Reading Space: Social Interaction and Identity in the Houses of Roman Pompeii: A syntactical approach to the analysis and interpretation of built space 9781841710860, 9781407352275

Table of contents :
Front Cover
Title Page
Copyright
TABLE OF CONTENTS
LIST OF FIGURES
LIST OF TABLES
PREFACE
INTRODUCTION. The Walls Will Tell Me Nothing
PART ONE. THEORY AND METHOD FOR THE INTERPRETATION OF BUILT SPACE
CHAPTER ONE. Space and Spatiality
CHAPTER TWO. Architecture, Knowledge and Power
CHAPTER THREE. The Language of Space
CHAPTER FOUR. Reading Spatial Layouts
PART TWO. THE HOUSES OF POMPEII: A CASE STUDY
CHAPTER FIVE. Pompeii and Access Analysis
CHAPTER SIX. Locating the Syntaxes
CHAPTER SEVEN. Space and Social Interaction
CHAPTER EIGHT. Social Networks, Power and Identity
CHAPTER NINE. Image and Reality
EPILOGUE. The Walls Will Tell Me Something
GLOSSARY
APPENDIX ONE. Ground Plans and Access Maps
APPENDIX TWO. Tables of Data
APPENDIX THREE. Results
REFERENCES

Citation preview

BAR S886 2000

Reading Space: Social Interaction and Identity in the Houses of Roman Pompeii A syntactical approach to the analysis and interpretation of built space

GRAHAME: READING SPACE

Mark Grahame

BAR International Series 886 2000 B A R

Reading Space: Social Interaction and Identity in the Houses of Roman Pompeii A syntactical approach to the analysis and interpretation of built space

Mark Grahame

[OJ

BAR International Series 886

2000

Published in 2016 by BAR Publishing, Oxford BAR International Series 886 Reading Space: Social Interaction and Identity in the Houses of Roman Pompeii

© M Grahame and the Publisher 2000 The author's moral rights under the 1988 UK Copyright, Designs and Patents Act are hereby expressly asserted. All rights reserved. No part of this work may be copied, reproduced, stored, sold, distributed, scanned, saved in any form of digital format or transmitted in any form digitally, without the written permission of the Publisher.

ISBN 9781841710860 paperback ISBN 9781407352275 e-format DOI https://doi.org/10.30861/9781841710860 A catalogue record for this book is available from the British Library BAR Publishing is the trading name of British Archaeological Reports (Oxford) Ltd. British Archaeological Reports was first incorporated in 1974 to publish the BAR Series, International and British. In 1992 Hadrian Books Ltd became part of the BAR group. This volume was originally published by Archaeopress in conjunction with British Archaeological Reports (Oxford) Ltd/ Hadrian Books Ltd, the Series principal publisher, in 2000. This present volume is published by BAR Publishing, 2016.

BAR

PUBLISHING BAR titles are available from:

EMAIL

PHONE FAX

BAR Publishing 122 Banbury Rd, Oxford, OX2 7BP, UK [email protected] +44 (0)1865 310431 +44 (0)1865 316916 www.barpublishing.com

TABLE OF CONTENTS

LIST OF FIGURES

iv

LIST OF TABLES

vi

PREFACE

Vll

PART ONE. THEORY AND METHOD FOR THE INTERPRETATION OF BUILT SPACE INTRODUCTION

The Walls Will Tell Me Nothing Buildings and power Architecture, space and society Reading space Pompeii Analysis and interpretation Conflicting images

1 1 2 3 4

4 5

CHAPTER ONE

Space and Spatiality Objectivity and subjectivity The body, space and social encounters Gatherings and occasions Space and spatiality Context and contextuality A hierarchy of discontinuity Presence-availability From objective perspective to subjective standpoint

6 6 7 9

9 11 11 14 14

CHAPTER Two

Architecture, Knowledge and Power Knowledge of body and self Familiarity and observation Architecture as a means to freedom Architecture as an instrument of repression Disciplining the body and the self Architecture and social inequality Inhabitants and strangers Architecture and social institutions Architecture and society

15 15 16 17 18 19 20 21 22 23

CHAPTER THREE

The Language of Space Language, space and culture The social logic of space Space syntax theory A hierarchy of spatial events Architecture as text Reading and writing a spatial text

24 24 24

25

26 27 27

CHAPTER FOUR

Reading Spatial Layouts Access analysis Spatial description Justification and quantification Dimensionless spaces and systems

29 29 30 32 33

Table of Contents Control values Relative asymmetry Real relative asymmetry Depth from the exterior The significance of quantification

33 34

35 35 35

PART TWO. THE HOUSES OF POMPEII: A CASE STUDY CHAPTER FIVE

Pompeii and Access Analysis Pompeii as a spatial entity Sampling strategy Region six Defining the analytical units Applying access analysis Architectural features Missing upper floors The analysis

37 37

38 38 39 40 40 41 42

CHAPTER SIX

Locating the Syntaxes Principles and pathways Distributed and non-distributed aspects of spatial order Locating order in configurations Interpreting the results A rule of addition Relations with the exterior Structure and contingency Politics and ideology

43 43 44 45

46 49 51 54 54

CHAPTER SEVEN

Space and Social Interaction Presence-availability and interaction potential Nodes and courtyards Defining interaction potential Interaction with the exterior Non-courtyard houses Single courtyard houses Double courtyard houses Multiple courtyard houses The reproduction of space

56 56 57 59

60 61

62 64 68 73

CHAPTER EIGHT

Social Networks, Power and Identity Individualistic and collectivistic societies Space, identity and architecture Buildings and identity Identity in Pompeian houses Architecture, identity and social hierarchies Houses, households and social power Social power in Pompeian houses Internal and external social networks Spatial patterns and external networks Social networks and identity in the non-courtyard houses Social networks and identity in the courtyard houses Houses and spatial dispersion Class, neighbourhoods and power Pompeii and society 11

74 74 74 75

76 77

78

79 80 81 81

83 84

85 86

Table of Contents CHAPTER NINE

88 88 89 90 92

Image and Reality The physical image of the house Textual and syntactical analysis The house as a reification of society Text and sense Text and reference Pompeian houses and Roman culture Style and identity Pompeian houses and identity Natural and embodied languages Spatial and literary texts

93

94 94 95 96 96

EPILOGUE

98

The Walls Will Tell Me Something

99

GLOSSARY

APPENDIX ONE

Ground Plans and Access Maps

101

APPENDIX Two

Tables of Data

172

APPENDIX THREE

Results

197

REFERENCES

200

111

LIST OF FIGURES

1. 1

Illustration of two individuals, A and B, involved in a social encounter. The solid boarder defines the limits of copresence and the 'front' and 'back' regions of the encounter are marked.

1.2

Illustration of the effect on space when a social encounter between two individuals, A and B, comes under the observation of a third individual, C, within the conditions of co-presence.

1.3

Illustration of how the boundary expands the perception of distance. A social encounter is more likely to occur in situation A than it is in situation B, despite the same physical distance being involved in both.

1.4

The affect of the boundary on space and social interaction.

1.5

Model of the defensible space paradigm. A, public space of the street; B, semi-public realm; C, semi-private realm; D, private realm (after Newman 1972 and Hanson and Hillier 1982).

1.6

Illustration of how building and settlement space are defined by the primary cell.

1.7

Two identical building plans showing the different possible interpretations of their spatial composition.

1.8

Illustration of how a hierarchy of discontinuity is produced by locating cells within cells.

2.1

Plan of a simple building consisting of a central space (X) and a series of surrounding spaces (Y).

