Vous êtes sur la page 1sur 219

Socia l and Political Theory from Polity Press

David Beetham, Max Weber and Ihe Th eory of M odem Politic.l
Richard Be rnstein, ed., Habemws alld Modernity
Ric hard Bernstein, Philosophical Profile:,'
The Constitution
John Burnheim, Is Democracy Possible?
Harold Garfinkel, Studies in Elhnomelhodology
Anthony G idde ns, The Natioll -Stale alld Violence (Vol. 2 of of Society
A Contemporary Critique of Historical Materiah~m)
John He ritage, Gat/inkel and Ethnomelhodology
1. N. Isbister, Freud: All Introduction to his Life and Work Outline of the Theory of Structuratio n
Martin Jay. Mar.ri.fm alld Totality
Hans Joas, G. H. Mead: A COlllemporOlY Re-examilllllion of
J/J:\' Th ol4ghl
Harvey J. Kaye, The British Marxi.rt Historian.f
Thomas McCa rthy, Th e Critical Theory of liirgen Habermas Anthonyt iddens
Claus Offe, DisorRanized Capitalism
Ca role Paleman. The Problem of Political Obligalion
Mark Poster, Foucault, Mar.ri.HlI alld History
Julia Swindells, Victorian Writ ing and Working Women
John 8 . Thompson , Studie.~ in the Theory o/lde% RY

A Selec tion o f Forthcoming Titles

Barry Barnes, The Basis of Power
Zygmunt Bauman, Legislators and Inlerpreler:,-
Roy Bhaskar, Reason, Emancipatioll and Being
Norberta Bobbio, What is S()cialism ?
Pierre Bourd ie u, Language and Symbolic Power
Cornelius Castoriadis, The Im aginary In sli/l/ rion of Society
W. Haug. Critique of Commodity A e.I'lherics
David Held, Model.~ of Democracy
C laude Lerort, The Politicol Forms of M odem Society
Niklas Luhmann. L(JI'e os PassiOIl
William Outhwa ite, Habemws
Alan Rya n, Political Philo.mphy: All Illtrodllction
Barbara Sichtermann. Femillinily: The Politic.~ nf the Per.wllal
Mic helle Stanworth. FeminisfII (jlld S(){'i()lo~y
Jo hn B. T ho mpson. The IlIl u /prell/lilJII of Ide() I()~y
Nigel Thrift, Slwiol Th erNY (/lId Hl/fIIml G ('(j~ml'lIy
Jamcs Tu lly. cd .. 1\1ellllill~ (/1/(/ (,u lIll',11
Jo hn Ul'ry and SCUll LISh . TI/(' Hlld o( ()r~a lliz ('d C(ll'ilall:l'II/
j ' :1II1 Willis, To I.ahll/l": Til, S"I/i"dil'" Sidl ' ,,/Capilul Polity Press FSZEK Kozponti K6n yvtar

"1111 11111111111111 II
Anthony G iddens.
First published 1984 by
Po lity Press, Cambridge, in association with Basil Blackwell. O):fo rd .
Reprinted 1985. First paperback edition 1986.

Editorial Office:
Polity Press.
Dales Brewery. Gwydir Street, Cambridge. CHI 2U , UK
Basil Blackwell Ltd ,
108, Cowley Road. Oxford , OX4 UF, UK Prerace ix
Abbreviat io ns
All rights reserved. E):cept for the quo ta tion of short passages for the pu rposes Introduction "
o f c riticism and review, no pari of this publication may be re produced. slo red
in a re trie val system. or transmitted. in any fann or by any means, e lec tron ic , Elements of tbe Tbeory of Structuration
mechanical , photocopying. recording or otherwise, without the prior
permission of the publisher. The Agent, Agency 5
Agency and Power 14
E):cept in the United States of America . this book is sold subject to the Structure, Structuration 16
condition that it shall nOlo by way of tmde or o therwise be circ ulated without T he Duality of Structure 2S
the publisher's prior consenl in an y fo rm of binding o r cover othe r than tha i in Foons of Institution 28
which it is published and witho ut a simila r condition including this conditio n
being imposed on the subsequent purc haser. Time, the Body, Encounte rs 34

British Ubrary Cataloguing in Publication Data 2 Consciousness, Self and Social Encounters 41
Giddens, Anthony Reflexivity, Discursive and Practical Consciousness 41
The Constitution of Society T he Unconscious, Time, Memory 4S
1. Sociology 51
Erikson: Anxiety and Trust
I . Title
30 1' .0 1 HM 24 Routinization and Mo tivation 60
Presence, Co-Presence and Social Integration 64
ISBN 0-7456-0006-9
ISBN 0-74S6-00077pb
r ",-" 1 3J 3 Coffman: Encounte rs and Routines 68
Seriality 73
Talk , Reflexivity 78
Positioning 83
Critical Noles: Freud on Slips 0/ the Tongue 93

3 Time, Space and Reglonallzatlon 110

Time-Geography 110
T ypscl II)' Piorlccr . East Susscx C ritical Comments 116
Pdnl etl ill {t rent Ilriw in Modes o f Regionalization 119
Front Regio ns, Back Regio ns . 22 Mutual Knowledge versus Commo n Sense 334
Disclosure and Self ' 26 Generalizatio ns in Social Scie nce 343
Regionalization as Generic .30 The Prac tical ConnOlatio ns of Social Science 348
Time. Space, Context .32
Against 'M icro' and 'Macro': Social and System Integration ' 39 en'tical No tes: Social Science, History and Geography 355
Critical Notes: Foucault on Timiflg and Spacing 145 G lossary 373
Bibliography 379
4 Structure, System, Social Reproduction .62
Index 392
Societies, Social Systems 103
Structure and Constraint : Durkhe im and Others .69
Three Senses of 'Constraint' .74
Constraint a nd Reificalion .79
The Concept of Structu ral Principles '80
Structures, Structural Properties .85
Contradiction .93
Mak ing History .99
Critical Notes: 'Strnctural Sociology' and
Meth odological Individualism 207
Blau: a Version of Structural Sociology 207
An Alternative? Methodologica l Individualism 2 11

5 Change, Evolution and Power 227

Evolution ism and Social Theory 228
Adaptation 233
Evolution and History 236
Analysing Social C hange 244
C hange and Power 256
Critical Notes: Parsons on Evolution 263

6 Siruclurlition Theory, Empirical Research .00 Socr.1 Critique 28.

A Reitera tio n o f Basic Concepts 28 .
The Analysis of Strategic Conduct 288
Uninte nded Consequences: Against Functionalism 293
T he Dualit y of Structure 297
T he Problem of Structural Constra int 304
Clml radic lion and the Empi rical Siudy of Conflict 3.0
Inslilulhmal Stabilit y and Change 3.9

Draw i ng Togelher the Threads: $ lru Cluralion Theory
nml Forms of Resea rch

For some while. and in a number of previous publications. I have

been seeking ( 0 establish an approach to social science which
departs in a substantial fashion fro m existing traditions o f social
th o ught. This volume provides a summation of those previous
writings. setting them o ut in what [ ho pe is a developed and
coherent manner. The vague term 'app roach' to social science
actually conveys very well what I take to be the method ological
implicatio ns of structuratio n th eory. In social science, fo r reasons
expanded upon in considerable detail in what follows, conceptual
schemes that ord er and inform processes o f inquiry into social
life are in large part what 'theory' is and what it is for. I do not
mean by this, o f course, that it is not the aim of sociallheory to
illuminat e , interpret and explain substantive features of human
co nduct. I mean that th e task o f establishing and validating
generalizations - I shall not say 'laws' - is o nly o ne amo ng
various other prio rities o r aims o f social theory. The task of
co nstru cting set s o f stably established generalizations, which is
(perhaps) the lynch pin of the endeavours o f the natural sc ie nces,
is not an ambitio n of much relevance to social sci ence. Or so I
Many people have been good enough to loo k through and
co mm ent upon earlier drafts o f the book or have otherwise
co ntributed very directly to its final form . r would like to thank
th e fo llo wing persons in particular : Mrs D. M. Barry, John
Forrester, Diego Gambelta , Helen G ibson , Derek G regory , David
Held, Sam Hollick, Geoffrey Ingham, Ro bert K. Merton , Mark
Poster , W. G. Run ciman , Quentin Skinner, John B. Tho mpson
;tIld Jo nathan Zeit lin .
January 1984

CCHM A COl/temporary Critique of Historical Materialism, vol. I

(Londo n: Macmillan/Berkeley: Unive rsity of California
Press. 1(81)
CPST Central Problems in Social Th eo ry (London: Macm illanl
Berkeley: University o f Califo rnia Press, 1979)
CSAS The Class Sfru cture uf the Advanced Societies, rev. edn
(London: Hutchinson/ New York : Harper & Row. 1981 )

NR5 M New R ules of Sociological Method (London:

Hutc hinson/ New York: Basic Books, 1976)
peST ProfIles and Critiques Ii I Social Th eory (Londo n :
. Macmillan/ Berkele y: University of California
Press, 1982)
SSPT Studies in Social and Political Theory (London:
HUlchinson/ New York : Basic Books. 1977)

All by A nthony G idde ns.


The backdrop to this book is to be found in a series o f significant

developments which have taken place in the social sciences over
the past decad e and a half. These have been concentrated in
substantial part in social theory, and bear especially upon that
most maligned and most provocative of the social sciences,
socio logy. Socio logy is by its very nature controversial. However,
for a considerable perio d after the Seco nd World War ,
particularly in the English-speaKing world , there was a broad
consensus about its nature and tasks and about those of the social
sciences as a whole. There was, it could be said , a middle ground
shared by otherwise competing perspectives, a terrain o n which
intellectual battles could be fought o ut. During that period
sociology was a n academic growt h area , a subject with a
burgeo ning reputation, even if it remained distinctly unpo pular in
man y circles. It was dominated internationall y by American
sociology, and in social theory th e influence o f Talcott Parsons
was marked. I The prestige enjoyed by Parsons's ideas can be
exaggerated retrospectively - many found his taste fo r abstrac-
tion and obscurity unattractive, and he had his fair share of critics
and detractors. Ho wever. The Structure of Social A ction, first
published in the late 19305 but widely known o nly in the post-war
period, was in more than o ne way a key work in the form ation of
modern sociology. In it Parsons established a systematic pedigree
for social th eory , based upon an interpretat ion of European
thought in the nineteenth and early twentie th centuries. The
work of Durkhe im , Max Weber and Pareto loomed large, but
Marx played a ve ry slight role in deed . The writings of th e
1890 - 1920 generation had supposed ly gone beyond Marx in all
important respects. sifting o ut wha t was val uable and discarding
th e d ross.
xiv Introduc tion Introduc tio n xv

The book also set up an approach to social theory of a very important innovations within Marxism during this period - such
definit e type , combining a sophisticated version o f functionalism as the revival of interest in the 'young Marx', attempts to merge
and a naturalistic conception of sociology. Parsons's subsequent Marxism and ph enomenology, and su5sequently Marxism- and
writings elaborated these views in considerable detail, emphasizing structuralism - these were no t well known to those who called
that although human action has very special and distin ctive themse lves 'socio logists' , even in Europe. Th ose who regarded
attributes , social science by and large shares th e same logical themselves as both sociologists and Marxists tended to share the
framework as narural science. Himself writing and working in an basic assumptio ns of fun ctionalism and natura lism , which is o ne
American context , Parsons's attempt to pinpo int th e origins o f his reason why mu ch co mmo n ground for debate was fou nd .
thought in European social th eory actually served to reinforce Th e fissures in this common ground opened up remarkably
the dominant position of Am erican sociology. For Durkheim , suddenly in the late 1960s and early 1970s. and they went very
Weber and Pareto were regarded as forerunners of the deep. There is no doubt th at their origins were as much po litical
development of the 'action frame of refere nce', to be given its full as intellectual. But whatever their provenance, they had th e
ex pression by Parso ns and his colleagu es, Sociology may have its effect of largely d issolving whatever conse nsus had existed before
main theoretical origins in Europe , but the further ela boration o f about how social theory should be approached . In its place there
the subject was a task that had been largely transferred across th e appeared a baffling variety o f competing theoretical perspectives,
Atlantic. Curiously, this resul t was achieved at the expense of a none able fully to recapture the pre-eminence formerly enjoyed
concomitant recogni tion of the importance of indigeno us by the 'orthodox consensus', It became apparent to those working
American contributions to social theory ; G. H. Mead received in sociology that all along there had in fact been less of a
sho rt shrift in Th e Strucwre of Social A ction, as Parsons came conse nsus about the nature o f social theory than many had
later to acknowledge. To this day, however, there are textbooks imagined. Some traditions of tho ught. such as symbolic
on social theory, or 'sociological theory', emanating from the interactionism, had all th e while been accorded considerable
United States, which begin with the classic Europ ean thinkers but support withou t storming the c itadel of the orthodox consensus,
the n convey the impression that social theory in Eu rope Other schools o f thought that had developed in large pan
subseq uently came to a stop - any further progress is taken to be separately from the main body o f the social scie nces were taken
a purely American affair. seriously for th e first time, including phenomenology and the
But even within the confin es of the debates deriving directly criti cal theory of the Frankfurt phil osophers, Some traditions
from Parsons's writings, some o f the leading comribu[ors were which had seemed mo ribund were given a new impetus, Although
European. Marxism has long been a much more important Weber had been influenced by the hermeneutic traditio n and had
influence in European than in American intellectual culture, and incorporated its main concept of verslehen into his work, most of
some of Parsons's most perceptive critics drew inspiration from those co nnected with sociology would certai nly not have regarded
Marx as well as from readings o f Weber rath er different from 'hermeneutics' as part of the ir lex icon. But, partly in conjunc tion
those which Parsons had made. Dahrendorf, Lockwood, Rex and with phenomeno logy, interpretative traditions in social thought
others of a similar standpoint took the theoreti cal content of again came to th e fore. Finally, other styles of thought. such as
Parsons's work much mo re seriollsly than did his American radical ordinary language philosophy, were adopted int o social th eory in
critics (c. Wright Mills and , later, Gouldner). The former group vafl Ous ways.
regarded Parsons 's contributions as of major importance but as With these developments the centre of gravity in respect o f
one-sided in neglectin g phenomena they saw as primary in Marx innovative contributions to social theory moved back towards
- class division , conflict and power. They were no t themselves Europe. l It became obvious that a great deal of the more
Marxists, hut th ey envisaged something o f a fu sion bet ween
Parso nial1 and Marxist <..: on ce pts. Whil e th ere were many 'R eferences may he f(lund on pp. uxvi- uxvii.
x....i Introduction Introduction x....ii

interesting theoretical work was going o n there - and for the issues are to do with th e nature of human action and the acting
most part in languages other than English. European social theory se lf ; with how interac ti o n shou ld be conceptualized and its
was, and is, not o nly alive but kicking very vigoro usly. But what is relatio n to institutions; and with grasping the practical connota-
th e o utcome of these stirrings? For th e loss of the centre ground tions of social analysis. I understand 'sociology ', by contrast, to
formerly occupi ed by the orthodox consensus has seemingly left be not a generi c discipline to do with the study of human societies
social theory in a hopeless disarray. Notwithstanding the babble as a whole , but th at branch of social science which foc uses
of rival theoretical voices, it is possible to discern certain common part icularly upo n (he 'advanced ' or modem societies. Such a
themes in this apparent confusion. One is that most of the schools disciplinary characterization implies an inte lleclU al divisio n of
of thought in question - with notable exceptio ns, such as labou r. nothing more. While there are theorems and concepts
structuralism and 'post-structuralism' - emphasize the active, which belong distinctively to th e industrialized wo rld , there is no
reflexive character of human condu ct. That is to say, they are way in which something called 'sociol ogical theory' ca n be clearly
unified in their rejection of the tendency of the orthodox distinguished from the more general concepts and concerns of
consensus to see human behaviour as the resuitof force~ -that social theory. 'Socio logical theory' , in o ther words. can if o ne
actors neither cont rol no r comprehend . In additio n (andt his does likes be regarded as a branch of social theory more generally, but
include both structuralism and 'post-structuralism '), they accord it ca nnot sustain a wholly separate identity. T his book is written
a fundam~ntal role to language , and to c9gnitive fa culties in the with a definite socio logical bias, in the sense that I tend to
explication of social life. Language use is embedded in the concentrate upo n material parti cu larly relevant to modern
concrete activities of day-to-day life and is in some sense partly societies. But as an introductio n to structuration theory it is also
constit utive of those activities. Finally, the declining importan ce intended in substantial degree as a fo rmulation of th e tasks of
of empiricist philosophies of natural science is recognized to have social theory in general a nd is 'theory' in the same sense. That is
profound implications fo r the social sciences also. It is no t just to say, the focus is upo n the understanding of human agency and
the case that social and natural science are furth er apart than of social institutions . .
advocates of the orthodox consensus believed. We now see that a 'Social theory' is not a term which has any precision. but it is a
philosophy of natu ral science must take account of just those very useful one fo r all that. As 1 represent it , 'social theory'
phen omena in which the new schools of social theory are in volves the analysis of issues which spill over into philosophy , y
interested - in particular, language and the interpretatio n of but it is no t primari ly a philosophical endeavour. T he social
meaning. sciences are lost if they are not directly related to philosophical
It is with these three core sets of issues, and their mutual pro blems by those who practise them. To demand that social
con nections, that the theory of structuration, as I represent it in scientists be alive to philosophical issues is not the same as
this book, is concerned. 'Structuratio n' is an unlovely term at driving social science into the arms of those who might claim that C-:"
best, although it is Jess inelegant in the Gallic context from whi ch it is inherently specu lative rath er than empirical. Social theory
it came. I have not been able to think of a more engaging word has the task of providing conceptions of the natu re of human
for the views I want to convey. In elaborating the concepts of social activ!!y and of the hUl"!lan agent whic fi car"-6e placed inthe
structuration theory, I do not intend to put forward a potentially service of empirical work . T he main concern of social theory is
new orthodoxy to replace the old o ne. But structuratio n theory is th e same as that of"the social sciences in general : th e illumination
sensitive to the shortcomings of the orthodox consensus and to of concrete processes of social life. To hold that philosophical
th e significance of th e convergent develo pments noted above. debates can contri bute to this concern is not to suppose that such
In case there is any doubt about terminology here, let me debates need to be resolved conclusively before worthwhile social
emphas ize that I lise the te rm 'socia l theory' to e ncompass issues research can be in iti ated. On (he contrary, the prosecutio n of
Ihal I hold to be th e concern of all the social sc iences. These social resea rc h ClIO in prin cipl e cast light on phi losophical
xvii i Introduction Imroduc rio n xix

controversies just as much as the reverse, In particular, 1 think it suc h a view has little to commend it , since it does not help to
wrong to slant social theory too unequivocally towards abstract cl arify the explanatory import of muc h of what social scientists
and highly generalized questions of epistemology, as if any (or natural scientists either, for that matter) do. Most 'why'?'
significant developm ents in social science had to await a clear-cut questio ns do not need a generalization to answer them, nor do
solutio n to these. '""" the answers logically imply that there must be some generaliza-
A few remarks are necessary about the 'theory' in social theory.
~ .
tions lurking around whi ch could be invoked to back up the
There are certain senses o ften attributed to 'theory' in the social answers. Suc h observations have become fairly commo nplace in
sciences from which J want to maintain some considerable the philosophical literature. and I shall not try to ex tend them
distance. One conception used to be popular among some of further. Much more contentious is a second claim I defe nd , and
those associated with the orthodox consensus, although it is no elaborate in the book, that the uncove ring of generalizatio ns is
lo nger widely held today. This is the view - influenced by not th e be-all and end-all of social theory. If the proponents of
certain versions of the logical empi ricist philosophy of natural 'theory as explanatory generalization' have too narrowly confined
science - that the o nly form o f 'theory' worthy of the name is the nature o f 'explanat ion', they have compounded the error by
that expressible as a set of deductively related laws or failing to inquire closely e nough into what generalization is , and
generalizations, This sort of not ion has turned o ut to be of quite shou ld be, in social science.
limited application eve n within the natural sciences. If it can be Generalizations tend towards two poles, with a range and
sustained at all, it is only in respect of certain areas of natural variety o-f possible shadings between them. Some hold because
science. Anyone who would seek to apply it to social science actors themselves know them - in some guise - and apply them
must recognize that (as yet) there is no theory at all; its in the e nactment of what they do. T he social scientific observer
constru ction is an aspiration deferred to a remme future, a goat does no t in fact have to 'discover' these generalizatio ns, although
to be striven for rather than an actual part o f the current-pu"l-suits [hat observer may give a new discursive fo rm to them. Other
of the social sciences. generalizations refer to circumstances. or aspects of circum-
Although this view does have some adherents even now , it is stances. of which agents are Ignorant an -Wh"ich effectively 'act'
far removed fro m anyt hing to which I would hold that social on them, independent of whatever the agents may believe they
theory could or sho uld aspire - for reasons which wi ll emerge are up to. Those I shall call 'structural sociologists' tend to be
clearly enough in the body of the book which follo ws. But there is interested o nly in generalization in this second sense - indeed ,
a weaker versio n of it which srill commands a very large following this is what is meant when it is claimed that the 'theory' in social
and which invites rather longer discussion even in this prefatory theory sho uld co mprise ex planato ry generalizations. But the first
con text. This is the idea that the 'theory' in social theory must is just as fundamental to social science as the second, and each
consist essentially of ge neralizations if it is to have ex planatory form o f generalization is unstable in respect of the other. The
content. According to such a standpoint, much of what passes for circumstances in which generalizations about what 'happens' to
'social theory' consists of conceptual schemes rather than (as agents hold are mutable in respect of what those agent s can learn
sho uld be the case) 'explanatory propositions' of a generalizing knowledgeably to 'make happen'. From this derives the (logically
type. open) transformative impact whi ch the social sciences can have
\ Two problems have to be separated here. One concerns the upon their 'subject matter'. But from it also comes the fact that
natu re o f explanation in the social scie nces, I shall take it for th e discovery of 'laws' - i.e ., generalizations of typ e 2 - is only
grant ed that explanation i ~t ex tual , th e clearing up o f queries. o ne co nce rn among ot hers that are eq ually important to the
Now it III/Rill be held that the only qu eries worth th eir salt in social th eoretical content o f social scien ce. Chief among these o ther
sc ience ilre those of a very generalized kind. which can therefore concerns is the provision o f conceptual means for analysing what
be answered o nly by reference to abstract genernlizalions. But ac tors know about why th ey act as they do. panicularly either
xx Introduction Introduction xxi

where they are not aware (discursively) that they know it , or has to be reconceptu alized as a duality - the duality of structure.
where actors in o ther contex ts lack such awareness. These tasks Altho ugh recognizi-ng the significance of the 'lingu istic turn ', it. is
are primarily hermeneutic in c haracter, but they are an inherent not a version of hermene utics o r interpretative socio logy. Whde
and necessary part o f social theory. The 'theory' involved in acknowledging that society is not the creation of individual
'social theory' does not consist only, or even prim arily, of the subjects, it is distant from any conception of structural sociology.
formulation of generalizations (of type 2). Neither are the concepts T he attempt to formulate a coherent accou nt of human agency
developed under the rubric 'social theory' made up only of those and of structure demands, however, a very considerable
which can be fed into such generalizations. Quite to the contrary. conceptual effort. An expositio n of these vi ews is offered in the
these concepts must be related to others referring [0 th e ope ning chapter and is further developed throughout the book. It
knowledgeability of agents, to whic h they are inevitably tied. leads on directly to o ther main themes , especially that of the
Most of the controversies stimu lated by the so-called 'lingu istic study o f time-space relations. T he structural properties o f social
turn' in social theory, and by the emergence o f post-empiricist systems exist o nly in so fa r as forms of socia l conduc t ?re
philosophies o f science, have been strongly epistemological in reproduced chronically across time and space. The structurauon
character. T hey have been concerned , in other words, with o f institutions can be understood in terms of Ilow it comes about
questions of relativism, problems of verification and falsification that social activities become 'stretched' across wide spans of time-
and so on. Significant as these may be, concentration upon sp'ae. Incorporating - time-space in the he~rt .Of. social. t~~ory
epistemological issues draws attention away from th e more means thinking again abou t some of the disciplinary diVISions
'<?_n~ological' concerns of social theory, and it is these upo n which which separate sociq!Qgy' from ~ist0f2:'. and fr~&.~~~y. T~e
struc tu ration theory primarily conce ntrates. Rather than be- concept and analysis o f history is particularly problematiC. ThIS
comi ng preoccupied with epistemological disputes and with the book, indeed , might be accurately described as an extendej
question of whether or no t anything like 'epistemology' in its refl ection upon a celebrated and oft-quo ted phrase to be fo uoom
time-ho nou red sense can be formulated at all , those working in Marx: Marxcomments th at 'Men (let us immediately say human
social theory, I suggest, should be concern ed first and foremost beingsf make histo ry, but not in circumstances of their own
with reworking conceptions of human being and human doing , choosing.'. Well . so they do. But what a diversity of compl ex
social reproduction and social transformation. Of prime impor- problems of social analysis this apparently innocuous pronounce-
tance in this respect is a du alism that is deeply entrenched in ment turns out to disclose!
social th eory, a division between objectivism and subj ectivism.
Objectivism was a third -ism characterizing the o rthodox The phrase is to be found in Ihe introduct?ry paragr~phs o~ The Eighteenth
consensus, together with n <!~ural ism apd functionalism. In spite Brumaire of Louis Bonaparte. It was made In a polemical vem: those who are
of Parsons's terminology of 'the action frame of reference' , there ignorant of history. Marx says, may be condemned to repeat it. perhaps even
fa rcically. The exact q uo ta tion in the o riginal goes as follows: 'Die Menschen
is no doubt that in his theoret ical scheme the object (society) machen ihre eigene G eschichte. aber sie machen sie nic ht aus fre ien Stucken,
predominates over lhe subj ect (the knowledgeable human agent). nich t unler selbstgew.!ihlten. sondem unler unmiuelba r vorgefunde nen,
O thers whose views could be associated with that consensus were gegebenen und uberlieferte n Umstanden. Die Tntditio n aller tote,n
very much less sophisticated in this respect than was Parsons. By Geschlechler lastet wie ein Alp auf dem Gehime der Lebenden. Und wenn sle
attack ing objectivism - and structural sociology - those eben damit beschliftigl scheinen. sich und die Dinge urnzuwalzen, noch nicht
Dagewesenes zu schaffen. gcrade in solchen Epochen ~evolu.tionarer .Krise
influ enced by h ~.rmen e uli cs or by phenomenology were able to
beschw6ren sie angsllich die Geister der Vergangenhell zu Ihrem Dlenste
lay bare major shortcomings o f th ose views. But they in turn herauf, entlehnen ihnen Namen. Schlachtparole. KoslUm. urn in dieser
veered sharp ly towards subjectivism. Th e conceptual d ivide ultehrwUrdigcn Verk[eidung und mit dicscr erburgten Sprac he die neue
bel ween subject and soc ial object yawned as widely as ever. Weltgeschichts.-;:tenc auhufUhrc n: ~Marx and Engels: Werke, Vol 8. Berlin:
Slructuratio n theory is based on the prem ise thal th is dualism Diet1- Verlag 1960. p. 1151.
xxi i Introduction Introduction )(xiii

In formulating this account of structuration theory I have not involved in a continuo us mann er with the fl ow of day-to-day
been reluctant to draw upon ideas from quite divergent sources. conduct in the contexts o f social activity. But reflexivity operates
To some this may appear an unacceptable eclecticism, but J have only partly on a discursive leveL What agents know about what
never been able to see the force of this type of objection. There is they do, and why they do it - their knowledgeability as agents -
an undeniable comfort in working within established trad itions of is largely carried in practical consciousness. Practical c~m
thought - the more so, perhaps , given the very diversity of sciousness consists of all the things which actors know taCitly
a.pproac h ~ .that currently confro nts anyone who is o utside any about how to 'go on' in the contexts of social life without being
smgle tradition . The comfort of established views can however able to give them direct discursive expression. The significance of
~asil~ be.a cover for intellectual slot h. If ideas are im~rtant and practical consciousness is a leading theme of the boo k. and it has
Illummatmg, what matters much more than their origin is to be to be distinguished from both consciousness (discursive con-
able to sharpen them so as to demonstrate their usefulness. even sciousness) and the unconscious. Whil e accepting the importance
if within a fram ework which might be quite different from that of unconscious aspects of cognition and motivation, I do not
which helped lO engender them . Thus. fo r example. I acknow. think we can be content with some of the more conventionally
led~e the call for .a decentring of the subject and regard this as established views of these. I adopt a modified versio n of ego
basiC to structurauon theory. But I do not accept that th i~ implies psychology but endeavour to relate this directly to what , I suggest,
the evaporation of subjectivity into an empty universe of signs. is a fundamental concept of structuration theory - that of
Rather, social practices, biting into space and time , are considered routinization.
to be at the root of the constitution of both subject and social The routine (whatever is done habitually) is a basic element of
?bject. I admit the Central significance of th e 'linguistic turn', daY-lo-day social activity. 1 use the phrase 'day-to-day social
mtr?duced especiall.y by hermeneu tic phenomeno logy and act ivity' in a very literal sense , not in the more complex. and I
or?mary language philosophy. At th e same time. however, I ho ld think more ambiguous , way which has become familiar through
this term to be in some part a misleading one. The most important phenomenology. The term 'day-to-day' encapsulates exactly the
developments as regards social theory concern not so much a routinized character which social life has as it stretches across
turn towards language as an altered view of th e intersection time-space. The repetitiveness of activities which are undertaken
between saying (or signifying) and doing, offering a novel in like manner day after day is the material grounding of what I call
'I coQ,ception of praxis. The radical transmutation of hermeneutics the recursive nature of social life. (By its recursive nature I mean
~nd phenomenology initiated by Heidegger, together with the that the structured properties of social activity - via the duality
mnovations of the later Wittgenstein , are the two main signal of structure - are constan tly recreated out of the very resources
markers on the new path. But to pursue this path further means which constitute them.) Routinization is vital to the psychological
precisely to shake off any temptatio n to become a full. blown mechanisms whereby a sense of trust or ontological security is
disciple of either of these thinkers. sustai ned in the daily activities of social life. Carried primarily in
Let me offer here a short summary of the organization of the practical consciousness, routin e drives a wedge between the
book. Having given in the fi rst chapler an oulline of the chief potentially explosive content o f the unconscio us and the refl exive
concepts invo lved in structuration theory, in the second I begin monitoring of action whi cb agems display. Why did Garfinkel's
the more substantive part of the volume with a discussio n of 'experiments with trust' stimulate such a very strong reaction of
consci.ousness, the unconscious and the constitution of day.to- anxiety o n the part of those invo lved, .seemingly out of a.lI
day life. Hum an age nts or actors - I use these terms proportion to th e trivial nature of the circumstances of their
inl crchan~eabJy - have , as an inherent aspect of what they do, origin'! Because, I think , the apparently minor conventions of
the capacll y 10 undersland what they do while they do it. The daily social life are o f essen tial significance in curbing the sources
reflex ive capacili es of the human acto r Me c haracter istically of un conscio us tensio n tha t would ot herwise preoccupy most of
xxiv Introduction Introduc tion ~
o ur waking lives. the duration of 'institution al time', the 'supra-individual' structura-
Th e situated character of actio n in tim e-space th e routinization tion of social institutions. Fin ally , each person is positio ned , in a
of ac tivity and the repe titive nature o f day-to-da~ life - these are 'mUltiple' way , within social relatio ns conferred by specifi c social
phenomena which connect discussion of th e unconscious with iden tities; this is the main sphere of applicatio n o f the concept of
G~f~man 's analyses, o f :~-presence. In spite of th eir manifest social role. The modalities o f co-prese nce , mediated directly by
bnlhance, ~offma~ s w!Jtmgs are usually thought of as perhaps the sensory properties of the body, are clearly different from
s~mewhat lightweight In respect of their theoretical content , social ties and forms o f social interactio n establ ished with o th ers
either because he is re~arded above aU as a sort o f sociological absent in time or in space.
racon te~r - th e eq Uival e nt of a socio logi cal gossip whose It is not only in d ividuals who are ' position ed ' relative to o ne
observatlo~s ent~rtain and titillate but are no ne th e Jess superficiaJ another: the contexts of social interactio n are also. In examining
and ~en~JaU.y picayune -. or because what he portrays is specific these connectio ns, to do with the (..'O ntext ua lity o f social
to soctal hfe m modern, mIddle-class society, a cynical society of interaction, the techniqu es and approach of time-geography, as
amoral role players. There is something in each o f these views deve loped by Hagerstrand, are highly illuminating. Ti me-
and to a certain degree GoHman is vuln erable to them becaus~ geography also has as its principal concern th e location of
~e r~fr~ins from drawing ou t, in a fully systematic way , the indi vidu als in time-space but gives partic ulaL a lt entio n (0
l~phcatl?nS o f his standpo int. Where he does do so he tends to constraints over activity derivi ng from p hysical properties o f the
link the rituals of day-to-day social life to ethological accounts of body and of envi ro nments_illwllic~agents move. ReterenCe to
the behaviour of the high er an imals and to explicate them in these is bu t one o f the respects in which sociology can profit from
those terms. This may indeed be instructive, but it is not the most the writings of geograph ers. Ano ther is the interpretatio n o f
usefu l w.ay of relating his work to problems of social theory urbanism, which , I argue, has a basic part to play in social th eory :
because It.does not plug the right gaps in what,' he has to say. One and , of course, a genEral senSitivity to space and place is of even
such gap IS the absence o f an account of m0tivatio n , the main greater importance.
reasc.n why his writings are o pe n to the second interpretatio n --COffman gives some cons iderable att entio n to the regio naliza-
mentio ned .aoove. I try to show how an analysis of mot ivation, as tion of encounters, and I take the notion of regionalizatio n to be
de.veloped In relation to routinization an d the un conscious, can a very significant o ne for social theory . It has always been a main
brmg out, the syste~ati c character o f Goffman 's work more fully. concern of the writings of geograp hers, but I want to regard it as
Goffm~n s emphaSIS o n trust and taci strikingly echoes the mes less o f a purely spatial concept (han they o rdinarily do . The
fou nd m e~o psychology and generates an analytically powerful situated nature of soci al interaction can usefully be examined in
~nderstandmg of (he reflexive monitoring o f the flu x o f encounters relation to the different locales through which the daily activiti es
Invo lved in daily life. o f individuals are co-ordinated. Locales are not just places but
~undamental 10 ,soc~a.1 li ~e !s th e pOSitio ning o f the body in settings o f interactio n ; as Garfinkel has demons trated partic ularly
soc~<t! enco ~nters ..Pos [lI~nmg ~ere IS a rich term . T he body is persuasively, setti ngs ~re used c hronically - and largely in a tacil
pos lt.lOned to the Immedl3te Circumstan ces of co-presen ce in way - by social actors to sustain meaning in communicative acts .
relatl o~ to o th ers: Goffman provides an extraordinarily subtle But settings are also regio nalized in ways that heavily infl uence.
but [e.llmg set o f observations about face work , a bout gesture and and are influenced by , the serial charac ter o f e ncount ers. Time-
rcnex l~e c~>ntrol o.r .t>oc.tily ~ovement as inherent in the continuity space 'fixity' also normally means social fix ity: the substantially
o f socra l hfe. Posltlo nmg IS, however, also to be understood in 'given' c ha racter o f the physical milieux of day-to-day life
~cl ;~ti.on 10. th e serialit y ?~ e ncounters across time-space. Every interlaces with ro utin e and is d eeply influential in th e contours of
IIld l ~ld uaJ IS al ?nc~ positioned in the fl ow of day-to -day life; in ins titutio nal re produ c lio n . Regio nalization a lso has stro ng
the life-s pan wlllc h IS th e duration o f his o r her existe nce; and in psych o logical and soc ial resonance in respect of the enclosu re'
xxvi Introduction Introduction xxvii

