The social logic of space Le fait humain par excellence est peut-etre moins la creation de l'outil que la domestication du temps et de Tespace, c'est-a-dire la creation d'un temps et d'une espace humaine. Andr6 Leroi-Gourhan: La Geste et la Parole TO OUR STUDENTS The social logic of space BILL HILLIER JULIENNE HANSON Bartlett School of Architecture and Planning University College London 1 CAMBRIDGE UNIVERSITY PRESS CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo Cambridge University Press The Edinburgh Building, Cambridge CB2 2RU, UK Published in the United States of America by Cambridge University Press, New York www. org Information on this title: www.org/9780521233651 © Cambridge University Press 1984 This book is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press.
First published 1984 First paperback edition 1988 Reprinted 1990,1993, 1997,2001, 2003 A catalogue recordfor this publication is available from the British Library Library of Congress catalogue card number. 83-15004 ISBN-13 978-0-521-23365-1 hardback ISBN-10 0-521-23365-8 hardback ISBN-13 978-0-521 -36784-4 paperback ISBN-10 0-521-36784-0 paperback Transferred to digital printing 2005 Contents Preface ix Introduction 1 The problem of space 26 Society and space 26 The problem of space 29 The logic of discrete systems 33 The inverted genotype 42 Morphic languages 45 The logic of space 52 Introduction 52 Compressed descriptions 53 Some examples 55 Elementary generators: an ideographic language 66 The analysis of settlement layouts 82 Individuals and classes 82 A model for syntactic representation, analysis and interpretation: alpha-analysis 90 A procedure for analysis 97 Some differences 123 An excursion into interpretation: two social paradigms of space? 140 Buildings and their genotypes 143 Insides and outsides: the reversal effect 143 The analysis of the subdivided cell 147 Some examples of domestic space 155 Two large complexes from the ethnographic record 163 The elementary building and its transformations 176 Elementary buildings 176 Reversed buildings and others 183 The spatial logic of arrangements 198 From structures to particular realities 198 vii viii Contents Abstract materialism 201 The semantic illusion 206 7 The spatial logic of encounters: a computer-aided thought experiment 223 A naive experiment 223 Societies as encounter probabilities 234 8 Societies as spatial systems 242 Some societies 242 Notes towards a general theory 256 Postscript 262 The social logic of space today 262 Notes 269 Index 276 Preface However much we may prefer to discuss architecture in terms of visual styles, its most far-reaching practical effects are not at the level of appearances at all, but at the level of space. By giving shape and form to our material world, architecture structures the system of space in which we live and move. In that it does so, it has a direct relation - rather than a merely symbolic one - to social life, since it provides the material preconditions for the patterns of movement, encounter and avoidance which are the material realisation - as well as sometimes the generator - of social relations.
In this sense, architecture pervades our everyday experience far more than a preoccupation with its visual proper- ties would suggest. But however pervasive of everyday experience, the relation between space and social life is certainly very poorly understood. In fact for a long time it has been both a puzzle and a source of controversy in the social sciences. It seems as naive to believe that spatial organisation through architectural form can have a deter- minative effect on social relations as to believe that any such relation is entirely absent.
Recent reviews of sociological research in the area (Michelson, 19761) do not really resolve the matter. Some limited influences from such generalised spatial factors as density to social relations are conceded, subject to strong inter- action with such sociological variables as family (p. But little is said about the ways in which strategic architectural decisions about built form and spatial organisation may have social consequences. The puzzle is made more acute by the widespread belief that many modern environments are 'socially bad5.
Again, there is a tendency to discuss these in terms of simple and general physical variables, such as building height. However, the inference that more fundamental spatial factors are involved is strongly sup- ported by the failure of recent low-rise, high-density schemes to provide a convincing alternative following the debacle of high- rise housing. Modern high- and low-rise housing have in common that they innovate fundamentally in spatial organisation, and both produce, in common it seems, lifeless and deserted environments. IX x Preface It has become clear that a lack of understanding of the precise nature of the relation between spatial organisation and social life is the chief obstacle to better design.
The obvious place to seek such an understanding is in the disciplines that are concerned with the effect of social life on spatial organisation - how spatial organisation is in some sense a product of social structure. This has long been a central concern for geographers, but recently anthropologists (Levi-Strauss, 1963; Bourdieu, 1973, 1977), theoretical sociologists (Giddens, 1981) and archeologists (Ucko et., 1972; Clarke, 1977; Renfrew, 1977; Hodder, 1978) have become aware of the spatial dimension in their subject, and its importance to questions of social morpholo- gy and structure.2 This has created the early stages of a new interdisciplinary literature on the study of space and society. The first result of this attention, however, has been to show how little effective theory and methodology there is in understanding the society-space relation, in spite of two decades or more of the 'quantitative revolution'. But while academic disciplines may simply deplore the lack of theory, for architects and planners the problem is a more pressing one, since as things stand there is no way that scientific theory of the society-space relation can either help to understand what has gone wrong with contemporary design or suggest new approaches.