2.2

The building plan from figure 2.1. The black dots represent individuals and the shaded area shows the space that can be observed from X.

2.3

The building plan from figure 2.1. The black dots represent individuals and the shaded area shows the space the individual at X can observe.

2.4

The building plan from figure 2.1 showing individuals located within the shaded area.

2.5

The simple building discussed above showing its composition in terms of primary and secondary cells.

2.6

Diagram showing the opposition between inhabitants and strangers and familiarity and unfamiliarity.

3 .1

Diagram of the hierarchy of spatial events, according to how durable they are.

4.1

Building plan with accompanying access map.

4.2

Building plan and access map with each space described according to the bounded space method (after Hillier et. al 1984, figs. la and b).

4.3

A building with low definition. The various ways its spatial layout may be described are shown by the access maps at A, B, C andD.

4.4

Two spaces with a tangent drawn in each between two points. Space A is a convex space, but B is not.

4.5

Building plan shown in figure 4.2 divided according to the rule of convexity with access map (after Hillier et al. 1984, figs. 2a and b ).

4.6

The access map from figure 4.5 justified with respect to the exterior.

4. 7

Justified access map from figure 4.6, showing how the control value for Xis calculated.

4.8

An example ofa symmetrical (A) and an asymmetrical (B) configuration.

4.9

Configuration showing the number of boundaries that have to be crossed in order to reach any space from space X.

5.1

Ground plan of Pompeii with the regiones (regions) and main streets marked.

5.2

Detailed plan of Regio (region) VI.

5.3

Diagram of a hypothetical peristyle. The black dots indicate the individual columns and the shaded area shows the space they define. The dotted lines indicate the possible routes through the colonnade.

6.1

Model showing the transpatial realisations of the global-to-local and local-to-global pathways.

6.2

Diagrammatic representation of the global and local axes. The approximate area occupied by the houses of region VI is shaded.

6.3

A non-distributed configuration with control values calculated for each space.

6.4

Scatter plot of mean RRA values against control ratios for the houses of region VI.

IV

List of Figures

6.5

Scatter plot of control ratios against number of spaces for the houses of region VI.

6.6

Scatter plot of mean RRA values against number of spaces for the houses of region VI.

6. 7

Configurations formed by adding one space at a time to the previous configuration according to the rule of addition.

6.8

Two configurations both consisting of four spaces formed by adding one space to each of the joining possibilities from configuration 3 in figure 6. 7.

6.9

Recursive global outcomes from spaces joining to configuration 3 according to the rule of addition.

6.10

Joining probabilities for the spaces for the most frequently occurring outcome from figure 6.9.

6.11

Configurations formed by the rule of addition taking into consideration relations with the exterior.

7.1

Diagram showing a circulation space (A) and a movement space (B).

7.2

Diagram showing a space with high relative convexity (A) and one with low relative convexity (B).

7.3

VI xv 9 with areas of different interaction potential stippled.

7.4

Ground plan of VI xv 22 with stippled maps showing the local and global interaction potentials and the interaction potential with respect to strangers.

7.5

Ground plan of VI i 18, 20 with a stippled map showing the local and global interaction potentials and the interaction potential with respect to strangers.

7.6

Ground plan of VI viii 8 with stippled maps showing the local and global interaction potentials and the interaction potential with respect to strangers.

7.7

Ground plan of VI xv 9 with stippled maps showing the local and global interaction potentials and the interaction potential with respect to strangers.

7.8

Ground plan of VI xiv 37 with stippled maps showing the local and global interaction potentials and the interaction potential with respect to strangers.

7.9

Ground plan of VI vii 7 with stippled maps showing the local and global interaction potentials and the interaction potential with respect to strangers.

7 .10

Ground plan of VI ii 17, 20 with stippled maps showing the local and global interaction potentials and the interaction potential with respect to strangers.

7.11

Ground plan of VI iii 7, 25-26 with stippled maps showing the local and global interaction potentials and the interaction potential with respect to strangers.

7.12

Ground plan of VI v 9, 19 with stippled maps showing the local and global interaction potentials and the interaction potential with respect to strangers.

7.13

Ground plan of VI viii 23-24 with stippled maps showing the local and global interaction potentials and the interaction potential with respect to strangers.

7.14

Ground plan of VI xii 1-3, 5-8 with stippled maps showing the local and global interaction potentials and the interaction potential with respect to strangers.

7 .15

Ground plan of VI xi 9-10 with stippled maps showing the local and global interaction potentials and the interaction potential with respect to strangers.

8.1

Graph illustrating the relative strength of household (series one) and individual (series two) identity in the houses of region VI.

8.2

Ground plan ofregion VI with the blanked areas showing the location of the non-courtyard houses.

8.3

Bar charts showing the distribution of different house types in the insulae of region VI.

8.4

Data from figure 8.3 overlaid on to a ground plan ofregion VI.

9 .1

Idealised ground plan ofa Pompeian house (after Mau 1899: 24 7, fig. 115).

9.2

Comparison of two atrium houses with peristyles (drawings to scale).

9.3

Ground plans of VI i 9-10, 23 and VI xii 1-3, 5-8 with the Vitruvian spaces and servile areas marked.

9.4

Diagrammatic representation of the relationship between the various aspects of style.

V

LIST OF TABLES

1.1

Summary of the effects on interaction and the body of the division of space into front and back regions.

3.1

The equivalence between spatial and language events.

6.1

The effect of distributedness and non-distributedness at the global and local levels.

6.2

Joining probabilities for the next space to be added to configuration 3 from figure 6.14.

6.3

The effect ofnon-distributedness and distributedness on relations with the exterior.

6.4

The number of nodes at different depths from the exterior.

6.5

The combined joining probabilities for the spaces in configuration 2, figure 6.13.

6.6

The mean number oflinks to the exterior for all configurations with nodes.

7.1

How relative size and convexity creates spaces of different sizes and shapes. The numbers indicate their position in the hierarchy of interaction potentials.

7.2

Comparison of the number of nodes and the number oflarge circulation spaces at different depths from the exterior.

7.3

Frequency distribution of courtyards.

7.4

List of houses giving the minimum, mean and maximum RRA values for different numbers of courtyards.

7.5

The rank order of the RRA values for the spaces of VI xv 9 and their corresponding interaction potentials.

7.6

The rank order of the control values for VI xv 9 and their corresponding interaction potentials.

7. 7

The interaction potentials with respect to strangers for VI xv 9.

8.1

How the scope of a social network and the degree of freedom it affords affects collective and individual identities.

8.2

How different architectural arrangements can reinforce different aspects of identity.

8.3

The ideal layout most representative of the internal organisation of houses shown in table 7.4.

8.4

Illustration of how different layouts can preferentially reinforce either internal or external social networks, or both.

8.5

Table showing the ideal layout most representative of external networks of the houses shown in table 8.4.

8.6

How the distribution of households with different connections to external networks reinforce different facets of identity.