from view of some types of activities and some types of peopl e entities , and social systems as internally highly integrated unities.
and the 'disclosure' of others. Here we again find a major point of For such perspectives, even where direct organic metaphors are
conn ection between seemingly disparate ideas, those of Goffman rejected , tend to be closely allied to biological concepts; and
and Foucault; both accord great importance to the socially and these have usually been arrived at with reference to entities
historically flu ctu ating lines between enclosure and disclosure , clearly set o ff from the world arou nd them , baving an evident
confinement and display. internal unity. But 'societ ies' are very o ft en not lik e this at aH. To
I think it is a mista ke to regard encounters in circumstances of help take account of tha t, I introd uce the terms 'intersocietal
co-presence as in some way the basis upon which larger, o r sy&tems' and 'time-space edges' , referring to diffe.rent aspe,c ts 0L
'macrostructural', social properties are built. So-called 'micro- regio nalizatio Q which _cut _across social systems recogl1lza61y
sociological ' study does not deal with a reality that is somehow distinct as societies. I also use th ese no tio ns ex tensively in
more substantial than that with which 'macrosociological' analysis assessing interpretations of social change later in the book.
is co ncerned. But neicher. o n the contrary , is interaction in In formulating structuration th eory I wish to escape from the
situations o f co-presence simply ephemeral, as contrasted to the dualism associated with o bjectivism and subjectivism. But some
solidity o f large-scale or lo ng-established institutions. Each view criti cs have felt that not enough weight is given to factors
has its proponents, but I see this division of opinion as an empty emphasized by the first of these , particularly in respect of the
one and as a slightly more co ncrete version of the dualism in constraining aspects of the structural pro perties o f social systems.
social theory already mentio ned. The opposition between 'micro' To show that such is not the case 1 indi cate in some detail what
and 'macro' is best reconceptualized as concerning how 'constraint' can be taken to mean in social theory and how the
interaction in contexts of co-presence is structurally implicated in various senses that can be given to the term are understood in the
systems of broad time-space distanciation - in other words, how theory of structuration. Recogn itio n o f th e nature and significance
such systems span large sectors o f time-space. And this in turn is of structural constraint does not mean succumbing to the
best investigated as a problem of the connection of social with attractions of structural sociology, but neither, as ] try to make
system integration , as I define these terms. But a vital rider has to clear, do 1 accept a viewpoint close to methodological
be added to this. T he relatio n of social to system integratio n individualism. As conceptualized in st ru ct ura tio n th eory, 'stru c-
canno t be grasped o n a purely abstract level ; the theory of ture' means some thing different from its usual usage in the social
urbanism is essential to i(. For it is o nly with the advent of cities scie nces. I also introduce a cluster of o ther concepts centring
- and , in modern times , with the urbanism of the 'created upon thal of struc ture and e ndeavour to show why they are
enviro nment' - tha t a significant d evelopment o f system necessary. Most important among these is the idea o f 'structural
integration becomes possible. principles', which are ...s~ tur a l features f overall soc'eties or
We have to be very care ful indeed with the concept o f 'social societa1 totalities: 1 also see to show that it IS ttirough the notion
~[em' and the associated no tio n of 's<?ciety'. Th ey soynd of structural principles that the concept of contradiction can
mnocent terms, and they are probably indispensa61e if used with most usefully be specified as rel evant to social analysis. These
appropriate measures o f caution. 'Society' has a useful double no tions again cannot be expressed in purely abstract fo rm , and I
meaning, which 1 have relied upo n - signifyjng a bounded examine them with reference to three major types of society that
system, and social association in general. An emphasis upon can be distinguished in human history: tribal cultures, clJlli:r
regio nalizario n helps to remind us that the degree of 'systemness' divided societies and modern natio!!:slates <lssociated with the
in social systems is very variable and that 'societies' rarely have rise of Iilduslrial capitalism~''---
easily specifiabl e boundaries - until. at least, we enter the modern Mention of history reca ll s th e dictum that human beings make
world o f natio n-stat es. Fun ctio nalism and naturalism tend to hi story. What exactly is it that they mak e - what does 'history'
encourage lInl hinking accep tance of societi es as clearly delimit ed mea n here'! T he answer ca nno t be expressed in as cogent a form
xxvii i Introduction Introduction xxix

as the original maxim. There is, of course, a difference between directionality thro ugh the stages indicated , in respect of a given
history as events which elapse and history as writing about those criterion or criteria, such as increasing complexity or expansion
events. But this does not get us all that far. History in the first of the forces of production. A range of objectio ns can be brought
sense is temporality, events in their duration. We tend to associate against these ideas , both in respec t of their in trinsic demerits and
tempo rality with a linear seq uence, and thus history thought of in in terms of secondary implications which evolutio nism almost
this way with movement in a discernible direction . But this may inevitably tends to bring in its train , even if they are no t logically
very well be a cu lture-bound fashio n of thinking about time ; even entailed by it. 'Histo rical materialism', I think , is a version of
if it is not , we still have to avoid the equation o f 'history' with evolutio nism according to these crite ria , in a t least o ne of the
social change. For this reason it is worth speaking of 'historicity' main ways in which that contentio us term has been understood.
as a definite sense of living in a social world constantly exposed to If interpreted in this manner. historical materialism manifests
change , in which Marx's maxim is part of a general cultural several of the main and the secondary limitatio ns of evolutionary
awareness, no t a th eorem pec uliar to specialist social thinkers. theories more generally and has to be rejected for the same
History as the writing of history also poses its own dilemmas and reasons.
puzzles. All I shall have to say about these is that they are not Since I do not think it possible to compress 'history' into the
distin ctive ; they do not permit us to make clear-cut distinctions kinds of -scheme favou red by evolutionism in general , or by
between history and soc ial science. Hermeneutic probl ems historical materialism mo re specifically, I speak of geconstructing J
involved in the accurate deSCription of divergent forms of life , the rather than reconstructing them. By this I mean that accounts of
interpretatio n of texts, the ex plication of action, institutio ns and social change have to tak e a substanti ally different form from
social transformatio n - t.hese are shared by all the social sciences, evolutionism; there is no virtue in trying merely to remodel them
including history. somewhat. In addition to co ncepts already introduced , 1 make
How, then, should we approach the study of social change? J use of two others: those of 'episode' and 'world time' (the first due
try to show that the search for a theory of social change (where to Gellner, the second to Eberhard). All social life can be
'theory' means in this instance explaining social change by represented as a series of episodes; encou nters in circumstances
reference to a si ngle set of mechanisms , such as the o ld of co-presence certainly have an episodic form. But in this
evolutio nary favourites of adaptation and differential selectio n) is connection 1 am referring mainly to large-scale processes of
a doomed o ne. It is fl awed by the same kind o f logical change, in which there is some definite type of institutional
sho rtcomings that auach more generally to the suppositio n that reorganization , such as the fOmlation of cities in agrarian societies
the social sciences can uncover universal laws of human conduct. or the formati o n of early states. Episodes may certainly be
The sorts of understanding o r knowledge that human beings have fruitfully compared with o ne ano ther but no t in complete
of their own 'history' is partly constitutive of what that history is abstraction from the context of the ir origin . The influ ence of
and of the influences that act to change it. However, it is important 'world time' is relevant precisely to how far they are in fact
to give particular critical atte ntion to evolutionism because in comparable. 'Wo rld time' concerns the Y.Il[y!ng co njunctures in
one versio n or another it has been so influential in a variety of histqry that may affect the. cond itions ang o u ~~emingly
different areas of social science. I mean by 'evo lutionism ', as simil<!L episodes and the mfluence of wh'!l the agen,.!.~olved
applied 10 the social sciences, the explication of social change in know about such conditio ns and o utcomes. I seek to mdlcate the
terms of schemas which invo lve the following features: an analytic~l purchase of these notio ns by using as an illustration
irreversibl e series of stages through which societies move , even if th eo ries of state formation.
it is not held that all ind ividu al societies must pass through each Structuration theory will not be of much valu e if it does not
of them to reach the higher ones; some co nceptual linkage with help to illuminate problems of emp iri cal research, and in the
bio logica l th eories of evo luti on; and th e spec ifi ca ti on of concluding chapt er I take up thi s issue, which I hold to be
xxx Introdu ction Introduction xxxi

inseparable from the implications of structuration theory as a directly on them. The un conscio us, on the other hand, poses
form of critiqu e. I do not try to wield a methodological scalpel. altogether a different order of problem , certainly demanding
That is to say, I do not believe that there is anything in either the techniques of interrogatio n distinct from those involved in
logic or the substance of structuration theory which would descriptive social research.
somehow pro hibit the use of some specific research technique. Functionalism has bee n highly important in th e social sciences ,
such as survey methods, qu estionnaires or whatever. Some no t only because o f its prominence as a type of theorizing but
considerations brought into play are relevant to the mode of also because of the empirical stimulus it has provided. The origins
application of particular techniques to research questions and to of fi eldwork in anthropology are more o r less coterminous with
the imerpretatio n of results, but that is a rather differem maller. the impact of functi onalism , and in sociology also functionalist
The po ints of connectio n of structuration theory with empirical thought has helped to generate a significant body of research
research are to do with working out the logical implications of work. I think it essen tial to understand the attractions of
studying a 'subj ect matter' of which the researcher is already a functionalism in this respect, while still ho lding that conceptually
part and with elucidating the substantive connotations of the core its influence has been la rgely pernicious. Fun ctionalism has
notions of actio n and structu re . Some of the points I have made strongly emphasized the significance of unintended consequences
on the abstract level of theory apply directly on the level of of action, especially in so far as such conseq uences occur in a
research. A good deal of social theory, especially that associated regular way and are therefore invo lved in the reproduction of
with structural socio logy, has treated agents as much less institutionalized aspects of social systems. Functionalists have
knowledgeable than they really are. The results of this can be been quite right to promote this emphasis. But it is entirely
very easily discerned in empirical work , in respect of a failure to possible to study unintended conseq uences without the use of
gain informatio n that allows access to the full range of agents' functionalist concepts. Moreover. the designation of just what is
knowledgea bility in at least two ways. What actors are able to say unintentional in regard of th e consequen ces of action can be
about the conditions of their action and that of others is adequately grasped empiri cally only if the intentional aspects of
for eshortened if researchers do not recognize the possible action are identified, and this again means operating with an
significance of a range of discursive phenomena to which, as interpretation of agency more sophisticated than is no rmally held
social actors themselves, they would certainly pay close auemio n by those inclined towards functio nalist premiSes.
but which in social research are often simply discounted. These In structuration theory 'structure' is regarded as rules and
are aspects of discourse which in form are refractory to being resources recursi vely implicated in social reproduction; institu-
rendered as statements o f pro positional belief or which , lik e tio nalized features of social systems have structural properties in
humour o r irony, derive their meaning not so much from the the sense that relationships are stabilized across time and space.
contem of what is said as fro m the style, mode of expression or 'Structure' can be conceptualized abstractly as two aspects of
context of utterance. But to this we must add a second factor o f rules - normative elements and codes of significatio n. Resources
greater importance: the need to acknowledge the significance of are also of two kinds: authoritative reso urces, which derive from
practical co nsciousness. Where what agents know about what the co-ordination of the activity of human agents, and allocative
they do is restricted to what they can say about it, in whatever resources, which stem from contro l of material products or of
discursive style, a very wide area of knowledgeability is simply aspects of the material world. What is especially useful for the
occluded from view. The study of practical consciousness must guidance of research is the study of, first. the routinized
be incorporated in to research work. It would be an error to intersections of practices which are the 'transformation points' in
suppose that no n-discursive components of consciousness are structural relations and. seco nd, the modes in which institu-
necessarily nlo rc difficult to study empirically than the discursive. tio nalized practi ces connect social with system integration. As
cvcn thoug h agent s [hemselves. by definition. cann ot co mm ent regards th e first of th ese. to ta ke an example. it can be
xxxii Introdu c tion Introduction xxxii i

demo nstrated how private property, a cluster of rights o f separate from the universe of mea ning and actio n which th ey are
ownership , can be 'translated ' iOlO industrial authority, o r modes about. But, for their part , lay acto rs are social theorists. whose
o f sustaining managerial contro l. As regards the second, what has theories help (0 constitute the activities and institutio ns that are
to be ascertain ed empirically is how far the situated practices the object of study of specialized social observers o r soc ial
studied in a given range of contexts converge with o ne another in scientists. There is no clear dividing line between informed
such a way th at they enter directly into system reprod uction. An socio logical reflecti o n ca rri ed o n by lay actors and similar
alertness to the significance of locales as settings of interaction is endeavo urs on the part o f specialists. I do not want to deny that
impo rt ant here ; there is no reaso n why sociologists sho uld not there are dividing lin es, but they are inevitably fuzzy, and social
adopt some of the research techniques established by geographers, scientists have no absolute monopoly either upon innovative
in cluding the graphic techniqu es o f time'geography, in o rder to theori es or upon empirical in vestigations o f what they study.
study them. All this may perhaps be gra nted . But it still might no t be
IC the social sciences are understood as they were during th e accepted from these comme nts that we sho uld ta ke a different
period of dominance of the o rthodox consensus, their attainments view of the accomplishm ents ,tnd impac t of the soc ial sciences to
do not loo k impressive, and the relevance of social research to that indicated above. How cou ld it seriously be suggested that
practical iss ues seems fairly slight. For the natural sciences, or at social science has had as much influence, or more, upon the social
least the more advanced o f them, have precisely specified and world as natural science has had o n the material world? I think, in
generally accepted laws, together with a fund of uncontroversial fact, that this view can be maintained - although, of course, no
empirical observations which can be explicated in terms of those such com parison could be precise, in view of the very differences
laws. Natural science has become co upled to technological between what is involved in each case. The point is that reflection
capabilities of an awesome kind, d estructive as well as on social processes (theories . and observations about them)
constructive. In the eyes o f those who would mode l social scie nce cont inually enter into , become disentangled with a nd re-enter the
directly on natural science, th e former surely comes o ff a distant uni verse o f events that they describe. No such pheno meno n exists
second best. Both cognitively a nd practically, the social sc iences in the world of inanimate nature , which is indiffere nt to whatever
seem distinctly inferior to the natural scien ces. But if we accept human beings might claim to know about it. Cons ider , for
that social science should no longer be some sort of rep lica of example, theories of sovereignty formulated by seve ntee nth-
natural science and is in some respects a quite divergent ce ntury European thinkers. These were the res ults o f reflection
enterprise, a very different view of their relative achievements upon, and study of. social trends into which they in turn were fed
and influ ence can be defended. There are no universal laws in the back. It is impossible to have a mo dern sovereign state that does
social sciences, and there will not be any - not , first and foremost, not in corporate a discursively articulated theory of the modern
because methods of empirical testing and validation are somehow sovereign state. The marked tendency towards an expansion of
inadequate but because, as I have pointed out, [he causal political 'self-monitoring' o n th e part of the state is characteristic
conditio ns involved in generalizations about human social conduct o f mode rnity in th e West in general , creating the social and
are inherently unstable in respect of the very knowledge (or intellectual climate fro m which specialized , 'professional' dis-
beliefs) that actors have about the ci rcumstances of their ow n courses of soc ial science have developed but also both express
action. The so-called 'self-fulfilling prophecy', of which Merton and foster. One could certain ly make some sort of case for
and ol hers have written, is a spec ial case of a much more gene ric clai ming that these changes, in which social science has been
ph eno menon in the social sciences. This is a mutual interpreta tive centrally involved, are of a very fundamental c haracter. By the
inl t: rplay between soc ial scie nce and (h ose whose activiti es side of them the transformations of nature achieved by the natural
compose its subject matt er - a 'double hermeneut ic'. Th Q :> sciences do no t look so massive.
theories and findings o f th e social scicn~cs'can~ot be kept ~ holiy Refl ecting upon suc h co nsiderations a little furth e r. we ca n see
xxxiv Introduction Introduction xxxv

both why the social sciences may no t appear to generate a great state sovereignty, do seventeenth-century theories o f the state
deal o f original knowledge and also why theo ries and ideas pro- reta in a relevance to social or po litical refl ection today? Surely
duced in the past, apparently paradoxically, may retain a relevance exactly because they have contri buted to constitu ting the social
to th e present day which archaic conceptions of the natural world we no w live in. It is the fa ct that they are reflectio ns upon a
sciences do not have. T he best and most interesting ideas in the social reality which they also help to constitute and which both
social sciences (a) participate in foste ring the climate o f opi nion has a distance from , yet remains part o f, o ur social world that
and thl? social processes which give rise to t hem, (b) are in-greater engages o ur attention. T heories in the natural sciences which
or lesser degree entwined with theories-in-use which help to consti- have been repl aced by o thers which do the same job better are o f
tute those processes and (c) are thus unl ikely to be clearly dis1inct no interest to the current practice of science. Th is cannot be the
fro m considered refl ection whi ch lay actors may bring to b (~a~ in case where those theo ri es have helped to constitu te what th ey
so far as they discursively articulate, o r improve upon, theories- in terpret o r explicate. Th e 'history o f ideas' may perhaps
in -use. T hese facts have conseq uences, particu larly for socio logy justifiably be regarded as o f margi nal importan ce to the practising
(to which they are most distinctly relevant) which affect both the natu ral scientist , but it is much more than tangenti al to the social
prosecution o f empiri cal research a nd the fo rmul atio n and sCie nces.
reception of theories. In respect o f research they mean that it is If (hey are correc t, these ru minations lead o n in a direct way to
muc h more difficul l than is th e case in natural science to ;hold up ' a consideratio n o f social science as cr itiq ue - as invo lved in a
acceptance of th eories while searching for ways to test th em out practi cal fashion with social life . We cannot be content with the
appropriately. Social li fe moves on; appealing or potentiall y 'tech no logical ' version o f critiq ue proposed by th e o rtho dox
practical theories, hypotheses or find ings may be tak en up in consensus, a view deriving from a natural science model. Th e
social life in such a way tha t the o rigina l gro unds upon which th ey technOlogical view of critique supposes that the 'internal critiqu e'
cou ld be tested have altered anyway. There are many complex of social science - the critical assessments whic h those working
possible pe rm utations o f mutual 'feed- in' here, which co mbi ne in the social sciences make o f eac h o ther's views - uncompli-
also with th e difficul ties inh ere nt in controlling variables, cated ly generates an 'external crit ique' of lay beli efs that can be
repli ca ting observatio ns and oth er methodological q uandaries in th e basis of practi cal social intervention. But , given the
which the social sciences can find th emselves. Th eories in natural sign ificance of the 'double hermen eut ic' , matters are much more
science are o riginal, innovative and SO o n to the degree to which comp lex. The fo rmulation of critical theory is not an option;
they place in questio n what either lay actors o r pro fessio nal th eories a nd findi ngs in the social sciences are likely to have
scientists previo usly believed about the objects o r even ts to which prac tical (and polit ical) consequences regardless of whether o r
they refer. But theories in the social sciences have to be in some no t th e sociological observer o r pol icy-maker decides that they
part based upon ideas which (altho ugh not necessarily discursively ca n be 'applied' t.o a given practical iss ue.
fo rmulated by them) are already held by th e agents to whom they
refer . Once rei ncorporated withi n action, their original q uali ty This was not a pa rti cularly easy book to write and proved in some
may become lost; they may become all too fa mili ar. T he no tio n pa ri refractory to the normal o rde ring o f chapters. Structuratio n
o f sovereignty and associated theories of the state were stunningly theory was formulated in substantial parl through its own 'internal
new when first formu lated; today th ey have fn some degree c ritique' - th e cri tical eval uation of a variety o f currently
become a part of the very social reality which they helped to co mpe ting schools of social tho ught. Rather than allow some o f
estab lish. th ese critical confron tati ons to obtrude into the main secti ons of
l3u l why do some social th eories reta in their freshness lo ng the tex t, I have in cl uded them as append ices to those chapters to
afh.:r the co ndi tions that helped produce them arc past'! Why. whic h they most immed iately re late. (Notes associat ed with th em
now Ihat we arc well fam iliar with the concept and th e rea lity o f simi larly fo llow Ihe notes that belong to releva nt c hapters.) The
Introduc tion xxxvi i
xxxvi In trodu ction

reader who wants to foll ow the main line of the argument in an I' versio n of social science the ' action frame of reference'. But. as I
unen cumbered way can pass over them. They will , however, be have sought to show at some length elsewhere (s~e NRSM,.chapler
of int erest to anyone concerned either with how th e views I 3), what I would regard as a satisfactory ~ nc e~tl o n of action (an.d
other related notio ns, especially those of mtenllons and reasons) IS
defend differ from those of others or with the elaboration of
no tlO be found in Parsons's work. Th is is not, as some commentators
themes treated in a condensed way in the core of each chapter. A have suggested, because a later emphasis upon fu~ct,io n ali s m . an?
variety of neologisms are used in the book, and I have placed a systems theory tended to swamp an en.rlier concern With voluntansm .
glossary of these at the end. It is because the idea o f voluntarism was fla wed at source. In
Parsons's thought vo luntarism has always been linked with the
resolution of th e ' problem of order', conce ived of by him as the co-
Reference ordination of po tentially disruptive individual wills. It is resolved
through the demo nstration that actors inlernalize, as motives, the
Il would, of course, be a mistak e to suppose that the influence of shared valu es upon whic h social cohesion depends. The call for an
Parsons is confined to the past, to imagine that Parsons has been account of actio n becomes con fl ated with the demand to link. a
forgotten in the same way as he once suggested happened to Spencer ' psycho logicar theory o f motivatio n Wil~ a 'socio logic~r int e~pr eta
very S(X)n after his death. On the contrary, one of the most visible tio n o f the struc tural features o f SOCial systems . . Little. If any,
trends in soc ial theory to day is th e prim e part played by views drawn conceptual room is left for what 1 emphasize ~ th e kno.wledgeability
more or less directly from Parsons. One might inst ance the writings o f social actors, as constitutive in purt of SOCial practi ces. I do not
o f Luhmann and Habermas in Germany, Bourricauld in France and think that any standpoint which is heavily indebted to Parsons can
Alexander and others in the United States. 1 do not intend to discuss cope satisfacto rily with this issu e at the very core of the concerns of
any of this literature in detail, but it is probably worth spe lling out a social theory as I co nceive of it in this book.
little why I do not have much sy mpathy with those aspects o f the H those stro ngly indebted to Parsons today do no t regard
writings of suc h autho rs which are closely based o n Parsons's ideas. themselves as fun ctio nalists and have rejected the fun ctio nalist cast
All the writers in questio n are strongly critical o f Parsons's o f Parsons's thought in greater or lesser degree, th ey still t ~ke over
connections with fun ctionalism, of which Luhmann pro bably seeks oth er ideas related to most versions o f fun ctionalism. These mclude:
to retain more than the others. In this respect, I am in accord with a fascination with 'value-consensus' or s.ymbolic orders at the expense
them, as this text should make clear enough. But in oth er ways, for o f (he more mundan e, practical aspects o f social act ivity ; the
reasons which are also documen ted at some length in this book, I tendency to assume that societies are eas ily distin guis h~ b l e unities,
consider that a radical break. has to be made with Parsonian theorems. as biological organ isms are; and a fondness for e~ol utlon~ry- st~le
An important aspect o f this concerns t he filtering o f the influence of theories. I consider each o f these emphases to be seno usly mlsleadmg
Max Weber through the writings o f Parsons. 1 have o ft en been called and shall enter stro ng reservatio ns about them. There can be no
a 'Weberian' by critics who regard this as some sort o f irreparable doubt about the sophistication and importance of the work of some
fault. I do not see the term, as they do, as a slur, but neither do I uuthors currently endeavouring to develop Parsons's wo rk in novel
accept it as accurately applied to my views. If I draw upon Weber, it ways, particularly Luhmann and Haberma'i..Bu~ I think it as necessary
is from an angle different from that of the aforementio ned authors. to repudiate the newer versions o f. ParsOlllalllsm as ,I do t he longer
Thus Habermas's Weber (surprisingly perhaps) tends to be a established varieties o f no n-ParSOlllan structural SOCiology.
Parson ian-style Weber. concerned above all with the ratio nalization
o f values and with 'social differentiation', portrayed as generalized
processes of development. Social life is not depict ed here through
the lenses I would prefer to borrow from Weber, as concerned with
the multifarious practices and struggles o f concretely located actors;
with conflict and the clash o f sectio nal interests; and with the
territo riality and vio lence o f polit ical fo rmations or states.
I>arsons regarded himself as an actio n theorise and called his
Elements of the Theory
of Structuration

In offering a preliminary exposition o f the main concepts of

structuration th eoryJ it will be useful to begin from the divisio ns
wh ic h have separated functi onalism (i ncluding systems theory)
and struct uralism on th e one hand from henneneutics and the
various forms o f 'interpretative sociology' on the other.
Fu nctionalism and structuralism have some notable similarities,
in spite o f th e otherwise marked contrasts that exist be tween
the m . Bo th tend to express a naturalistic standpoint , and both are
inclined towards objectivism. Functionalist thought , from Comte
o nwards , has looked particularly towards biology as the science
providing th e closest and most compatible model for social
sc ience. Biology has been taken to provide a guide to
conceptualizing the structure and the functioning of social systems
and to analysi ng processes of evolution via mechanisms of
adaptatio n. Structuralist thought, especially in the writings of
l evi-Strauss, has been hosti le to evolutionism and free from
hiological a nalog ies. Here the ho mo logy between social and
natural sc ience is primarily a cognitive on e in so far as each is
s upposed to express si milar features o f the overall constitution of
mind. Both structuralism and functionalism strongly emphasize
Ihe pre-em inen ce of the social who le over its individual parts
(i.e .. its constitue nt actors , human subjects).
In herm eneutic traditions o f thought, of course, th e social and
natural sciences are regarded as rad ically discrepant. Hennen eu-
li es has been (he ho me o f that 'humanism' to which structuralists
have been so strongly a nd persistentl y opposed. In herme ne uti c
Ihought, such as presented by Dilthey . the gu lf between subject
and social o bject is al its widest. Subj ecti vi ty is the preconsLituted
Ikf.., r.., n cc.~ umy h.., [" llllli un 1111..11 - '1.
2 Elements o f the Th eory o f Structurarion Elements of the Th eory of Structura tion 3

centre o f the experience o f culture and history and as such socio logies. ]n struc turatio n theory a he nneneutie starting-point
pro vides the basic foundatio n of the social or human scie nces. is accepted in so fa r as it is ack nowledged that the descriptio n of
Outside the realm of subjective experience , and alie n to it , lies human activities demands a fami liarity with the forms of life
the material world , governed by impersonal re lations of cause expressed in those activiti es.
and effect. Whereas for those schools of thought whi ch tend It is the specifi cally reflexive form of th e knowledgeability of
towards naturalism subj ectivit y has been regarded as some thing hum an agents that is most d eeply involved in the recu rsive
o f a mystery, or almost a residu al ph eno menon , for hermeneutics o rdering of social practices. Continuity of practices presumes
it is th e world of nature whi ch is o paque - which , unlike human refl exivity, but reflexivity in turn is possible o nly because of the
activity, can be grasped on ly from th e outsi de. In in terpre tative continuity of practices that makes them distinctively 'the same '
SOC io logies, action and meaning are accorded primacy in the ac ross space and time. 'Re flexivity' hen ce sho uld be understood
explication of human conduct; structural concepts are not no tably no t merely as 'self-consciousness' but as th e monito red character
pro minent , and there is no t muc h talk of constra int. For o f the o ngoing now o f social life . To be a huma n being is to be a
fun c tionalism a nd struc tura lism , howeve r, struc ture (in the purposive agent , who bo th has reasons for his or her activities
di vergent senses attributed to th at concept) has primacy o ver and is able , if asked , to elaborate discursively upon those reasons
action , and the constrain ing quali ties of structure are strongly (i ncluding lying about th em). But terms such as 'purpose' or
accentuated . 'intention' , 'reason' , 'motive' and so on have to be treated with
The differences between these perspectives on social scie nce ca utio n, since their usage in the philosophical literature has very
have o ft en been taken to be epistemo logical, whereas they are in o ften been associated with a hermeneutica l voluntarism , and
fact also o nto logical. What is at issue is how the concepts of beca use they extricate human action from th e contex tuality of
action , meaning and subjectivity sho uld be specifi ed and how timespace. H uman action occ urs as a duree. a co ntinuo us flow of
they might re la te to no tio ns o f stru cture and constraint. If conduct, as does cognitio n. Purposive act io n is no t composed of
inte rpre tative socio logies are fo unded , as it were, upo n a n an aggregate or series of separate intentio ns, reasons a nd motives.
imperialism of the s ubject, fun ctio nalism and struc turalism T hus it is useful to speak o f refl exivity as grounded in the
propose an imperialism of the social o bject. One o f my principal
ambitions in the formulation of st ru cturation theory is to put an
end to each of these empire-building endeavours. Th e basic
domain o f study of the social scie nces , according to th e th eo ry o f
struc turatio n, is neither th e experie nce of the ind ividu al ac tor,
nor the existence of any form o f societal totality , but soc ial
I continuo us monitoring o f action which human beings display and
expec t others to displ ay. T he reflexive mo nitoring of action
clepends upon ratio nalization , understood here as a process rather
than a state and as inhe re ntly involved in the competence of
age nts. An onto logy o f ti me-space as const it ut ive o f soc ial
practices is basic to th e conce ption of structuratio n, which begins
prac tices ordered across space and time. Human social ac ti vi ties , from temporality and thus, in one sense , 'history'.
like some self-re producing ite ms in nature, are recursive. T hat is T his approach ca n draw o nly sparingly upon (he a nalytical
to say, they are no t brought into being by socia l actors but philosoph y o f ac tio n, as 'action' is o rdinarily portrayed by most
continually recreated by them via the very means where by they co ntemporar y AngloAmerican writers. 'Action' is no t a combina
express themselves as actors. In and through their activities agents tion o f 'acts': 'acts' are constituted o nly by a discursive moment of
reproduce th e conditi ons that make th ese activities possible. ull ention to th e dllree o f lived -through experience. Nor can
However, the sort of 'knowledgea bility' displayed in nature, in 'ac ti o n' be discu ssed in sepa rati on from th e body, its med iatio ns
th e fo rm o f coded programmes, is distant from th e cognitive skills wi th th e surro undin g wo rld a nd the coheren ce o f a n ac ting self.
disp layed by hUlman age nts. It is in th e co nceptualizing o f human Wha t 1 call a sfl'afl'/i'c liliofl model of the ac ting se l[ invo lves
knowledgeah ility and its in vo lveme nt in ac ti on that I see k to Irt;aling th e reflex ive mo nitoring, ra tio nali zatio n and motivation
appropriat e some o f the majo r contributio ns of in te rpre tat ive ur ac tio n as embedded se ts of processes. l The n\l io na li zat io n of
4 Ele ments of the Theory of StrUCWra ti on Th e Agent, Agency 5

action , referring to 'intentionality' as process, is, like th e other in consciousness only in distorted form. Unconscio us motivational
two dimensio ns, a ro utin e charac teristic of human condu ct, components of acti on, as psychoa nalytic th eory suggests, have an
carried o n in a taken-fa r-grant ed fashion. In circumstances o f internal hierarc hy of the ir o wn , a hi erarchy whic h e xpresses th e
interactio n - encounters and episcxles - the reflexive monito ring 'depth ' of the life histo ry of the individual acto r. In saying this I
of actio n typica lly, and again routinely, inco rpo rates the do not imply an uncritical acceptance of the key theorems of
mo nitoring o f the selling o f such interaction _ As I shall indicate Fre ud's writings. We s ho uld guard against two form s of
subsequently, th is pheno meno n is basic to the inte rpolatio n o f reduc tionism whi ch those writings suggest o r foste r. One is a
actio n within the time-space relations o f what I shall call co- reduc tive conception of institutio ns which , in see king to show the
presence. The ratio nalizatio n o f action , within the di versity o f fo undation of in stitutio ns in the unconscio us, fails to leave
c ircumstances o f interac tion, is the principal basis upon which sufficient play for the o peratio n of au to nomo us soc ial forces. The
the generalized 'co mpe tence' o f actors is evaluated by o th ers. It second is a reducti ve th eory o f consc io usness whic h, wanting to
sho uld be clea r, howeve r, that the tendency of some philosoph ers show how much of social life is govern ed by da rk c urrents outside
to equate reaso ns with 'no rmati ve commitments' sho uld be th e scope of actors' aware ness , ca nno t adequa tely grasp the level
resisted: such commitment s comprise only one sector o f th e of control which agents are charac teristically able to sustain
rationalizatio n o f ac tio n. If this is not understood, we fail to reflexively over their conduc t.
understand that no rms fi gure as 'factual' boundaries of social life,
to which a vari ety of manipulative attitudes are possible. On e
aspect of such attitud es , altho ugh a relatively superficial o ne, is to Th e Agent, Agency
be found in the commo nplace observation that the reason s acto rs The stratification model of th e agent can be represented as in
offer discursive ly for wh at they do may diverge fro m th e fi gure 1. The reflexive monit oring o f activity is a chronic feature
rati onalizatio n of actio n as actually involved in the stream o f o f everyday action and invo lves the conduct not just of the
condu ct of those actors . individual but also of o thers. Th at is to say, actors not only
This circumstance has been a frequent source of worry to
philosophers and o bservers of the social scene - for how can we
be sure thai people do no t dissimulate concerning the reasons fo r
their ac tivities? But it is o f relatively little interest co mpared with
lll1,Kk n"wl~ged
( n nd ilions of
rdlc"ivt~ m on itorin g o f act ion > i unintcndL-d
: CO~SL-q U''''CCS
I act,on
,l(l,on 1
A, .,"

ra t,ona ''liI ioon
aC ioon
the wide 'grey areas' tha t exist between two stram of processes , 'it,
not accessible to the discursive consciousness o f acto rs. The vast I, mot iva t ion of ac tion
bulk of [he 'sto ck s o f knowledge', in Schutz's phrase, or wha t I ~- -- - -----------------'
prefe r to call th e mut ual kn owledge incorporated in enco unte rs,
Figure 1
is no t direc tly accessible to the consciousness of ac tors . Most
such knowledge is practical in character: it is inherent in th e
capability to 'go o n' within th e routines of social life. Th e lin e monitor continuously th e fl ow of their activities and expect others
between discursive and practical consciousness is fluctuating and 10 do the same for their own; th ey also routin ely monitor aspects,
permeable , bo th in th e experi ence of the individual agent and as soc ial and physical, of the contexts in whi ch they move. By the
regards compari sons between actors in different contexts of social nllionalization of action , I mean tha t actors - al so routinely and
activity . Th ere is no bar between these, however, as there is ror the most part without fuss - maintain a co ntinuing 'theoretical
betwee n th e unco nscio us and disc ursive consciousness. Th e 1I1lderstanding' of th e ground s o f the ir activity. As I have
un co llsc io us includ es those fo rms of cognition and impul sio n Ill cntioned, havin g such un und erstanding sho uld not be equated
whic h arc e ith er who ll y repressed from con sc iousness o r a ppea r with th e disc ursive giving o f rcuso ns fo r parti cular it ems of
6 Elements of the Theory of SlruclUration Th e Agent, Agency 7

conduct, nor even with the capability of specifying such reasons Only in phenome no logy and ethno methodology , within socio-
discursively. However, it is expected by competent agents of logical traditions, do we find detailed and subtle treatments of the
o thers - and is the main criterio n of competence applied in day- nature of pract ical consc iousness. Indeed, it is these schools of
to-day conduct - that actors will usually be able to explain most tho ught , together with ordinary language philosophy, which have
of what they do, if asked. Questions often posed about inten tio ns been responsible for making clear the shortcomings of orthodox
and reasons by philosophers are normally only pu t by lay actors social scient ific th eori es in this respect. I do not intend the
either when some piece of cond uct is specifically puzzling or d isti nction between discursive and practical consciousness to be
when there is a 'lapse' o r fracture in competency which might in a rigid and impermeable o ne. On the contrary, the d ivision
fact be an intended one. Thus we wi ll no t ordinarily ask another between the two can be altered by many aspec ts of the agent's
person why he or she engages in an activity which is conventional socialization and learning experiences. Between discu rsive and
for the group or culture of whi ch that individual is a member. practical consciousness there is no bar; there are o nly the
Neither will we ordin arily ask for an explanati on if there occurs a differences between what can be said and what is characteristically
lapse for which it seems unlikely the agent can be held responsible, sim ply don e. However, there are barriers, centred principally
such as slips in bodily management (see the discussio n of 'Oops!' , upon repression, between discursive conscio usness a nd the
pp. 81- 3) or sli ps of the tongue. If Freud is correct, however, such unconscious.
phenomena might have a rationale to them, although th is is only
rare ly realized eith er by the perpetrato rs of such slips o r by others
who witness them (see pp. 94 - 104).
di ~c ur s;w c Ollsc i o u ~ n ess

