roman Britain. Leicester; unpublished BSc Dissertation University of Leicester.

Scott, S. 1991 ‘An outline of a new approach for the interpretation of Romano-British mosaics, and some comments on the possible significance of the Orpheus Mosaics of fourth-century Roman Britain’ Journal of Theoretical Archaeology 2.

Shanks, M. and C. Tilley 1987 Social Theory and Archaeology. Cambridge; Polity Press.

Smith, D.J. 1969 ‘The Mosaic Pavements’ in A.L.F. Rivet (ed.), 71-125.

Taylor, L. 1987 The Same but Different: Social reproduction and innovation in the art of the Kunwinjku of Western Arnhem Land. PhD thesis Australian National University.

Tilley, C. (ed.) 1990 Reading Material Culture. Oxford; Blackwell.

Todd, M. (ed.) 1978 Studies in the Romano-British Villa . Leicester; Leicester University Press.

Toynbee, J.M.C. 1962 Art in Roman Britain.

Toynbee, J.M.C. 1964 Art in Britain under the Romans.

Webster, G. 1969 ‘The future of villa studies’ in A. L. F. Rivet (ed.), 217-249.