9 .1

Frequency of occurrence of various combinations of the 'traditional' spaces found within the houses of region VI.

vi

PREFACE

Research can be a peculiar enterprise: you begin with one set of objectives and end up doing something else. This, in essence, is how this study came into being. It began in 1991 with my Ph.D. research, which started out as an examination of space in Pompeii. The sheer complexity exhibited by the architectural fabric of the city seemed to silently contradict the rather simplistic explanations offered for it in the academic literature. In particular, the constant reproduction of a standardised plan for the Pompeian house did not accord well with the intricate mosaic of housing that survives to this day. The first step was to find a more rigorous way of describing built space. By using a more systematic method I hoped to highlight the diversity of housing in Pompeii and offer a new interpretation of its meaning. Almost at once I was faced with a problem: how to describe space in a more precise manner. A solution presented itself when I came across Hillier and Hanson's (1984) seminal work, The Social Logic of Space. Access analysis seemed to provide a readymade method of spatial description, which was both rigorous and objective. What is more, the houses of Pompeii provide an almost perfect opportunity to apply this technique to archaeological data.

question of how we can come to understand human social action from a contemplation of built space. Consequently, to make the results of access analysis meaningful there seemed no choice but to build my own interpretative framework. Thus the direction of my research began to change. From a study of Pompeian houses it inexorably moved towards a theoretical understanding of the relationship between built space and human society. The houses of Pompeii remained, but as a case study, rather than the focus of research. The first statement of my theory came in my Ph.D., The Houses of Pompeii: Space and Social Interaction (1995). Despite the title, the houses of Pompeii were not the central concern. Much of the thesis was taken up with reflection on theoretical and methodological issues. Despite an attempt to focus the later chapters on Pompeii, the greater part of my intellectual effort was clearly directed towards resolving theoretical issues. Like all studies, my Ph.D. was helped by many who took the time and trouble to support my endeavour. First amongst these must be my thesis supervisor Simon Keay. He guided me through some turbulent times and constantly reminded me that whatever the distractions of theoretical reflection, the objective of archaeological research was to make concrete interpretations of evidence. I am grateful for his attempts to keep my feet on the ground. A debt of gratitude is also owed to my personal advisor, Steve Sherman. Not only did he provide a sympathetic ear, but he also directed me towards Hillier and Hanson's work. Without that important signpost, I might never have brought the thesis to a conclusion. In addition to their above endeavours, I must also thank Simon Keay and Steve Sherman for their critical comments on the various drafts of my thesis. Their guidance was essential. I would also like to acknowledge the debt I owe to the stimulating academic environment provided by the staff and postgraduates of the Department of Archaeology at Southampton. Others who took the trouble to read and comment on parts of the thesis were, Ray Laurence, Brian Sparkes, Julian Thomas and Andrew Trend. They all deserve thanks for without their comments both the thesis and this study would have been poorer. Of the above, I must single out Ray Laurence for a special mention. He took the time to discuss various aspects of my work with me and it is not an exaggeration to say that his comments were decisive in helping me clarify my research methodology. I also owe Andrew Wallace-Hadrill an equal debt of gratitude. The interest he has taken in my research both during and since my Ph.D. has been encouraging. I am particularly grateful for his comments on my work and for his offers of help. I would also like to thank him for agreeing to act as my external examiner. His comments, along with those of my other examiners, Simon Keay and Julian Thomas, made my viva a stimulating experience. Many of the suggestion they made have been incorporated into the current work. A debt of gratitude is also owed to Jo Berry, Ray Laurence and Andrew Wallace-

Nevertheless, the question of interpretation still had to be resolved. Hillier and Hanson relate the findings of access analysis to the functions of the spaces described by it and the archaeologists who have employed the technique have largely followed in their footsteps. The difficulty is that in situations where room functions are not well understood, it is not easy to make sense of the results of access analysis. Indeed, they do not appear to 'say' anything in their own right. This perhaps explains why enthusiasm for access analysis has recently waned. In the case of Pompeian houses this was not a problem, after all, scholars have for many years been attributing functions to the rooms in Pompeian houses, using evidence from the Roman literary sources. However, the difficulty with marrying this evidence with the results of access analysis was the glaring difference in methodological rigour. While access analysis is rigorous, systematic and objective, the interpretation of house plans by reference to the literary sources is imprecise, anecdotal and subjective. Remarks made by the Roman authors about their domestic environments tend to be in passing and do not pertain to Pompeian houses in particular. It seemed contradictory to advocate the use of a rigorous method of spatial description, on the one hand, and then to subject its findings to an interpretative procedure that conspicuously lacked it, on the other. At this point, I turned to the disciplines of architecture, sociology and anthropology for a more robust interpretative framework. However, I was disappointed, since all alternative approaches depended on an a priori understanding of the relationship between space and human social action. None of the above disciplines, perhaps for understandable reasons, had broached the archaeological vii

Preface

Had.rill, for their frequent invitations to the Department of Classics at Reading. These gave me an opportunity to present my ideas to a friendly, but still critical audience. Finally, I would like to thank my father, Norman Grahame, for undertaking the thankless task of proof reading the final draft of my thesis. Of course, my research would not have been possible in the first place had it not been for a postgraduate studentship from the British Academy and a grant from the Robert Kiln Trust. I thank both institutions for their financial support.

In helping me prepare this study there are others who deserve my thanks. First I would like to thank David Davidson and his colleagues at Archaeopress for agreeing to publish this work as a British Archaeological Report. I would also like to thank David for his advice and help. In addition to those already mentioned, there are those whose work the reader will see in the forthcoming pages. I would first like to thank Pete George for scanning my original hand drawings for me and saving them as bitmap files. His efforts can be appreciated in appendix one, where the ground plans and access maps for each house in this study can be found. His work saved me much time and energy. I owe a particularly large debt of gratitude to Karen Stocker who helped me format appendix one. She undertook the laborious task of renumbering all the spaces on the access maps and positioning them on the page. The professional look of appendix one is largely down to her efforts. Karen also undertook the thankless task of reading the entire draft of this study and making comments. Her critical powers and keen eye for detail improved the final result immeasurably. Again, I must thank my father for proof-reading the final draft of this study. Having to go through it all again must have been tedious. Finally, I would like to thank all those not already mentioned who have taken an interest in my work or in some other way commented on it. All comments have been welcome and I have tried to take them on board. Needless to say, all those who have contributed to this study cannot be held responsible for its deficiencies, nor are they accountable for any errors or omissions. These are my responsibility.

The current work is primarily a published version of my thesis. However, it builds on the thesis and much new material has been included. While chapters one to six are revised versions of thesis chapters, chapters seven to nine are new additions. This new material draws further meaning out of the results of the original analysis and tries to create a work which is more 'balanced'. The bias towards theoretical reflection in the thesis has hopefully been corrected by the inclusion of a much closer study of the houses themselves. The division of this work into two parts reflects the distinction between the more general, theoretical chapters and those that focus on the houses of Pompeii. In particular, chapter nine relates the findings of this study to broader debates about Pompeian houses. A consideration of the broader relevance of my work was absent from my thesis and I would like to thank Pedro Funari in particular for reading my thesis and arguing cogently that it should speak to a wider audience. Hopefully, there will be much in this work to interest those who study Roman houses, or ancient urbanism in general. However, none of this should detract from the fact that the primary objective of this study remains the advancing of a new theoretical framework for interpretation of built space. For this reason, this work should speak to an audience beyond the confines of Roman Archaeology and be relevant to anyone interested in the analysis and interpretation of built space. Because it details a method of 'reading' space, this study has relevance to the interpretation of buildings from all archaeological situations and not just those from historic contexts.