pr actic al consc iousne.s

I distinguish the refl ex ive monitoring and rationalization of
unconscious motive sj (Olln ilinn
act io n from its moti vatio n. If reasons refer to the grounds of
action, motives refer to the wants which prompt it. However,
mo tivation is not as directly bound up with the co nti nuity of As explained elsewhere in the book, I offer these conce pts in
actio n as are its reflex ive monitoring or rationalizat ion. Mo tivation place of the traditional psychoanalytic triad of ego , super-ego and
refers to potential for action rath er than to the mode in which id. The Freudian distinction of ego and id cannot easily cope with
action is chronically carried o n by the agent. Motives tend to th e analysis of practical consciousn ess, which lacks a theoretical
have a direct purchase o n actio n o nly in relatively unusual home in psychoanalytic theory as in the o ther types of social
ci rc umstances , situatio ns wh ic h in some way break wirh the tho ught previously ind ica ted. The concept of 'pre-conscio us' is
routine. For th e most part motives supply overall plans or perhaps the closest no tion to practical conscio usness in the
program mes - 'projects', in Schutz's term - withi n whic h a co nce ptual repertoi re of psychoanalysis but , as o rd inarily used ,
range of conduct is enacted. Much of our day-to-day conduct is cl early means something different. In place of th e 'ego', it is
not directly motivated. pre ferable to speak of the '1' (as , of course , Freud did in the
While competent actors can nea rly always report d iscursively original German ). This usage does not prevent anthropomor-
about their intentio ns in , and reasons for, acting as they do , they phism, in which the ego is pictured as a sort of mini-agent; but it
cannot necessarily do so of their motives. Unconscio us motivation does at least help to begin to remed y it. T he use of 'I' develops
is a sign ificant feature of human co nduct, although 1 shall later o ut o f, and is thereaft e r associated with, the positioning of the
indicate some reservations about Freud 's interpretat io n of th e age nt in soc ial encou nters. As a term of a predicative sort, it is
nature of the un conscious. The no tio n of practical co nsciousn ess 'e mpty' of conte nt , as co mpared with the richn ess of the actor's
is fundamental to structuration theo ry. It is that characteristic of st.: lf-descriptions invo lved with 'me'. Mastery ofT , 'me' , 'you'
the hum an agent o r subject to which stru cturalism has been re latio ns, as appli ed refl ex ively in d iscourse , is of key importance
parti cularly blind.J BUI so have o ther types of o bjec tivisltho ught. 10 Ihe e me rging compe tence of agents learning language. Since I
8 Elements of the Theory of Structuralion Th e Agent Agency 9

do not use the te nn 'ego', it is evide ntly best to dispense with coffee , thinking mistake nl y that it is tea , spilling the coffee is an
'super-ego ' also - a clumsy term in any case. The te rm 'moral act o f that person, even tho ugh it has not been do ne inte ntio nally;
conscience' will do perfectly well as a replacement. under another descriptio n, as 'spilling the tea', it is in tentio nal. 4
These concepts all refer to the agen t. What of the nature of (I n most instances, 'spi lling' something tends to have the
age ncy? Thi s can be connected with a further issue. The duree of implica tion that the act is uninten tional. It is a slip intervening in
day-to-day life occ urs as a flow of intenti onal action. However, a course of action in which the person is intending to do something
acts have unintended consequ ences; and, as indicated in figure 1, different altogeth er. na mely pass the cup ( 0 anothe r person.
unintended consequences may systematically feed back to be th e Freud claims that nearly all such behavio ural slips, lik e slips of
unacknowledged conditions of furth er acts. Thus one of th e the tongue, are actuall y unconsciously motivated. This , of course.
reg ular consequences of my speak ing or writing English in a brings them under inten tional descriptions from another angle. )
correct way is to contribute to th e reproduction of the English But even the view that for an e vent to count as an instance of
language as a whole. My speaking English correctly is intentio nal ; agency , it must be inte ntio nal o nly under some description or
the contributio n I make to the reproduction of the la nguage is a no ther is wrong. It confuses the designation o f agency with the
not. But how should we fonnul ate what unintended consequences givi ng of aCl-d esc ripti o n s; ~ and it mista kes th e continued
are ? monito ring of an action whi ch individuals carry o ut with the
11 has freque ntly been supposed that human agen cy can be defi ning properties of that act io n as such. Age ncy refe rs not to !
defin ed only in terms of intentio ns. That is to say, for an ilem of the intentions peopl e have in doing things but to their capability
behavio ur to co unt as action , whoever perpetrates it must intend o f doing those things in the first place (which is why agency
to do so, o r else the behavio ur in question is just a reactive implies power: cr. th e Oxfo rd English Dictio nary definition of an
response. The view derives some plaUSibility, perhaps, from the agent , as 'one who exerts power or produces an effect'). Agency
fact that there are some acts which ca nno t occur unless the agent conce rns events o f which a n individual is th e perpe trator , in the
intends them. Suicide is a case in poi nt. Durkheim 's conceptua l sense that th e individual could , at any phase in a given seque nce
efforts to the contrary, 'suicide ' canno t be said to occur unless o f conduc t, have acted differe ntly. Whateve r happe ned would
there is some kind of intent [0 prec ipitate self-destru ctio n. A no t have happened if that indi vidual had not inte rve ned. Action
person who steps off the curb and is knocked down by an is a contin uous process, a flow, in which the reflexive mo nitoring
oncoming car cannot be said LO be a 'suicide' if the event is whi ch the individual maintain s is fundament al to th e control of
accidental; it is something that happens to the individual, rath er the body that actors o rdinarily sustain throughout their day-to-
than something the individu al does. However, suicid e is not day lives. I am th e author o f many things I do not inte nd to do,
typica l of most human acts , in respect of inten tions, in SO far as it and may not want to bring about, but none the less do. Conversely,
can be said to have occurred o nly when its perpetrator in tended th ere may be c ircumstances in which I intend to achieve
it to occur. Most ac ts do no t have this characteristic. some thing, and do ac hieve it , altho ugh nO( directly thro ugh my
Some philosophers have argued . however, tha t fo r an event in agency. Take the example o f the spilled coffee. Supposing an
which a human being is in volved [0 count as an example o f indi vidual , A . were a malicio us spirit and played a practical joke
agency , it is necessary at least that what the perso n does be by placi ng the cu p on a sa uce r at suc h an angle that, when pi cked
intentiona l under some descriptio n, even if the agent is mistaken up, it would be very likely to spill. Indi vidual B picks up the
about th at desc ription. An offi cer o n a submarin e pulls a leve r co ffee, and it duly spills o ver. It would be right to say chat what A
int ending to cha nge course but in stead, having pulled th e wrong did bro ught the in cident about, or at least contributed to its
lever. sink s th e Bismarck. He has done something intentio nall y, comi ng about. But A d id nOI spill the coffee; B did. Individual S,
albeit no t what he imagined . but thus Ih e Bi.Hnarck has been sunk who did no t intend to spill the coffee. spilled the coffee; individual
thro ugh his agency. Aga in . if someo ne inte ntionally spills some A, who did inte nd thai the coffee shou ld be spilled , did not spill it.
10 Elements of the Th eory of Structuration The Agent, Agency 11

But what is it to do something unintentionally? Is it different reason, while knowing the prowler was there , the agent did not
from bringing about consequences unintentionally? Consider the seek to use this knowledge to alert the intruder. Unintentional
so-called 'accordi on effect' of action .6 An indi vidual fli cks a doings can be separated conceptually from unintended
switch to illuminate a room. Although this is intentional , the fact consequences of doings , although th e distinction will not matter
that the turning o n o f the switch alerts a prowler is not. Supposing whenever the foc us of concern is the re latio n between the
the prowler flees down the road , is ca ught by a policeman , and in tentio nal and unintentional. The consequences o f what actors
after due process spends a yea r in gaol o n the basis of being do, intentionally or unintentio nally , are events which would not
convicted of the burglary. Are all these unintended consequences have happened if that acto r had behaved differently, but which
of the act of flicking the switch? Which are things the individual are not within th e scope o f the agent's powe r to have brought
has 'done '? Let me me ntion an additional example , taken from a about (regardless of what the agent's inte ntions were).
theory of ethnic seg regat io n. 7 A pattern of ethnic segregation I think we can say that all the things that happened to the
might d evelop, without a ny of those involved intending this to prowler following th e fli cking of th e switch were unintended
happen, in the following way, which can be illustrated by analogy. consequences of the act, given [hat th e individual in question did
Imagine a chessboard which has a set of 5-pence pieces and a set not know the prowler was there and th erefo re initiated the
of lO-pence pieces. These are distributed randomly on the board, sequence unintentionally. If there are complexities in this, they
as individuals might be in an urban area. It is presumed tha t, are to do with how it comes abou t that a seemingly trivial act may
while they feel no hostility towards th e other group, the members trigger events far removed from it in time and space , not whether
of each group do not want to live in a neighbourhood where th ey or not those consequences were int end ed by the perpetrator of
are ethnically in a min ority. O n the chessboard each piece is the original act. In general it is tru e that the further removed the
moved around until it is in such a position that at least 50 per ce nt consequences of an act are in lim e and space from the original
of the adjoining pieces are of the same type. The result is a context of the act , the less lik ely those consequen ces are to be
pattern of extreme segregation. The 'O-Cent pieces end up as a intentio nal - but this is, of co urse, influ enced both by the scope
sort of ghetto in th e midst o f th e 5-cent ones. The 'composition o f the knowledgeability that actors have (see pp. 90-2) and the
effect' is an outcome o f an aggregate of aclS - whether those of power they are able to mobilize. We wou ld o rdinarily think of
movi ng pieces o n the board o r those of age nts in a ho using what the agent 'd oes' - as contrasted with the consequences
marke t - each o f which is intentionally carried out. But the e ns uing from what has been done - in terms of pheno mena the
eventual o utcome is ne ither intended nor desired by anyone. It is , agent has more or less within his o r he r contro l. In most spheres
as it were , everyone's do ing and no one's. o f life, and in most forms of activity , the scope of control is
To understand what it is to do something unintentionally, we limited to the immediate contexts of action or interac tion. Thus
have first of all to be clear how 'intentional' should be understood. we would say that turning o n the light was something the agent
This conce pt r define as c haracterizing an act which its did , and probably a lso a lerting th e prowler , but not causing the
perpetrator knows, or beli eves, will have a particular quali ty or prowler to get caught by the policeman or to end up spending a
outcome and where such knowledge is utilized by the autho r of yt:ar in gaol. Although it might be th e case that th ese events
the act to ach ieve this quality or o utcome. s If the characteri zation wo uld not have happened wh en and where they did without the
of agency given above is co rrec t, we have to separate o ut the :1(,;1 of flicking the switch, their occ urrence depended on too
question of what an agent 'does' from what is 'intended' or th e many other cont inge nt outcomes fo r th em to be some thing the
int entional aspects o f what is done . Agency refers to doing. o riginal ac tor 'did' .
Switching o n th e light was something the agent did. and al erting Philosophers have used up a greal deal of ink attemp ting to
the prow ler was also so methin g that agent did. It was unintend ed :lIlal yse th e nature of int entio nal activil y. But fro m th e point of
if th e actor d id no\ know th e prowl er was th ere and if for somc vil.;w o f the soc ial sc iences. it is hard to exagg~ra l e th e importance
12 Elem ents of the Theory of Sl ructuration The Agen t. Agency 13

of the unintended consequ ences of intentional conduct. Merto n needs and intentions in the ind ividu al actOr. rn the indi vidual ,
has provided perhaps the classical discussion of th e issue. 9 He wants tha t are constitutive of the mo tivatio na l impulses of the
points o ut , entirely correctly, that the study of uninte nded actor gen erate a d ynamic relat io n betwee n mo ti vatio n and
consequences is fund amen tal 10 the sociological enterprise. A intentionality. This is no t the case wi th social systems, except
given item o f acti vity may have eith er (a) non-significant o r (b) where acto rs behave in cognizance o f what they tak e to be social
significant co nseque nces; and e ither (c) singly signiricant needs. I I
consequences o r (d ) multiply significant consequen ces. What is This po int having been mad e, the re can be no quarrel with
judged 'significa nt ' will depe nd upo n the nature o f the study Merton's emphasis upo n the significa nce of connecting unin-
being undertak en o r the theory being develo ped. lo However , tended co nsequences of actio n with institutio nalized practices ,
Merto n then goes o n to couple unintended consequences with those deeply embedded in time and space. This represents the
fun ctional analysis, a conceptual move which , although conven- most important of three main researc h contexts - separable
tionally made in the sociologica l literature , I wish to rejec t. In from one an o th er o nly analytically - in whic h the influence of
particular , it is important to see that the analysis of unintended unintended consequ ences ca n be analysed. One is th e turning on
co nsequences does not (as Mert on claims it does) make sense of th e light/ alerting the prowler/ causing th e prowler to fl ee/etc.
seemingly irrati onal fo rms o r patterns of social conduct. Merton type of example. The interest of the resea rche r here is in the
contrasts inte ntio na l activity (manifest functions) with its c umulation of events deriving from a n initiating circ umstance
unintended co nsequences (latent functions). One of the aims o f without which that cumulatio n wo uld no t have been found. Max
identifying latent fun ctio ns is to show that apparently irra tio na l Weber's analysis of the effects o f the Battle o f Marathon on the
social activiti es may not be so irrational after all. T hi s is subsequent developm ent o f G reek culture. a nd th ence o f the
particularly likely to be the case, according to Merton, with formation of European c ulture in ge neral, is a case in point , as is
enduring activities o r prac tices. These may often be dismissed as his discussion of the conseq uences of the firing of the bullet that
"'superstitio ns" , "irratio nalities", "mere inertia of traditio n " , etc', killed Archduke Ferdinand at Sarajevo. l.I T he concern is with a
However, in Me rton's view, if we discover that they have a late nt singul ar set o f events, traced through and analysed co unter-
functio n - an unintended consequ ence, o r set o f consequ ences , factually. The researche r asks, 'Wha t would have happened to
which help to secure the continued re productio n of the practice events B, C , D , E .. . if A had no t occurred?' - thereby see king
in questio n - th en we demo nstrate [hat it is not so irrational at all . to identify the role o f A in th e cha in o r sequence.
Thus a ceremo nial, fo r example , 'may fulfil the latent fun ction A second type o f circumstance upo n which [he social analyst
o f reinfo rCing the gro up identity by providing a periodic occasion might focus is o ne in whic h, instead o f a pa tte rn o f uninte nded
on which the sca ttered membe rs o f a group assemble to engage in co nseque nces initiated by a single event , there is a pattern
a commo n activity'.II But to suppose that such a demo nstratio n of resulti ng from a complex o f individua l acti viti es. T he discussion
a fun ctional relation provides a reason for th e existence o f a o f ethn ic segregatio n mentio ned above is an example o f this.
practi ce is mistaken. What is being more or less surreptitiously Here a definite 'end result ' is taken as the phenomeno n to be
smuggled in here is a conceptio n of 'society's reasons' o n the basis ex plain ed, and that end result is shown to derive as an unintended
of imputed social needs. Thus if we understand that th e gro up (;onsequ ence from an aggregate o f courses of intentional conduct.
'needs' the ce remonia l to enable it to survive , we see its T he th eme of rationality te nds to surface again here, although
continuatio n as no lo nge r irratio nal. But to say that the existence Ihis time [h ere is no logical o bj ecti o n to be made to it. As game
of a soc ial slate A needs a social prac tice B to help it to survive in Ih eo ri sts have convincingly po inted o ut. th e o utcome of a series
recogni zably similar form is to pose a question that then has to be or ratio nal ac tio ns, unde rtaken se parat ely by indi vidua l actors,
answered ; it does nOI itse lf answer it. The relatio n betwee n A and 1I1ay he irra tio na l for a ll o f Ih e m.' 4 'Pe rve rse effec ts' are o nl y one
B is no t analogo us 10 Ih e re lalio n th aI o bta ins bet wee n wants o r Iy pe o f uninte nd ed co nseque nces, altho ugh il is no do ubt tru e
14 Elements of the Th eory of Structuration Agency and Power 15

that situations where they occur are of particular interest. 15 the individual is confined by a range of specifiable circumstances. 16
The third type of context in which unintended consequences But it is of the first importance to recognize that circumstances of
may be traced out is that pointed to by Merton: where the social constraint in which individuals 'have no choice' are not to
interest of the analyst is in the mechanisms of reproduction of be equated with the dissolution of action as such, To 'have no
institutionalized practices. Here the unintended consequences of choice' does not mean that action has been replaced by reaction
action form the acknowledged conditions of further action in a (in the way in which a person blinks when a rapid movement is
non-reflexive feedback cycle (causal loops). I have pointed out made near the eyes). This might appear so obvious as not to need
that it is not enough to isolate functional relations in order to saying. But some very prominent schools of social theory,
explain why such feedback occurs. How, then, does it happen associated mainly with objectivism and with 'structural sociology' ,
that cycles of unintended consequences feed back to promote have not acknowledged the distinction. They have supposed that
social reproduction across long periods of time? In a general way, constraints operate like forces in nature, as if to 'have no choice'
this is not difficult to analyse. Repetitive activities, located in one were equivalent to being driven irresistibly and uncomprehen-
context of time and space, have regularized consequences, dingly by mechanical pressures (see pp. 211-13).
unintended by those who engage in those activities, in more or Expressing these observations in another way , we can say that
less 'distant' time-space contexts. What happens in this second action logically involves power in the sense of transfoqnative
series of contexts then, directly or indirectly, influences the further capacity, In t~is sense, the most all-embracing meaning of 'power',
conditions of action in the original context. To understand what power is logically prior to subjectivity, to the constitution of the
is going on no explanatory variables are needed other than those reflexive monitoring of conduct. It is worth emphasizing this
which explain why individuals are motivated to engage in because conceptions of power in the social sciences tend faithfully
regularized social practices across time and space, and what to reflect the dualism of subject and object referred to previously.
consequences ensue. The unintended consequences are regularly Thus 'power' is very often defined in terms of intent or the will, as
'distributed' as a by-product of regularized behaviour reflexively the capacity to achieve desired and intended outcomes. Other
sustained as such by its participants. writers by contrast, including both Parsons and Foucault, see
power as above all a property of society or the social community.
Agency and Power The point is not to eliminate one of these types of conception
at the expense of the other, but to express their relation as a
What is the nature of the logical connection between action and feature of the duality of structure. In my opinion, Bachrach and
power? Although the ramifications of the issue are complex, the Baratz are right when, in their well-known discussion of the
basic relation involved can easily be pointed to. To be able to 'act , matter, they say that there are two 'faces' of power (not three, as
otherwise' means being able to intervene in the world, or to Lukes declares). 17 They represent these as the capability of actors
refrain from such intervention, with the effect of influencing a to enact decisions which they favour on the one hand and the
specific process or state of affairs. This presumes that to be an 'mobilization of bias' that is built into institutions on the other.
agent is to be able to deploy (chronically, in the flow of daily life) This is not wholly satisfactory because it preserves a zero-sum
a range of causal powers, including that of influencing those conception of power. Rather than using their tenninology we can
deployed by others. Action depends upon the capability of the express the duality of structure in power relations in the following _
individual to 'make a difference' to a pre-existing state of affairs
or course of events. An agent ceases to be such if he or she loses
way , Resources (focused via signification and legitimation) are
stru ctured properti es of social systems, drawn upon and
the capability to 'make a difference', that is, to exercise some sort reproduced by knowledgeable agents in the course o~ interaction. U
o f power. Many interesting cases fo r social analysis centre upon Power is no t intrinsi cally conn ected to th e achIevem ent of
th e margins o f what can count as actio n - wh ere the power of secti o nal interests. In this conception th e use of power
Struc ture , Structuralion 17
16 Elemen ts of the Theory of 51rucruration
Th ese two ideas o f stru cture might seem at first sight to have
characterizes not specific types of conduct but all action , and nothing to do with on e ano ther, but in fac t each relates to
power is not itself a resource. Resources are media through which important aspects o f the structuring o f social re lations, as~e.ct s
power is exercised , as a ro utine eleme nt o f the instantiatio n of which , in the th eory of structuration . are grasped ~y rec~gm z.m~
condu ct in social re produ ction. We should not conceive o f the a differentiation between the concepts o f 'stru cture and system.
structures of domination built into social instituti ons as in some In analysing social relatio ns we have to a~knowl~dge . bo~h a
way grinding out 'docile bodies' who behave like the automata syntagmatic dime nsion , the pa tterning o f SOCial relations m lIme-
suggested by objectivist social scie nce. Power within social space invo lving the reproduction o f situated practices , and a
systems which enjoy some continui ty over time and space paradigmatic dime nsion, involving a virtual order of 'f'!2~des of
presumes regularized relations of aut o nomy and dependence structuring' recursively implicated in such re produ ctio n . In
between ac to rs or co llectivities in contexts of social interactio n. d structuralist traditions there -iSUSu ally ambiguity o ver whether
But all fo rms of de pe ndence o ffer some resources whe reby those stru ctures refer to a matri x o f admissible transfo rmations within a
who are subordinat e can influ ence th e acti vities of their superiors, set or to rules of transfo rm ation governing th e matrix. I treat
This is what I call th e dialectic of control in socia l systems. structure, "in its most el emental meaning at least . ~s referring to
such rules (and resources). It is misleading, however, to speak. of
Stru ctu re, Structurat io n 'rules o f transfo rmati on' because all rul es are inh erently
transfo rmational. Structure thus refers, in social analysis , to the
Let me now move to the core o f structuration theory : the co ncepts stru;luring properties allowing th e ' bindin~' o~ time-~pace in
o f 'structure', 'system' and 'du ality of structure'. The notio n o f social systems , the properties which make II pOSSible for
structure (or 'social structure'), o f course, is very promine nt in discernibly simil ar social practices to exist across varying spans o f
the writings of most functi o nalist authors and has lent its nam e to time and space and which lend them 'systemic' form. To say that
the traditio ns of 'stru cturalism'. But in neither instance is this structure is a 'virtual order' of transfo rmative relations means that
conceptualized in a fashion best suited to the de mands o f social social systems, as re produced social practices , do no t have
theory . Function alist authors and their critics have give n mu ch 'stru ctures' but rather exhibit 'stru ctural p rop erties' and that
more atte ntion to th e idea of 'function' than to that o f 'structure', stru cture exists, as time-space presence , only in its instantiations
and conseq uently the latter has tended to be used as a rece ived in such practices and as memory tr.!ces o rientin_ the conduct o f
notion. BU[ there can be no doubt about how 's truc lUre' is usually knowledgeable human age nts. This cf~es no t. preve nt u~ fro~
understood by fun ctio nalists and , indeed, by the vast majo rity of ) conceiving of structural properties as hierarchIcally orgaOlz~d III
social analysts - as some kind of 'patterning' of socia] ~el a ti o ns lenns of the time-space exte nsion of the practices they rec ursIvely
or social phenome na. This is o ft e n naively conceived of in terms o rgani ze. The most deeply embedded .structural . rro pe rties ,
o f visual imagery, akin to the skeleto n or mo rpho logy o f an implicated in the reprodu ction o f sOCietal to tahtles , I call
o rganism or to the girders o f a building. Such conceptio ns are structural principles. Those practi ces which have th e greatest
closely connected to the du alism of subject and social object: tim e-space extension within such to talities can be referred to as
'structure' here appears as 'ex te rnal' to human actio n , as a source institutions.
o f constraint -on the (ree initiative of the independently constituted T o speak o f structure as 'rules' and resources , an~ ~f s tru~tures
subj ect. As conceptualized in structuralist and post-structuralist as isolable sets o f rules and resources, runs a dlstm ct fisk of
thought, o n the o th er hand , the notio n of stru cture is more misint erpretatio n because of certain dominant uses of 'rules' in
int eresting. Here it is ch aracteristi ca lly tho ught o f no t as a th e philosophical literature.
patterning o f presences but <IS an intersectio n o f presence and
a bse nce: underl ying codes have to be inferred from su rfa ce (I) Rul es ure o fI en tho ught o f in co nn ectio n with games. as
manifestatio ns.
18 Elem ents of th e Th eor y of Structuratjon Structure. Structuratio n 19

formalized prescriptions. The rules implicated in the called upon to perform. In proposing a usage of 'structure' that
reproduction of social systems are not generally like this. might appear at first sight to be remote from conventional
Even those which are codified as laws are characteristically interpretations of the term, I do not mean to hold that looser
subject to a far greater diversity of contestations than the versions be abandoned altogether. 'Society', 'culture' and a range
rules of games. Although the use of the rules of games such of other forms of sociological terminology can have double usages
as chess, etc. as prototypical of the rule-governed properties that are embarrassing only in contexts where a difference is made
of social systems is frequently associated with Wittgenstein, in the nature of the statements employing them. Similarly , I see
more relevant is what Wiugenstein has to say about children's no particular objection to speaking of 'class structure', 'the
playas exemplifying the routines of social life. structure of the industrialized societies' and so on, where these
(2) Rules are frequently treated in the singular, as if they could terms are meant to indicate in a general way relevant institutional
be related to specific instances or pieces of conduct. But this features of a society or range of societies.
is highly misleading if regarded as analogous to the operation One of the main propositions of structuration theory is that the
of social life, in which practices are sustained in conjunction r~l es and_r~ c ~ rawn upon in _the__ !"~_duction and
with more or less loosely organized sets. reproduction of social action are at the same time the means of
(3) Rules cannot be conceptualized apart from resources, which syste~ reQ!..o duction (the duality of structure) . But how is one to
refer to the modes whereby transformative relations are interpret such a claim? In what sense is it the case that when I go
actually incorporated into the production and reproduction about my daily affairs my activities incorporate and reproduce,
of social practices. Structural properties thus express forms say , the overall institutions of modern capitalism'? What rules are
of domination and power. being invoked here in any case'? Consider the following possible
(4) Rules imply 'methodical procedures' of social interaction, as instances of what rules are:
Garfinkel in particular has made clear. Rules typically
(1) The rule defining checkmate in chess is .. .' ;
intersect with practices in the contextuality of situated
(2) A formula: an ::::: n 2 + n-l;
encounters: the range of 'ad hoc' considerations which he
(3) 'As a rule R gets up at 6.00 every day';
identifies are chronically involved with the instantiation of
(4) 'It is a rule that all workers must clock in at 8.00 a.m.'
rules and are fundamental to the form of those rules. Every
competent social actor, it should be added, is ipso facto a Many other examples could of course be offered, but these
social theorist on the level of discursive consciousness and a will serve in the present context. In usage (3) 'rule' is more
'methodological specialist' on the levels of both discursive or less equivalent to habit or routine. The sense of 'rule' here is
and practical consciousness. fairly weak , since it does not usually presuppose some sort of
(5 ) Rul es have two aspects to them, and it is essential to underlying precept that the individual is following or any sanction
distinguish these conceptually, since a number of philoso- which applies to back up that precept; it is simply something that
phical writers (such as Winch) have tended to conflate them. the person habitually does. Habit is part of routine, and I shall
Rules relate on the one hand to the constitution of meaning, strongly emphasize the importance of routine in social life. 'Rules' ,
and on the other to the sanctioning of modes of social as I understand them, certainly impinge upon numerous aspects
conduct. o f routine practice, but a routine practice is not as such a rule.
Cases (1) and (4) have seemed to many to represent two types
have introduced the above usage of 'structure' to help break o f rule , constitutive and regulative. To explain the rule governing
with th e fix ed o r mechani cal character which the ter-m tends -to checkmate in chess is to say something about what goes into the
h,lve in o rt hodo x soc iologi ca I Usage.~Th e concepts of system and very making of chess as a game. The rule that workers must clock
stru cturati o n do mu ch o f th e wo rk that 'structure' is o rdinarily in at a certain ho ur, on th e o ther hand , does not help define what
20 Elements of the Theory of SUuc turation Struc ture, Struc /uration 21

:-V0rk is: i~ specifies how work is to be carried on. As Searle puts because it allows fo r the method ical continuation of an established
Il , regulative rules can usually be paraphrased in the form 'Do X', sequence. Are linguistic rules li ke this? I think they are - much
o r 'If Y, do X : Some constituti ve rules will have this characte r, more than they are like the sorts of rule o f whic h Chomsky
but most will have the form 'X counts as Y ', or 'X cou nts as Y in speaks. And this seems also co nsonan t with Wittgenstein's
co nt e~ t C'. '8That there is something suspect in this distin ction , as arguments. or a possible construal of th em at any rate.
refern.ng to two types of rul e, is indicated by the etymo logical Wittgenstein rem ark s, 'To understand a language means to be a
clumSiness of the term 'regu lative rule'. After all, the word master of a techniqu e.'20 This can be read to mean th at language
'regu lati ve' already impli es 'rule': its dictionary defin ition is use is primarily methodo logical and that rul es o f language are
'contro l by rules'. I would say o f (I) and (4) that they express two methodically applied procedures implicat ed in the practical
aspects o f rules rather than two variant types of rule. ( I ) is act ivities o f d ay-la-day life. T his aspect of language is ve ry
certainly part of what chess is, but fo r those who play chess it has important , although not o ft en given much promi nence by most
sanctioning o r 'regulati ve' pro perties; it re fers to aspec ts o f play fo llowers o f Wittgenstein . Rules which are 'staled ', as (1) and (4)
that must be observed . But (4) also has constitutive aspects. It above, are interpretatio ns of activity as well as relating 1O specific
does not perhaps enter into the d efinition of what 'work' is, but it SOrts o f activiti es: all cod ified rules take this form , since they give
does ent er into that of a concep t like 'industrial bureaucracy'. verbal ex pression to what is supposed to be don e. But rules are
Wh at (I) and (4) direct our attentio n to are two aspects o f rules: procedures of action, aspec ts of praxis. It is by reference to this
their role in the constitutio n o f meaning, and th eir close that Wittgenstein resolves what he first of all sets up as a 'paradox'
connectio n with sanctio ns. o f rul es and rule-fo llowin g. This is that no courSe of action can be
Usage (2) might seem the least promising as a way of said to be guided by a rul e because every course o f actio n can be
co nceptual izi ng 'rule' that has any relation to 'structure'. In facl , I made to accord with that rule. However , if such is th e case, it is
shall argue, it is the most germane o f all o f them. I do not mean to also true that every course o f actio n can be made to conniet with
say that social li fe can be red uced to a set o f mathemat ical it. T here is a misu nde rsta nding he re . a confu sing o f the
prin ciples, which is very far fro m what I have in mind . I mean that inte rpretatio n or verbal ex pressio n of a rule with fo llowing (he
it is in the nature of formula e that we can best discover what is rule .2l
th e most analytically effective sense of 'rule' in social theory. The Let us regard the ru les of social life , th en, as techniques or
formula a" ;;;; 1121+ n-l is fro m Wittgenstein 's example of number generalizable proced ures applied in the enactment/reprod uction
gam es. 19 One person writes down a seq uence of numbers; a o f social practices. Formul ated rules - those th at are given
second wo rks out the formula supplying the numbers wh ich ve rbal expression as ca no ns o f law, bureau cratic ru les , rules of
follow. What is a formula of this kind, and what is it to understand games and so o n - arc thus codified interpretations o f rules
o ne ? To understand the fo rmula is not to utter it. For someone rather th an rules as such. They shou ld be taken no t as
co ul~ utter it and not understand the series; altern atively, it is
exemplifying rules in general but as spec ific types o f fo rmulated
poSSible to understand the series wi tho m being able to give verbal rule , which , by virtu e o f th eir overt formul atio n, tak e o n various
expressio n to the formula . Understanding is no t a me ntal process specific qualilies. n
accompanying th e solving of the puzzle that the sequence of So far these consideratio ns offer o nl y a preliminary approach
numbers presents - at least, it is not a mental process in the to the problem. How do formulae relate to th e practices in which
sense in which the heari ng of a tunc o r a spoken sentence is. It is social acto rs engage, and what kinds of formulae are we most
simply being able to appl y the formu la in the right cont ext and interested in for genera l purposes of social ana lys is? As regards
way in orde r to continu e the seri es. the first part of the qu estio n , we ca n say that aware ness o f social '
A formula is a generalizabl e procedure - generalizable because ru les, ex pres.<;ed first and fo remos t in practi ca l conscio usness , is
il applies ove r a range o f co nl ex ls and occasio ns. a procedure Ih e very core o f th at 'kn ow ledgea bilit y' whic h speci fi cally
22 Elements of the Th eory of St ruc turation Struc ture, StructuratiotJ 23

characterizes human agen ts. As soc ial actors, all human beings 'go on'. The discursive formulation of a role is already an
are highly 'learned' in respect of knowledge which they possess , interpretation of it, and , as I have no ted, may in a nd of itself alter
and apply , in th e productio n and reproduction o f day-to-day the form of its appli cation. Amo ng rul es that are not just
social encounters; the vast bulk of such knowledge is practical d iscursively formulated but are formally codified , the type case is
rather than theoretical in charac ter. As Schutz and many o thers that o f laws. Laws , of course , are among th e most strongly
have pointed o ut , actors employ typified schemes (formulae) in sanctioned types of social rules and in mo dern societies have
the course of th eir dai ly activities to negotiate routinely the fo rmally prescribed gradatio ns o f retribu tion. However , it would
situations of social life . Knowledge of procedure, or mastery of be a serious mistak e to underestima te the strength of info rmally
the techniqu es o f 'do ing' social activity , is by de finiti o n applied sanc tions in respect o f a variety o f mundan e daily
methodological. That is to say, such knowledge does not specify practices. Whatever else Garfinkel's 'ex periments with trust' might
all the situations which an actor might meet with , nor could it do be thought to demonstrate, they do show th e extraordinarily
so; rather , it provides for the generalized capacit y to respond to compelling for ce with whi ch apparently mino r features of
and influ ence an indeterminate range of social circumstances. conversational response are invested. 24
Those types of rul e which are of most significance for social The structuring qualities o f rules can be studied in respect, first
theory are locked into the reproduction of institutio nalized o f all, of the formin g, sustaining, termination and reforming of
practices, that is, practices most deeply sedimented in time- encounters. Although a dazzling vari ety of procedures and tactics
space. 2) T he main characteristics of rules relevant to general are used by agents in the constitutio n and reconstitution of
questions of social analysis can be described as follows : e ncounters, probably particularly significant are those involved
in th e sustaining of ontological security. Garfink el's 'experim ents'
i n tensi~ taci t info rmal wea kl y san( t;'med are certainly relevant in this respec t. They indicate that the
prescriptions involved in the stru cturing o f daily in teraction are
Mlallow fo rmali~ed
much more fixed and constrai ning than might appear from the
ease with whi ch they are ord in ari ly fo llowed. T his is surely
By rules that are intensive in nature, I mean formulae that are because the deviant respo nses or acts that G arfinkel instructed
constantly invoked in the course o f day-to-day activities, that his 'experimenters' to perform disturbed th e sense of o nto logical
enter into the stru cturing o f much o f the texture of everyday life. security of the 'subjects' by undermining the intelligibility of
Rules o f language are o f this character. But so also , fo r example . discourse . Breaking o r igno ring rules is no t, of course, the o nly
are the procedures utilized by actors in organizing turn-taking in way in which the constitutive and sanctio ning properties of
conversations o r in interaction. T hey may be contrasted with intensively invoked rules can be studied . But there is no do ubt
rul es which, altho ugh perhaps wide in scope, have o nly a that Garfinkel has helped to disclose a remark ably ri ch field of
superficial impact upon mu ch o f the texture of social life. The study - performing the 'sociologist's alc hemy', the 'transmutation
contrast is an important one, if only because it is commonly taken of any patch of ordinary social activity in to an illuminating
for granted among social analysts that the more abstract rules - publication'. H
e.g., codified law - are the most influential in the structuring of I distinguish 'structure' as a generi c term from 'structures' in
social activity. I would propose, however, that many seemingly the plural and both from th e 'stru ctural properties of social
trivial procedures foll owed in daily life have a more profo und systems'.26 ' S~re' refers not only to rules implicated in the
influ e nce upon the generality of social conduct. The remaining production and reproduction o f social systems but also to
categori es sho uld be mo re or less self-explanatory. Most of th e resources (aOout wFilcl1 I have so far nOI said mu chb ut will do so
rules implica ted in the produ ctio n and reprodu ction of social shortly ). As ordinarily used in Ih e social sciences, 'st ructu re'
prac tices arc o nly tacitl y grasped by actors: th ey know how to tends to be emp loyed with th e morc enduring aspects of social
24 Elements of th e Th eory of SUt/c lt/ration The Dliality of Structure 25