The aim of this book is to reverse the assumption that know- ledge must first be created in the academic disciplines before being used in the applied ones, by using architecture as a basis for building a new theory - and a new approach to theory - of the society-space relation. This is possible, we believe, because theories of the relation between society and its spatial form have encountered two fundamental difficulties. First, there is no con- sistent descriptive account of the morphological features of 'man- made' space that could be lawfully determined by social proces- ses and structures. Second, there is no descriptive account of the morphological features of societies that could require one kind of spatial embodiment rather than another.
The reason for this lack of progress is at root to do with the paradigm within which we conceptualise space which, even in its most progressive forms postulates a more or less abstract - certainly a-spatial - domain of society to be linked to another, purely physical domain of space. The paradigm in effect conceptualises space as being without social content and society without spatial content. Yet neither can be the case, if there are to be lawful relations between them. The aim of The Social Logic of Space is to begin with architecture, and to outline a new theory and method for the investigation of the society-space relation which takes account of these underlying difficulties.
First, it attempts to build a concep- tual model within which the relation can be investigated on the basis of the social content of spatial patterning and the spatial Preface xi content of social patterning. Second, it tries to establish, via a new definition of spatial order as restrictions on a random process, a method of analysis of spatial pattern, with emphasis on the relation between local morphological relations and global patterns. It establishes a fundamental descriptive theory of pattern types and then a method of analysis. These are applied first to settlements and then to building interiors in order to discover and quantify the presence of different local and global morphological features.
On this basis, it establishes a descriptive theory of how spatial pattern can, and does, in itself carry social information and content. The argument then turns to society, and extends the same morphological argument into the domain of social relations, by considering them as restrictions on random encounter patterns. From this naive spatial view of society, a theory is developed of how and why different forms of social reproduction require and find an embodiment in a different type of spatial order. This 'spatial logic of society' is applied first to some well-documented examples to establish the theory in outline, then is applied to try to give some account of the variability in spatial form in contem- porary industrial societies.
The book is thus a statement of a new theory and sketch of new methods of spatial analysis. It should be emphasised, however, that a considerable number of studies have now been carried out at University College London using this framework, and it is intended that further volumes of case studies using the theory and method should follow The Social Logic of Space as soon as possible: these include the social logic of settlements, the social logic of housing, and the social logic of complex buildings. Because it represents a new theoretical departure, however, The Social Logic of Space embeds itself only tangentially in the established frameworks and methods of the subject. Even fields of research that might appear, at first glance, to be close to our approach, turn out eventually to have limited relevance.
For example, the 'pattern language' of Christopher Alexander and his colleagues at Berkeley (1977),3 while appearing at first to be close to our notion of fundamental syntactic generators, is in fact quite remote, in intention as well as in its intrinsic nature. For our purposes, Alexander's notion of a pattern is too bound to the contingent properties of configurations to be useful for us; while at a more abstract level, his preoccupation with hierarchical forms of spatial arrangement (surprising in view of his earlier attack on hierarchical thinking in 'A city is not a tree' (1966)4) would hinder the formation of non-hierarchical, abstract notions of spatial relations which, in our view, are essential to giving a proper account of spatial organisation. The more recent development of 'shape grammars' by Stiny and Gips (1978)5 would again, at first sight, appear to be close to the xii Preface notion of 'space syntax' as formulated in this book, the more so since 'shape grammars' are firmly concerned with the abstract generative principles of spatial patterns. But while conceding their superior mathematical refinement, we have found that shape grammars are in general too over-refined to model the untidy systems which are found in the real world of settlements and buildings.
Our notion of 'syntactic generators' is insufficiently formalised for a full mathematical treatment, yet syntactic gener- ators are right for the job that they are intended to do: capturing the formal dimensions of real-world spatial systems in terms of the social logic behind them. Syntactic generators are simpler than shape grammars. Moreover, they are shape free. We are convinced that it is unnecessary to specify shape in order to model real-world generative processes; indeed, that the concept of shape obscures the fundamental relational notions that underpin human spatial order.
Moreover, with the limited role assigned to randomness in shape grammars - as opposed to the foundation of space syntax on the notion of randomness - we find that in their very foundations they tend to overdetermine the realities that we are trying to model. At a more general level, we can properly be accused of ignoring the considerable development of mathematical methods of spatial analysis in quantitative geography. The reason for our lack of continuity with this work is more fundamental.