Mark Grahame Southampton July2000

email. [email protected]

Vlll

INTRODUCTION

The Walls Will Tell Me Nothing Buildings are everywhere. They are an integral and omnipresent part of our lives. Buildings are so commonplace, we tend to accept their presence without question. Naturally enough, our attention is focused on our daily lives. However, the routines that go to make up our lives require us to constantly move from one room to the next and from one building to another. Yet, so effortlessly do we negotiate space that we hardly give it more than a passing thought. For this reason, built space takes on the appearance of a passive backdrop to our activities. It seems to provide a series of settings for them, but little more. This perception of buildings is not just ingrained in public consciousness, it also informs academic opinion as well. Many disciplines, including, architecture, history, anthropology, sociology, and, of course, archaeology, all approach buildings from this standpoint. This perspective has many names: 'functionalism' and the 'Man-Environment paradigm' amongst them, but whatever it is called, the approach follows the assumptions of the 'dramaturgical analogy' (Goffman 1959, 1963). The dramaturgical analogy underpins a branch of sociology known as Ethnomethodology (Garfinkel 1967). Ethnomethodology deals with the minutiae of social life and commonly employs theatrical metaphors to describe social situations. Social life itself is seen as analogous to a 'drama' with each individual being an 'actor' who plays out a specific 'role'. By extension, buildings are analogous to 'theatres' because they provide 'stages' for the drama of social life. Like a theatre, the building itself has little bearing on the performance that takes place within it. If the stage is empty of actors and props, the theatre is simply an empty space, almost without meaning. Buildings are therefore little more than 'containers' for social life and without people to give them 'life', they are nothing but empty shells. If so, it follows that the only way buildings can be understood is in terms of the activities they contain. This study can be read as a lengthy critique of the dramaturgical analogy. However, I should make it clear that my purpose is not to deny the validity of the model, but rather to point out that it has a serious limitation. This limitation has a profound bearing on the way buildings from archaeological or historical contexts are understood. Of course, there is a dynamic relationship between buildings and people, and, naturally, it is valid to explore it. However, for archaeologists and historians the limitations of the dramaturgical analogy raise an acute problem that can be understood by performing a simple thought experiment. Let us assume that we have excavated a simple rectilinear building, which has been swept clean of all artefacts. Let us also assume that there are no written records, nor any ethnographic data on which we can draw. What can we say about this structure? It should be apparent that under the terms of the dramaturgical analogy the answer is very little. Outside of statements about the manner of its construction, nothing can be said at all. To make any substantive

statements about its social meaning, we need to repopulate the building with the people and activities it originally contained. Only by doing so can we 'bring it alive'. If this type of information is not readily available, we are left with little to say. It can be so difficult to derive meaning from buildings that many archaeologists are pessimistic about what can be learned from them about past societies. Linda Donley-Reid neatly encapsulates this view when she says, We cannot know the relationship between the architecture and social organization without knowing how that specific culture lived in those spaces. If I know nothing of a culture and am placed in its architectural remains, the walls will tell me nothing (1990: 115, my emphasis).

However plausible this position might seem, it is unsatisfactory in one crucial respect and this is that it renders built space meaningless. If buildings cannot be understood other than through the activities they contained, it follows that they cannot have any meaning in their own right. However, is this really a credible position? There are reasons to think that it is not. Buildings are amongst the largest and most complex artefacts created by a society. Surely, such elaborate artefacts cannot be meaningless. Buildings and power

we accept that buildings are meaningful, we are immediately confronted by the question of how. To answer this question we need to think more carefully about what kind of an entity a building is. Buildings, as Hillier and Hanson remind us, are peculiar artefacts in that while they present themselves to our experience as objects, they also create and order empty volumes of space (1984: 1). This aspect of built space is often neglected, but it is fundamental. Unless this is remembered, it is all too easy to focus on the fabric of the building itself and we can be easily seduced into thinking that what defines a building is the material used in its construction and the nature of the decor applied to its walls. Although, these are important, if we concentrate on them we lose sight of its primary purpose. If

If it is true that buildings function to order empty volumes of

space, we might wonder what space so ordered signifies. Answering this question is not at all straightforward because of the nature of space itself. Space is a void and it seems contradictory to apply meaning to a void. However, we can glimpse an answer by thinking about the effects buildings have on people. Rather than just being 'containers', there is evidence to suggest that buildings have the power to shape us as individuals and the society to which we belong. The most telling evidence comes from planning disasters of the 1960s and 1970s, when supposedly Utopian housing estates and high rise developments generated more social problems than they solved (Coleman 1985). Analysis has indicated that these developments fragmented communities by isolating individuals from one another. This disrupted the traditional

Introduction networks of mutual support that had prevailed in terrace slums and made surveillance difficult. The consequences were squalor, crime and, finally, social breakdown. The example of modem planning demonstrates that in order to function, communities need to inhabit certain kinds of spaces. However, the fact that they cannot simply 'adapt' to different built environments implies that ordering of space itself has a 'power' to shape society. The kind of society engendered by the new estates and tower blocks contradicted the social organisation of the communities that were supposed to live in them.

Wilson argues that architects must produce buildings that are both imaginative and sympathetic to the needs of their users and to do so much deeper theoretical reflection as to their nature is required. The functional and aesthetic approaches to built space have featured prominently in disciplines other than architecture. If architecture needs to rethink its approach to built space, then other disciplines need to revise their thinking as well. In archaeology, there are signs that attitudes are changing, since there has been an attempt to see buildings as material manifestations of ideologies and cosmologies (e.g. Parker Pearson and Richards 1994; Samson 1990). However, after a brief flurry of activity, these attempts have largely faltered, because, methodologically, they still depend on the dramaturgical analogy. Those buildings that have been most satisfactorily interpreted tend to be from those societies which have written records, or for which ethnographic data exists. Prehistorians have been much less successful in reaching the meaning of built space, except in those situations where a direct parallel may be drawn with another type of artefact. Neolithic houses, for instance, have been plausibly compared with tombs (Hodder 1994; Richards 1990).

The power that buildings have to shape people and the communities they live in has been theorised at some length by Michel Foucault (e.g. 1977). His thesis is that buildings are instruments that act upon the body and literally transform the character and personality of the individual. Although Foucault is concerned with buildings designed to 'discipline' (e.g. asylums, prisons and hospitals), it is reasonable to extrapolate from his thinking and see all buildings as having this kind of instrumentality (see chapter two). The linkage between buildings and power has also been made by Thomas Markus (1993). He discusses the power that buildings have to classify and order social relations. He concerns himself with the considerable expansion in the range of building types that began in the eighteenth century. Unlike traditional architectural studies that focus on questions of style and technology, Markus sees the emergence of new building types as heralding the constitution of a new society. This society, which was realised during the Industrial Revolution, required buildings that controlled people and knowledge with greater efficacy than before. New building types were therefore not just the consequence of the development of new materials, or the refinement of building techniques. Nor, did they appear simply because of changes in fashion. Instead, they arose because they were necessary to the functioning of an industrial society.

Architecture, space and society So, despite there being some imaginative approaches to built space, there is still much to do. Buildings may influence society, but it is not entirely clear how. If we are to interpret built space without recourse to textual or ethnographic data we need a much more precise understanding of the relationship between architecture and society. I would argue that the flaw in the dramaturgical analogy is not just methodological, it is also theoretical. By seeing built space as analogous to a series of settings for activities, the dramaturgical analogy implicitly drives a wedge between space and society. Rather than being integral to society, space is merely its product, or outcome. This would imply that society can somehow function independently of space, but, as we have just seen, the conclusions of recent research do not support this view. Indeed, social theorists have rejected the notion that space merely forms a surface upon which human action takes place, but instead they view it as part of the way which society is constituted (e.g. Gregory and Urry 1985). This idea is difficult conceptually because of the nature of space. It is not easy to see how a void can become an active component in social life. Clearly, without understanding how space affects social life, it will be nigh on impossible to interpret built space.