system s in mind , and I do not want to lose this connotation. The co ncept of the duality of structure as suc h. It is to do with how
most important aspects of struct ure are rules and reso urces soc ial systems, especially 'societies', should be conceptualized.
recursively involved in institutions. Institutions by definitio n are
th e more enduring features o f social life. In speaking of the The Duality of Structure
structural pro perties o f social systems 1 mean their inst itu -
tionalized features, giving 'solidity' across time and space. I use SY5Iem( s) Slruc luriJ lin n
the concept o f 'structures' to get at relations of transformatio n
Rul{'s a nd '{'sou.ces, or Ik pfOd uc..d ,,,13tio n. Conditions Rove.n ing the
and mediatio n which are the 'circ uit switches' underlying observed st"ts o f tra nsforma fio n he"....e n iJc to , s or c on tin ui ty o r transm uta tio n
conditions of system reprod uctio n. 'elalions. o rga nized a s collec 'i~~ 'it's, of st ...':lurt'5. and
p ro llt'.lie s o f soc iiJ l o .giJ ni1e<i iJS .e Rula. therdo <\: the ..., productio n
Let me now answer the question 1 originally posed: in what systems 'iOC ial prac tict's 01 socia l syste ms
manner can it be said that the conduct of individual actors
reprodu ces the structural properties of larger collectivities'! Th e Let me summarize the argument thus far. Structure, as recursively
question is both easier and more difficult to answer than it organ ized sets of rules and resources, is out of time and space,
appears. On a logical level, the answer to it is nothing more than a save in its instantiations and co-o rdination as memory traces, and
truism. That is to say, while the continued existence o f large is marked by an 'absence of the subj ect'. The soc ial systems in
collectivities or societies evidently does not depend upon the which structure is recursively impli cated, on the contrary,
I activities o f any particular individual, such collectiviti es o r co mprise the situated activities of human agents, reproduced
societies manifestly wo uld cease to be if all the agents invo lved across time and space. Analysing th e struc turation of social (
disappeared. On a substantive level, the answer to the question syste ms means studying the modes in whi ch such systems, i
depends upo n issues yet to be broached - those concerning the gro unded in the knowledgeable acti vities of situated actors who \
mechanisms of integration of different types of societal totality. It draw upon rules and resources in th e diversity of action contexts, )
is always the case that the day-ta-day activity of social actors are produced and reproduced in interact io n. Crucial to the idea
draws upon and reproduces structu ral features of wider social o f structuration is the th eorem of th e duality o f structure , which
systems. But 'societi es' - as I shall make clear - are not is logically implied in the arguments portrayed above. The
necessarily unified collecti viti es. 'Social reproduction' must not {;o nslitution of agents and S!!u ctures are no t two i"-~ependently
be equated with the consolidation of social cohesion. The locatio n given sets of pheno mena , ~~m ,15~ re present a ~uali.!}'.
of actors and o f co llect ivities in different sectors or regio ns of Accord ing to the notion o f [he duahty o f struc ture, the structural
more e!1compassing social systems strongly influences th e impact properties of social systems are bot h med ium and outcome of the
of even their habitual conduct upon the integration of socie tal pract ices they rec ursively o rganize, Stru cture is not 'ex ternal' to
totalities. Here we reach the limits o f linguistic examples which individuals: as memory traces, and as instantiated in social
might be used to illust rate the concept of the duality of structure. pfClctices, it is in a certain sense more 'internal' than exterior to
Considerable illuminati on o f problems of social analysis can be I heir activities in a Durkheimian sense, Structure is not to be
deri ved from studyi ng the recursive qualities of speech and l:qu<l led with constraint but is always both constraining and
language. Wh en I produce a grammatical utterance , I draw upon l: IHlbling. This, of course , does not prevent the- structured
the same sym actical rules as those that utterance helps to produ ce. properties of social systems from stretching away, in time and
But I speak th e 'same' language as the other speakers in my ~ p:l ce, beyond the control o f any individual actors. Nor does it
language co mmunity; we all share the same rules and lingui stic compro mise the possibility that actors' o wn theo ries of the social
practices, give or tak e a range o f relatively minor variations. Such s y ~ t l: m s which they help to constitute and reconstitute in their
is not nccess .lfil y the case with th e structural properties of social activities may reify those systems. The re ifi ca tion of social '
systems in general. But this is not a problem to do with th e I'c llliions. or th e discursive 'nat uraliza tion' o f th e historically
26 Elements of th e Th eory of Slructuration The Dua li ty of Structure 27

cont ingent circumstances and products of human actio n, is o ne continuities in social reproduction across time-space. It in turn
of the main dimensions of ideology in social life.27 presupposes the reflexive mo nito ring of agents in , and as
Even the crudest fo rms of reified tho ught. however, leave co nstituting, the duree of daily social activity. But human
unto uched the fund amental significance of the knowledgea bility knowledgeability is always bound ed. T he flow of action
of buman actors. Fo r knowledgeability is fo unded less upon continually produce~ ~on~q!I~nces ~hich~:..~ uninteijOeal?t
discursive than practical conscio usness. The knowledge of socia l actors. ana these un intended conseq ue nces a lso may fo rm
conventio ns, of o neself and of ot her human beings , presumed in unacknowledged conditjo ns of action in a feed back fashion .
being able (Q 'go o n' in th e diversity of con texts of social life is Human- history is created by inte ntio nal actiVi ties but is no t an
detailed and dazzling. All compete nt members of society are intended project; it persistently e ludes eHorlS to bring it under
vastly skilled in the prac tical accomplishments of social acti vities conscious direction. However, such attempts are continually made
and are ex pert 'socio log ists' . The knowledge they possess is no t by human beings , who o perate unde r the threat and the pro mise
inciden tal to the persistent patterni ng of social life but is integral of the circumstance th at they are the o nly creatures who make
to it. This stress is absolute ly essenti al if the mistakes of their 'history' in cognizance of that fact.
functionalism and structuralism are to be avoided, mistakes which. The theorizing of human beings abo ul their actio n means that
~uppressrng rud lsCoun fi ng ae nts' reasons - th e rationalizatio n just as social theory was not a n inventio n of professional social
of action as chronically invorvecrmtne structuration of soc ial theorists, so the ideas produ ced by those theorists inevitably tend
practices - look for the origins of tlJeir activities in phenomena to be fed back into social life itself. One aspect of this is the
of which these agen ts are i gn ora nt. l~/But it is equally important to attempt to monitor , and th ere by contro l, highly generalized
avoid tumblin.s into,the op~s ing error of hermeneutic approaches co nditions of system reprodu ctio n - a phenomeno n of massive
and of variou ~ versl0l"!s ~f eh eno ~e no}0ID:.., whiChJ~_n d . to r.egard importance in the contemporary world . To grasp such monitored
society as the plastic c.reation_of hJl_~an sub.i!cts. Each of th ese is processes of reJ~roducti on conceptually, we have to make certain
an ill egi tim ate fo rm of reduction , deriving from a failure di sti nctio ns relevarit -towhat sOcial systems 'are' as rePIoduced
adequately to conceptualize th e duality of structure. According practices in interaction settings. The relations im pl ied or
to structuration theory, the moment of the produ ction of ac tio n is actualized in sociafSysternsare. of course, widely variabl e in
also o ne of r~ rodu ction ~__ in e- contexts of the day-to-day lerms of the ir degree of 'looseness' and permeability. But, this
ena-ciiTie n t of social life. This isw even dunng the most vlolem heing accepted, we can recogn ize two levels in respect of the
upheavafs or most rad ical fo rms of social change. It is no t accurate mea ns whe reby some eleme nt o f 'syste m ness' is achieved in
to see th e structural properties of social systems as 'social int eraction. One is that generally prominent in functio nalism, as
produclS' because this te nds to imply that pre-constituted actors refe rred to earlier, where inte rde pendence is conl2e ive_d of_as a
somehow come together to create them.1'I In repro duc ing ho meostatic process akin to mechanisms- of self-regul ation
structu ral properties to repea l a phrase used earli er. agents also tl perating within an organism. There can be no objection to this
reproduce the co nditio ns (h at make such actio n possible. as lo ng as it is acknowledged that the 'looseness' of most social "... /,...:-
Structure has no ex istence independent of the know ledge that .-;yste ms makes tiLe o rganic parall el a very remote one and that
agents have about what they 00 in their day-to-day activity . I his re latively 'm ec hanized' mode of system reprod uctio n is pot
Hum an age nts -ahvays-know what th-e-y are doing on the level of I he o nly o ne fo und in human societies. Ho meostatic system
disc ursive co nsciousness under some description . However, what reprodu ction in human society can be regarded as involving the
th ey do may be quite unfa miliar under other descripti ons, and o peration of ca usal loops. in whi c h a range o f unintended
they may know little of th e ra mified consequ ences of th e activiti es co nsequences of act io n feed back to reconstitute th e initiating
in which th ey engage. circu mstances. But in many co ntexts of soc ial life there occur
Th e dua lit y of stru c ture is a lways th e main gro undin g o f proces>es of se lective 'info rmatio n filt ering' whe reby strat egically
28 Elemen ts of rh e Theory of St ruc rvration Fo rm s of Institution 29

pl aced actors seek refl exively to regu late the overall condit io ns o f de ments and power. Th is is most eviden t in the no t infrequent
system reproduction eith er to keep things as they are or to co ntexts o f social 1ife where what social phenomena 'are', ho w
change [hem ,Xl they are ap tly described , is contested . Awareness o f such
T he distinctio n between homeostatic causaUoops rind refl exive co ntestatio n, of divergent and overlapping characterizatio ns of
self-regulatio n in system reprOd uction must be co mplemented by activity, is an essential part o f 'knowing a fo rm of life', although
o ne fu n he r, and fin al, o ne: tha t between social a nd ~ste m this is no t made clear in the writings o f a utho rs s uch as Winch ,
i nt~grali o n ,JI 'Integratio n' may be understood as in volving who treat forms o f life as both unified a nd co nsensuaJ.l5
~roci t1 o.!p ractices (o f a uto no my and dependence) between
actors or collectivities,J! Social integratio n then means systemness
o n the leve,l o f f~~~~ interactio n, System integra tion r.ci~rs sign ification (- _ - -~ domi na rio n 1<- - --~ leg it imatio n I
to co nnections wltli those who are physically absent in time o r
space, T he mechanisms of system integration certainly presuppose (,,,,xJalilyl
,- .--- -- -.--.,
, rnt.... prel allve ,
: ,c herne :
!- -f;ci llt; - i
,------ -,
,-- -- -,
those of social in teg ratio n. but such mechanisms are also d istinct '- -- -;,;; - - -- ' l' 1-
, , ,
in some key respects fro m those in volved in relario ns o f co-
prese nce , 1"'<"dOlon <:tJrnrntJn ic~tior1 -E - - --)[ po:'er 1<---- --)[ san;;io n I
Social I n/(~!lra l i()n 5v5 t,' m integ,a tion

Ik.: ip roc ity bctwl<l;tn actor s in Re c iproc it v h~twee n actors o r (o ll.. c tivi ti t S Figure 2
co n lcxt ~ of co-p ,*~se nct acro eXlcnd ... d tim l< -span'

The d imensions o f th e d uali ty o f st ructure are portrayed in

Fig ure 2,16 Hu man actors are not o nly able to moni tor their
Fo rm s of In stitutio n activiti es and those o f o thers in th e regulari ty of day-to-day
cu ndu ct ; they a re also able to 'monitor that monitoring' in
T he d ivisio n o f r ules into modes o f signifying o r mea ning discursive conscio usness. 'Interpre tative schemes' are the modes
co nstitutio n a nd no rmative sanc tio ns, together with the concept o f typifica tio n incorporated within actors' slocks of knowledge,
o f resources - fundamental to the conceptualizatio n o f power _ applied refl exively in the sustaini ng o f communica tion. T he stocks
carries va rio us implicatio ns whic h need to be spe lled o uLJJ What o f kno wledge which ac tors d raw upon in the prod uc tion and
I call the 'modalities' o f struc turation serve to clarify the main r..: productio n o f in terac tio n are the same as those where by they
dimensio ns of the d uality o f struc ture in interactio n, rela ting the afe ~I b l e to make acco unts, o ffer reasons, etc.37 The communi-
knowledgeable ca paci ties o f agents to structu ral features. Ac to rs l:a l io n o f meaning, as wi th all aspects of the co ntextuality of
draw upo n the modalities of struc turation in the reproductio n o f a": lio n. does not have to be seen mere ly as happening 'in' time-
systems of interac tion, by the same token reco nstituti ng their spaCl:. Agents ro uti nely inco rporate temporal a nd spatial featu res
structural propert ies. T he communicatio n of meaning in inter- Il l' e ncounters in processes o f meaning co nsti tution. Communi-

action, it sho uld be stressed , is se parable o nly analytically fro m L'<I!i o n. as a ge neral element o f interactio n, is a more inclusive
~ h e opera tio n o f normative sanctio ns, T his is obvio us, for example , ~ o I H.;e pt tha n commun ica tive intent (i. e, what an actor 'means' to
In so far as language use is itself sanctio ned by th e very nature of ~ a y o r do). Th ere are o nce mo re two fo rms o f red uc ti onism to be
its 'public' charac ter. 34 T he very ide ntification of ac ts o r of aspects avo ided he re. So me phil osop he rs have tried to deri ve overall
o f int erac tio n - their acc urate descriptio n. as gro un ded !heuries o f mea nin g or co mmuni ca tio n fro m communicative
hermeneutica lly in the capabi li ty of an o bserver to 'go o n' in a ill le n!; o th ers. by co nl raSl . have supposed that communicati ve
fo rm o f li fe - implies the interlaci ng o f meaning. no rma tive ilil e n! is at best marginal 10 th e constit uti o n o f th e mea ningful
30 Elements of the Th eory of Str uctura tio n Forms o f Ins titutio n 31

qualities of interaction, 'meaning' being governed by the structural the analysis of these stru ctural properties are indicated in th e
ord ering of sign systems. In th e th eory of structuration. however , table below. The th eory of coding presum ed in the study of
these are regarded as of equivalent interest and importance, structures of significatio n must look to the extraordinary advances
aspects of a duality rather than a mutually exclusive dualism . in semiotics which have been pio neered in recent decades. At th e
The idea of 'accountability' in everyday English gives cogent same time we have to guard against the associatio n of semio tics
expressio n to the intersection of interpretative schemes and with struc tuntlism and with the sho rtcomings of the latte r in
nonns. T o be 'accountable' fo r o ne's activities is both to explicate respect of the analysis of huma n agency. Signs 'exist' o nly as the
the reasons fo r the m and to supply the nonna tive gro unds medium and o utcome o f communica tive processes in interactio n.
where by they may be 'justified'. Normative compo nents of Structuralist co nceptio ns of language , in commo n with similar
interactio n always centre upo n relations between the rights and di sc ussio ns of legitimatio n , tend to ta ke signs as the given
obligations 'expected' of those participating in a range of pro perties of speaking and wri ting rather tha n examining their
inte ractio n contex ts. Fo rmal codes of conduct , as, fo r example , recursive grounding in th e communicatio n of meaning.
those enshrined in law (in contemporary societies at least), usually
express some so rt of claimed symmetry between rights and
obligations , th e o ne being the justification of the other. But no ~t rtl c ture{5) Th eoreticil /Oom-ain IM !i!utiona I Order

such symmetry necessarily exists in practice, a phenomenon which

\ ll(n ifiratio n Theory of c odin g Sy mtmlic f1fd" rS/"'od..,~ of d isco u"e
it is important to emphasize, since both the 'normative
functionalism ' of Parsons and the 'structuralist Marxism' of Theory of re ~ urCe autho ri ntion Po litical in stitution>
I),,,,,ind tion
Theo ry of res.o urce allocatio n Eco no m ic i n s titut i on ~
Althusser exaggerates th e degree to which normative obligations
are 'internalized' by the members of societies ..'l!! Neither standpoint l"l(i t imatio n Theo ry of normati "'l regulation L!~l!al i n stitution ~

incorporates a theory of action which recognizes human beings

as kno wledgeabl e agen ts, refl ex ively monitoring the fl o w of
interaction wi th one another. When social systems are conceived St ru ctures of significatio n always have to be grasped in
of primarily from the point of view of the 'social object', th e co nnectio n with do minatio n and legitimatio n. Once more this
emphasis comes to be placed upon the pervasive influence of a hears upo n the pervasive influe nce of power in social life. There
no rmatively e<rordina ted legitimate o rder as an overall de ter ;tre certain positions which have to be carefull y skirted here.
minant or 'programmer' of social conduct. Such a perspecti ve T hus some re levant issues have been brought to the fore by
masks the fact that the no rmative elements of social systems are I-Iahermas's critique of Gadamer and ensuing debates.:l9 Amo ng
co ntingent clai ms which have to be sustained and 'made to count' o lh er things, Habe rm as f ritic ized Gadam.er's_ con~e~ion_~ f
through the effective mo bilizatio n of sanctions in the contexts of linguistically saturated 'traditions' for fa iling to demonstrate that
actual e ncounte rs. No rmative sanctions express structural fram es of me~ftinglncorp_o~~!-.e _c!i (fer~ ntral~_~r power.- T he
asymmetries of dominatio n, and the relations of those 1l0 mina liy I.:rili cism is valid eno ugh, but Habe rmas sought to develop the
subj ect to them may be of vario us sorts other than expressions of point in th e directio n of showing the significance of 'systematically
the commitme nts those no rms supposedly engender. di slO rted' fo rms of communi catio n. He has not been able on this
Concentration upon the analysis of the structural properties of hasis, ho wever, satisfa cto rily to integrate the concept of power
social systems, it sho uld be stressed , is a valid procedure only if it with an institutional theo ry. 'Do minati on' is not the same as
is recognized as pl acing an epocht upon - holding in suspension 'syste matically disto rt ed' stru ctures of signifi cation because
- refl exively mo nito red social conduct. Under such an epoche do minatio n - as I co nceive of j( - is th e very condition of
we may distinguish three stru ctural dimensions of social systems: e xiste nce of codes of signifi ca tio n.o4I1 'Dominatio n' and 'power'
signifi catio n, domin atio n and legitimatio n. The conno tatio ns of canno t be tho ught of only in te rms of <Isy mme tries of distri bution
32 Element5 of the Theory 0 1 SrwCWrafion Forms of Institution 33

b ut have to be recognized as inh erent in social association (or, 1 signs; they conj oin those intersectio.ns of c<:>d~s whi ch ~re
would say, in human action as such). Thus - and here we must especially rich in diverse fo nns of meanmg assocI3u<:m , operat ing
also reckon with the implicatio ns of the writings o f Fouca ult - alo ng th e axes of metapho r and metony~y ..Sy":,bo~lc orders and
power is not an inh erently noxious ph enomenon, not just the 'Issociated mod es of discourse are a maj or InSlitUl lOnal locus of
capacity to 'say no'; nor can dom ination be 'transcended' in some ideology. However, in the theory of structuration i~eology is not
kind of putative society o f the future , as has been the a partic ular 'type' of symboli~ ord~r o r for~ o f dl~~u rse., O ne
characteristic aspirati on of at least some strands of socialist ~annot separate off 'ideologIcal dIscourse from sClenc.e, for
thought. IJxa mple. 'Ideo logy' refers o nly to those asy ~~etfL.es of
What are the conno tatio ns of the claim that the sema ntic has dom in at io n which connect signification to the legItimatio n of
priority over the semiotic rather than vice ve~? Th ey ca n be sectional in terests.4J
spell ed out, I think, through a comparison of structuralist and We can see fro m the case o f ideo logy that structures of
post-structural ist conceptions o f mean ing on the o ne hand , and significatio n are separable only analytically either from do~~nat~on
that which can be deri ved from the late r Wittgenstein o n the ,Ind fro m legitimation. Domination depends upon th~ moblh zatlo ~
other." The foundat ion of a theo ry of meaning in 'difference' in Il f two distingu ishable types of resource. ~ llo cat lve reso urces
which. following Saussure , there are no 'positive values' leads refer to capabilities - or, mo re accurately, 10 fon~s of
almost inevitably to a view accentuating the primacy of the I ransfo rmative capacity - gen ~rating co~ n4.. ove... oblects ,
semi otic. The fi eld o f signs, th e grids of meaning, are created by gllods o r materi al p.E~nomena. ;Autho!itativ.e resou rces refer to
the ordered nature of d ifferences which comprise ('.odes. Th e Iypes o f trans formative capacity generaung co mmand o ver
'retreat into the code' - whence it is difficult o r impossible to re- persons o r '!.ctors. Some f~rms o f alloca tive ~esourc.es (suc,~ as
emerge into the world of activity an d even t - is a characteri stic raw mat erials, land, etc. ) might seem to have a real eXistence m a
tactic adopted by structuralist and post-structuralist authors. Such way which I have claimed that structural properties as a whole ~o
a retreat, however, is not necessary at all if we understand the nol. In the sense o f having a time-space 'presen ce', in 3 certam
relational character of the codes th at generate meaning to be way such is obviously the case. But their 'materia lity' do~s not
located in the o rdering of social practices, in th e very capacity to 'lHect th e fact that such phen omena become resources, m the
'go on' in the multiplic ity o f contexts of soc ial activity. This is a lIIallllt;!r in whic h I apply that term he re , o nly whe~ incorporated
discovery which Wittgenstein himself surely made, albeit against wit hin processes o f structuration. The transform~tl on~1 c haracter
a very different philosophical backdrop, when he abandoned i. 1" reso urces is logically equival en t to, as well as mh erentl y bou.nd
some o f the main parameters of his earl y writings. Wh ereas his lip wi th the instant iat io n _of , that of codes and normative
earlier analysis o f language and meaning terminates in paradox - salle tions.
a sorl of Ind ian rope trick , pulling up the ladder after it has been T he classificatio n of in stitutional orders offere d above depends
climbed - hi s later view hugs the gro und of routine social llpllll resisting what has sometimes been ~aU~d ':ubstantivist'
practices. Even the most complicated semiotic relatio ns have a ...o llcepts o f 'economic', 'political' and o the r institu tio ns. We can
grounding in the semanric properties generated by th e ru le- cOllceive of th e relationships involved as fo llows:
governed properties o f daily acti vities .
In the terminology indicated in the table above the 'signs' SymboliC orderlo/modc) of d ; ~ our!o\:

implied in 'signification' sho uld not be eq uated with 'symbols' . Politic .. ! u.,t itution ,
Many writers treat th e two terms as eq uivalent, but I regard
l) (. 011, ~ ) <;1 (e" " om;, ""tit ut "J '"
symbols , interpolated wi thin symbolic orde rs, as one main
dimensio n o f the 'cl ustering' of institutio ns. 42 Symbols coagulate I \) ~

the 'surpluses of mean ing' impli ed in the polyva lent charac ter of wI,,"'. ~ " .:",1" .<1"" ,. I) d"",,,,.,I',,". I '" 1"-.:it ,,,,.I1"'"
34 Elements of the Theory of Struc turation Tim e, rhe Body, Encounte rs 35

'S ubstantivisr' conceptio ns presume concrete institutional ba nal and evident feature of human day-to-day life. It is in some
d iffere ntiation of these various o rders. That is to say, it is held , pan the lac k of 'fit ' between o ur unproblematic coping with the
for example, that 'politics' exists o nly in societies having distinct co ntinuity of conduct across time-space, and its ineffable
fo rms o f state apparatus and so on. But the work o f character when confro nted philosophically, that is the very
anthropologists demonstrates effectively enough that there are essence of the puzzling nature of time. I make no particular claim
'politi cal' phenomena - to do with the ordering of authority 10 elucidate this ma tter, 'S t Augustine's problem'. But the
relations - in all societi es. The same applies to th e o ther fundamental questi on of social th eory , as 1 see it - the 'problem
instituti o nal orders. We have to be partic ularly careful in o f order' conceived of in a way quite alien to Parsons's formulation
conceptualizing the 'economic', e ven having made th e point that when he coined the phrase - is to explicate how the limitatio ns
this does not presuppose the ex istence of a clearly d ifferen tiated nf individ ual 'presence' are transcended by th e 'stretching' o f
'econo my'. T here has been a stro ng tendency in some of the social re lations across time and space.
lite rature o f economics to 'read back ' into traditional cultures T he duree o f daily life, it is not too fanciful to say, operates in
concepts that have meaning o nl y in the contex t o f ma rke t so me thing akin to what Levi-Stra uss calls 'reversible time'.
economies. T he 'economic' cannot properly be defined, in a Wheth er or not time 'as such' (whatever th a t wo uld be) is
generic way at least, as concerning struggles for scarce resources.'" reversible, the even ts and routines of daily life do no t have a o ne-
T his is somewhat like defining power solely by reference to way fl ow to them. T he terms 'social reproductio n' , 'recursiveness'
sectional struggles. It is not scarcity of resources as such . far less and so on indicate th e repetitive character o f day-to-day life, the
struggles o r sect ional divisions centred upon distri but ion, tha t is rllutines o f whi ch are fo rmed in terms of the intersection o f the
th e main feature o f th e 'econo mi c'. Rather, the sphere of the
'economic' is given by the inhe rently constitutive role of allocative
resources in th e struc turation o f socie tal to ta lities. O the r Iii.. Spiln nl the individual , 'irre" e' sible tim."
ca utio nary no tes sho uld be added here. If it is held that all
/ongll" d",i'" of in~t itu' ,ons; 'f(>Vero; ible l ime'
socie ties are haunted by the possibilit y o f material scarci ty, it is
o nly a short step to the suppositio n that conflicts over scarce
resources make up the fundamental motor of social change, as is pass mg (but continually returning) days and seasons. Daily life
presumed in at least some versions of historical materialism and has a duration , a fl ow, bu t it does not lead anywh ere; the very
in many no n-Marxist theories also. But this presumptio n is both :ldjcctive 'day-to-day' and its synonyms indi cate that time here is
logically wanting, usually depending upon a specious form of constituted o nly in repeti tion. The life o f the individual, by
functiona l reaso ning, and empirically false .4~ con lrast. is no t o nly finite bu t irreversible , 'bei ng towards death'.
T his is death, to die and know it. This is th e Black Widow, death '
Time, the Bod y, Encounte rs 'Lowell ). Time in this case is th e time o f the body. a frontier o f
presence quite diffe re nt fro m the evaporatio n o f time-space
In concluding this abbreviated opening exposition , we may retu rn illherenl in the duratio n o f day-to-day activity. Our lives 'pass
to the theme of time and hi story. As the finitude o f Dasein and as away' in irreversible t.ime with the passing away o f the life o f the
'the infinity of the emergence of be ing from nothingness', time is orgalli sm. T he fa ct t.hat we speak of the 'life cycle' implies that
perh aps the most enigmatic feature o f human experience, Not for IlIc rc are elements of repe titio n here too. But the life cycle is
no thin g (sic) was that philosoph er who has attempted to grapple rC:l ll y a concept that belo ngs to the successio n o f ge nerations and
in the Ill os t fundamenta l way with the problem , He idegger, Illu s 10 the thi rd dime nsio n o f tempo rali ty ind ica ted above. This
compell ed to use termino logy of the most daunLing obscurity. But is Ih e 's upra-indi vidual' duree of the lo ng-te rm ex iste nce of
time . o r Ihe constitutio n o f expe rie nce in time-space , is also a illslihllio ns, Ihe }Of/RIW ell/rae of institutio nal time.
36 Elements 01 the Theory of Structuration References 37

The reversible time of institutions is both the condition and th e have made th eir own ind epend ent contributions. Thu s 1 shall
o utcome o f the practices organized in th e continuity of daily life , pro pose not only that the time-geograph y o f Hiigerstrand (with
the main substantive form o f the duality of structure . It would not appropriate critical emend atio n) offers forms of analysis of
be true , however, as I have already men tioned , to say tha t th e signifi ca nce for structuratio n (heory but a lso tha t some o f the
routines o f daily life are th e 'foundati o n' upon which institu tio nal ideas in volved comple me nt Goffman 's conceptio ns rather
forms of societal orga nization are built in time-space . Rather, direc tly.
each enters into the constitution of the other, as they both do into Relatio ns with those who are physically absent, as I have said ,
the constitution of the acting self. A ll social systems, no matter invo lve social mechanisms d istinct fro m what is invo lved in
how grand or far-flun g, both exp ress and are expressed in th e contex ts o f co-prese nce. We have to deal he re with some basic
ro utines o f daily social life, mediating the physical and sensory questio ns about the structuring o f institutio ns. T hese have a
prope rti es of the human body. '!ateral' aspect to the m - particularly in th e modern wo rld , given
These considerations are of very considerable impo rtan ce for th c tremendo us expansion of the time-space distan ciatio n o f
the ideas se t o ut in th e succeeding parts o f this book. The body is social activity in the contemporary era. But they also raise once
the 'locus' of the ac ti ve self , but the self is obvio usly not just an mo re the proble m of 'history' , since the absen t others include
ex tensio n of the physical charac teristics o f the organ ism that is its past ge neratio ns whose 'time' may be very differe nt from that o f
'ca rrier'. Th eorizing th e self means fo rmulating a conception o f 1hose who are in so me way influen ced by residues of their
mo tivatio n (o r so T shall argue) and relating mo tivation to the ;lc ti viti es. These ma tters will be my concern in the co ncluding
co nnections be tween unconscious and conscious q ualities of the dwptcrs.
age nr. The self ca nno t be unde rstood o utside 'histo ry' - 'history'
meaning in this case the tempo ralit y o f hum an practices.
expressed in the mutual interpolat ion o f the th ree dimensions I I{pfe re nces
have distinguished.
ror more detai led discussions of the basic concepts of structuration
I ea rli er introduced the notion of co-presence, with speci fi c
theory, the reader should turn to NRSM. especially chapters 2 and
reference to social integrati o n. The study o f in teractio n in 1: CPST: and CCHM. chapters 1 and 2.
circ umstances o f co'presence is o ne basic component o f the 2 CP5T pp.56- 7.
'bracketing' of time-space that is both condition and ou tcome o f .' CPST. chapter l.
human social associatio n. 'S ystem ness' here is achieved large ly ,I Don ald Davidson, 'Agency', in Essays on Actions and Events
through the ro utine reflexive mo nitoring o f conduct a nchored in 10xford: Clarendon Press, 1980), p. 45.
prac tical consciousness. Relatio ns in conditio ns o f co-presence ." NRSM. chapter 2.
consist of what Goffman has aptly called encounters, fading away h klcl Feinberg, 'Action and responsibility', in Max Black, Philosophy
across time and space. No one has analysed encounters more ill America (hhaca: Cornell University Press, 1965) . On the problem
\If wha1 are 'co nsequen ces" see Lars Bergstrom, Th e Alternatives
perceptively than Goffma n himself , and I shall draw heavily upo n
his work in part of what fo llows. The importan ce o f Goffma n's lIlId Consequences of A ction.l (S tockholm: Almqvist, 1966).
7 Thomas Sc helling. 'On the ecology of micromotives', Th e Public
work is due in no small degree to his preoccupat io n with the
Il/lere.~/. vol. 25, 197 1: 'Dy nam ic models of segregation', Journal of
temporal and spatial o rdering of social acti vity. He is o ne of the Mathematical Sociology. vol. 4, 1971. See also th e discussion in
few sociological wri ters who treat time-space relation s as H:lymond Bo udon, The Vnillt ended COl/sequences of Social Action
fundamental to the prod uction and rep roduction of social life, (Lo ndon: M'lcmillan. 1982), pp. 43ff.
ra ther than as making up 'bo undaries' to soc ial act ivity which can K NRSM. p. 76.
be safely leh to 'specialists' - geographers and histo rians. Bu t '1 Me ri OIl. however, fa vours the term. 'unllnlicipated' rather than
those working in the no min all y separate subject area of geogm phy IIl1irll ended consequ ences. In my ;lIIillysis 'in tention' presumes
38 Reference . References 39

knowledge of the lik ely consequ ences of action and therefore 'structure' from 'stroctures' and used the latt er term too casually as
anticipation. Of course, one can anticipate that something will synonymous with the former.
happen without int ending it to happen, but o ne cannot intend 27 CPS T. pp. 195-6.
something to happ en without anticipating that it might happen. 28 Cf. Roy Bhaskar. The Possibility 0/ Naturalism (Brighton: Harvester,
R. K. Merton, 'The unanticipated consequences of purposive social t 979), chapter 2.
action', Amen"call Sociological Review. vol. I, 1936; idem, 'Manifest 29 Ibid. , p.48.
and latent fun ctio ns', in Social Theory and Social Structure 30 Cr. ibid . pp. 78-9. There I distingoished three levels of 'systemness'
(Glencoe: Free Press, 1963). which here. for purposes o f sim plification, are reduced to two.
IO Merton. 'Manifest a nd latent fun ctions', p. 51. 31 This distinc tion was introduced into t he literature by David
II Ibid., pp. 64- 5. Lockwood who, however, employed it rather d ifferently from the
12 For a full er discussion, see CPS T. chapter 6. way I do : David Lockwood, 'Social integrat io n a nd system
D Max Weber, The Methodology 0/ the Social Sciences (G lencoe: integration', in George Z. Zollsc han and W. Hirsch. up/orations in
Free Press, 1949). Social Change (London: Routledge, 1964).
14 Mancur Olson, The Logic a/Collective Action. (Cambridge, Mass.: .12 My for mulation of the concept of 'system integration' in CPST,
Harvard Un iversity Press, 1965); Boudon, The Unintended p. 77. was ambiguous. I did not make it clear whether the separation
Consequences 0/ Social A ction; Jon Elster, Logic and Society, o f social from system integration depended upon a distinction
Contradictions and Possible Worldi' (Chichester: Wiley, 1978); Jo n between co-presence and absence in social relations, or between the
Elster, Ulysse.s and the Sirens (Cambridge: Cambridge University ties linking actors as comrasted with those linking collectivities. As
Press, 1979). I use it now, the notion refers to the first of these two sets of
15 Baudon, The Unintended Consequences 0/ Social A ction, contrasts, but they are in a ny case closely overlapping, so th e faulL
chapter 2. was not too consequential.
16 For a fu rther development of this point, see 'Power, the dialectic of 1.\ CPST, chapter 2.
control and class strUCluration', in Anthony Giddens and Gavin .14 Cf. Paul Ziff, Semantic Analysis (Ithaca: Cornell University Press,
Mackenzie, Social Class and Th e Division 0/ Labour (Cambridge: 19(0) .
Cambridge University Press, 1982). .l) Cf. Hanna F. Pitkin , Wittgenstein and Justice ( Berk eley: University
17 Peter Bachrach and Morton S. Baratz, 'The two faces o f power', o f California Press. 1972), pp. 24 1- 64 .
American Political Science Revie w. vol. 56, 1962 ; Power and .lh For this style o f re presenting these relations 1 am indebted to Derek
Poverty (New York : O xford University Press, 1970) ; Steven Lukes. G regory ; see his Regional Trans/ormation and Industrial Revolution
Power, a Radical View (London: MacmiJlan, 1974). Fo r further (London: Macmillan, 1982), p. 17 .
discussio n of these points, cf. CPST. pp. 88-94. .\7 Peter Marsh et 01., The Rules 0/ Disorder (Londo n: Ro utledge,
18 Jo hn R. Searle. Speech A cts (Cambridge: Cambridge University 1978), p. 15 and passim.
Press, 1%9), pp. 34- 5. .lX NRSM. pp. IOS - IO.
19 Ludwig Wittgenste in, Philosoph ical Investigations (Ox fo rd : .N Jtirgen Habermas. Zur Logik der Sozialwissenscha/ten (Ttibingen;
Blackwell, 1972), p. 59. Siebeck & Mohr, 1967); 'On systematically distorted communica-
20 Ibid . p. 81. tion', Inq uiry, vol. 13, 1970.
21 Ibid. '- It) C f. my ' Habermas's critique o f he rmeneutics', in S5PT.
22 Ibid. 'II See CPST, pp. 33- 8.
23 CPST. pp. SOff. 12. Paul Ricoeur. 'Existence and hermeneutics', in The Conflict 0/
24 Harold Garfinkel, 'A conception o f, and experiments with, "trust" /flf erpretations (Evansto n: Nort hwestern University Press, 1974).
as a cond ition o f stable concerted actions'. in O. J. Harvey, '11 For an elaboralion of this posilion, see CPS T. chapter 5. Symbolic
Motivation and Social Int eraction (New York: Ronald Press. 1963). orders and modes of discourse const itute the 'cultural' aspects of
25 Erving Goffm an. Frame Analysis (New York: Harper, 1974), p. 5. soc ial syste ms. But. as with 'society' and 'h istory', I call upon the
26 In NRSM I had not appreci ated th e need to distinguis h teflll 'culture' to fulfil a double dUly. Thus I s hall speak o f 'cultures'
40 References

~n a general way. as a term interchangeable with 'societies'. although

In some co ntex t~ these terms have ( 0 be accorded mo re precision.
Cf. Karl PoianYI et al. , Trade and Market in the Early Empires
(New York : Free Press, 1957) , pp. 243- 70 and passim.
My reasons f~r m~ing ~hese claims are given at some length in
Consciousness, Self and
CCHM, especially In the In trod uction and in chapter J.
Social Encounters

I n this ch apter I shall seek to fulfil several o bjectives. First of all. I

shall discuss some basic concept ual pro blems posed by connecting
the main concepts o f structuratio n theory to an interpretation o f
the nature of the unco nscious. This turns upon q uestio ns o f how
th e sel f, especially the 'I' of the re flexive agent, shou ld best be
conceptualized. 1 shall then move o n to a portrayal of how the
psychological found atio ns of the interweaving of conscious and
unconscio us can be represented , utilizing in particular the writings
o f Erikson. But it will be a majo r part o f my argument that s uch a
po rtrayal immediately raises questio ns of a social nature to do
with the routinized c haracter of d ay-to-day life. Via an analysis of
'critical situations', in which ro utines are radically d isru pted, I
sha ll try to indicate how the reflexive mo nitoring of encounters in
circumstances o f co-prese nce ordinarily co-ordinates with
unco nscious compo nents of personality. This will lead direc tly
Ihro ugh to an examina tio n of some o f the insights which ca n be
draw n fro m Goffman about inte rac ti on between co-prese nt
age nts. Concern with the body , as the loc us of the acting self and
, IS positioned in tim e-space, is th e key linking theme of the
mat erial discussed and analysed.