The power of buildings is not just limited to modem society. Pierre Bourdieu (1977) has described how the house in traditional, small scale societies served as more than just a shelter. In his perceptive analysis of the Kabyle house, Bourdieu shows that the house provides a resource from which children learn their vision of the world. By moving through it, the house instils a certain 'way of being' or 'habitus', as Bourdieu calls it, into those who use it. Consequently, the house serves as an instrument that aids the creation and recreation of society. Enough should have been said to make it clear that buildings do have power to shape us both socially and psychologically. The realisation that they do has become central to recent theoretical debates in the discipline of architecture itself For example, the renowned architect, Colin St. John Wilson (1992), has argued forcefully that architecture must transcend the sterile opposition between function and aesthetics. The former sees buildings as utilitarian objects, designed to perform specific tasks, while the latter sees buildings as analogous to works of art, designed to make statements about society. The functionalist approach has usually resulted in sterile and unimaginative buildings, while the aesthetic approach has often produced bizarre structures completely out of tune with the feelings of those who have to use them.

In chapter one, I explore this question by beginning with the most fundamental element of social life, the encounter. Without people encountering one another, society could not exist. I argue in chapter one that when encounters take place the space between and around those participating in the encounter becomes shaped by virtue of the asymmetry between the front and the back of the human body. The social properties of the space in front of and behind the body are altered by this action. From the point of view of the participants in the encounter, the area of space between them is 'public' because each is aware of the other's speech and actions. By contrast, the space around them is 'private' 2

The Walls Will Tell Me Nothing

because as they attend to one another, it 'fades out'. The situation can become increasingly complex if an encounter comes under the observation of others. Space is again transformed into many different public and private zones. Using this idea, I argue that architecture functions in an analogous way. It too creates differentiated spatial domains that have distinct social properties. The primary difference between architecture and the body is that architecture provides some permanence to those properties. In the case of an encounter, once the episode of social interaction is over, space loses its social properties. However, architecture allows those properties to be preserved, at least in part.

advance their own interests than others will. Such individuals will gain power by using it to control knowledge about themselves and to deprive others of their personal autonomy. The same arrangement of space will therefore 'work' to the advantage of some and to the detriment of others. Because the same architectural arrangement can do both at the same time, it cannot ever be a passive 'stage' for activities. Therefore, architecture shapes social relations by constantly exerting a subtle, if unconscious, 'pressure' on all those who use it. By doing so, it helps produce and sustain the social inequalities that define individual identities. Identity is a relative construct: who one is depends one whom one knows. It is only though relations with others that individual identity has any meaning. Because it functions to aid the creation of identities, architecture has institutional effects in that it helps sustain those identities over long periods of time. If so, it is clear what the walls can tell us: they can tell us about social identity.

This analogy, though, does not explain why architecture is necessary in the first place. If space can be shaped with the body, why is it necessary to have architecture? I address this question in chapter two, by introducing into the discussion the concept of knowledge. The purpose of a social encounter is to allow the exchange of knowledge. This knowledge is not just confined to what each participant in the encounter can articulate discursively, it also pertains to knowledge of the self. The exchange of knowledge is necessary for the development of trust between individuals. Without trust, social bonds cannot develop and without social bonds, there can be no society. However, the knowledge communicated by an individual is not given indiscriminately, but is targeted towards specific individuals. Studies in psychology have shown that human beings need to retain control over knowledge about themselves in order to avoid anxiety. Architecture comes into play as a means of creating intimacy. It allows individuals to retain control over knowledge about themselves by retreating into private situations. Buildings thus help individuals sustain a sense of personal autonomy by providing concealment from the prying eyes of the world.

Reading space

This raises the question of how we learn about social identity from architecture. In chapter three, I address this methodological question by suggesting that we can see architecturally divided space as having properties analogous to those of a written text. The reasoning behind this assertion is that because architecture is the outcome of human action it must 'document' that action in some way. How are we to 'read' this document? When we confront a written text, our ability to read it is predicated on our knowledge of the language in which it is written. If a text is written in a language we do not understand, it is unintelligible. We also have to know how to read. Unless we possess this skill, even a text written in a language we speak will still be unintelligible. So, to read space we must first determine the 'language' in which it is written and then discover how to read this language. This can be done by comparing speech and writing with social interaction and architecture. A speech act is a transient event. Once it is complete, it is lost. Writing helps to preserve speech acts by encoding what has been said. This makes knowledge of what was said in the speech act available later. A social encounter can be thought of as being analogous to a speech act. Just like speech, an encounter is a transient event. Architecture is therefore analogous to writing because it helps 'fix' the social properties of space in much the same way as writing fixes speech. If social interaction is analogous to speech and architecture is analogous to writing, what spatial activity counts as reading? The answer to this question is simple: movement through architecturally divided space. Just as we learn to read by encountering texts, we learn to negotiate space by moving through it. As Piaget's (1956) pioneering research into the development of spatial competency in children has shown, the negotiation of space is just as much a learnt skill as is the ability to read.

If so, architecture can be seen as empowering: a means to

freedom. However, it follows that if architecture can empower it can also control. Although it may provide concealment, it can just as easily enforce exposure. By being forcibly exposed to scrutiny, especially to the scrutiny of those in whom a sense of trust has not been built, individuals lose some of their personal autonomy. To avoid shame, that is, anxiety caused through the loss of personal autonomy, individuals are forced to conform to whatever behaviours are considered appropriate in the circumstances. In this way, individuals can be 'coerced' into behaving in ways contradictory to their desires and interests. The act of exposure suggests an inequality between observer and observed. This is understandable when we recognise that inequalities in power are built up through inequalities in knowledge. Individuals gain power over others by ensuring that they have more knowledge about those they observe than those they observe have about them. Those who are observed consequently become 'pliable', meaning that their actions can be more readily directed.

If space is read through movement, it means that if we want to read architectural remains we require a method that helps us recreate the possibilities for movement through a building. In chapter four, I explore one such method: access analysis. Access analysis is a topographical method that describes the relations of permeability in a building and offers indices that allows those relations to be quantified. It was developed

In consequence, architecture appears to have 'two faces', but when does architecture become repressive and when does it act as a means to freedom? The answer to this question is that it always functions as both. Any arrangement of architectural boundaries will tend to simultaneously afford opportunities for privacy and exposure. However, some individuals will be more able to take advantage of this arrangement of space to 3

Introduction primarily by Hillier and Hanson (e.g. 1984) as a design method, but I argue in chapter four that it can also be profitably employed as an analytical technique. Most of the chapter is taken up with a discussion of the methodological issues raised in creating formal analytical plans, or access maps, for buildings and the significance of the different indices that can be calculated from them.

interaction. In slightly larger houses, where there was space for a courtyard, the pattern of interaction was much more centralised, with encounters habitually directed towards the courtyard. In these single courtyard houses, there was a clear cut opposition between the 'public' space of the courtyard and the more 'private' rooms that surround it. In houses with more than one courtyard, the pattern of interaction became more dispersed because more courtyards provided 'rival' centres for interaction. Although each courtyard still dominated interaction at the local level, globally their influence was diminished. This was particularly noticeable in those exceptionally large houses with three or four courtyards. It was clear from these findings that the houses engendered a variety of social patterns. Just as there was no standardised ground plan, there was no standardised pattern of interaction either.

Pompeii Having established how space may be read, I move on to consider a case study, namely that of the houses of Roman Pompeii. Chapter five presents a brief overview of Pompeii and discusses the selection of a case study from the many surviving houses. The specific issues raised in applying access analysis to these buildings are also considered. Pompeii presents an almost unparalleled opportunity to undertake an extensive case study of buildings in an archaeological context, using the theory and methods advanced in the first part of this study. In addition, Pompeian houses have long been interpreted by reference to the Roman literary sources. This approach, although not explicitly acknowledged as such, uses the dramaturgical analogy. This makes it possible to compare the results of this study with those derived from a more orthodox approach. This is done in the final chapter.

Such heterogeneous social patterns suggest the formation of heterogeneous social identities. By this, I mean that the houses help set up and sustain an extensive range of different identities. In chapter eight, I explore the nature of these identities by employing the opposition between individualistic and collectivistic identities (Duncan 1985). While not being able to reach the substance of any particular identity from a study of built space, it is possible to reach the sense of identity that individuals may have had. Crucial to accomplishing this is the recognition that our modem, Western perception of identity is individualistic: each person is seen to be endowed with certain rights, prerogatives and responsibilities, which he or she holds as an individual. However, not all societies conceive of individuality in this way. In many non-W estem societies, individuals define themselves in terms of a wider social group. The rights and obligations held by a single individual depend on the prerogatives and responsibilities of the group as a whole. In this way, individual identity is 'diffused' throughout a collective. While acknowledging that societies cannot be easily compartmentalised, the opposition between individualistic and collectivistic identities does provide a useful framework for interpretation.

Analysis and interpretation The remaining three chapters are concerned with the analysis and interpretation of the data generated by access analysis. One hundred and forty-four houses were analysed to provide the data for this study (Grahame 1995) and access maps and tables of data for all the houses have been included in the appendices. In chapter six, I use these data to locate the syntaxes, which are the rules that govern the ordering of space. Using probability theory, I explore how complex spatial configurations can be built up from much simpler patterns. Crucially, I show that order can arise from the following of only a few rules. By applying this thinking to the data from the case study, I demonstrate that house plans in Pompeii are highly contingent. I mean by this that their final forms were conditioned by a series of factors that operated locally and not by a set of rules common to all. For this reason, each house plan is idiosyncratic and nothing emerges from the analysis that could be construed as a standardised plan.

Using this paradigm, I argue that all houses in Pompeii would have generated a strong sense of individual identity. However, Pompeian society was not as individualistic as modem W estem society because individual identity was rooted in a collective identity. This was based on the household. It may seem contradictory to have a strong sense of individuality and a clearly delineated community, but the two can be reconciled by returning to the idea that architecture is an instrument of social power. Both individual and household identity can be strong if we conceive of each house as being the locus for a social hierarchy. To be stable, a hierarchy has to be clearly defined, which presupposes a strong communal identity. In addition, each person will occupy a specific 'position' within that hierarchy and so will have a strong sense of their own individuality. Each house in Pompeii can therefore be thought of as having generated and contained a defined social hierarchy. Power was thus localised into the houses themselves, rather than being distributed throughout the wider community. Competition between households to create and sustain social hierarchies would have generated a mosaic of houses of different sizes

Chapter seven moves from the abstract level of space syntaxes to the concrete level of movement and encounter. In this chapter, I analyse how a representative sample oflayouts, drawn from the larger study, conditioned the potential for social interaction. By potential for interaction, I mean how likely it was that one person would have encountered another in any given space. To arrive at an estimate for the interaction potential of a space, access analysis had to be extended to include an assessment of the effects of physical dimensions on the probability of an encounter. By analysing how different houses shaped interaction, it was possible to identify a trend in the social patterns generated by them. The smallest houses of all, characterised by simple arrangements of space, usually generated a diffuse pattern of interaction. In other words, no one space acted as a 'centre of gravity' for 4

The Walls Will Tell Me Nothing factors. If so, it seems that we cannot see Pompeian housing as being 'Roman' in any direct or unambiguous sense.

and with different spatial compositions. The formation of hierarchies thus helps to explain the spatial heterogeneity discovered in chapter six.

This conflict of images raises the inevitable question of which is preferable. My argument is that the desire to interpret the houses of Pompeii by reference to the Roman literary sources is flawed; not I should stress, because the sources are somehow at fault, but rather because the assumptions underpinning the traditional methodology are unsound. The flaw is precisely the limitation of the dramaturgical analogy discussed at the head of this chapter. We cannot know what went on in a space unless we have documentation that refers to that space. The Roman literary sources do not speak directly of Pompeian houses, nor do they discuss the people and practices located within them. Consequently, the way of living in Pompeian houses is usually inferred indirectly from often casual comments by the ancient authors. To use the sources in this way, it has to be assumed that texts and archaeology emerge from the same cultural context. However, this merely begs the question of what constitutes the same cultural context and as we have seen it is not easy to link the houses of Pompeii with a wider Roman culture. If so, the literary sources only 'touch' Pompeian housing in a vague and rather indirect manner.

Conflicting images As was mentioned earlier, in the final chapter I contrast the findings of this study with the conclusions drawn from the approaches conventionally used by historians and archaeologists to interpret Pompeian houses. I do so by contrasting the 'image' of the Pompeian house offered by this study with its orthodox 'image'. Quite remarkably, the two are almost diametrically opposed. Pompeian houses are usually seen as having relatively standardised plans, but this study suggests the opposite. Pompeian houses are also usually understood to have sheltered a known set of individuals: the master of the house, his wife, their offspring and slaves. However, this study indicates that the houses were socially much more heterogeneous than that. Finally, Pompeian houses are usually seen as belonging to a much broader cultural horizon, that we call 'Roman', to the point where to speak of one is to imply the other. Given the spatial and social heterogeneity found in Pompeian houses, this linkage must be called into question. It was suggested earlier that it is difficult to sustain the argument that the builders of Pompeian houses all followed a common set of rules. If so, the surviving houses cannot be seen to indicate the presence of cultural homogeneity and if cultural homogeneity cannot be found in Pompeii, it is difficult to give credence to the idea that Pompeii was part of a much broader cultural horizon. Rather, it makes more sense to suggest that Pompeian housing was primarily conditioned by local

This study, I argue, generates a more realistic image of Pompeian housing because it sees space as analogous to a text that is read with the body. By using access analysis to simulate movement, this text can be read and the social significance of space revealed. Because it is in situ, this text documents Pompeian society more reliably than the literary sources do.

5

PART ONE. THEORY AND METHOD FOR THE INTERPRETATION OF BUILT SPACE CHAPTER ONE

Space and Spatiality interpretation. Indeed, some archaeologists have suggested that unless such evidence is available, nothing can be learned from the walls alone (Donley-Reid 1990). This view is not only overly pessimistic, it is also flawed (see Introduction), since the implication is that buildings are only meaningful in relation to the people and activities they shelter. This must be in error, since it implies that buildings are not in any way constitutive of the activities they contain. Therefore, although a unification of the objective perspective and the subjective standpoint may be justifiable on theoretical grounds, from an archaeological standpoint, it would be more satisfactory if we could begin with the objective perspective and work towards the subjective standpoint. Despite criticism of it, the objective perspective is easier to apply to archaeological remains and this no doubt explains archaeological interest in the techniques of spatial analysis (e.g. Clarke 1977; Hietala 1984; Hodder and Orton 1976; Wagstaff 1987). However, reaching the subjective standpoint from objective conditions is not at all easy. Indeed, knowing what might have been in the minds of past people is a contentious issue even amongst those who have been critical of the objective perspective (Hodder 1992. Cf. Barrett 1988). However, we need not go this far. Subjectivity is not just confined the feelings or ideas individuals have, it also involves the way in which they exist in the world. Or, to put it another way, it concerns their 'way of being'. How individuals exist in the world is partly determined by the objects they encounter. For example, clothes condition bearing and dining utensils, eating practices. However, objects do not just affect the way these activities are performed, they also condition the prevailing attitudes towards them. In other words, subjectivity arises, at least in part, from an encounter with the real world. If so, it follows that we can learn something about the way in which subjectivity was constituted in the past from a study of the objects that survive from it. We can, in short, work from the objective perspective to the subjective standpoint.