Reflex ivity, Discursive and Practical Consciousness

Freud d ivides th e psychic o rgan izatio n of the individual into

thn:c, divisions represe nted in English by the unfo rtunate terms
' id ', 'ego' and 'super-ego'. I do no t beli eve these terms are
parti cularly useful and shall in stead substitut e th e threefo ld
di visio n s ugges ted in the s tratificat io n mo de l: basic security
system , practi ca l and discursive conscio usness. I do no t mean
42 Consciousness, Self and Socia l Encoun ters Reflexivity, Oiscursive and Prac t ica l Consciousness 43

these to parallel the Freudian notio ns directly . The intersecting Freud , of course, does himself speak. of hydraulic fl ows , blockages
planes of [he interpretative schemes and norms which ac tors o f energy and so o n. But these then co njure up the sort of
utilire in the constitutio n o f their conduct are embedded in all mechanical conceptio n o f the origins o f human conduct
three dimensions of personality. But certainly the 'I' (das Ich ) is at associated with th e most naive forms of objectivism . Part of th e
th e core o f what is invo lved in discursive consciousness and pro blem is the use o f the terms ego, super-ego and id (whether in
demands considerable attentio n conceptually. We can approach [h eir original German formulation or in their English version),
the issues involved by tracing some of the difficulti es posed by each of which has some connotation of agency ; each is a mini-
Freud's division of the person ality , especially in so far as these agent within the age nt as such. Discarding the terms 'id' and
bear upo n problems of agencyY" 's uper-ego' helps , but this has to be complemented by recognition
Freud , of course , regarded the individual as an agent but also o f the distinctive charac ter of das lch. the '1'.
o ften spoke of the id , ego and super-ego as agencies within the We might suppose tha t the T is the agent. However, this is
individual. In his writings prio r to the f920s Freud freque mly .'i urely mistake n, even tho ugh it figur es as the central assumption
used the term das l ch to refer to the whole person, as well as to o r pro position o f who le schools o f philosophy, including
designate a part of the mind . These shifts of usage also apply to Carlesianism and the latter-day philosophy o f G. H. Mead. Mead's
'super-ego' , sometimes differentiated from another no tio n, that of writings certainly help to elucidate th e processes leading to the
'ego-ideal '. Terminological in consistencies and transitio ns seem l! mergence of a 'self' as a 'me '. But th e 'I' appears in Mead's
to indi cat e here some rather more significant conceptual tro ubles. writings as the given co re o f agency, and its o rigins hence always
Suppose das lch is a subdivision o f mind. How can Fre ud then say re main obscure. T o relate th e 'T' to agency, it is necessary to
such things as that the ego 'dec ides on the repudiatio n o f the fo llo w the detour s uggested by structuralists in respect o f the
incompatible idea"!2 Is th e ego's deciding some sort o f process in dece nt ring of th e subject. without reaching concl usio ns which
miniature o f the agent's deciding? This, surely, does nOI mak e IrCal the subject simply as a sign within a significatio n structure .
much sense. Freud also writes. for example, of the ego's 'wish to T he constitution o f the T comes about only via the 'discourse of
sleep', although while sleep occurs it 'stays on duty' to pro tect Ihe Other' - that is, thro ugh the acquisition o f language - but
against the worst emanations of th e unconscious, 'guarding' the Ihe T has to be related to the body as the sphere o f action. The
sleep o f the d reamer. Th e sam e sort of questions arise. Whose h.:rm T is in linguistic terms a 'shifter' : the contextuality of social
sl eep is it that the ego desires? Th e agent's'! Its own '! Whose pos itioning' determin es who is an 'I' in any situat ion of talk.
waking does the 'guard ' protect '? And so on. Consider, finally, Allho ugh we might tend to think of T as bearing upon the richest
Freud's most general characterizatio n of the tasks o f the ego. T he :lIld most intimate aspects of our experience, it is in a way one of
ego has the task qf 'self-preservation ', which it executes 'by Ihe emptiest terms in language .4 For the 'I' re fers o nly to who is
learning to bring about changes in the external wo rld to its o wn speaking, the 'subject' o f a sentence or utterance. An agent w.ho
advantage'.J But which 'self does th e ego de fend ? Is its ad vantage has maste red the use o f T , as Mead says, has also mastered the
also my advantage? II.'iC o f 'me' - but o nly via concomitant mastery o f a syntactically
Now one traditional tactic amo ng interpreters o f Freud is ( 0 d iffe rentiated language. Fo r I have to know tha t I am an T when
accept that there are misl eading anthropomo rphi c usages in I .'i peak to 'you', but thai yo u are an 'I' when you speak to 'me',
Freud's writings , but to claim th at these can be dispell ed if we a ll d that I am a 'you' wh en you speak to me .. . . and so o n. The
understand id, ego and super-ego as referring to 'processes' or pl )inl is not just thai these usages presume lin gui stic skills of a
'fo rces'. But this is not really very mu ch help, for su ch concepts ve ry complicated kind but al so that they entail a ramified control
do no t allow us properly to grasp th e nature of human agency. t Il' [h e body and a d evelo ped knowledge of how to 'go o n' in the

pl uralit y o f cont exts o f socia l life.

Rcfcrcllt:,,:1< may be fOll nd o n PI'. 10; - '1. Recogniti o n o f th e esse ntial impa rLance o f the refl ex ive
44 Consciousness, Self and Social En counters The Uncons c io us, Tim e, M emory 45

mo nitoring of conduct in th e day-to-day continuity of social life becomes the instrument of the o ther's financial d iscomfiture.
does not mean disavowing the significance o f unconscio us sources Here the agent has to 'think' about what he o r she is do ing fo r that
o f cognitio n and motivatio n. But it does involve giving some activity to be carried o ut 'conscio usly'. 'Consciousness' in this
attention to the differentiatio n whic h separates 'conscious' from se nse presumes being able to give a coherent account of one's
'u nconscious'. activities and the reasons for them.
Ordinary English usage gives us at least a general guide to this.
Sometimes we speak of consciousness as equivalent to what
might be called 'sensibility'.5T hus someone who falls asleep or is rh e Unconscious, Time, Memory
knocked over the head 'lapses into unconsci ousness' or is
'rendered unconsc ious'. 'Unconscio us' here means something It is clear that the psychoanalytic sense of 'unconscious' has
di fferent from its orthodox Freudian usage, and the 'conscious- something to do with a contrast drawn between it and this third
ness' with which it is contrasted has a very broad sense. To be meaning o f 'conscio us', a contrast with what I have term ed
'conscious' in this meaning is to register a range of s urro unding discursive consciousness. Discursive conscio usness means being
stimu li . T here is norhing specifically reflexive about consciousness able to put things into words. T he 'unconscious' in psychoanalytic
understood in this way . Th e sense in which hum an beings 'lose' theory has reference to the opposite of this - no t being able to
and 'regain ' consciousness is directly applicabl e to the higher give verbal expression to the promptings of action.
animals also. T his notion of conscio usness evidently refers to th e To fu rther explicate the notion of 'un conscio us' as 't he
sensory mechanisms of the body and to their 'normal' modes of un co nscious" howeve r, it is necessary to make some co mments
(l l1 memory, since memory and language are paten tly very close. T
o peratio n and is presupposed by the concepts of both practical
and discursive consciousness. propose to argue that 'the unconscious' can be understood on ly in
'Conscio us' is somet imes used to refer to circ umstances in [e nns o f memory and that this in turn means examining rather
which peopl e pay attention to evenrs going o n arou nd the m in I.: arefully what me mo ry is. He re all the iss ues of theor izing
such a way as [ 0 relate their activity to those events. In o ther h;mporality whose significance I have insisted upon before
words, it refers to the refl ex ive moni toring of conduct by human reappear.
agents, largely in the se nse of what 1 have call ed practica l 11) Prima facie, one might suppose that memory refers simply to
consciousness. Thus, for examp le, a school teacher may be the past - to past ex periences, traces of which somehow
'conscious' o f what the children in the front rows of the classroom remain in the o rga ni sm. Action then occurs in th e spatiality
are do ing bu t 'unconscious' o f ot hers near th e back who have o f the present, drawing upon memori es of the past whenever
started gossiping with o ne another. The teacher may be being such are needed or desired. A moment's reflection will
inattentive, but is not un conscious in the same sense as an demonstrate the inadequacy of such a view. 'Present' cannot
individual who has 'lost consciousness'. If this sense o f 'conscio us' be said or written witho ut its fadin g into the past. If time is
has its counterpart among ani mals, it is not as unambiguo usly no t a successio n o f 'presents' but 'presencing' in the sense
defin ed as in the more elemental sense o f conscio us ness no ted attributed to this by He idegger, then memo ry is an aspec t o f
above. A third sense o f 'conscio us', labelled by Toulm in presencmg.
'articulateness' , corresponds roughly to discursive consciousness. 6 \2 ) O ne might imagine that memory is above all a recall device
To use Toulmin's example, a businessman who obtains money on - a mode of re trieving in for matio n or 'remembering'. Such
false pretences from a client can be said to have engaged in :t view is quit e co nsistent with the idea that th e past is clearly
'conscious and deliberate fraud'. On the other hand , if the same seve red from th e presen t because memory can then be seen
consequence fo llows quite inadvertently from the activities o f the as the recall of the past int o the present. But o nce we discard
businessman , without his being aware of it , he 'unco nscious ly' sllc h a sl andlX'int , it is no longer plausible to defi ne memory
46 Con sc iousne5s, Self and Social En coun ters The Unc onsciOU5, Tim e, Memory 47

as the remembrance of things past. Proust's titl e sho uld actively organized as such by the perceiver. The main po int. of
surely be read as an iro nic comment on just this type of reference has to be ne ither the single sense no r the contemplatIVe
naive conception . Recall is o bviously not irre leva nt to perceiver but the body in its active engagements with ~he material
memo ry , but it does nOt designate what memory is . a nd social worlds. Perceptual schemata are neurologically based
fo rmats whereby the temporality of experi ence is continually
These o bservations indicate that memory and perceptio n are processed. Such processing may .in turn. be. unde rsto?d ~s
very closely linked. It is of some interest to point out th at th eories inh erently involved with th e refleXive mOnitoring of actl.on m
of perception tend to divide around an axis of subjectivism versus general. It seems impossibl e to deny that th e new-bo rn l!1fant
o bj ec tivism. One type of standpo int tends to emphasize , in qu asi- possesses an innate perceptual equipment. In o th er words, It has
Ka ntian fashion, the role of th e perceiver as the processor of not o nly the sense o rga ns but also neurologically esta blished
what wo uld otherwise be a fo rmless void .1 An o pposing view sc hemata that allow it to respond selectively to the surro unding
ho lds that perception is organized by [he pre-given fo rm of the wo rld . even if that selectivity is relatively gross compared with
o bject-world.' Attempts to o vercome this division have stressed wha t is developed later. A good deal of evidence exists to the
the importance of time , a nd of spatial differentiatio n, in effec t that infants respond with movements of the head LOwards
perceptio n. Like intentions, reasons, etc. , percepti on is no t an t he direction of sounds, follo w moving objects visually and reach
aggregate of discrete 'perceptio ns' but a flow of activity integrated o ul towards them. 'Looking towards sounds', of course , already
with the movement of the body in time-space. Percepti on is invo lves integration of th e senses. I! Neonates already assess this
organi zed via anticipato ry schemata whereby th e individual in terms of a time diffe rence between acousti c respo nses in the
anticipates new incoming info rm atio n whil e simulta neo usly !wO ears , leading to the mo vement of the head in o ne direction.a r
me ntally digesting old. Pe rceptio n normally in volves the the o ther. Such responses do, of course , become mo re precise
continued active movement of the eyes, and usually of th e head , with further psychological and mo tor develo pment ; it takes a
even when the body is at rest. Because schemata are anticipatio ns, lung while for children to learn the ~rts of co~ing co.n cep!u~lIy
th ey are , as one author puts it , 'the medium where by the past with o bjects that have gone o ut of Sight. Nammg o r Identlfymg
affects the future', which is 'id entical with th e und erlying objects is evidently not just a matter of attaching a label to
mechanisms of memory.' It may very well be th at to uch , ph enomena wh ose qu aliti es are already k.n own. To na~e
ordinarily regarded as the most humble of the senses, and certainly slImething correctly is to be able to talk about It correctly , whIch
th e least studied, provides most clues for und e rstanding means typifying its properties : relating it to a class of comparable
perception in general. T o uc h has no clear-cut perceptual locus , ! )hjects differentiating it fro m other c1asses.
' 2 [n this respect we

like the eye ; incoming haptic info rmation is not ord ered thro ugh can see both the attractio ns and the limitatio ns o f Gibson's
an y single mechanism within the nervous system ; th e use of co uch co ncept of 'affo rdance'. According to Gibson, all the uses or
is seU-evidently part of the manipulatory movement of the body ac t ivities which objects mak e possible - which they afford to the
in the COnlexts of its actio n. A striking feature of most of the human actor - are directly perceivable. Such a view has the
lite rature o n perception , mo reover, is that it treats the senses as ad vantage of stressin g the prac tical charac te r of perceptual
tho ugh they operated in separatio n from o ne an other. It has been <l ei iviti es, but it does not indicate their conn ectio n with
observed that virtually all ex perim ental studies of perception co nce ptual designatio ns o f o bjects, which are lik ely to be
have invo lved only a single sense. IO That this is artificial is shown !.: llilurally variable. .
by th e most cursory examinatio n of the nature of day-to-day life, If perception be unde rstood as a set of tempo ral.orde:tng
in which the continuity of ac tiviti es persistently int egra tes the devices, shaped by , yet sha ping, the movements and o n entatlons
va rio us senses. (If th e bod y in th e cont exts of its behavio ur, we can understand
Perceptio n. then. depends upon spatil.ll and temporal continuity. lhercby th c significan ce of selec ti ve attentio n in day-to-day
48 Consc io(lsn ess, Self and Soc ial Encounters The UnCOfl!ic ious, Tim e, Mem ory 49

conduct. In every context of activity there is far mo re going on If the 'present ' is not c ut off from the fl o w o f act ion, 'memory'
than th e actor actually attends to, events or qualities that remain can be no thing other than a way of desc ri bing the knowledge-
unnoticed. How does this happen ? The usual answer is that ability of human agents. If memory does not designate 'past
~edundant material is filtered o ut. But this is quite misleading. for experie nce', neither does consciousness (in any of the three senses
It suggests an ac tive attempt to reject red und ant mate rial. mentioned above) express the 'present'. What a person is 'aware
Selection is , how~ver, a positive rather than a negati ve process; it of' canno t be fixed at a particular point in tim e. We need to
ex presses the active engagements of agents wi th their e nviron- d istinguish . the refore , between consciousness as sensory aware
ments .. Consider the foll owing much debated ex perim ent. JJ Tape ness (the first and most general sense of th e term mentio ned
recordln~S of two separate and different spoken messages were above); memory, as the temporal constitutio n of consciousness ;
played simultaneously to experimental subjec ts, one in each ear and recall , as [he means of recapitulating past e xperie nces in such
and at equal volume. Subjects were instruc ted to listen to o nly a way as to focus them upon the continuity o f actio n. If memory
one r.n~ge ~nd to repeal it as they heard it. They experienced refers to this temporal mastery so inherent in human experience,
no dlffl 7ulty III doing this and by and large did not 'hea r' the then disc ursive and prac tical co nscio usness refer to p~ycholog ;cal
alternative message at aiL The exp erimental situation is an mechanisms of recall. as utilized in co ntexts of acti on. Disc ursive
i~teresting o ne because it mirrors what agents do most o f the co nscious ness connotes those forms o f recall which the actor is
time when co-present with others in situations where more than able to ex press verbally. Practical conscio usness involves recall
o~e con~ersation is going on. The experimental results have been 10 whi ch the agent has access in the duree of action without being
Widely Interpreted in te rms o f negat ive info rmalio n fih ers. '4 a ble to express what he or she thereby 'kno ws'. Th e unco nscious
Redundant information , in other words, is supposed ly blocked refe rs to modes o f recall to which the agent does no t have direct
o ff from reaching the higher cortical centres - definite neural access because there is a negative 'bar' of some kind inhibiting its
mecha.nisms have been suggested as controlling su ch a process. unmed iated in corporation within the refl exive mo nitoring of
But th~s type o~ th eo ry. not on~y treats the individual as essentially co nduct and , more pa rticularly , within discursive conscio usness.
a. pass!v~ receI ver o f mput; It also depends upon an untenable The origins of the 'bar' are o f two related sorts. First, since the
dISSOCI8~lon between .perception and memory . For it is supposed ca rliest experiences o f the infant , shaping the basic security
t~at whil e we perceive everything in our environm ent at any sys tem whereby anxiety is canali zed or contro lled , predate
g lv ~ n moment , much of what is perceived is 'blocked o ff' _ very differen tiated linguistic competence , they are lik ely to remain
rapidly forgotte~'. 15 As Neisser has pointed out, the assump tio n is th e rea frer 'outside the bounds' of discursive consc iousness.
that any use of Informati on a few milliseconds after it has been Second, the unconscious contains repressions which inhibit
regist ere~ is ? epe.ndenr upon memory rather than perception. discursive formulation .
Such.a view IS neither conceptually compelling nor empiri cally As a matter of conce ptual definition , these remarks are
pla~slble. If perception is regarded as what agen ts do , as part of moderately consonant with Fre ud 's characteristic usage o f the
their temporally and spatially situated activit ies, th ere is no need 't..:o nscio us and 'the unconscio us'. But the thesis that most day-to-
to posit any blocking mechanisms at an. day activities are not directly motivated means placi ng in question
Organisms are active: they do some things and leave others undone. th e mod el of motivati o n with whi ch Freud chara cterist ica lly
To pick one apple from a tree you need not filter ou t all the others: o pe rated . For Freud a ll human act ivities are moti vat ed , incl uding
you just don't pick them. A th eory of apple picking would have I for exa mple ) apparent trivi ata or 'e rrors' such as slips o f the
much to ex plain (How do you decide whieh onc you want't Guide tongue. Freud was o ft en con cerned precisely to demonstrate thaI
you r ha.lld to it '? Grusp il'!) bu t it would not have 10 speci fy a phe no me na whic h might be s upposed to be 'accid e nta l' do, in
mec han ism to keep unwant ed appl es ou t nf your hand. ". r i l C I, have the ir origin in (un co nsc io us ) Illo tives. Th ere is 11 0
partic ular reason 10 question th e illumina ting qualit y of Fre ud 's
50 Consciousness, Sel f and Social Encounters Erikson : Anxiet y and Trust 51

insights in such matters. But it makes no more sense to claim that be adequately contained by the basic security system is specifically
every act or gesture is moti vated - meaning that a definite a feature of critical situations.
'motive' can be auached to it - than it does to treat action as Criticizing Freud 's terminology of agency and self carries with
involving a string of intentions or reasons. There is a logical flaw it several implicatio ns. The 'T' is an essential feature of the
here in the simplifi ed view of th e nature of human actio n. Action , refl exive monitoring o f action but sho uld be identified neither
as I have said oft en, cannot satisfa ctorily be conceptualized as with the agent nor with the self. By th e 'agent' or 'acto r' I mean
an aggregate o f ac ts. Concentrating mainly upon specific Ihe overall human subject located within the co rporeal time-
demarcated 'segments' of behavio ur (ne urotic symptoms), Freud's space o f (he living o rganism. The T has no image. as the self
writings inevitably tend to express such a deficient conception of does. The self , however, is not so me kind of mini-agency within
acti on. But rath er than supposing that every 'act' has a Ih e age nt. It is the sum o f those forms of recall whereby the agent
corresponding 'mo tive', we have to understand the term refl exively characte ri zes 'what' is at the o rigin of his or he r action.
'mo ti vation' to be a processual one. What thi s means concretely T he self is the agent as characterized by the agent. Self, body and
is that the unconscio us only rarely impinges directly upon the me mo ry are therefo re intimately related.
renexive monilOring o f conduc t. No r are the connectio ns involved
sole ly depend ent upo n psycho log ica l mechanisms within the
Erikson : Anxiety a nd Trust
personality of th e individual ac to r; th ey are medi ated by th e
social relation s which individuals s ustain in the routine practices T heori es which give prominence to unconscious elements of
of their daily lives. human behaviour oft en tend to go along with o bj ectivist
Elaborating a little o n this po int provides some thing of a pe rspectives. It is no t too difficult to see why. Fo r objectivism,
transition between the discussion so far in this chapter and that li ke many accounts o f the unconscio us, treats the refl exive
which follows later. The main theorems I wish [0 pro pose run as mo ni toring of actio n as mere fro lh o n the surface o f human
fo llo ws. Ordinary day-to-day life - in greater o r less degree activit y, whose tru e o rigins lie elsewh ere. In setting o ut an acco unt
acco rding to context and the vagaries o f individual personality _ (I f (a few features o f) the unconscio us a nd social relations, I shall
involves an ontological security expressing an autonomy of bodily 110 1 foll ow those versions of structuralist psychoanalysis, asso-
control within p redictable routines. T he psychological o rigins of c iat ed particularly with Lacan , that are currently fa shio nable in
o nto logical sec urit y are to be fo und in basic anxiety-controlling some quarters. Altho ugh Lacan's writings undeniably contain
mechanisms (as indicated by Erikson, whose ideas r discuss in som e ideas o f great interest , in my opinion they express an
w~at (ollows), hierarchically ordered as compone nts o f person- impove rished conce ptio n of the agent similar to that generated
allty. The generation o f feelings of trUSt in others, as th e deepest- hy 'structuralist Marxi sm' .r7 Lacan has been one o f the fi gures in
lying element of the basic security system, depends substantially I hI.! fo refront of th e attacks upon th e work of the so-call ed 'ego
upon predictable and caring routin es establish ed by parental psyc hologists' within psychoanalysis. Th ese polemi cs have been
fi gures. The infant is very early o n both a giver as well as a ill substantial d egree s uccessful, sin ce the work o f Sullivan ,
receiver o f trust. As he or she becomes mo re auto nomous Ilo rn ey. Erikson, Kardine r and o thers no w lies under so mething
ho wever, the child learns the importance of what a re in Goffman '~ \I f a shadow. I consider that some o f the contributio ns o f these
term 'protective devi ces', which susta in the mutuality implied in aUlh ors. ho wever , reta in a very co nsid erable impo rtan ce and
tru st via tact and o th er formulae that preserve th e face of others. ." hall draw upon th em in so me part here.
Ontological security is protected by such devices but maintained Critiqu es. 'revi sio ni sms' and self-professed 'o rthodo xi es' have
in a more fundam ent al way by th e ve ry predictability o f ro utine , hl:c n as pro lific in psycho analyti c th eo ry sin ce th e early years of
something whic h is ntdically disrupted in critical situ<t tio ns. The thi s ccntury as Ih ey have bee n wi thin Ma rx ism. Th e ego
swamping o f habilual modes of ac ti vity by anxie ty whic h canno t psycho logists. howcver. ha ve hee n as...ocial ed wilh two principal
Erikson : Anxiety and Trust 53
52 Consciousness, Self and Social Encounters
Iransition in th e life of the individual, as Chomsky has
lines of development as regards the 'classical' formulatio ns of demonstrated , whose consequences can be fairly readily identified
psychoanalysis in Freud's writ ings. On the one hand , they have hu t (he origins of which remain tantalizingly obscure.
taken up the perspective fostered by Anna Freud, That is to say , In all societies th e early nurture of the infa nt is dominated by a
they have a~gued that. Freu d's preoccupation with repression and single mothering agent, nearly always the biological mother of
the unconscIOus led him to underplay the more cognitive , rational the child , T he initi al phases of personality develo pment may be
~mponents of th e agent. On the other hand , they have been c haracteristically associat ed with resolutions of needs or tensions
mfl,uenced. by the writings of social analysts, especially anthropo- der iving from th e physical traits of the organism. But it seems
logl.sts, .whlch d emonstrate the sheer diversity of human modes of almost certain that Freud squeezed these in to ( 00 deterministic a
SO~I~I hfe: ~re ud 's cultu~al writings - however much th ey may sc heme, and a more flexible one is required to make sense of
retam, theIr Importance m some ways - were essentially bound variat ions between and with in societies. We may say that the
up . wIth the evo ~uti~mi s ~ o f nineteenth-century anthropology. earl iest interaction between infant and mother is layered into the
Bem ~ aware of thiS diversIty means also ack nowledging the variety development of the 'unconscious': neither 'bodily movement' nor
of ~I f.fe r~ nt fo rms ~f family o :~anization, and hence of early 'hod ily comrol' is very similar to the senses in which they are
SOCialization, that eXIst. Recognttlon of these two se ts of factors involved in 'actio n' in the case of the adult member of society. If
take.n. togeth~r, means making substantial departures from mor~ we fo llow Erikson , we can distinguish three successive polarities
tradl.tlo nal ~Iews of psychoanalytic theory, although it does not ;I.,\sociat ed with th e t ransformation of the body into an instrument
entml, adoptmg a full-blown cultural relativism; there are processes nr act ing-in-the-wo rld. T he first, and earliest , is that o f 'basic
of child development and adult personality common to a ll human trust' versus 'basic mistrust', The new-born infant is a bundle of
societies. Erikson expresses this in Childhood arId Socielv in the impulses, which have certain genetically given homeostatic
following way : ' ILl l.!c hanisms of adjustment, existi ng in an ali en environment; the
:lcl ivities o f the mother provide care a nd protectio n, 'Trust' (here
Psyc,hoa,nalysis )od,ay is implementing the smdy of the ego. ' , , It is
shIfting Its emphasIs from the concentrated study of the conditions ~'o llceived o f as a trait of perso nality) is understood as
which blunt and distort the individual ego to the study of the ego's psyc hologically 'binding' time-space by the initial awakening of a
roots in soc ial organiz.at ion, . ,. Long childhood mak es a technical ...e llse that absence does no t signify desertion. The psychological
and menia l virtuoso out of man, but it also leaves a lifelong residue tl yamics underlying the intersection of presence and absence
of emotional immaturity in him ,' ~ 11; I VC their point of origin in th e body, bod ily needs, their modes

" r satiatio n and con troL

, Erikson, together with Sullivan, are perhaps the two o utstanding As Erik son comments, 'T he infant's first social achievement,
fIgures among those writers who have preserved certain universal IlI en, is hi s willingness to let the mother o ut of sight without
elements of Freud's original account of the stages of psychosexual IIndue anxiety or rage, because she has become an inner ce rtainty
development , while at the same time adopting contributio ns from a ... well as a n o uter pred ictabil ity.' Predictabilit y, continuity,
the social sc i ~n~, 1 s~ a ll draw - although sparingly and critically sa illen ess , provid e 'a rudimentary sense of ego id entity whi ch
- upo n th eir Ideas III what fo llows, On th e basis of both his dc pends ... on the recognition th at there is an inner population
clin ical work and the study of a range of cult ures, Erikson has " r remembered and anticipated sensations and images which are
distingu!shed a se ~i es of stages of personality developmem over firm ly correlated with th e outer population of fami li ar and
the penod from mfancy to adu lthood, His discussion of th e prcdictabl e things and people', ' ~ 'Trust' here equ als confi de nce,
natu,re of l,he motivational inclination s and mental capacities of ami vcry ea rly 011, Erikso n suggests, it has a definite mutuality to
the m(a~t, IS extremely p~rsuasive. But I do no t think he brings it ; Ihere is at least an inc ipie nt feel ing o f 'bein g trustworth y'
o ut,sufflclent ly the essenllal threshold in ch ild development that II ss()(;ial Cd wit h Ihe generali 7.ed ex tension of tru st to th e o ther.
derives from the phase o f th e sy nt actica l mastery o f lang uage, a
54 Consciousness, Self and Social Encounters Erikson. Anxiety and Trust 55

Not, of course, that the initial formation of trust occurs without generic way through the hands and arms. Holding on and letting
conflict or strain. On the contrary, it operates against the go are the behavioural correlates of the main polarity on which
background of diffuse anxiety, control of which suggests itself as this stage is centred, autonomy versus doubt or shame. As with
the most generalized motivational origin of human conduct. The the prior phase, with which it can stand in a relation of generalized
interaction between infant and mother embeds the growing tension, the polarity can be resolved in a relatively benign or
human individual in a nexus from which, for better or for worse, more disruptive way. To hold on as a greedy mode of retention
there is thereafter no escape. The mother is an agent (already a can represent a cruel self-absorption or can be a pattern of care
representative of the 'generalized other') who, in caring for the expressing autonomy. Letting go can similarly be a hostile
infant, lays a social claim upon it that presages the normative expression of aggressive impulses or a more relaxed attitude to
sanctions associated with the later formation of social 'letting things pass'. It seems important to emphasize the
relationships. The anxiety of absence is defused through the significance of the psychodynamics of shame as contrasted with
rewards of co-presence, setting the ground for the dialectic of guilt. Many psychoanalysts, following hints given by Freud, have
engagement and disengagement on which the diversity of treated shame as specifically connected to fear of genital
encounters is based. The expansion of the autonomy of the exposure. This certainly helps to indicate one aspect of anxiety ,
infant, anchored in control of the body as a medium of action about bodily 'appearance' , which (as will shortly be indicated)
(which undergoes a massive transformation with the mastery of Goffman shows to be so important. But the phenomenon of
language), simultaneously widens and integrates this dialectic. shame is surely much more pervasive than Freud's comments
Each individual has the right - varying in content in manifold would lead us to believeJl
ways in different contexts - to maintain a distance from others The prevalence of feelings of shame or self-doubt is indicated
by preserving bodily privacy and an integrity of self. But the self by the frequency with which being 'ashamed' and comparable
has to submit to social engagement, given that this is done with terms (, mortified', 'humiliated', etc.) appear in ordinary talk. The
proper deference to the tactful recognition of the needs of others. idea, suggested by some writers, that guilt is 'private' while shame
The infant does not yet know this, nor its connection with face. is 'public' seems difficult to sustain. Shame bites at the roots of
Face, as Becker puts it, is 'the positive feeling of self-warmth se lf-esteem and clearly is closely related to the rather milder
turned to the world for others' scrutiny and potential sabotage'.w ex perience of 'embarrassment'. Both shame and embarrassment
As the foundation of a tension-management system, the are located psychologically in the intersection of engagement and
trust/mistrust polarity is organized around relations between disengagement, the failure to 'bring off' certain aspects of
projection and introjection as mechanisms of personality. Infantile performance through being 'caught out' in various ways. Unlike
introjection, as Freud holds, assimilates outer goodness and inner 'g uilt', 'shame' and 'embarrassment' capture both sides of
certainty; projection treats an inner harm as external male- encounters: that is to say, the latter two terms can be used by the
volence. 21 Themselves based on identification, these mechanisms ind ividual about his or her own conduct or that of others. I can be
become overlain by a variety of more mature psychic forms. But ashamed of myself, of something which 1 have done, or
they come to the fore again in situations of extreme threat or embarrassed about it. But I can also be ashamed of the conduct
crisis. The physical maturation of the body subsequently sets the (If someone else, as well as embarrassed for him or her. Here we

stage for the transition to a new phase of development. Erikson seem to detect a difference between the two emotions. To be
suggests that this is not best understood in terms of a shift ashamed of somebody else's behaviour indicates a tie with that
between pleasure zones on the surface of the body, as Freud I II her, signalling a certain recognition of association with, or even

holds, although fixations may become centred on these. 'Holding responsibility for, the other. To be embarrassed for someone,
on' and 'letting go' are obviously applicable to control of the rat her than ex pressing an alienation from his or her conduct,
waste products of the body but are expressed in a much more reveals a certain complicity with it , a sympathy for someone who
56 Consciousness, Se lf and Socia l Encounters Erikson ' Anxiety and Trust 57

has been unn ecessarily 'exposed', But this is purchased at th e pri ce o f repression, which in some
It is especially interesting, in the light o f Goffman's pre- individuals and in some circumstances can have cri ppling costs in
occupation with like happenings , to note that Erikson links shame fo rms of anxiety stemming from guilt.
in (he infant (having strong residual traces in the security system
o f the adult) to bodily posture and to 'fro nt' and 'back' regio ns o f For here the child becomes forever divided in itselr. T he instinct
the body. Here we can see a mode in which Freud's theory o f anal fragments which before had enhanced the growth of his infantile
body and mind now become divided into an infantile set which
re te ntio n can be expressed in a much more socialized form. The
perpetuates the exuberance of growth potentials., and a parental
'front' and 'back regio ns' in which encounters occur, and in the
set which supports and increases self-observation, self-guidance,
contex t o f which social occasio ns are staged . perhaps re la te
and self-punishmenl. N
directly to the more primal ex perience o f the fro nt/ back
regionaliza[ion o f the body. To sustain 'fro nt' in social life is to Put together , th e three phases re present a progressive
avoid the anx ieties provoked by shame. and loss of front leads movement towards autonomy, which shou ld be understood as the
precisely to shame or embarrassment. For the infant 'be hind' foundatio n of the capabi li ty fo r the re fl exive monitoring of
means 'the behind'; co nd uct. But 'autono my' does not mea n the shedding of the
anxiety-provoking stimuli o r the modes o f coping with anxiety
the small being's dark continent, an area of the body which can be
which comprise the security system of the adult personality. The
magically dom inated and effectively invaded by those who would
attack one's power of autonomy, , , This stage, therefore, becomes motivational components of th e infantile and the adult personality
decisive for the ratio of love and hate, co-operation and wilfulness, derive from a generalized o rientatio n to the avoidance of anxiety
freedom of selfexpression and its suppression. From a sense of and the preservation of self-estee m against the 'fl ooding through '
self-control without loss of self-esteem comes a lasting sense of of shame and guilt, We may presume that the mechanisms of the
good will and pride: from a sense of loss of self-control and of security system remain o n an unco nscio us level because they are
forei gn overcontrol comes a lasting propensity for dou bt and pre-linguistic - although the Oedipal phase is the very time at
shame.l.I which the ch ild learns to constitute itself as an T .