Objectivity and subjectivity If we want to interpret built space, how should we go about

it? In trying to answer this question we are immediately faced with a complex theoretical issue, namely, the nature of space itself Quite obviously, we cannot make sense of space unless we have some idea of what sort of entity it is. The problem for us, as Werlan (1993) has recently shown, is that the nature of space is hotly debated. However, broadly speaking, we can discern two main theories. The first we might call the 'objective perspective'. This approach has an impeccable intellectual pedigree, emanating as it does from the writings of Descartes and Kant. The objective perspective conceives of space as a continuous physical domain upon which individuals are located. Each individual can be assigned a position in space according to a system of grid co-ordinates and by measuring the time taken to move from one point on this grid to another, the degree of 'friction' or 'resistance' to movement can be calculated. The need to expend energy while crossing space has implications for how frequently individuals can encounter one another, or acquire the resources they either want or need. The criticism most frequently levelled at the objective perspective is that it dehumanises space (e.g. Harvey 1973, 1989; Relph 1976; Gregory and Urry 1985; Soja 1989). Individuals are not just positioned in space, they also have feelings about it. Space is a meaningful experience and how individuals 'feel' about the spaces they inhabit affects how they act within them. Human action in space cannot therefore be reduced to measurements of the energy expended to cross it. To interpret space, we consequently have to have knowledge of the symbolic meanings attached to it. Often the revised designation 'social space' is used to capture this subjective concern for, and involvement with, space. This, the second main theory of space, we might call the 'subjective standpoint'. Debate in the social sciences over these two paradigms has been lively and has recently spilled over into archaeology (e.g. Hodder 1982a and b, 1987, 1991, 1992; Miller and Tilley 1984; Shanks and Tilley 1987, 1992; Thomas 1993; Tilley 1994).

In the case of space, how we go about doing this is not easy to conceptualise, since space is a void. Indeed, the objective perspective conceives of space as nothing more than a surface and it is difficult to see how a surface can condition subjectivity, except in the most rudimentary of ways. The solution to the predicament lies in recognising that when people are placed in space, their presence 'transforms' it, and it is this transformation, rather than space itself, that operates to condition subjectivity. To explore how, let us begin from first principles. It is obvious that space becomes transformed when it is divided through the erection of boundaries. Through boundaries, the continuum of physical space is fragmented into smaller units, but from the subjective standpoint much more is involved than just the division of

Various attempt have been made to unify the objective and subjective paradigms (e.g. Lefebvre 1991; Werlan 1993), but they all are difficult to apply in archaeological situations, since they normally require an appreciation of the dynamic relationship between human beings and the spaces they inhabit. This usually necessitates an a priori understanding of the subjective meanings assigned to space. In some archaeological situations, this type of analysis is possible, but in most, it is not. Without evidence as to how a culture lived in the spaces it created there are clear limits to this type of 6

Space and Spatiality

space. To appreciate what, we first need to understand more about how that entity we call 'society' comes into being. To do this, I would like to draw extensively on the sociological theory of Anthony Giddens (1979, 1981, 1984), and, in particular, his commentary on the writings of Erving Goffinan (1959, 1963).

on a face-to-face basis - the turning of ones back being synonymous in many, if not all, cultures with an insult. When two, or more, individuals face one another the architecture of the body differentiates the space between them from that surrounding them. Figure 1.1 illustrates the situation in diagrammatic form. The space enclosed by the fronts of the two bodies is the front region to the encounter, while that outside ofit is the back region.

The body, space and social encounters According to Giddens, the most basic elements of social life are encounters. Encounters are, 'the guiding thread of social interaction', the succession of engagements that we have with others, 'ordered within the daily cycle of activities' (Giddens

According to Goffinan 'front' is, that part of the individual's performance which regularly functions in a general and fixed fashion to define the situation for those who observe the

CO-PRESENCE

Back region

Back region

Figure 1.1. Illustration of two individuals, A and B, involved in a social encounter. The solid boarder defines the limits of co-presence and the 'front' and 'back' regions of the encounter are marked.

performance (1959: 32).

1984: 72). Without regular contact with others, social life would be impossible and society could not exist. Encounters take place in space, but much more is involved than simply being located at a particular point relative to others. Through a social encounter the continuum of physical space is 'shaped' and the 'instrument' that performs this task is the human body. It has long been recognised that the way the body is physically structured has an important bearing on the way in which we experience and orient ourselves in the world. As Tuan has observed:

Hence, a front region is the space within which a 'performance' is given and received. Giddens, however, is critical of Goffinan because of his use of the dramaturgical analogy, which conceptualises social action in terms of 'performances' given by social 'actors' to 'audiences'. Giddens instead stresses the mutual monitoring of conduct that occurs during periods of social interaction. This is an important point because social interaction is not just restricted to face-to-face encounters, in the manner of an actor facing an audience, but to a much more general 'being together', which Giddens has termed, following Goffinan, copresence (1984: 69-72). Co-presence is achieved when individuals, 'sense that they are close enough to be perceived in whatever they are doing ... and close enough to be perceived in this sensing of being perceived' (Goffinan 1963: 17).

Among the mammals the human body is unique in that it easily maintains an upright position. Upright, man is ready to act. Space opens out before him and is immediately differentiable into front-back and left-right axes in conformity with the structure of his body(1977: 35).

These axes of differentiation can be thought of as the architecture of the body. However, of all these axes the frontback opposition is the most significant one socially. While the body is symmetrical according to the left-right axis, the forward orientation of the senses creates an asymmetry between 'back' and 'front' and this has distinct consequences for social interaction. The face, in particular, is, 'the dominant area of the body across which the intricacies of experience, feeling and intention are written' (Giddens 1984: 67) and when human beings engage in social interaction it is

Broadly speaking, we can discern two ways that individuals can interact when co-present. The first occurs when individuals engage in a face-to-face encounter. This is known as focused interaction because during the encounter each participant focuses on what the other participants are saying or doing (Giddens 1989: 101). In addition to listening to what is being said and observing what is being done, the 7

Chapter One

participants also pay attention to a host of much subtler signals, which include tone of voice, gestures, facial expressions and body language. These are all monitored by each participant as part of the on-going flow of interaction. In contrast, unfocused interaction only requires that individuals be aware of one another. Individuals choosing not to pay attention to others may, in fact, consciously avoid an encounter. Even so, those individuals will still be aware that they are co-present with others, despite their failure to acknowledge them explicitly.

back region is the space of enclosure of the body because others are 'hidden away' (table 1.1). Nevertheless, it is still possible for those engaged in focused interaction to become aware of any disturbance in the back region, especially if another individual enters the conditions of co-presence. From the point of view of this individual, the bodies of those participating in the encounter form a 'wall' that bars his or her participation. This may be beneficial, permitting this individual to go about his or her business unnoticed and uninterrupted, but it might also be disadvantageous if this individual is seeking social contact. Obviously enough, to be included in the encounter the individual has to 'break through' the 'wall'. The new participant may, of course, be welcomed, but the need to break into the encounter may also be used as a strategy by those already engaged in focused interaction to exclude.

Table I.I. Summary of the effects on interaction and the body of the division of space into front and back regions. Region

Front

Back

Interaction type

Focused Interaction

Unfocused Interaction

Attitude of the body

Disclosure

Enclosure

When an individual comes across others engaged in focused interaction, and is excluded, what options are available? The individual could move on, but if he or she remains within the conditions of co-presence and observes the social engagement without actually participating in it, space is further transformed by this act of observation. This is represented in diagrammatic form in figure 1.2. A new front region opens out between the observer, C, and those being observed (A and B). Although, the observed still exclude the observer, they are now within the front region defined by the observer's body. This front region is obviously different from that defined by the bodies of A and B. From the perspective of C, the front region defined by A and B is a domain from which he or she is excluded. It is, for C, a space of enclosure, a back region, to which access is denied. However, the exclusion of C does not mean that A and B are not affected by C's presence. The differentiation of space into front and back regions was the means by which A and B attempted to

Consequently, a front region is the spatial domain formed by the front of the body during focused interaction. It is a 'public' space within which the participants in focused interaction are both observing and observed. Outside the front region, behind the face-to-face engagement, the back region is the space of unfocused interaction. While engaged in focused interaction, it becomes difficult to be aware of what is going on in this region of space, despite attempts by the participants to monitor it (Giddens 1984: 72). The back region thus forms a 'background' to a social encounter and while it is on going this background 'fades away'. This differentiation of space has consequences for the visibility of the body. From the point of view of the participants in the encounter, the front region is the space of disclosure of the body: a space of 'seeing' and 'being seen'. In contrast, the

CO-PRESENCE

Back region (A+B)

C

Back region (A+B+C)

Figure 1.2. Illustration of the effect on space when a social encounter between two individuals, A and B, comes under the observation of a third individual, C, within the conditions of co-presence.

8

Space and Spatiality

guarantee seclusion from others. However, by being observed by C their privacy dissolves in that they become objects of 'public' scrutiny. Even though they may not attend to C, C attends to them and so gains knowledge of their social encounter. Indeed, the very presence of C may be sufficient to bring the encounter to an end. Consequently, however much A and B may try to maintain the conditions of exclusion, the presence of C will inevitably influence the course of their encounter in some way or other.

the same way as the body 'protects' individual instances of focused interaction. However, if we were to leave matters here we would be working with a conception of architecture as a technology that functions to define contexts for social activities. However, the effect of architecture is much more profound because when a region of space is enclosed within a boundary its social character is altered. The movement from one space to another is not just a change of locales, it is also marks a change of social situation. It has been usual to see the social character of a space as being defined purely by the functions (occasions) it contains. However, I wish to argue that just as space is transformed into front and back regions by the body, it is also similarly transformed by architecture. The difficulty, however, is explaining how the spaces transformed by architecture manage to retain their 'front' or 'back' characteristics when no one is in them. To overcome this difficulty we need to recognise the crucial distinction between the space of, and the spatiality of, social life.

The situation would become more complex still if another individual were to enter the conditions of co-presence and observe C observing A and B. Without detailing what would happen, it should be apparent that space would be transformed yet again and new front and back regions would be defined. It follows from these simple examples that 'front' and 'back' are not absolute categories, but are relative to the position of the body with respect to others. Consequently, any configuration of front and back lasts only so long as individuals maintain their positions in space, but given that human beings are mobile, we can see that front and back regions are transient and are subject to continual redefinition.

Space and spatiality

To make sense of the difference between these concepts I would like to draw on Dreyfus' (1992) commentary on Heidegger's existential philosophy of Being. Dreyfus has argued that Heidegger's seminal work Being and Time (1962) may be read as a critique of Husserl's phenomenology. Husserl, in keeping with the tenets of Cartesian philosophy (the objective perspective), believed that the world of experience is independent of our feelings about it. Consequently, in order to 'know' the world we first have to transcend our feelings about it, so that we may see it as it 'really is'. Heidegger inverted this proposition by showing that any knowledge that we may have about the world depends in the first instance on a 'background' of social practices into which we are socialised from birth. These provide us with a foundation for making sense of the world in the first place. Heidegger describes this condition as being-in-the-world. The 'world' in Heidegger's sense is both the physical and social environments we inhabit. The important contribution of Heidegger's philosophy was to recognise that our subjective standpoint arises out of the objective conditions that we confront. If this is so, then it follows that the social relations in which we engage must also be conditioned by the physical spaces in which we live.

Gatherings and occasions

What is important about front and back regions is that they are concepts that transcend the dichotomy created between the objective perspective and the subjective standpoint. These regions are partly objective in that they are constructed by the positioning of physical objects (i.e. bodies) in space and partly subjective in that they have different sensations attached to them. However, if space is transformed through the architecture of the body, what effect do architectural boundaries have? It follows logically that if the body transforms space, then so must architecture. To help us explore this issue we need to introduce the distinction Goffman has made between two types of social encounter: gatherings and occasions. Although both involve focused interaction, they have different temporal rhythms. Gatherings are encounters that 'have a very loose and transitory form, such as that of a fleeting exchange of "friendly glances" or greetings in a hallway' (Giddens 1984: 71). Occasions, on the other hand, habitually involve a plurality of individuals and are often clearly bounded in space and time. Giddens has called the places at which occasions occur, locales (1984: 375). A locale is a physical region of space with definite boundaries, which functions to provide a setting for an occasion. Occasions tend to be organised as sequenced phenomena in that during the daily cycle of activities, certain individuals will frequent particular locales at certain times. Who is required to be present at an occasion, and, more importantly, who needs to be excluded, is usually well understood by the participants. Occasions are therefore less transient in nature than gatherings and for this reason it would be impossible to sustain them if they were continually disrupted by the scrutiny of others. We have already seen that the front and back regions formed by the architecture of the body are ephemeral and so a more robust form of shielding is necessary. This is where architectural boundaries come into play. They help 'protect' the integrity of occasions in much

To understand how, we can draw again on Dreyfus (1992: 42-5), who points out that in English we can discern two senses of the word 'in'. The first is a purely spatial sense, implying location (e.g. It is in the box; He is in the house; She lives in Italy). This use of 'in' expresses inclusion, but this is different from the second, existential, sense of 'in' that expresses involvement (e.g. 'in love', 'in business', 'in a bad mood'). Heidegger's term being-in (with a hyphen) relates to this latter sense, in that being-in-the-world does not just mean a location in space, but involvement with the world. Being-in is consequently characterised by 'concern', something that can only be a subjective property of individuals. Being in (without a hyphen), on the other hand, is characterised by indifference. Spatial relations are an objective property of 9

Chapter One

both people and objects, but only people, of course, can be involved and have social relations. However, this does not mean that inclusion is irrelevant to social life. We have already seen that social relations depend on encounters and that encounters cannot take place without individuals being co-present. It follows that inclusion is a necessary precondition for involvement. Of course, involvement is not simply a single state of being-in, in that one is either 'involved', or one is 'not'. Social relations are expressed linguistically in terms of distance (e.g., She is a distant cousin. He is a close friend. We have drifted apart.). In other words, we feel 'closer' to some people because we are more involved with them and more 'distant' from others because we are less involved with them. Spatial metaphors are thus used to express the degree of caring we have for others. It is the sensation that we are closer to some and further from others that constitutes the spatiality of social life.

in which our sense of spatial distance has been 'annihilated' through a reduction in the time taken to communicate across it. The sense that the resistance to movement inherent in physical space can be overcome to accommodate social interaction has been termed distanciation (Giddens 1984: 258-5, 377; Harvey 1989: 222). What constitutes a 'long way' is therefore not just a matter of the quantity of physical space that has to be traversed, rather it depends on how easily and quickly distances can be overcome. What I would like to do here is suggest that the reverse is also true: a sensation of distance may be created in a physical region of space that may actually be quite small. This can be done by enclosing space within an architectural boundary (figure 1.3). The effect of a boundary is to create a distinction between the space screened off by the partition and that

A