The third phase , the o ne that c ulminates in , and coincides

wi th , the mastery of syntact ically developed language, focalizes a II I locomolOr ;n'(' ... (" .....
G.'ni (,,1 ...,rSU5
po larity of initiative versus guilt . This is the phase o f Oedipal ~u, 1t

transi tio n which, whatever its obscurities and complexities,

appears as a uni versal c risis phase in human psycho logical II "",,,-,culal a utonomy
developmem, So far as the body is concerned , it is marked by the vel sus
Ana l
~ha me. douh(
mastery o f an upright stance and ambulatory movement in thal
stance, and by the maturatio n o f infantile genitality. The dramat ic ha~ic Iru~1
potential of this phase for late r personality development is given Or... 1 v"r ~ u ~
Sen sory mi stru st
by the conjunction of the demand for repressio n of earl y
attachment to th e mother (i n both boys and girls) , co upled with , ,
the capabilit ies th at beco me part of this process as it coin cides
with a vast leap forward in linguistic skills. It is a phase of Fi gu re 3
initiative beca use th e accomp lishm ent of the Oedipal transition
allows the chi ld th e internal co ntrol necessary to venture forth Figu re 3 indicates tha t th e successive phases presume varying
from the immed iate confin es o f the famil y into pee r relati onships. ra tios of ind ependence and dependen ce, co m binatio ns of bodily
SII Consciomness, Se lf and Social Encounters Erikson . Anxiety and Trust 59

modes and psychological mechanisms. If tracing out individual at the key phase of development. It is striking how closely some
differences were at issue, it would imply thinking through the o f Vygotsky's observations about what to an adult would appear
empty boxes, which would become filled in so far as infantile 10 be a 'dissociation' between speech and conduct resemble those
fixations or modes of regression exert a pervasive influence over made by Merleau-Ponty in respect of brain-damaged patients (see
the motivation of behaviour. pp . 65 - 7). For instance, a child may be able to carry out a fairly
Research into child development suggests rather strongly that co mplex task only on condition that it verbally describes each
the formation of capabilities for autonomous action meshes movement as it goes along. Children , like many of the 'mentally
closely with understanding others to be agents. Three main steps ill', are not reluctant to talk to themselves in public - a
in the formation of concepts of agency can be distinguished, phenomenon which has to be distinguished from Piagel's
coinciding with the stages described by Erikson. One is the ide ntification of 'egocentric speech'.
recognition of what has been called 'simple agency' - that others Having appealed to Erikson a good deal, I should perhaps
can causally intervene in a sequence of events to as to change make it clear that my appropriation of some of his ideas is
them.25 The infant's awareness that its body is a locus of action Intended to be strictly limited and qualified. I consider the least
goes along with the attribution of like qualities to the bodies of illteresting areas of Erikson's work to be those for which he is
others. At quite an early age infants react differently in their pro bably most famed - to do with the formation of 'ego-identity'
interaction with 'agent-like' others, although the aspects of the and with the importance of developmental stages in personality
conduct of such figures to which response is made are relatively 1hat stretch up to adolescence and beyond. Erikson is critical of
simple and clear-cuL 16 Other agents are, however, still treated Fre ud's formulations about the 'ego' and its relations to society.28
instrumentally, as a special type of object in the environment, Th is is partly because of their sociological inadequacies. Freud
rather than as physically separate beings from the self, who can drew upon highly inadequate sociological texts (such as
go away and return. The emotional competence associated with contemporary discussions of crowd psychology) in his writings,
trust seems closely connected with the cognitive understanding At the same time , psychoanalytic method was based on individual
of agency as a property of distinct beings. But specifically 'human' case histories. Between these there is a large gap. No satisfactory
properties, generalized to human agents rather than attributed to account of a differentiated society was worked out by Freud or
particular parental figures, mark a transition to a third stage. ma ny of his epigones; 'the concept of social organization and its
Vygotsky, among others, has demonstrated the close relation hearing on the individual ego' was 'shunted off by patronizing
between locomotor skills (the mastery of the body as a locus of lributes to the existence of "social factors''' ,l'I The concept of the
action) and the syntactical mastery of language. His work scarcely ego was thus established by Freud, Erikson points out, in relation
answers the 'Chomskyan problem' - how does the child, 10 its opposites in the lawless nature of the crowd and the
relatively suddenly, manage successfully to co-ordinate syntactic primeval instincts of the id. In order to try to take acco unt of the
structures? - but it does elucidate important aspects of the emba ttled moral sensibility of human beings, Freud introduced
association of agency and speech. Language use, in differentiated 1he super-ego or ego-ideal - also, however, thinking of it in terms
form, depends upon the expansion of the 'practical intelligence' primarily of a burden which the ego has to bear. Erikson wants to
of the child - in other words, upon definite aspects of practical co mpensate for this one-sided emphasis. Rather than concen-
consciousness.27 The development of 'practical intelligence' lrating upon what is denied to the infant by social organization,
accelerates, it can be suggested, from the period of the resolution we should be (;oncerned also with how the child benefits from it ,
of the third phase in Erikson's scheme, since it involves the OIlld we should give greater consideration to the influen ce of
exploration of the body as a medium of action. But the initial differentiated types of soc ial o rganizati on. Erikson's noti o n of
emergence of 'practical intelligence' dates from the first ego-identity is intend ed to co mplem ent the traditionally esta b-
exploratory mo vements of the very young infant; mastery of li shed psychoanalytic conce pt s. JO
syntactical spee(; h CO li verRes with th e growth of practical mas tery I am large ly in accord with Erik son 's criti ca l co mments on
60 Consciousness, Self and Socia l Encounters Routiniza/ion and Motivatio n 61

Freud. But th e term 'egoidentity' is not a satisfacto ry one. Th e sense in which critical si tu ations, for specific individuals or clusters
term 'ego', as 1 have ind icated, does too much conceptual work in of ind ividuals, are themselves built into the regularity of social
psychoanal yti c th eo ry. Tha t o f 'ego-identity' tends only to life by the very nature of th e intersection betwee n the life process
compound the con fusions tha t already exist. Even Erikson admits or 'cycle' of the individu al, the duree o f activity o n the o ne hand
that it has at least fo ur connotations. Sometimes it refers to a and the longue duree of institutio ns o n the o ther. These are the
'conscio us' sense o f indi vidual identity, It can also mean 'an crises typically marked by rites o f passage , beginning for the
unconscio us stri ving fo r a continuity o f personal c haracter'. A ind ividual with birth and terminat ing in death. However, forming
third mea ning is 'a criterion fo r the silent doings o f ego syntheSis'. as they do an intrinsic pa rt of the continuity o f social life , even
A fo urth se nse is 'a maintenance of an inner solidarity with a Iho ugh they are discontinuities for ind ividuals, such si tuations
group's ideals and identity',JI None of th ese single uses, it might le nd themselves to have a definitely routinized charac te r.
be remarked, is particularly luc id , let alone the concept that By 'critical si tuat ions' I mean c ircumstances of radical
em braces the m ali t disju nctu re of an unpred ictable kind which affect substantial
numbe rs of individuals, sit uations that threate n o r destroy the
Rou t inization and Motivation certitudes o f institutionali zed ro utin es. I am conce rned at this
poi nt not with analysing the social o rigi ns of such circ umstances
Rather than employing the concept of ego-identity, in what follows hut with their psychological conseq uences, a nd with what those
I shall make use of Erikson's ideas of the origins and nature of co nsequences indicate abo ut the generality o f ro utine social life.
bod il y auto no my and of trust. A sense of trust in th e continuity o f Since I have discussed critical situ atio ns in a certain amount of
the object-world and in the fabric of social activity, Lshall suggest, detail elsewhere.-'2 I shall mention here only o ne - a famous
depends upo n certain specifiable connections be tween the portrayal of a wholly infamo us episode in recent history. This is
indi vid ~al age nt and the social contexts through which that agent Bettelh ei m's discussion in The i nformed Heart. a description and
moves In th e course of day-to-day life. If the subject cann ot be analysis of the experiences of the author and o th ers in Dachau
grasped save thro ugh the reflex ive constitution of daily activiti es and Buchenwald. In the camps, he writes, 'I ... saw fast changes
in social prac tices, we ca nn ot understand the mechanics o f lak ing place , and not o nly in behavio ur but personality also;
personality apart fro m the ro utines o f day-to -day life thro ugh incred ibl y faster and o ft en much mo re radical changes than any
whic h the body passes and which the agent produ ces a nd Ilml were possible by psychoa nalytic treatme nt : J3 The concen-
reproduces. T he concept o f roulinizalion. as grounded in practical lralio n-camp experience was marked no t o nly by confinement
conscio usness, is vita l to the theory o f struc turatio n. Routin e is hu l also by ex treme disruptio n o f accustomed fo rms o f daily life ,
integral both to the continuity o f the personality o f the agent , as de ri vi ng from the brutalized conditions of ex istence, ever-present
he or she moves alo ng the pa ths o f daily activities, and to the threat o r actuality of vio le nce from the camp guards, scarcity o f
insti tutio ns o f soc iety, whi ch are such o nly thro ugh th ei r rllod a nd other elementary provisions for the sustenance o f life.
co ntinu ed reprod uction. An exam ination of routinizatio n, I shall Th e cha nges in person ali ty described by Bettelh eim -
claim , provides us with a master key to explicating th e expe rienced by all prison ers who were interned in the cam p over
characteristic form s of relatio n between the basic security system a period o f years - follow ed a ce rtain seq uence of stages. The
on the one hand and the reflexively constituted processes inherent sequence was quite eviden tl y a regressive o ne. The very process
in the ep isodic characte r of enco unters on the other. o f initial imprisonm ent was traumatic for most of the inmates.
We can probe the psychological nature of the routin e by T orn away from family a nd fri e nd s, us uall y with little or no prior
consid ering th e results of situ atio ns where the establi shed modes warning, many prisoners we re subjected to torture during their
o f acc ust omed dai ly life are drasti ca lly undermined o r shatt ered Ir; l1Isportali o n 10 th e cu mps. T hose fr o m middl e-class or
~ by stud ying Wh<Jl ma y be ca ll ed 'critical siluati ons'. The re is a p1'llfess io nal backgrou nds, who mostly had hud no previous
62 Consciousness, Self and 50cial Encounters Routinizat;on and Motivation 6]

contact with the police or (he prison system , experienced the human agents, avo iding eye contact with o thers. making o nly
greatest dislocation in the ini tial stages of transportation and gross movements of the body and shu f1ing their legs whe n they
'initiation' into camp life . According to Bettelheim, the suicides walk ed. T hese men and wo men soon di ed. Only prisoners who
that took place in prison and transportation were confined mainly managed to maintai n some small sphere of control in the ir d?ily
to this group. T he vast majority of new prisoners, however, lives, whic h they still regarded as their 'own', were able to su ~vlve.
sought to distance themselves psychologically fro m the dreadful T hey preserved , as Berte lheim says, 'th e mainstay of a rad ically
pressures of camp life and tried to maintain the modes of conduct reduced but still present humanity'. None the less, they were
associated with their previo us lives. But this proved impossible to un able to avoid a range of c hildlike atti lUdes, a very marked
do. T he 'initiative' o f which Erikson writes as lying at th e core of diminu tio n in time se nse, in th e capacity to 'thin k ahead', and
human autonorny of action was very rapidly corroded; the vo latil e mood swings in response to entirely trivial happenings .
Gestapo in some degree deliberate ly forced the prisoners to All these things refer to the behaviour of prisoners who had
adopt childlike behaviour. heen in the camps for no more than a year (whiC ~ inc.luded
Bettelhei m). The 'o ld prisone rs' , those who had surVived 10 the
The vast majority of prisoners wen t through the camp without a ca mps fo r several yea rs, behaved different~y. T hey had lo~t
public flogging, but the screamed threal thai they were going to get altogether any orientatio n to the world o utSide and had , as It
twenty-five on the behind rang in their ears several times daily.... were reconstitut ed themselves as agents by integrating themselves
Threats like these, and also the curses thrown at prisoners by both
into ~amp life as participants in th e very rituals o f d egradation
the SS and prisoner foremen, were almost exclusively connected
whic h as new pri so ners, they had found so offensive. T hey were
with the anal sphere. 'Shit' and 'asshole' were so standard that if
llft en ~nable to recall names, places and eve nts in thei r previous
was rare when a prisoner was addressed otherwise.30
lives. T he end result , fo und in most but no t all old prisone rs, was
The guards exerted strict but wilfully erratic contro l over toilet, ;:1 reconstructed personality based upo n identificatio n with th e
in the sense both of eliminatio n and of general cleanliness. All o ppressors themselves, the camp guards. Old priso ners ~ped the
these activities were carried on in public . The camps destroyed activities of their captors, not merely to curry favour With them
virtually all differentiation betwee n 'front' and 'bac k regions', hut also. Be ttelhe im suggests, because o f an introjection of the
making the latter physically and socially a central preoccu pation normative values of the 55.
of camp life. How should we interpret these events? The sequence o f sta~es
Beuelheim places particular emphasis upon the general seems fairly clear (altho ugh not set o ut in this way by Be tte lhelm
unpredictability o f evenlS in the camps. T he feeling of au tonomy himself) . T he di sruptio n and the deliberate ly sustained attack
of action that individuals have in th e ordinary routin es of day-to- upon the o rdinary routines of life produce a high degr~e of
day life in orthodox social settings was almost completely a ll xiety, a 'stripping away' of the socialized responses ass~cJated
d issolved. The 'fu tu ral' sense in which the duree of social life with the security of th e management of the body and a predictable
o rdinarily occurs was destroyed by the manifestly contingent rmmework of social life. Such an upsurge of a nxiety is expressed
character of even the hope that the next day would arrive. The ill regressive modes of behaviour , attacking the founda tio n o f the
prisoners , in o ther words, li ved in circumstances o f radical hasic security system grounded in trust man ifested towards o thers.
o ntological insecurity: 'it was the senseless tasks , the lack of Those who are ill-equipped to face these pressures succumb and
alrnost any time to oneself, th e inability to plan ah ead beca use of go und er. Some are able to sustain a mi~imal sphere of con.trol
sudden changes in camp policies, that was so deeply destructive.'JS and self-estee m that a llo ws them to survive for a longe r peflod.
So rn e prisoners became 'walking corpses' (Muse/manner, so- Bu t eve ntually , in most of the o ld prisoners at least, a pro~es~ of
called) because they su rrendered fatalistically to whatever the 'rcsocialiw tion' ta kes place in which an a ltitude o f trust (hmlt~d
future might hold. T hey no lo nger behaved as tho ugh they were ano highl y a mbivale nt) ,.lI> in volving ide ntifica tion with authOrity
64 Consciousne5S, Self and Social Encounters Presence, Co Presence and Soci al Int egration 65

figures: is re-established. Such a sequence of he ightened anxiety , perhaps the most telling re Oectio ns on the matter are to be fo und
regressio n, fo llowed by a reconstruction of typical patterns o f in Merleau-Po nty. I shall begin by considering these: they lead us
a~tion , appears in a range o f critical situatio ns in o therwise very d irectly into Goffman's o bservatio ns. The body, Merleau-Po nty
different contexts, such as respo nses to being unde r fire o n th e points o ut , does no t 'occupy' time-space in exactly the same sense
battlefi eld for prolonged periods o f time, forced interrogat io n as material objects do. As he puts it, 'The outline o f my body is a
and torture in prisons and o the r conditions of ext reme stress.:!? fro ntier which ordinary spatial relations do no t cross.' .}II This is
Ordina ry day-to-day social fife, by contrast - in greater or because the body, and the expe ri ence of bodily mo vement, is the
lesse r degree, according to context and the vagaries o f individual centre o f fo rm s of action and awareness which rea lly define its
personality - involves an o nt ological security founded o n an unity. The time-space relations of presence , cen tred upon th e
autonomy of bodily control within predictable routines and body, are geared into not a 'spatiality of position' , in Merleau-
~ n ~o~nte rs. The ~outinized c harac ter of the paths a lo ng which Ponty's words, but a 'spatiality o f situation '. The 'here' of the
mdl vldu als move In the reversible time of daily life does no t just body refers no t to a determinate series of coord in ates but to the
'happen '. It is 'made to happen' by the modes o f reflexive situa tio n o f the acti ve body o rie nted towards its tasks. Much as
mo nitoring of action which ind ivid uals sustain in circ umstances Heidegger says: 'if my body can be a "form" and if the re can be ,
of c.o-presence . The 'swamping' of habitual modes o f activity by in front of it , important fi gures against indifferent backgro unds ,
anxiety that cannot be adequately contained by the basic sec urity [his occurs in virtu e of its being polarized by its tasks, of its
syste m is specifically a fea tu re o f critical situations. In ordinary existence towards them , of its collecting together o f itself in
social life actors have a mo tivated interest in sustaining the forms pursuit of its aims; the body image is finally a way of stating that
of tact and 'repair' which Goffman analyses so acutely. However , Ill y body is in-the-world .'YI
this is no t because social life is a kind of mutually protective T he observations of Goldstein and others on brai n-damaged
contract into which individuals vOluntarily enter , as Goffm an o n pa tients provide graphic illustratio n of how this is SO.40 Thus some
occasion suggests. Tact is a mechanism whereby agents are able s uc h individuals are no t able to carry o ut moveme nts which
[ 0 reproduce the conditions o f 'trust" or omo logicaJ security ahstrac t from th e visually present milieu. A person can point to a
wjt~i~ which more primal tensio ns can be canalized and managed. part o f the body on ly if he o r she is able to watch the movement
ThIS IS why o ne can say th at ma ny o f the specific featu res o f day- carried out and actually touch that part of the body. From
to-day encounter are no t direc tly motivated. Rath er, there is a Il bservations such as these it becomes apparent th at. while both
generalized motivation al com mitment to the int eg rati o n of are seemingly 'positio nal' phenomena, 'touching' is no t the same
ha bitual practices across time and space. ,IS pointing' . The difference indicates th e importance of bodily
space as an extraordinarily complex field of matrices o f habitual
Presence, Co- Presence and Social Integration actio n. Th e brain-damaged pati ent , asked to perform a given
movement of [he body , assumes a general positio n o f the whole
T he routines o f day-to-day life a re fundamental to even the most hody to carry o ut the task. It is no t c ut dow n, as in the normal
ela borate forms of societal o rga nizatio n. In the course o f their individual, to a minimal gesture. Thus , asked to salute , the patient
daily ac ti vities indi viduals e nCO unter each o ther in si tua ted takes up a formal sta nce o f the who le body - the ind ividu al
contexts o f interaction - inte raction with others who are ' manages to make th e gesture o nly by adopting the generalized
physica lly co-present. situatio n to whi c h th e move ment correspo nd s. Th e normal
The social characteristics o f co-presence are anchored in th e individuaL by contrast. sees the situation as a test o r as play. He
spatiality of the body , in orientati on to othe rs a nd to th e Ilr she is, as Merl eau -Ponty says. 'using the body as a means to
ex periencing self. G offman has d evoted considerable ca re to pla y acti n g .~ ' II is the di lemma o f the pati ent whic h provides
analysing this ph enomenon, parti cularl y with rega rd to 'face'. but most insight into the o rd in ary in tegration o f the body into the
66 Consciousness, Self and Social Encounters Presence, CoPresence and Soc ia/ In tegration 67

dllree of activit y. For th e body operates , and is understood as a of biological evolution. We need no t tran spose biological into a
'body' by its owner, o nly in the contextualiti es of ac ti o n. presumptively parallel form o f social evolution to see the
Wittgenstein 's question, 'What is the differen ce between my implications of this fo r human soc ial processes in circumstances
raising my arm and my arm going up'?'. has here c reated ma ny o f co-presence, In human beings the face is not simply the
difficulties , whatever he may have wanted the inquiry to draw proximate physical o rigin of speech but the do minant area o f the
o ur attention to. For it seems to treat as typical just that case of a body acro ss whi ch the intri cacies of experien ce , feeling and
test or a playful command ; and the theory of actio n then can be intenti o n are written. In banal but very significant ways the face
taken, misleadingly, to hinge on contrasts between 'movem ents' in human social relatio nships influences the spacing of individuals
a nd 'actio ns', as disc rete o pe ratio ns, rather than on the time- in circ umstances o f co-presence. Positio ning 'fac ing' the o ther or
space contexwality of bodily activity in the flow o f daily conduce o thers who are being addressed assumes a distinctive impo rtance
Such activity of th e body , in th e fl ow of actio n, is immediately as co mpared with positioning in most animal societi es. The
involved in th e onlological security or attitude o f 'trust' towards !lumbe rs of people who can directly parti cipate in fac e-to-face
the continuity of the wo rld and of self implicated in the duree of e ncounters is inhe re ntly stric tly limited , save in those types of
day-to-day life. For the brain-<iamaged patient a tho ro ugh physical a
situatio n where o ne o r few individuals address a crowd or an
examinatio n o f an object is required befo re it can be identified as , audience fac ing them. But such circ umstances , of course , demand
say, a 'key'. Normal individuals wo uld engage in such a scrutin y o f th aI those in th e crowd or audi ence sacrifice continu ous face-to-
an object o nly in unusual circumstances - wh ere, for example , face contact with on.e another, iThe prima~y ~f thle face as a I
they were playing a party game in which th ere were definite medium of expressIOn and of commum catlon I has mo ral
reasons to suppose that o bjects might not be as they appear. The implica tio ns , many o f which a re very ac ute ly teased o ut by
continuity of o rdinary life wo uld be impossible were we to attempt Go ffm a n. To turn o ne's back o n another while the o the r is
to submit all o bjects to such d etailed inspection. Fro m this we see speaking is in most (perhaps aU '!) societies a gesture of indifference
that Garfink el's 'etcetera clause' applies not just to language o r , I I' contempt. Moreover, most (aJl'?) societies tend to recognize a
co nversati on but also to bodily acti viti es in physica l relatio n to ling uisti c similarity between the face as a term referring to
the externa l world. All this is in turn intrinsically involved with physiogno my and face as concerning the maintenance o f self-
time and time-sense. Let me quote again from Merleau-Ponty: estee m. No doubt there are a range o f cultures, such as traditional
C hin ese culture o r sectors of it , which place an especial emphasis
Whereas in the normal person every event related to movement or lipa n th e preservatio n of face in most settings. No doubt also this
sense of touch causes co nsc iousness to put up a hos t of intentions may have something to do with the famou s differentiatio n made
which run from the body as the centre of potential action either hy Be nedict and o thers between 'shame' and 'guilt' cultures , even
towards the body itself or towards the object, in the case of the ir this differentiatio n seems to have been drawn much too crudely.
patient, on the other hand, the tactile impression rema ins opaque But aspects of the preservation and 'saving' of face are almost
and sealed up . . .. The normal person reckons with the possible, n :rtainly generic to a whole diversity of transcultural contexts of
which thus, without shifting from its position as a possibility,
acqui res a sort of actu ality. In [he patien t's case, however, the field Stl!.:ia l encounters.
of actuality is limited to what is met with in the shape of a real The twin th emes o f the control o f the body in fields o f actio n in
contact or is related to these data by some explicit process of \..o-prcsence and the pervasive influence o f face are essential to
deduction. ' 1 rh e whole o f Go ffm an's writings . How sh ould we understand th e
rCnll 'co-prese nce"! As Goffman Llses it, and as 1 employ it here
The body , o f course, is not an und ifferentia ted unity . Wha t a lst} . co-prese nce is an cho red in th e perce ptual and co mmuni-
Ge hlen calls the eccentri c posture o f human beings - standing cat ive modaliti es o f th e bod y. What G o ffman calls 'the full
upright <.I nd 'o ut wa rd ' towards the world - is no do ubt the result \..o lH.l itio ns o f co -presence' arc fo und whe neve r agents 'sense tha t
68 Co nsciousness, Self and Social Encounters Co ffm an ' Enco unrers and Routines 69

they are close eno ugh to be perceived in whatever they are doing , another. G offman's wri tings are cert ain ly relevant to both, even if
including their experi encing of others, and close enough to be he has had a guarded stance towards problems o f long-tenn
perceived in this sensing of being perceived ',4J Although th e 'full institutional process or development .
conditio ns of co-presence' exist o nl y in un med iated co ntac t Finally, it is frequ ently supposed that no t o nly are Coffman's
be tween those who are ph ysica lly present , mediated contac ts that wri tings confined in their relevance to contemporary societies
permit some o f the in timacies o f co-presence are made possible in but they directly ex press features o f conduct which are peculiarly
th e mode rn e ra by e lectro nic commun ications, most no tably the mo dern , even d istinc tively Am erican. Thus Gouldne r,
telepho ne_ 4oI In contempo rary societies, and in diffe ring fo rm ats commenting upon Goffm an's work , says:
in ot he r c ultu res , th e space contained in a room - with
exceptions , such as parties, in which the whole house may be it dwells upon the episodic and sees life only as it is Lived in a
'opened up' - ordinarily defines expected boundaries o f co- narrow interpersonal circumference, ahistorica l and non-institu-
presence _ Of course, th ere a re many 'p ublic places', in jostling tional, an existence beyond history and society.... [h I reflects the
crowds o n the stree ts and so on, in which there is no clea r new world, in which a stratum of the new midd le class no longe r
physical circumscribing o f th e co nditions of co-presence. believes that hard work is use ful or that success depends upon
diligent application. In this new world there is a kee n sense of the
irrationality of the relationship between individual achievement
Goffman: Encounters a nd Rout in es and the magnitude of reward , between actua l contribution and
soc ial regulation. It is the world of the high-priced Hollywood star
Because Goffman has so persistently devoted himself to analysing and of the market for stocks, whose prices bear little relation to
the ro utin es of day-to-day life, his writings offer many l heir earn ings.""
illuminati ons about the character of social integration. Several
misunderstandings about Goffman 's writings need to be countered Gould ner expli citly contrasts this standpoint with what he calls a
before these insight s can most profitabl y be developed. He has to 'structural' approach , to th e detriment of th e former . The social
be rescued here from the im portunate embrace of his admirers. world Coffman portrays is not simply highly culturally specific
Goffman is o ft e n thought o f as an idiosyncra tic o bserver of hut deals o nl y with the transient, no t with the enduring
social life. whose sensiti vity to the s ubtleties of what I have called institutional fonns that mould people's lives. One could not say
practical and discurs ive conscio usness derives mo re fro m a I hat suc h an indictment o f Goffman - in so far as it is an
combinatio n o f an acute in telligence and a playful style th an from indictment - is who lly unjustified . But Gouldner's c ritique also
a co-ordina ted approach [0 social a nal ysis.4~ This is very reveals o nce mo re just that dualism which I have previo usly
misleading and o ne reason why Goffman has no t generally been ~ lI gges t ed is so pervasive in the soc ial sciences. T he fixit y o f
recognized as a social theorist o f considerable stature. I want to institutio nal forms does nO[ exist in spite o f, o r o utside , the
say, in any case, that Go ffm an's writings have a highly systematic e ncou nters of day-to-day life but is implicated in those very
character, and this is in no small degree what gives them their I'llco /ullers.
intell ectu al power. Another misunderstanding, whi ch Goffman T he evanescence of enco unters expresses the temporali ty of
himself has hardly bee n concerned to forestall, is thal his writings th e duree of dail y life and th e co ntingent character of all
are rel evant o nly CO a form of 'microsociology', which can be structu ration. But Goffman makes a very persuas ive case for
cleanly severed from 'macrosociological' issues. A much mo re arguing that the 'fadin g away' inherent in the syntagmatic ordering
interesting way LO approac h Coffman's works is treat th em as , Il l' soc ial interactio n is consistent with a very marked fixity of
being concerned to map o ut the intersections of presence and ' form in social reprodu c tio n. Alth o ugh he does not, to my
absence in social int eractio n. Th e mechanisms of soc ial and kl lo wledge, anywhere cla im thi s, I th ink th at his writings disclose
system int cg rati o n, to repeat, necessaril y int erlace with o ne features o f co-prcse nce thaI are fo und in all societi es, however
70 Co nsciousnes s, Self and Social Enco unters Goffman . Encounters and Routines 71

relevant those same writings indeed may be to identifying novel Gatherings refer to assemblages of people comprising two or
characteristics of the contemporary era. Goffman's work holds more persons in contexts of co-presence. By the term 'context'
up a mirror to many worlds, not just to one. In using ideas (Goffman prefers that of 'situ ation') I mean those 'bands' or
formulat ed therein , nevertheless , I do not want to endorse all of 'strips' of time-space with in which gatherings take place. Anyone
Goffman's own emphases. entering such a band of time-space makes himself or herself
Goffman's writings comprise a major contribution to an 'available' for moving into that gathering or may actually form it
exploration of the relations between discursive and pract ical if it is dyadic in character. Gatherings presume the mutual
consciousness in the contexts of encounters. However, he has reflective monitoring of conduct in and through co-presence.
little to say about the unconscious and may, indeed , reject the T he contextuality of gatherings is vital, in a very intimate and
idea that such a phenomenon has any importance at all in social integral fashion , to such processes of monitoring. Context includes
life. Moreover , Goffman's analyses of encounters presume the physical environment of interaction but is not something
motivated agents rather than investigating the sources of human merely 'in which' interaction occurs (see pp. 118). Aspects of
motivation, as many of his criti cs have complained. The lack is a context, including the temporal order of gestures and talk , are
serious one and one of the main reasons (the other being a routinely drawn upon by acwrs in constituting communication.
disinterest in long-term processes of institutional transformation) T he importance of this for the formulation of 'meaning' in gestures
why Goffman's work has something of an 'empty' feel to it. For and in talk, as Garfinkel has done more than anyone else to
why do the agen ts whose reflective monitoring of conduct is elucidate, can scarcely be exaggeratedY T hus linguists have very
described with so much subtl ety follow the routines that they do? often sought to analyse semantic problems either in terms of the
The question could be answered, up to a point, if it were the case 'internal' linguistic competence of individual speakers or by
that the individuals portrayed by Goffman were represented in a examining the properties of isolated speech acts. But the 'closure
voluntaristic fashion as cynical agents who adapt to given social of meaning' of the polyvalent terminologies of everyday language
circumstances in a purely calculated and tactical way. But achieved in discourse can be grasped only by studying the
although many have interpreted Goffman in such a fashion, this contextual ordering of whole conversations.
is not the main implication which I wish to draw from the terrain Gatherings may have a very loose and transitory form, such as
of study which he has opened up . A stress upon the prevalence of that of a fleeting exchange of 'friendly glances' or greetings in a
tact in social encounters, the repair of strains in the social fabric hallway. More formalized contexts in which gatherings occur can
and the sustaining of 'trust' suggest, rather, a predominant concern be called social occasions. Social occasions are gatherings which
with the prOlection o r social continuity, with the intimate involve a plurality of individuals. They are typically rather clearly
mechanics of social reproduction. bounded in time and space and often employ special forms of
Goffman develops a typology of the contours of interaction, fixed equipment - formalized arrangements of tables and chairs
and I shall employ several of his concepts, modifying them and so on. A social occasion provides the 'structuring social
somewhat, in what follows. The range of concepts can be set out co ntext' (Goffman's term) in which many gatherings 'are likely to
as follows: rorm, dissolve and re-form, while a pattern of conduct tends to be
recognized as the appropriate and (o ft en) official or intended
[CINlrel>en(e 1
one'.48 A whole variety of routinized aspects of daily life, such as
the work day in a factory or office, are of this sort. But there are
$OC ia l occas io ns also many more irregular social occasions, including parties,
dances, sports events and a diversity of other examples. Of course,
unfoc used intera ction
a sector of physical space may simultaneously be the site or locale
focll)t.-d in terac ti OIl : of several social occasions, eac h involving multip le gatherings.
. . ",t""" Bul more ohen than not there is a normatively sanctioned
72 Com ciousne55, Self and Soc ial Encounters Seriality 73

'overriding social occasio n' to which others are supposedly discursive consciousness. This in turn has to be ex plicated in
subordinated in a particular sector of tim e-space. terms both of the control of the body and of the sustaining or
The contex tual characteristi cs of gatherings, wheth er or not rules or conventions. Third, encou nt ers are sustain ed above all
these occur o n social occasions, can be divided into two ma in through talk, through everyday conversation. In analysing the
forms. Unfocused interactio n relates to all those gesrures and communication of mean ing in interactio n via th e use of
signals wh ic h can be communicated between individu als simply interpretative schemes, th e phenomenon o f talk has to be taken
because o f th eir co-presence within a specific contex t. The very seriously, as constitutively involved in encounters. Finally.
physical pro pe rties o f the body and the limited scope o f th e the contextual organization o f encounters must be examined.
positio ning o f the face are major constraints here, Actors' since the mobilization o f time-space is the 'gro unding' of a ll the
generalized awareness o f the presence of o thers may range s ubtly above elements. I shaH undertake this latter task in terms of
over a wide spatial ex tensio n, even including those standing seve ral basic notio ns , those of 'presence-availabili ty' , 'local e' and
behind them. Bu t such 'cueings of the body' are very diffuse the relation of 'enclosure/ disclosure'. Rather than discussing these
compared with those that are possible, and are chron ically latter three concepts in this chapter, however, I shall defer them
utilized, in face-to-face int eractio n. Focused interaction occ urs l until later.
where two or more individuals co-ordinate their activities through
a continu ed intersect io n o f facial expression and voice . However Se riality
much the participants might monitor whatever else is going on in
the wider gathering, focused interaction in some part introdu ces Encounters are sequ enced ph enomena, int erpolated within. yet
an enclosure of those invo lved from o thers who are co-present. A giving form to, the seriality of day-to-day life. The systematic
unit of focused interaction is a face engagement or an encounter. properties of encounters can be traced to two principal
Encounters are the guiding thread of social interaction, the characteristics: opening and closing, and turn-taking. Let me
succession o f engagements with others ordered within the daily look briefly at each of these. Th e duree o f d aily life, as lived by
cycle of activity. Although Gorfman does not include this fo rmally each individual, is a continuous fl ow o f activity, bro ken only (but
within his schema o f concepts , I think it highly important to regularly) by the relative passivity of sleep. The duree of activity
emphasize the fact that encounters typically occur as routines. ca n be 'bracketed' o r 'conceptually segmented', as Schutz says,
That is , what fro m the angle o f the Deeting moment might appear by a reflexive moment o f attentio n o n the part o f the subject.
brief and trivial interchanges take o n much more substan ce when This is what happens when someone is asked by another to
seen as inherent in th e ite rative nature of social life. The supply 'a reason' or 'reason' fo r, o r o therwise to explicate, certain
routinizatio n o f enco unters is o f major sign ificance in binding the features of his or her activity. But the duree o f daily life is also
Deeting encounter to social reproduction and thus to the seeming ' bracketed' by the o pe ning and clos ing o f encounters. In
'fixity' o f institutio ns, Goffman's words , 'One may speak, then, o f opening and closing
I have defined social in tegration as systemness in circumstances temporal brackets and bo unding spat ial bracke(s.'~ Fond as he is
of co-presence. Several phenomena suggest themselves as being o f dramaturgical metapho rs and ana logies, Goffman gives as an
most immediately relevant to the constitution of social integration aample the devices whi ch are e mployed in the opening and
thus defin ed. First, in o rder to grasp the connection of encounters closing of theatrical spectacles. To signal the opening of a play, a
with social reproductio n stretching away over time and space, we hell rings, the lights go down and the curtain is raised. At the
must emphasize how encounters are formed and reformed in the co nclusion the auditorium lights go on again as the curtain falls.
duree o f daily ex istence. Second, we should seek to identify th e Most soc ia l occasion s use so me type o f formal c ueing devices for
main mechanisms of th e duality of structure whereby encounters opening and closing - a characteristi c o f ritual occasions as
are organized in and through the int ersections o f pract ical and mu ch in traditional cultures as in the variety o f more secular
74 Consciousness, Self and Social Encou nters Serialit y 75

social occasions characteri sti c of contemporary socIetIes. Th e are characteristically used to produce a 'conventio nal engagement
bracketing of initiation ceremoni es, for example, typically cues a c1osure'.52 T hat is to say, a no rm atively sanctio ned 'barrier'
dramatic change in th e manner of conduct within the fram e of separates tho se engaged in the encoun ter from o thers who are co-
the occasio n - markers indicating, as it were , a shift from the present. This is a collaborative work . in which participants in the
profane to th e sacred. Caillo is has demonstrated in this regard face engagement and bystanders - o fte n. o f course, in vol ved in
the parallels between, as well as the directly historical influences their o wn engagements with o ther parties - sustain a sort of 'civil
upo n, the spheres o f religio n and 'play'.50 inattentio n' towards o ne ano ther. Goffm an indicates various ways
One might hazard the guess th at bracketing markers tend to be in which this may be achieved and how it may be dislocated. As
regarded by eve ryday acto rs as particularly importa nt when the in all areas of the mutual mo nitoring o f interactio n. there are
activiti es th at occur during th e encounter , or upo n a social ex trao rdinarily complex features even to the manifestatio n of
occasio n, are treated by the parries involved as particula rly 'inattention '. Thus bystanders a re usually expected no t only not
divergent fro m the normal ex pectations of everyday life. G offman to ex plo it a situatio n o f proximity o f prese nce, whereby they
gives this exampl e. In a medi cal examination of the naked body, could fo llow what is going o n in o th er face engage ments, but also
or in the drawing o f the same object in an art class , the individu al actively to demonstrate in attentio n. This can be pro blematic. For
does no t usually shed his or her clo thes in the presence o f the if inattention is too studi ed , the effect may be to suggest that the
other or oth ers, o r dress again in their presence at the conclusio n individual is in fact eavesdro pping.
of the enCOUnLer. Undressing and dressing in private allow the All sorts of complications o f th ese ph enomena are possible.
body to be sudd enly exposed and hidden, both marking the T here may be many circumstances in which an individual may be
boundaries of the episode and conveying that the actions stand interested in overhearing the content of an encounter and may
separate from sexual or oth er connotations that might otherwise very deliberately simulate inattentio n. Ho wever , this runs the risk
be read into th em. This is part o f what G offman calls th e 'keying' of being noticed because o f an artifi ciality of posture o r because
of encounters and suggests a close connection with Wittgenstein's of a host of other traits that can give away what is going o n. The
discussions o f the int erweaving of form s of life. The occurrence point o f this should not be taken to suggest , as many interpreters
of encounters, marked and given a d efinite social 'hu e' o r 'ethos' , of Goffman have te nded to do, that most o f the marvelously subtle
allo ws fo r transfo rm atio ns o f a multiplicity of episodes into intricacies of interactio n are studied o r cynically manipulative .
divergent 'types'. T he o pposite is the case. Wha t is striking about the interaction
skills that acto rs display in the production and reproductio n o f
We (and a considerable number of theys) have the capacity and encounters is their anc ho ring in practical conscio us ness. Tact
inclination to use concrele, actual activ ity - activity thai is rathe r than cynic ism is inhe rent in the structuratio n of
meaningful in its own right - as a model upon which to mark encounters. While the content o f what counts as 'being tactful'
transformations for fun, dece ption, experiment. rehearsal, dream.
may vary widely, the signi ficance o r tact in otherwi se very
fan tasy. rilUal, demonstration, analysis and charity. These lively
shadows of eVents are geared into th e ongoing world but not in different societies or cultures is impossi ble to dispute. Tac t - a
quite the close way that is true of ordinary, literal activity.51 lalent co nceptual agreemen t amo ng participa nts in interaction
co ntexts - seems to be the main mechanism th at sustains 'trust"
Most of the enco unters th at comprise the seriality of social life or o ntological sec urity over lo ng time-space spans. Tact in the
take place either o utside (i n lime-space) or against the backdrop suslaining of conventional engagement enclosure becomes clearly
of the gath erings fo und o n social occasions. Face engagements in point ed up in circumstan ces which threaten to fracture such
many of these co ntex ts do no t invo lve clear enclosures which cut clos ure . Thus in very constricted spaces, such as lifts, it is virtually
off th e inl eraction from no n-parti cipants. In such circumsta nces impossible to sustain a posture o f not liste ning. In Anglo-American
the refl ex ive mon ito ring o f th e bo dy, o f gesture and pos itio ning, society, at least. th e te nd ency in such a situat ion is to suspend
76 Consciousness, Se lf and Soc ial EncounteLs Seria lity 77

commun ication, with perhaps only the occasional comment that same tim e it is made clear to oth ers that th is is in o rder to sustain
indicates that an encounter is suspended rather than broken off. eye contact in an engagement where the positioning of others
Similarly. if three people are talking and one is interrupted to threatens to block the view. Such movements may be carried on
take a pho ne call , the others cannOI feign complete inattention in an exaggerated fas hion , in fa ct. thus indicating to others thal
and may carry o n a sort of hesitant , limp conversation.S.J Contexts the actor making them is aware that such body mo tion wou ld
o f encounters such as these may directl y express asymmetries of us ually be looked upon as odd.
power. Thus if, say, two individu als in a lift continue ( 0 carryon Turn -taking in encounters has been much studied by writers of
their talk regardless of their surroundings of overly close prox im ity an e thnomethodological bent. ~~ Their work is orten decried as
to o thers, it may very well be that they thereby demonstrate to trivial. But this is a short-sighted assessment indeed. For tum-
those who are their subordinates or inferiors their indifference to taking is rooted in the most general propert ies o f the human body
the sustaining o f civ il inattention in such a context. However , and hence expresses fund amental aspects o f th e nature of
they may nevertheless betray a certain concern about deviating interactio n. Moreover, turn-taking is one major feature of the
from a norm that ord in ari ly would be observed, and hence th ey serial character of social life, he nce connect ing with the overall
may talk even more loudly than they would in other character of social reprod uction. Turn -taking is one form of
ci rcumstances. 'co upling constraint' . deriving from the simpl e but elemental fact
Encounters invo lve 'spacing', as regards both the position of that the main communi ca tive medium of hum an beings in
bodies in relation to o ne another, inside and outside the region of situations of co-presence - talk - is a 'single-order' medium.
face engagement. and the serial spacing of contributions to the Talk unfolds sy ntagmati call y in th e fl ow o f th e duree of
encounter in terms of seriality or turn-taking. Co llabo rative intcractio n, and since on ly one person can speak at one time if
spacing within locales is Obviously relevant to the bracketing of communicative intent is to be realized, contributions to encounters
encounters (and. I shall try to indicate later. is subject to what are in evitably serial. It should be said that the empiri cal study of
Hagerstrand calls 'coupli ng constraints' and 'packing constraints'). conversations shows that they have a much less symmetrical form
The generalized normat ive sanctions influencing acceptable Ihan might be supposed. The managing of turn-taking rarely
proximity of individuals in public places does vary cross-cuiturally, happens in such a way that participants finish sentences. There is
as do sanctio ns affecting the limits of acceptable bod ily contact a pletho ra o f hesitation phenomena ; speakers break into what
between persons in varying contexts. S4 But spac ing can be ano ther is saying , such that there are no clear divisio ns in the
effectively organized only within the limits o f 'easy talk' - not so taking of turns and so o n . ~
far apan that participants have to shout and no t so close that the Turn -taking may apply to the seriality of encounters as well as
ordinary cues of facia l expression , which help to monitor the to the interaction between agents within encounters and may be
si ncerity and authe nticity of what is said, cannot be observed. again closely bound up with differen tials o f power. All organiza-
Face engagements, when o th ers are co-present, are almost always tions invo lve the co-ordination of interaction in nows of time-
carried on wi th some turning of the body away from those who space re lations 'c hannelled ' through regularized contexts and
are not party to th e engage ment. and the arrangement of bodies locales (see pp. I 19ff). Thus the process of o rganizing trials in the
is such that there is no physical barrier to the free exchange of daily life of the courtroom has a formalized serial character, in
glances or visual contact. This may be difficult to achieve in which one case is heard. and bracketed as a definite social
crowded sit uations in whi ch th ere is quite a lot of movement - at occasion, while the parti es invo lved in the next are lin ed up in the
a party. fo r instan ce, or in a crowded train. In such contexts there adjo ining waiting room. There are very many similar ex amples in
may be some trans itory re laxat io n of the sa nctions whic h soc ieties of broad lim e-space distanciation. Sa rtre's discussion of
ordinarily co ntro l excessive mobility of the limbs. A person may se ria lity here has a direct co nn ection with the seeming triviata of
quite acceptably sway thc body about in this situation, if at th e conversational turn-laki ng. Sart re poin ts o ut thai a banal example
78 Consciousness, Self and Social Encounters Talk, Reflexivity 79

o f seriality , a que ue for a bus, can be used to demonstra[e the promised , in a generic way amo ng the 'mentally ill' and transitorily
mutua] coupling o f time-space relatio ns o f presence and absence: in bodily and verbal lapses o r slips.
For Goffman 'me ntal illness', even the most serio us fo rms o f
these separate people form a group, in so fa r as they are all
'psycho tic disturba nce', are exempli fied above all in inability, or
sta nding on the same pavemen t, which protects th em from the
traffic crossing the square, in so far as they are grouped around the unwillingness. to accept the d iversity of m inu te (altho ugh wholly
same bus stop, etc... . T hey are all, or nearly all, work ers, and untrivi al) forms of mo nitoring of bodily movem ent and gesture
regular users of the bus service; they know the timetable and which are the normative core of day-to-day interaction . Madness
frequency of the buses; and conseq uently they all wait for the same is a cl uster of 'situatio nal impro prieties',S8 Psyc hotic behaviour
bus: say, the 7.49. This object in so fa r as they are depe ndent upon d iverges from, or actively clashes with, the publi c o rdering of
it (breakdowns, failures, accidents) is in their p resent interest, But time-space relatio ns, via the body and its media. where by human
this present interest - since they all live in the district - refers be ings 'get o n with o ne ano ther' in circumstances o f co-presence.
back to fuller and deeper structures of their general interest: T he 'mentally ill' do no l confo rm to the extremely ught (and
improvement of public transport, freezing of fares, etc. T he bus continuo us) bodily contro l dem anded of 'normal individuals';
they wait for unites them, being their in terest as individ uals who they do nor respect the intricacies o f the for mul ae governing th e
this morning have business on the n've dro ite; but, as the 7.49, it is formation , maintenance, breaking off or suspension of encounters;
their interest as commuters; everything is temporalized: the and they fail to contribute the manifold forms o f tact that sustain
traveller recognizes himself as a reside nt (that is to say, he is 'trust'.509 Individuals are very rarely expected 'just' to be co-present
referred to the five or ten prev ious years), and then the bus in gatherings and never are pe rmitted to act thus in encounters.
becomes characterized by its daily eternal return (it is actually the T he reflexive monito ring o f actio n, in contex ts of co-presen ce ,
very same bus, with the same dri ver and conductor). The object demands a sort o f 'controlled alert ness': as Go ffm an expresses it ,
takes on a structure which ove rflows its pure inert existence; as actors have to 'exhibit presence'. This is exactly what many
such it is provided with a passive future and pasl, and these make it 'me ntal patie nts' - fro m those in a state o f a pparent catato nic
appear to the passengers as a fragment (an insignifica nt one) of stu por ro those who move o nly mechanically, as if driven by some
their deslinyY force, rather than being ordinary human agents - do not do. 60
The exhibiting o f presence takes quite artfully deliberate forms
Tal k, Reflex ivity but is undeniably exem pli fied first and fo remost in practical
co nsciousness. Consid er perso nal appearance and th e visible
Goffman's most telling co ntributio ns to understanding the marks of dress and bodil y adornment. Concern with a ppearance
susta ining and reproduction of e nco unters are to do with the is manifest , for example. in th e care wi th which an individual
relatio n between the reflexive control of the body - that is to selects and arranges types of clo thing or adornment in relation to
say, the refl exive self-mo nito ring o f gesture, bodily movement participatio n in partic ular COnlexts o f activity. But it would be
and posture - and the mutual co-ordinarion o f interactio n very misleading to suppose that such car~ is the proto ty pic~1
through tact and respect fo r the needs and demands o f o thers , mode o f sus taining bodily idio m. More basiC, mo re complex , IS
The prevale nce of tact , trust o r ontological security is achieved the chronic monito ring of the arrangeme nt o f clothing. in relation
and sustained by a bewildering range of skills which agents de ploy to bod ily posture, in th e presence of others. Thus 'm ental patients'
in the production and reproduction o f interaction. Such skills are may sit slackly, their clothing disarranged or cru mpl ed; women
founded first and forem ost in the normatively regulated co ntrol may not observe th e usual ex pectation in Western societies, to
of what might seem, even more than turn-taking, to be th e tiniest, kee p th e legs closely toge th er whe n wearing skirts, and so on.
most insignificant details o f bodily moveme nt or ex pression, T his T here is a fund ame ntal di fference betwee n bohemians o r hobos,
is readi ly demo nstra ted whe n these are lacking o r are com- who fl o ut the conve ntio ns o f Ihe wider soc ie ty in their modes o f
80 Comciou.me55. Self and Soc ial Encou nters Tal/< , Reflexivity 81

dress and other modes of conduct, and the 'mentally ill'. For the grasp. Such events, which are characteristic features of daily life
normative expectat ions in which bodily control and appearance on the wards, tend to run counter to the presumption of general
are grounded concern not merely the trappings of adornment or communality of interest which staff ordinarily wish to foster. The
gross param eters of moto r behaviour but precisely the kind of final precipitous depanure of the staff member exemplifies
'sustained control' which simultaneously 'carries' and demonstrates circumstances which , in the world o utside , are likely to occur
agency. only where the individual attempting to leave in such a manner is
That such c hro nic self-monitoring is not undemanding is demonstrating rejection o f a strong mo ral tie - e.g. a love
indicated by the pervasive importance of 'back regions' - found relatio nship - to which th e purs uer lays claim. Such an
in varying contexts in all societies - in which control of bod ily implication, of course, is not necessarily lost upon the 'mental
posture, gesture and apparel can be in some degree relaxed. But patient' on the ward of a hospital. Indeed. many apparently
even whe n a lo ne an individual may maintain presentability. Fo r bizarre elements of encounters between the sane and the mad
someone who is discovered inadvertently 'unassembled' cedes to seem to represent 'experiments' wh ich the latte r carry out upon
others aspects o f se lf that are perhaps only visible at such the usual frameworks of encounters. 'Schizophrenics', as Laing
moments. 61 The po int is that the sustaining of 'being seen as a says, are perhaps aptly regarded as taking seriously, on the level
capable agent' is intrin sic to what agency is, and that the motives of practical consciousness and in their actual conduct, so me of
which prompt a nd rein fo rce this co nnection as inherent in th e the questions that philosophers pose hypothetically in the solitude
reproduction of soc ia l practices are the same as those which of their studies. They really worry about, and build their activities
order such reproduction itself. The strongly sanctioned character around, heterodox solutions to qu estions such as ;In what sense
of these ph eno mena is we ll brought out in the following am 1 a person?', 'Does the wo rld on ly exist in so far as I perceive
observations: it ?' and so on. 63 But most of the 'experimental activities' of the
mad, significantly, are to do with the cueings and the normative
Bodily idiom, then, is conventionalized discourse. We must see ' sanctions associated wi th the complexities of bodily con trol within
that it is, in addition, a normative one. ThaI is, there is typically an the immediacies of encounters. Garfinkel's 'experiments with
obligation to convey certain information when in the presence of trust' duplicate some of th e jarring feelings o f diSQuietude which
others and an obliga tion not to convey other impressions ..
'normal ' individuals experience when the routines o f daily life are
Although an individual can stop talking, he cannot stop communi'
cat ing through body idiom ... Pamdoxically, the way in which he called in question .60
can give leasl in formation about himself - although this is still Many o r these considerations apply to talk as the discursive
appreciable - is to Iii in and act as persons of his kind are medium of communicative intent in contexts o f co-presence.
expected 10 aCLM Discussion of 'response cries' (fo rms o r utterance that are no t
talk ) can provide an appropr iate transition to the study of talk.
Many 'mental patients' have difficulty with , or flout , the norms Such cries demonstrate once more that what may seem entirely
associated with th e o pening and closing of encounters. Thus a trivial and wholly 'spontaneous' characteristics o f human conduct
person on the ward o f an asylum may hold one of th e staff in an are tightly ordered nomatively. Respo nse cries transgress the
encounter no matter ho w many indi cations the staff member may no rmative sanctions against no t talking to o nese lf in public.
give that he or she wi shes to move o n. The patient may pursu e Co nsider 'OOpS!,6S ;Oops!' might be th o ught of as a pure reflex, a I
the other closely, regardl ess of how rapidly the person walks, and mechanical response like blinking the eyes when someone moves
might then try to acco mpany th e orderly through the door at the a hand sharply towards another's fac e . But this seemingly
end of the ward, even if it is a locked ward. At such a point th e involuntary reaction lends itself to det ail ed analysis in terms of
staff member may hav e physically to restrain the patient from agency Hnd th e body. Wh en so meo ne exclai ms 'Oops !" on
followin g, perhaps tearing himself or herself away from the other's dropp ing somet hing or knocking somethin g o ver it might appear
82 Con sciousness, Se lf and Social Encounters Positioning 83

at first sight as jf the so und advertises a loss of control. thu s 'contextuali ty': the context ualit y o f talk, lik e the contextuality of
drawing attention to an inference which the person would wish to bodily post ure, gesture and mo vemen t, is the basis upon which
avoid , a dislocation in the routine forms of control that indi cate such ph eno mena are co-ordinated as enco unters extending in
refl ex ively mo nitored agency. But the exclamation in fac t shows time-space. Talk is an intrinsic feature o f nea rly all encounters_
to Olhers that the occurrence in qu estion is a mere accidenr , fo r and also displays similarities o f systemic form. Talk ordinarily
which the indi vidual ca nnOt be held responsible. 'Oops!' is used manifests itself as conversation. 'Conversatio n' admits of a plural ,
by the agent to display that the lapse is only thm , a mo mentary which indicates that conversatio ns are episodes having beginnings
and contingent event , fa ther than a manifestation o f either a and endings in time-space. Norms o f talk pertain not only to what
more generalized incompetence o r some opaque intent. But this is said , the syn tactical and semantic fo rm o f utterances , but also
also hides a range of other subtle shadings and possibilities. Thus , 10 the routinized occasio ns o f talk. Conversations, or units of
for example. 'Oops!' is used - and is known to be used - o nly in talk , involve standardized opening and closing devices, as well as
situations o f minor fai lu re rather than in those of majo r calamity. devices for ensuring and displaying the crede ntials of speakers as
Hence 'Oops!', spo nta neo us and immediate though it may be, having the right to contribute to the dialogue. The very term
demonstrates care and attention to the implications of the sudden ' bracketing' represents a stylized insertion o f boundaries in
occurrence and th erefore indicates overall competence which writing. Let me give Goffman th e las t wo rd in th e bracketing that
overrides what is there by ex hibited to be only a minor slip . constitutes this section. What is talk , viewed interact ionally? 'It is
There is mo re. 'Oops!' can be constru ed as a warning to others. an example of that arrange me nt by whi ch individuals come
A hazard exists in th e milieu of co-presence, and others in th e toge ther and sustain matters having a ratified , jo int, curren t, and
vicinity would do well to take care. When someone has a minor running claim upon attention, a claim which lodges them together
mishap th e excla matio n 'Oops!' may sometimes be offered by a in some sort of intersubj ective , mental world,'67
participant rath er than by the individual experiencing it. The
'Oo ps! ' perhaps sounds a warning to the other at the same tim e as Pos ition ing
conveying the assurance th at the slip will not be treated by th e
observer as compromising the o th er's competence as a responsible Social systems , I have emphasized , are o rganized as regularized
agent. 'Oops!' is normally a c urt sound. But the ' 00' in it may be social practices, sustained in encounters dispe rsed across time-
more pro lo nged in some situatio ns. Thus someone may extend space. The actors whose conduct constitutes suc h prac tices are
the sound to cover a part o f a task or enterprise in which a positio ned ', however. All actors are positio ned o r 'situated' in
particular hazardo us mo ment has to be overcome for its successful time-space , li ving along what Hagerstrand ca lls their time-space
execution. Or a pare nt may utter an ex tended 'Oops !' or paths, and they are a lso positio ned relatio nally, as the very term
Oopsadaisy!' when playfully tOSSing a child in th e air, the sound social position ' suggests. Social systems on ly exist in and through
covering th e phase when the child may feel a loss of control , the continui ty of social prac tices, fadin g away in time. But some
reassuring it and perh aps at the same time helping to fa cilitate a of th eir structural properties are best charac terized as 'position-
developing understand ing of the nature of response cries.66 practice' relations. 6tI Social positions are constituted structurally
. .',?ops!' thus turns o ut to be not as distant from talk as might as speci fi c intersections of signification, domination and legitima-
mltlally be supposed, sin ce it participates in that very publi c tion which relates to the typificatio n of agents. A soc ial position
character of communi catio n, intersecting with practices, which involves the specification of a definit e 'identity' within a network
Wittgenste in id entifi es as th e fo undation of language use. In th e o f social relations, that identity, however, bei ng a 'category' to
light of the preced ing discussion in this chapter, it should be clear which a particular range of no rmative sanctio ns is relevant.
that th e ind cx icaJit y o f o rdinary language is a 'problcm' neith er Sin ce Linton the concept o f soc ial position has ordinarily been
for lay speak crs nor for philosop hi cal analysis. 'Index icaJity' means associated wit h that of rol e .lOd the latt er has received far more
84 Consciousness, Se lf and Socia l Encuunters POsitioning 85

discussion and analysis than the former. 69 J do not intend to Significance of POSitioning in this most rudimentary sense is
survey this discussion, only to emphasize some reservations about obviously closely bound up with th e level of time-space
the notion of role. The concept is connected with two apparently distanciation of societal totaliti es. In those societies in which
opposed views , about each of which I have some unease. One is social and system integration are more o r less equivalent ,
that of Parsons, in whose theory role is fundamental as the point positioning is only thinly 'layered '. But in contemporary societies
of connec tio n between mo tivation , normative expectations and individuals are positioned within a widening range o f zones, in
'values', This version of th e ro le concept is much too closely ho me. workplace, neighbo urhood. c ity, nation-state and a
bound up with the Parson ian theorem of the dependence of worldwide system , all displaying features o f system integration
soc ieml integratio n upon 'value consensus' to be acceptable. The which increasingly relates the minor d etails of daily life to social
other is the dramaturgical viewpoint fostered by Goffman , about phenomena of massive time-space ex tension.
which mo re will be said in th e next chapter, for here we reach [h e Positioning in the time-space paths of day-to-day li fe , for every
limits of his views. The two conceptions might seem to be contrary individual. is also positioning within the 'life cycle' o r life path.
to one another but actually have a definite affinity, Each tends to T he formation of an 'I' is perh aps found ed on the original
emphasize the 'given' character of roles, thereby serving to express narcissism of a ;mirror phase' in personality d evelopment. The
the dualism of actio n and structure characteristic of so man y child forms the capability o f becoming a reflexive agent through
areas of social theory. The script is written, the stage set, and the positioning of the body in relation to its image. Th e very
actors do th e best th ey can with the parts prepared for them. co nnotation of T as a shifter necessarily relates self to positioning
Rejecting such standpo ints does not mean dispensing with the within the seriality of disco urse and action. Positioning along the
concept of role entirely , but it does imply regarding the life path, of course, is always closely related to th e ca tegorizing of
'positioning' of actors as a more important idea. For definitiona l social identity, 'Childhood' and 'adulthood ', among a number of
purposes 1 shall adopt the formulation I have offered in a previous o ther possible forms of age grading, always mingle biological and
work. A social posi tion can be regarded as 'a social identity that social criteria of ageing, Differential positioning on th e life path is
carries wi th it a certain range (however diffusely specified ) of the major constraining condition influencing th e fundamenta l
prerogatives and obliga tio ns that an ac tor who is accorded that Significance of the family in conjoining ph ysica l and social
identit y (or is an "inc umbe nt" o f that position) may ac tivate or reproduc tion , A human socie ty in which all members were born
carry out : these preroga ti ves and obligations constitute the ro le- as a single age cohort wou ld be impossible , since th e human
prescriptio ns associated with that position. 1O infant has such a long period o f more or less complete dependency
'Position' is best understood as 'positioning', allowing the second upon the ministrations of its e lders.'
of these terms to mine a ric h vein of meanings. Ac tors are always But it is the intersection between these fo rms of positioning
positio ned in respect o f the three aspects of temporality around and that within the longue duree of institutions which creates the
which the theory o f struct uration is builL The positioning of ove rall framework of soc ial positioning, Only in th e con text of
agents in circumstances of co-presence is an elemental featur e of such intersection within institutionalized practi ces can modes of
the structuration o f encounters. Positioning here involves many lime-space positioning , in relation to th e duality o f structure, be
subtle modalities of bod ily move ment and gesture, as well as the properly grasped. In all societies it see ms to be the case that age
more general motion of th e body through the regional sectors of (or age grade) and gender are the most all-embracing criteria of
daily routin es. The positioning of actors in the regions of their attributes of social identity, But altho ugh it is common in the
daily time-space paths, of co urse, is their simultaneous positioning sociological literature to speak of age ro les, gender roles and so on
within the broader regionalization o f societal totalities and within in a ge neric way, I sha ll not foll ow such usage. Social identity
intersoc ie tal systems wh ose broadcast span is convergent with th e conferred by age or ge nd e r - and o th e r supposedly 'ascriptive'
geopo litica l distributio n of social systems o n a global sca le. T he characteristics, such as skin pigmentatio n - tend to be the focus
86 Co nsc io usne55. Se lf and Soc ial En counters POSitioning 87

of SO many aspects of conduct that to employ the term 'role' to It is of the first impo rtance to emphasize that a theory of
describe them is both misleading and superficial. 71 T he notio n of ro utine is not to be equated with a theory of social stability. The
role , as many critics of its pro nigate use in the social sciences concern of structuratio n theory is with 'order' as the transcending
have po inted out, has some conceptual precision only if applied of time and space in human social relationships; rou tiniza tion has
in contexts of social interaction in which the normative rights and a key role in the explication of how this comes about. Routine
obligations associated with a specific identity are relatively clearly persists through social change of even the most dramatic type,
formulated. As its dramaturgical origins indicate, it is useful to even if, of course, some aspects of taken-for-granted routines
speak of role only when th ere are definite settings of interaction may be compromised. Processes of revolution , fo r example , no
in which in the normative definition of 'expected ' modes of doubt usually dislocate the daily activities of multitud es of people
conduct is particularly strongly pronounced. Such settings of who eith er are caught up in the fervour of revolt or are the
inte raction are virlUally always provided by a specific locale or luckless victims of social events which they have had no part in
type of locale in which reg ula rized encounters in conditio ns of initiating. But it is in circumsta nces in which the texture of day-
co-presence tak e place.7J Settings of this sort tend to be associated to-day life is attacked froma lly and systematically deformed - as
with a more clear-cut closu re of re lationships than is fo und in in the concentration camps - that the ho ld of routine is more
social systems as a whole. substantively broken. Even here , as Bettelheim demonstrates so
'Positio ning' gets at wh at I shall call the contextualities of we ll , ro utines, including those of an obn oxious sort, are re-
interaction and allows us to spe ll ou t, in a direct way, the established.
relevan ce of Goffman's work fo r stru cturation theory. All social It is instructive to see the rules implicated in e ncounters, as
interaction is situated interaction - situated in space and time. It Goffman suggests, as being clustered in fram eworks or 'frames '.
can be und erstood as the fitful yet routinized occurrence of Framing may be regarded as providing the ordering of activities
e ncounters, fading away in time and space, yet constantly and meanings where by ontological security is sustained in the
reconstituted within differe nt areas of time-space. The regular o r e nactment of daily ro utines. Frames are clusters of rules whic h
routine features of e ncounters, in Lime as well as in space, help to constitute and regulate activities , defining th em as
represe nt institutionalized fea lures of social systems. Ro utine is activities of a certain sort and as subject to a give n range of
founded in tradition, custom or habit , but it is a major error to sanc ti o ns. Whenever individuals come togethe r in a speci fic
suppose that these phenomena need no explanation , that they a re context they confront (but, in the vast majority of circumstances,
simply repetitive forms of behaviour carried out 'mindlessly'. On a nswe r without any diffic ulty whatsoever) the question 'What is
the contrary , as Goffman (together with ethnomethodo!ogy) has going on here?' 'What is going on?' is unlikely to admit of a simple
helped to demonstrate , the routinized character of most social answer because in all social situations there may be many things
activity is something that has 10 be 'worked at' continually by 'going on' simultaneously. But participants in interaction address
those who sustain it in their day-to-day conduc t. One of the most this question c haracteristica lly o n the level of prac tice , gearing
striking gaps in Goffman 's writings is the absence of an account the ir conduc t to that of o thers. Or, if they pose such an question
of motivation. In the preceding sectio ns I have sought to remedy discursively, it is in rela tio n to one particular aspect of the
this by suggesting that trust and tac t, as basic pro perties which situation that appea rs puzzling or disturbing . Framing as
participants bring to encounters, can be interpreted in terms of co nstitutive of, and constri cted by , encounters 'makes sense' of
the relation between a basic sec urity system, the sustain ing (in th e activities in which participants engage , bo th for themselves
praxis) of a sense of ontologi ca l security, and the ro utine nature and for others. This includes the 'literal' understanding of events
of social reproduction whi ch agents skilfu lly organ ize. The but also the criteria by which it is made plain that what is going
moni tori ng of th e body, the control and use of fa ce in 'fa ce work ' o n is humo ur , play, th eatre and so on.
- these are fundam ental to social integratio n in time and space. Primary frameworks of daily ac ti vity can be seen as those
aa Consciousne ss, Self and Socia l Encounters Positioning 89

generating 'literal' languages of description both for lay participants formed a 'vocabulary of motive' wh ereby both staff and inmates
in encounters and for social observers. Primary frameworks vary in terpreted actions , especiall y deviant or problematic ones . It
widely in their precision and closure. Whatever its level of was not treated simply as a description o f what was tacitly
organization. a primary framework allows individuals to categorize acknowledged; rather, [he circ umstances in which the code was
an indefinite plurality of c ircumstances o r situations so as to be called upon could be altered by the fa ct o f in vok ing it. 'Telling
able to respond in an appro priate fashion to whatever is 'going the code' meant , as the phrase sounds, not o nly reporting upon
o n'. Someone who finds that what is going o n at a particular time what the code is but re primanding those who contravened it ; it
and place is, say , a party , may be able to bring into play conduct exhibited the code as a contro l device . that exhibiting being part
o f an apposite kind even if some aspects of the contexts a re o f how it in fa ct operated as suc h. I wou ld s uggest that this is
unfamilia r. Most of Goffman 's work is to do with rules which characteristic of 'rule interpretations' discu rsively offered in many
allow for transi tions to be made between primary and secondary social contexts.
fram eworks. Thus the 'keys' in transformations are the formul ae Rules applied reflexive ly in circumsta nces of co-prese nce are
whereby an activity that is already meaningful in a primary neve r limited in their implications to speci fi c enco unters but
fram ewo rk is given a meaning in a secondary on e 74 For example, apply to the reproductio n of the pa tte rning of e ncounters across
a fight can be 'play', an apparently serious comment a joke. But time and space. The rules of language , o f primary and secondary
exactly the same kind of analysis could be carried out to indicate framing, of the conduct of interpe rso nal interac tion all apply
the rules involved in transitions between different primary o ver large arenas of social life, although th ey canno t be taken as
framewo rks. necessarily coextensive with any given 'society'. Here we have to
It wo uld no t be releva nt to pursue the detail of Goffman's give some attention to conceptually differentiating between 'social
analysis of framing any furth er in this context. Let me instead interaction' and 'social relations' (altho ugh I shall no t a lways be
briefly consider the significance which the discursive fonnulation parti cularly careful to separate them subseq ue ntly). Social
of rules can have by taking a different piece of work, that of interaction refers to encounters in which individuals engage in
Wieder o n 'telling th e code'.n Wieder's research reports th e situations of co-presence, and hence to social integratio n as a
results of a part ic ipant observatio n study in a residential unit for level of the 'building blocks' whe reby th e institutions o f social
rehabilitating paro led prisone rs. The inmates spo ke o f the systems are a rtic ulated . Social relations are certainl y in volved in
existe nce o f rules o f conduct which they called the 'code'. The the structuring of interaction but are also the main 'building
code was explicitly verbalized but not, of course, fo rmalized in bloc ks' arou nd which institutio ns are a rti culated in system
written form as it was established and co-ordinated by inmates, integration . In teract io n d epends upo n th e 'positio ning' o f
not [he staff. No inma te could apparently rec ite all the maxims individuals in the time-space contex ts o f activity . Social relatio ns
making up the code, but all could mention some , and the code concern the 'positioning' of individuals within a 'social space' of
was freq uen tly discussed. It was made up of such rules as: do not sy mbolic categories and ties. Rules invo lved in social posit ions
'snitch' (inform about o th er inmates to staff); do not 'cop o ut' are normally to do with the specificatio n o f rights and o bligations
(Le., admit guilt or respo nsibility for an act defined by staff as relevant to persons having a partic ular social identity, or belonging
illegitimate); do not steal from other inmates; share with o th ers in a particular social category. Th e normative aspects o f such
any unexpected gifts or benefits which might be received; and so rules, in other words, are particularly pronounced, but all the
on. Staff knew the code too and made use of it in their dealings previously stated characteristics o f rules apply to them too. They
with inmates. As Wied er says , 'It was used as a wide-reaching may, for example, be tacitly followed rather tha n discursively
scheme of int erpretati on which "struct ured" their environm e nt.' 7~ fo rmulated. There are many such cases in the anthropological
But , as he also points o ut, its verbalization meant that it was lit e rature. An instance is cultures in which th ere is unilateral
in voked in ways that implicitly formulated rules cannot be. It cross-cousin marriage. Alth o ugh the members of th ese cultures
90 Consciousness, Se lf and Socia l Encounters Positioning 91

obviously have some ideas which they put into effect about who in contexts remote from his or her experience. How far the
marries who, th e rules of eligibility that th ey are in fact followin g agent's social skills allow immediate ease in c ulturally alien
in their behaviour are tacit rather than explicit. contex ts is obviously variable - as, of course, is the meshing of
Goffman demonstrates that social integration depends upon different forms of convention ex pressing dive rgent boundaries
(he renexive ly applied procedures of knowledgeable agents, but between cultures or societies. It is not just in knowledge - or
he does not indicate in any effective way what are (he limits or belief claims - whic h agents are able to fonnulate discursively
the bounds of suc h know ledgeability , nor does he indicate th e that they display awareness of broader conditions of social life
forms which such knowledgeability takes. I want to pose such a o ver and above those in which their own activities take place, It is
questio n he re: in what sense are agents 'knowledgeable' about often in the manner in which routi ne activities are carried o n, for
the characteristics of the social systems they produce and example, that actors in circumstances of marked social inferiority
reproduce in their action '? make manifest their awareness of their oppression. Goffman's
Let us presu me that 'knowledge' equals accurate o r va lid writings are replete with commentaries o n this type of pheno-
awareness - I do no t say 'belief, because beliefs are o nly o ne menon. But in other respects when we speak of 'the knowledge
aspect of knowledgeability. It does not make sense to treat actors have of the societies of which they are members' (and
pract ical consciousness as exhaustively constituted by propo- others of which they are not), th e reference is to discursive
sitional beliefs, although some elements could in princip le be conscio usness. Here there is no logical difference between the
thus formul ated. Practi ca l consciousness consists of knowing the criteria of validity in terms of which belief-cla ims (hypotheses,
rules and th e tacti cs wh ereby daily social life is constituted and theories) are to be judged in respect of lay members of society
reconstituted across tim e and space. Social actors can be wro ng and social observers.
some of th e time about what these rules and tactics might be - in What - on a general plane, at any rate - are the types of
which cases their errors may emerge as 'situational impropri eties'. circumstance that tend to influence the level and nature of the
But if there is any continuity to social life at all, most actors must 'penetration' actors have of the conditions of system reproduction?
be right most of the tim e; that is to say, they know what (hey are They include the following factors:
doing, and they successfull y comm unica te their knowledge to
( I) the means of access actors have to knowledge in virtue of
others. The knowledgea bility incorporated in the practica l
acti vities which make up the bulk of daily life is a constitutive their social location ;
feature (together with power) of the social world . What is known (2) the modes of articulation of knowledge;
(3) circumstances relating to the validity of the belief-claims
about the social world by its constiLuent actors is not separate
from their world , as in the case of knowledge of events o r o bjects taken as 'knowledge';
(4) factors to do with the means of dissemination of available
in nature. Testing out just what it is that actors know, and how
they apply that knowledge in their practical conduct (which lay knowledge.
actors engage in as we ll as soc ial observers), depends upon using Of course, the fact that all actors mo ve in situated contexts
the same materials - an understanding of recursively organized within larger totalities limits the knowledge they have of other
practices - fro m whi ch hypotheses about that knowl edge are contexts which they do not directly expe ri ence. All social actors
derived. The measure of th eir 'validity' is supplied by how far know a great deal more than they ever d irectly live through, as a I
actors are able to co-o rdinate their activities with others in such a result of the sedimentation of experience in language. But agents I
way as to pursue th e purposes engaged by their behaviour. whose lives are spent in one type of milieu may be more or less
There are, of cou rse, pote ntial differences between know ledge ignorant of what goes on in others. This applies not only in a
of the rules and tac Li cs of practical conduct in th e milieux in 'Iateral'sense - in the sense of spatial se paratio n - but also in a
whi ch the agent moves and knowledge about those which apply 've rtica l' o ne in larger societies. T hu s those in elite groups may
92 Co nsciousness, Se lf and Social Encounters

kn ow very littl e about how others in less privileged sectors live,

and vice versa. However, it is worth mentioning that vertical
segregation of milieux is nearly always also a spatial segregation.
In category (2) abo ve I mean to refer both to how far belief Criti ca l Notes: Freud on Sli ps of t he Tongue
claims are o rdered in terms of overall 'discourses' and to the
nature o f different discourses. Characteristic of most commo n-
sense, everyday claims to knowledge is that they are fo rmulated As an example of some o f th e no tions analysed in this c hapter I
in a fragmentary, disloca ted way. It is no t only the 'primitive' who propose to consider interpretat io ns o f sli ps of the to ngue in
is a bricoleur: much day-to-day talk among lay members o f all discourse. What Freud calls 'parapraxes' (Felrlleistungen) refer
societies is predicated upo n claims to knowledge that are disparate no t just to verba l infelici ti es but to misw riting. misreading,
o r le ft unexamined. Th e emergence of discourses of social mishearing and to the temporary forgett ing of names and other
science, however , clearly influences all levels of soc ial inte rpre- items. Freud treats these as belo nging together in some part
tation in societies wh ere it has become influential. Goffman has a because the terms designating them have a simil ar root in German,
large audience , no t limited to his professional sociological all beginning with the syll able Ver- (Versprechen, Verlesen,
colleagues. Verhdren, Vergessen ). All parapraxes invo lve errors, but most
So far as (3) is con ce red , it is enough to point out that refer to seemingly unimpo rtant o nes whic h are without lasting
individuals may opera te with false theories, descriptio ns o r significance in the activities o f th e individuals who commit them.
accounts both o f th e contexts o f their own action and of the 'O nly rarely', Freud writes, 'does one of them, such as losing an
characteristics of more encompassing social systems. There are o bject, attain some degree of practical importance. For that
obvious sources of possible tension here between prac tical and reason , too, they attract littl e attention, give rise to no more than
disc ursive consciousness. These can have psychodyn amic o rigins, feeble emo tions , and so on.'!* In fact , he tries to demonstrate ,
in repressions which separate off or muddle th e reasons why these minor infractions supply clu es to key characteristics of the
people act as th ey do and what they are inclined or able to say psycho dynamics o f perso nality .
about those reasons. But obviously there can be more systematic Whether o r not parapraxes do actually fo rm a single class o f
social pressures that can influence bow far false beliefs are held errors I shall not be concerned to discuss here. I shall concentrate
by the members o f a society about features o f that society. o nl y upon slips of the to ngue. Employ ing a classification
Particularly influentia l in respect of (4) , it is almost needless [ 0 established by the linguist Meringer and by Mayer , a psychiatrist
say, are the relat io ns, histo rically and spatially, between o ral (w ith whose views he o the rwise disagrees). Freud mentio ns tbe
c ulture and the media o f writing , printing and elec tro nic following types o f verbal erro r: transpositions (the 'Milo of Venus'
communica(ion. All o f the latte r have made a difference not o nly instead o f the 'Venus o f M ilo'); pre-sonances o r anticipations ('es
to stocks o f avai lable knowledge but also to types of kn owledge war mir auf der Schwest .. . auf der Brust so schwer' - 'Schwest'
produced. is a nonexistent word); post-sonance.\ o r perseveralions (' ich
fordere Sie auf auf das Wohl unseres Chefs aufzutossen', rather
than 'anzustossen '); contaminatio ns ('er setzt sich auf den
Hinterkopf' , a combination o f 'er setzt sich ein en Ko pf auf' and
'c r stellt sich auf die Hinterbe in e'); and .wbstitutioTls ('ich gebe
di e PrHparate in de n BriefkaSlen', instead o f 'Briitkasten').2
Meringer tried to explain th ese in terms o f phases of neutral
94 Con sciousness, Se lf and Socia l Encounters Critica l No tes: Fre ud on Slips of the Tongue 95

excitation. When a speaker utters the first word of a sentence, a mentioning Orvieto, Freud and his travelling compani?~ h~d
process of excitation , connected with anticipating the form of th e been talking about the customs of the Turkish people ilvmg III
utterance, is set in being. This process sometimes has the effect of Bosnia and Herzegovina, Fre ud was teUing the other of the
disturbing later sounds in the utterance, Some sounds are fatalistic attitude with which th e Turks approach sickness and
physically more intense than others, and these can affect other death . If a doctor tells the m nothing can be don e to save someone
sounds or words. T o discover tbe source of slips of the tongue we who is ill, their response is 'Herr lSir j, what is there to be said? If
therefore have to look for those sounds or verbalizations which he could be saved , I know you wou ld have saved him:3 The
have the highest physical valence. One way of doing this , words 'Bosnia', 'Herzegovi na' and 'Herr' have an un consciously
according to Meringer, is to consider what is involved in searching charged association with 'Signorelli', 'Bouicelli' and 'Boltraffio'.
for a forgotten word , such as someone's name. The first sound to A second anecdote lay close to the first in Freud's mind . In
come bac k into consciousness is always the one of greatest co ntrast to their quiescence in the face of death , the Turkish
intensity before the word was forgotten. This is often, for examp le, people in question display great agitation when affl ic~ed by sexual
the critical sound in the word or the vowel which is particularly disorders. Thus one said: 'Herr , you must know that If that comes
accentuated. Freud will have little of this. In the case of forgotten to an end, then life is of no value.' Freud had suppressed this
words it is very rarely true that either the initial sound or th e anecdote from his account, since he was talking to a stranger. He
accentuated vowel is the first to be recalled. Speakers may thereby diverted his attention from thoughts which might h~ve
sometimes believe this to be the case but in fact are usually been provoked in his mind by the th emes of ~eath and sexua!J~y,
wrong; Freud asserts that in the vast majority of instances the He had recently received an unfortunate piece of news while
initial sound which the speaker utters in attempted recall is the staying at Trafoi, a small village in the TyroL One of his patients,
wrong one. to whom Freud had devoted considerable atte ntion and who was
As an instance of the latter phenomenon Freud's fam ous suffering from what Freud refers to as an 'incurable sexual
discussion of his own lapse of memory about the name of the disorder',~ had committed suicide. T he similari ty of the words
painter Signorelli can be mentioned. Talking about the frescoes 'Trafoi' and 'Boltraffio' indicated that this event had made itself
of the 'Four Last Things', Death , Judgement, Hell and Heaven, in felt psychologically in spite of Freud's decision not to men~io.n it.
Orvieto Cathedral. Freud found himself unable to recall the Having established this resemblance. Freud asserts , It IS. no
name of the artist. Rather tha n finding the name he was trying to longer possible to regard the forgetting of 'Signorelli' as a chance
remember , he could think only of the names 'Botticelli' and event ; it was something that was (un consciously) motivated, The
'Boltraffio', On being told the correct name by another person. he item which Freud deli berate ly chose not to mention became
recognized it without any hesitation. The forgetting is not to be displaced on to anoth er element . the painter's name.
explained in terms of anything distinctive about the painter's The connections establ ished hereS indicate that th e name
name itself or any definite psychological aspect of the context in 'S ignorelli' became divided in two. One of the pairs of syllables ,
which Freud was trying to recall it. Freud was as familiar with one 'elli' occurs in unaltered form in one of the two names which
of the substitute names, 'Botticelli' , as with 'S ignorelli', and more came to Freud's mind. The oth er has become involved in a
familiar with 'Signorelli' than with the other mistaken name that network of connections by means of the translation of ;Signor'
occurred to him . 'Boltraffio', Freud's inability to recall the word into 'Herr' . A displaceme nt has occurred between the names
happened in the course of a casual conversation with a stranger .Her.legovina and Bosnia - two pl aces o ften spoken of together
while driving from Ragusa in Dalmatia to a place in Herzegovin a. in the same phrase. Most of the co nn ections which produced the
Freud offers the following analysis of the phenomenon , The forgetting have been forg ed below the level o f co nsciousn~ss.
forgetting of the name was connected with the preced ing topic The suppressed topic and th e fa ctors that have brought to mmd
whi ch had been disc ussed in the conversation. Just prior to Ih e substitute names do not have any manifest conn ections, The
Critical Notes: Freu d on Slips o f the Tongue 97
% Consciousness, Se ll and Social Encounters
T assenme.fcher ia no nexistent wordl - I mean Ta.{chen-
messer (pocket-k nife).' Freud recogn izes that there are
diffi culties of articulation with th e word, but he po ints out
th e error to th e patient and associates it with a name that
impinges on un co nscious anxieties.
(2) Another woman patient, asked how her uncle is, answers: '1
do n't know. Nowadays I o nly see him in f lagrante: The
phrase she meant to use is en passanl . The term said in error
is shown to relate to an episod e in th e patient's past.
(3) A young man addresses a wo man in the street with the
words: 'If you will permit me, madam, I shou ld like to
begleit-digen you.' He wants to accompany (beg/eiten) her
but fears his o ffer would insult (beleidigen ) he r. As in th e
Figure 4
'Signorelli' case, a concealed intentio n - th e requ est no t
being a wholl y innocent o ne o n the man's part - leads lO an
s i ~ i1 ari ti es invo lved do depend partly upon common sounds unconsciously motivated slip o f the tongue .
which the words possess, but these ca n be pieced toget her only (4) During a disputatio us meeting the chairman says: 'We shall
when w,e understand that the forgetting is a result of repression. now streiten (qu arrel, instead o f schreiten, proceed) to point
Not all Instances of the fo rgetting o f names, of cou rse, are of this fo ur on the agenda .' The speaker's true view , whic h he
sort : ' ~y the side of simpl.e cas~ wh~re proper names are fo rgotten intends to suppress , manifests itself in his verbal mistake.
there IS a type o f fo rgeHmg which IS mQ[ivated by repressio n.'f> (5) Someone is asked, 'What regiment is your son with?' The
. A mechanis.m similar to his, Freud goes on to argue , exists in answer given is: 'With th e 42nd Murderers' (MiJrder. instead
Instances of slIps of the to ngue. Verbal errors may be of the type of M6rs er, 'Mortars').
analysed ~y Meringe r and Mayer, wh ere one compo nent of an (6) A guest at a social occasion advances the op ini on : 'Yes a
utterance mflu ences another, or they may be lik e th e 'S ignorelli' woman must be pretty if she is to please men. A man is much
eX8,:,ple, where the influences that produ ce the error co me from better o ff ; as lo ng as he has his five straight lim bs he needs
? u.tslde the utterance and the immed iate circ umsta nces in which no thing morel' T his is o ne o f numerous examples o f what
It IS made. Both have their origins in a kind o f 'excitatio n', but in Meringer and Mayer called conta minations but which Fre ud
the. o ne case this is internal to th e utterance o r to th e situation in regards as in stances o f the psychological process o f
:-vhlch t?e words .are said ; in the other it is external to th em. Only condensation. The utterance is a fusion of two turns of
m the f~rst type IS there any possibility of explaining slips of the phrase resembling each other in meaning: 'as lo ng as he has
to ngue m terms of a mechanism linking sounds and words to one his four straight limbs' and 'as lo ng as he has his fi ve wits
ano ther so that .they influence articulatio n. Moreover , su bjected about him' . Freud notes that , as in many slips o f the lOngue ,
to furth er scrutmy. t~ e first type in fact evaporates. Slips o f the the remark could pass as a jo ke. The difference lies simply in
tongue that seem at fi rst blush to be simply the result o f a 'contact whether or not th e speaker conscio usly inte nded th e words
effect o f so unds' .actually .turn o.ut o n further investigation to to come out as they did.
depend upon outside (that IS, motivated) influences. (7) Rea nalysis o f one o f the Meringer and Mayer examples: 'Es
Fr e ~d lists many examples of slips o f the tongu e, including the war mir auf der Sc hwest ... auf der Brust so schwer: This
fo llowmg: canno t be adequate ly ex plain ed by th e anti cipation o f
(I) O n the part o f a woman patie nt : '1 shut up lik e 8 sounds. The slip o f the lo ngue is probabl y to be int erpreted
100 Consciousness, Self and Social En coumers Crit ical Note 5: Freud on Slips of th e Tongue 101

(4) 'And here in Ho llywood it is rumo ured that the former to ngue stand o ut muc h more promine ntly in radio talk than in
movie starlet is expecting her fifth chi ld in a mo nth.' day-to-day conversations. First , th e discourse does not take place
between co-present communicants. Disentangled from other cues,
( I)'Turns will give you instant relief and assure you no what is said becomes a more 'witnessable' phenomenon than it is
indigestion or distress during the night. ... So try Turns and when embedded in everyday activities. This is also tru e of many
go to sleep with a broad . . , Iturns page I smile.' of Freud's examples o f slips of the tongue , culled as they are from
(2) ' It's time now, lad ies and gentle men , for o ur featured guest , the therapeutic situation. The the rapeutic encounter, after all ,
the prominent lecturer and social leader , Mrs Elma Dodge ... hardly exemplifies o rdinary talk a ny more than broadcasting
[S uperman c ut-inl who is able to leap buildings in a single does. The words of the patient are treated as having a special
bound.' significance, to be carefully scrutinized. Second , announcers are
(3) A local TV stalion showing a box ing match from Madison specia lists in the production of flawl ess speech and are expected
Square Gard en interrupted the programme to report the to be such by the nature of their pro fession. The main task of the
death of a loca l pOlitician . On c Ulling back to the fi ght , the performer is to present the script fluidl y and clearly. It is only
an nouncer was saying: 'That wasn't much of a blow , folks! ' when we recognize how distinctive a nd unusual this re latively
flaw less speech mode is that we can begin to appreciate the
In these cases no sl ip of the tongue is involved , but they do contingencies of o rdinary day-la-day talk. Both lay participants
otherw ise take th e form of parapraxes. Something has gone awry and linguists usuall y regard everyday talk as much more
with what the speaker intended to convey. The second set of 'perfected' and 'o rd ered' than in fa ct it is. Summarizing recent
examples is interesting because if we did not know the work on the empirical study of conversations, Boomer and Laver
ci rcumstances in wh ich they occurred , it would seem as though comment:
they contain typica l 'only Lao true' utterances. No motive fo r them
ca n be imputed , unless the producers responsibl e for c utting II is important 10 recognize that ill speech 'normar does not mean
from one programme to the other somehow (con sc iously or perfect'. The norm for spontaneous speech is demonstrably
otherwise) organized the sequencing to have the effec ts noted. imperfect. Conv ersation is characterized by frequ ent pauses,
The first category of slips are more difficult to interpret. It may hesitation sounds, false starts, misarliculations and correc ti ons.
be the case that th ese are unconsciously mo tivated ambiguities. In everyday circumstances we simply do not hear many of our own
But this seems unlik ely. It is more probable that their ambiguous tongue-slips nor those made by others. They can be discerned in
running speech only by adopting a specialized 'proof-reader' mode
character would pass unnoticed by speakers and listeners alike if of liste ning. D
they were uttered within ordinary , everyday conversations. The
po int is not just that their ambiguous meanings are not In most circumstances of day-to-day conversations it is, in fact,
immediately apparent but also that in everyday talk mean ings very difficult indeed to distinguish slips o f the tongue from the
other than those int ended by speak ers tend to be ruled o ut by fragmented nature of vi rtually all th e talk that goes on. As
COnlextual features of the conversatio n. Speake rs are able to GoHman points om, for a partic ular utterance to be tested as a
address themselves to the specific people with whom they are slip o r as 'faulty', it has to be of a sort which the speaker would
engaged, pre-se lecting words and phrases so that possible alter were he or she to begin the utl erance again (or, o f course,
alternative readings are excluded. Radio or TV announcers o ne that actually is altered or 're medied'). It will no t do to
cannot do this because they speak to a generalized audience, that identify slips of th e to ngue by reference to an idealized model of
audience not being co-present with th em. enunciation or disco urse. Moreover, to understand the character
Now. it would clearly be mistaken to regard radio talk as of day-to-day talk , we have to look at the other types of fault that
typical o f talk in genera l. There are (wo reasons why slips of the may intrude. Whm are th e implications o f this?
102 Consciousness, Self and Social En counten Criti ca l No tes : Freud on Slips of the Tongue 103

First, as regards sLips of the tongue , it may be argued that speech is like and confirm that verbal parapraxes canno t be
Meringer and Mayer were no t as far o ff the mark as Freud tended interpre ted against an idealized conception of 'correct' speech.
( 0 argue. Fromkin has demo nstrated that mispro nunc iation of Anno uncers' talk differs fro m the day-to-day use of language in so
words man ifests properties simi lar to those characte ristic of far as it does approximate to suc h a conceptio n. The talk and the
'correct' word production. L4 This does not show that suc h faults acti vities of annou nce rs when they are on set in fact co mes close
are not brought about by unco nscio us promptings, but it does to how human social life wo uld be if it were actually like the
suggest that there is usually no 'interruption' in th e reflexive portrayals given by objectivist social scientists. Most of what is
mon itoring of speech productio n that necessarily needs to be said is programm ed prior to transmission or screeni ng and can be
in voked to explain slips of the tongue. T he phenomena of pre- modified only in marginal ways by th e agent follo wing the script.
sonances and perseverations are also presumably directl y bound T he actor here does appear merely as a 'bearer' of pre-given
up wi th the refle xive mo nito ring of speech. Wo rds must patterns of social o rga nizatio n - or, as Goffman puts it , an
characteristically be transfe rred fro m the brain to speech as 'a nimator', a 'sounding box fro m which utte rances come' .15 The
sy ntagmatically ordered gro upings, or e lse such speech distur- vast majority of situa tio ns of talk (and of inte ractio n) are simply
bances would not occ ur at all. not like this. The 'loose' or nawed character of day-to-day ta lk , or
A second large category of faults concerns not individual speech what ap pears as such when compared with an idealized mode l, is
productio n as such but turn-taking. A speaker may begin to talk actually generic to its cha racter as enmeshed in human praxis.
before the utterance of another is co ncluded, either 'ove rlapping' What is remarkable, to put it another way, is no t lac k of technical
with or directly interrupting the o th er; two participants might polish in talk but the fa ct that conversations and the (always
begin speaking si multaneously; each may 'back off' from speak ing, co ntingent) reproduction of social life have any symmetry of
producing an unwanted gap in the conventional fl ow. Just as in form at all. In day-lO-day in teraction the normative e lements
the case of individual speech faults, most such disju nctions pass involved in communicatio n in talk as the productio n of 'good
comple tely unnoticed by speakers engaged in o rdinary conver- speech' are hardly ever the main impelling inre rest of participants.
sa tio n. They are 'heard' o nly when, fo r example , a strip of speech Rather, talk is saturated with the prac tical demands of the routine
is recorded so that they can be deliberately attended to. Here enactment of social life.
again day-te-day talk is no t like radio tal k, where overlaps, double Accepting this means recasting Freud 's view. According to
uptakes, etc., are very noticeable. It is more often than not the Freud , every slip of the tongue has a motiva ted origin and could
case in conversations that overlap occurs, so that one speaker is in principle be explain ed if su fficient knowledge of the psycho-
beginning an utterance while another is finishing. But participants logical make-up of the individual in question were avail able . Here
filt er these out so that contributions to the conversation are we clearly discern an implied picture of well-ordered speech ,
heard as separate strips of talk . fro m which slips of the tongue lead th e speaker to depart. The
Third , faulty talk which is recognized as such usually invo lves standpoint I am advocat ing in effect turns this around . 'Well-
remedial procedures initia ted eithe r by the speake r or by the orde red' speech , in the context of day-to-day conversatio ns at
listeners. Correction by o thers see ms relatively rare , partly least, is geared to the overall mo tivational involveme nts which
because many imperfections which are pho nological or syntactical speake rs have in the course of pursuing their practical ac ti vities.
slips wh en judged against an idealized grammatical model a re not 'Correct speech', in co mmo n with many o the r aspects of such
hea rd as such, but partly also because tact is exercised in respect activities, is not usually direc tly motivated - unless one is an
of what might be taken to be th e incompetencies of speakers. announcer. It should be po inted out in parenthesis that on
Remedial work done by speakers nea rly always concerns turn- occasion disturbed speech may be so motivated. Thus in
taking diffi culties rather than slips of the tongue. circ umstances of mo urnin g, a bereaved person who maintained
Th ese o bserva tions tell us a good deal about wha t everyday o rdinary sta nd ards of speech production might be thought hard-
References 105
104 Consc iou sness, Self and Social Encounters

hearted and unfeeling. Where there are sanc tions implying that References
people sho uld manifest emotiona l agi tatio n , speech disturbances,
o r alterations in normal modes o f speech , may be o ne way of COflSciousness, Self and Social Encounters
'bringing o ff such states. 16
If most particular forms of language use are not directly A particularly useful discussion of these difficu lti es is to be found in
motivated, then it follow s that most slips of th e tongue cannot be Irv ing Thalbcrg, 'Freud's anatomies of the self, in Richard
Wollheim, Freud. A Co llection of Critical Essays (New York :
traced to unconscious motivation. Where does this leave us, then,
Dou bleday, 1974). A rev ised version of this essay appears in
as regards Freud's th eory o f verbal parapraxes? I would make the Wollheim and James Hop kins, Philo.~ophical Essays on Freud
following suggestion. Freud's interpretation probably app lies only (Cambridge: Cambridge Un iversity Press, 1982).
in circumstances rather di fferent from those he had in mind when 2 Quoted in Thalbcrg, 'Freud's Anatomies of the self', p. 156.
formulating it. In Freud's view, slips o f the tongue te nd to be 3 Freud, An Outline of Psychoanalysis (London, Hogarth. 1969),
made above all in casua l or routine situations, whe re no thing pp.56-7.
muc h hangs on what is said . On such occasio ns, the unconscious 4 P. F. Strawson, Th e Bounds of Sense (London: Methuen, 19(6),
is likely to 'break through', as it were, and disturb th e utt erances pp. 162-70; G. E. M. Anscombe, 'The first person', in Samuel
that a speaker produces. I would hold that on these occasions - Guuenplan, Mind and Language (Oxford: Blackwell. 1972); 1 L.
which make up most of social life - unconscious elemen ts a re Mackie. 'The transcendal "]''', in Zak Van Struaten, Philosophical
actually least prone to influ e ncing directly what is sa id. Subjects (Oxford : Clarendon Press, 1980).
Ro utinization, involving the contin ual 'regrooving' o f th e familiar 5 Stephen Toulmin. 'The genealogy of "consciousness''', in Paul F.
Secord, Explaining Human Behaviour (Beverly Hitls: Sage, 1982),
in circumstances of substantial ontological security, is the main
cond ition of [he effec tive reflex ive monitoring by human beings
6 Ibid., pp. 60- 1.
of their activities, Anxiety concern ing the actual fo rm of speech 7 See 1 S. Bruner, Bey ond the Illformation Given (New York : Norlon,
will be heightened only when the acto r has a specific inte rest in 1973).
getting wha t he o r she says 'exactly right' , This is what radio and 8 J. S. Gibson, The Ecological Approach to Visual Pe rception
TV announcers have to do. It is likely to be the case in a (Boston: Houghton Mimin, 1979).
declaration of love, contrary to Freud's supposition. We can also 9 Ulric Neisser, Cognition and Reality (San Francisco: Freeman,
readily make sense of the 'Signore lli ' example and th e forgetting 1976), p. 22. See also idem, Memory Obsen'ed (San Francisco:
of proper names generally as a motivated phenomcnon. Proper Freeman, 1982); John Shotter, "'Duality of structure" and
nam es have a special significance which other words do no t. To "intentionality" in an ecological psychology', Journal for Ihe Theory
mispronounce so meone's name or to call someone by the wrong of Social Behaviour. vol. 13, 1983.
name causes personal affront in a way that other vagaries o f 10 Neisser, Cogn ition and Reality. p. 29.
I I M. Wertheimer, 'Psychomotor coordination of auditory and visual
pronunciation do not. Th ere is thus a special premium o n getting
space at birth', Science. vol. 134, 1962.
names right , which perhaps means that the recall o f na mes 12 Neisser, Cognition alld Reality, p. 72.
impinges more immediate ly o n sources o f anxiety tha n do other IJ E. C. Cherry, 'Some experiments on [he recognit ion or speech, with
linguistic ite ms. As I have pointed out, something sim ilar applies one and two ears', Journal of the A coustical Society of America.
to th e therapeu tic encounter as well . vol. 25, 1953.
14 A. M. Treisman, 'Strategies and models of selective attention'.
Psychological Revie w, vol. 76, 1969.
15 1. A. Deutsch and D. Deutsch, 'Attent ion': some theoret ical
considerations', P~')'c h o loJ!,ic (l1 Review. vol. 70, 1%3.
16 Ncisscr, C01.:nilioll ml(l Reel/ity. pp. 84- 5.
106 Consciousness, Self and Soc ial Encounters References 107

17 CPST, pp. 120- 3. 38 M. Merleau-Ponty, Phenomenology of Perception (London:

18 Erik H. Erikson, Childhood and Society (New York: Norton. 1963), Routledge, 1974).
pp.15 - 16. 39 Ibid., p. 1Ol.
19 Ibid., p. 247. 40 L Goldstein, Language and Language Distrlrba",:e,~ (New York:
20 Ernest Becker, The Birth and Death of Meanillg (New York : Free Grune and Stratton, 1948).
Press, 1962), p. 95. 4 1 Merleau-Po nty. Pheflomeflology of Perception, p. 104.
21 See also Erikson, Childhood and Society, p. 249; Harry Stack 42 Ibid., p. 109.
Sullivan, The Imerpersonal Th eory of Psychiatry (London: 43 Erving Go ffm an, Behaviour in Public Places (New York : Free
Tavistock, 1955), chapter 4. I do no l accept Erikson's claim that Press, 1%3), p. 17; idem, In teractiOIl Ritual (London: Allen Lane,
these psychological phenomena can be directly related to the fo rm 1972), p. I.
of social institutions. 44 Cf. hhiel De Sola Pool. The Social Impact of the Telephone
22 G. Piers a nd M. B. Singer, Shame and Guilt (Springfield: Addison, (Cambridge, Mass.; MIT Press, 1981).
1963). Here 1 repeal some o bservations originally made in relation 45 This seems to be the prevalent not ion, for instance, in most of the
to the theory o f suicide; cr. SSPT, p. 393, foolnole 32. contributions to Jason Ditton, Th e View from Goffman (London:
23 Erikson, Childhood and Society, p. 251. Macmillan, 1980). See also Alasdair Mac intyre, Afte r Virtue
24 Ibid., p. 256. (London: Du ckworth, 1981), pp. 108 - 9. C f. R. Harre and P. F.
25 Dennie Wo lf, 'Understand ing others: a longitudinal case study of Secord, The Explanal.ion of Social Behaviour (Ox ford: Blackwell,
th e concept of independent age ncy', in George E. Forman, Action (972). chapter 10.
and Thought (New York: Academic Press, 1982). 46 Alvin W. Gouldner, Th e Coming Crisis of Western Sociology
26 T. B. Brazelton et a/.. 'The origins of reciprocity', in M . Lewis and (London: Heinemann , 197 1), pp. 379 - 8 1.
L. Rosenblum, Th e Infa nt'S Effects on the Caregiver (New York: 47 CPS T. pp. 83-4, and passim.
Wiley, 1974). 48 Goffman, Behaviour in Public Places. p. 18.
27 L. S. Vygotsky, Mind in Society (Cam bridge: Harvard University 49 Erving Goffman, Frame Analysis (New York: Harper, 1974), p. 252.
Press, 1978), pp. 2Off. 50 Roger Caillois, Man , Play and Games (London: Thames & Hudson,
28 Erik H. Erikson, Identity, You th and Crisis ( London: Faber & 1962); see also the famous work by Jan Huizinga, Homo Ludem
Faber, 1968), c hapter 5; idem, Idemiiy and the LIfe Cycle (New (London: Routledge, 1952).
York : International Universit ies Press, 1967). 51 Goffm an, Frame Analysis. p. 560, I shall not discuss here the
29 Erikson, Identliy and the Life Cycle, p. 19. epistemo logical queslio ns which are broached, but hard ly resolved,
30 See ibid., chapter 3, 'The problem of ego-identity'. in Goffman's discussion in this book. T hey share a good dea l in
31 Ibid., p. 102. com mo n with SchUlz's ponderings over the nature o f ' multiple
32 See CPST, pp. 123 - 8. realities', a nd with many other c urre nts in modern philosophy
33 Bruno Bettelheim, Th e Iliformed Heart (Glencoe: Free Press, concerned with the apparent ly relat ivistic implications of the
1960), p. 14. Goffman'S work o n 'tolal institutions' overlaps at many mediation of frames o f meaning. See NRSM, chapter 4.
poin ts with the analysis given by Bethelheim: Goffman, Asylums 52 Goffman, Behaviour ill Public Places, pp. tS6ff.
( Harmondsworth: Penguin, 1% 1). 53 Ibid.
34 Bettelheim, Th e Illformed Heart, p. 132. 54 This theme, of course, has been much explored. Th e best-known
35 Ibid. , p, 148. work is Edward T. Hall, Th e Silellt Language (New York:
36 'Since old prisoners had accepted, or been forced to accept, a Doubleday, 1959); see also the same author's The Hidden Dimension
childlike depe ndency o n the SS, many of them seemed to want to (London: Bodl ey Head, 1%6).
feel thaI at least some of the people they were accepting as all Harvey Sacks and Emmanuel A. Scheglo rr, A simplest systematics
powerful fa th er images were just and kind', ibid., p. 172. for the organisalion o f turn-talking in conversation, Language. vol.
37 Sec the examples collected in William Sargant. Bartle for the Mill d 50. 1974.
(London: Pan. 1959) .
108 COflsciousncH. Self and Social Encounters References 109

56 Cf. George Psathas. Everyday Language: Studies in Ethflometho- 7S D. Lawre nce Wiede r, 'Telling the code', in Roy Turne r,
dology (New York : Irvingto n, 1979). E/hrlOmethodology (Ha r mondsworth: Pe nguin , 1974J.
57 Jean-Pa ul Sartre, Critique of DialeClical RelLmn (Londo n: New Le h 76 Ibid., p. 149,
Book ~ p. 259).
58 Goffman, Interaction Ritual. pp. 14 1rr. Critical Notes: Freud on Slips of the T ongue
59 Habermas, Theorie de.t kommUllikatiren Handelns ( Fra nkfurt:
Suhrkamp, 198 1), vol. L sect ion 3, Sigmund Freud, Int roductory Lectures 011 Psychoanalysis
60 Goffman, Behaviour ill Pu blic Places, p, 25. (Harmondsworth : Penguin, 1974), p, 5 1.
61 Cf. th e general discussion o f politeness in Penelope Brown and 2 R. Me ringer and C. Mayer, Versprechen ulld Ver/esen (Vienna,
Stephen Levinson, 'Universa ls in la nguage use: politeness pheno- 1895).
mena', in Esther N. Goody, QuestiorlS and Politeness (Cambridge : 3 Freud, The Psychopmhology of Everyday LIf e ( Harmondswor lh:
Cambridge University Press, 1978). Pengu in, 1975). p. 39.
62 Go Uman, Behaviour ill Public Places, p. 35. cr. Jo hn Blac king: The 4 Ibid" p. 40.
Allthropology of the Body. (Londo n: Academic Press, 1977). 5 O riginally published in Fre ud's art icle, T he physical mechanism o f
63 'I lake many bodily feelings to be private. If I have a burn o n my forgetfu lness' ( 1890); see the Standard Edition, vol, 3.
arm, I lake the pain [0 be private, the sight to bc public. This is not 6 FreUd, The Psy chopathology of Everyday LIfe, p.44.
a lways so. Some people feel t ha t th ey can actually feel anot he r 7 Ibid., p. [35.
person's pain, or think direc tly anoth er's thoughts, a nd may fee l 8 Boilea u, Art poeliqu e, quoted in ibid" p. 148,
{hat oth er people can feel their bod ily feelings, or ac tually be 9 Fre ud, Introdu ctory Lectures 0 11 Psychoanalysis, p. 7 1.
thin king their thoughts', R. D. Laing, Self and Others (London: 10 Eeving Goffman. 'Rad io talk: a study of the ways o f o ur errors', in
Penguin, 197 1), p, 34, Forms of Talk (Oxford: Blac kwell, 1981).
64 Haro ld Garfin kel, 'A conception o f, and experimenLS with , "trust" 11 Ibid., p. 242.
as a conditio n o f s ta ble concert ed ac tio ns', in O. J. Ha rvey, 12 They were no do ubt se lec ted for this reason, Most o f Goffman 's
Motivation alld Socia/ l meractiofi (New York : Ronald Press, 1963). ma terial comes fro m collectio ns o f ' bloopers' edit ed by Ke rmit
65 Erving GoHman, Forms of Talk (Oxford : Blackwell, 198 1), pp. 101 rr. Sc hafe r, suc h as Prize Blooper.~ (G reenwic h: Fawce tt, 1965).
66 I bid .. p. 103. 1:'\ Donald S. Boomer and Jo hn D. M. Laver, 'S lips o r the tongue',
67 Ibid" pp. 70 - 1. British Journal of Disorders of Communication, voL 3. 1%8, p. 2.
68 Roy Bhaskar, The Possibility of Naturalism (Brighto n: Harvester, [4 Victoria A. Fromk in, 'The no n-anomalous nature of anomalous
1979), pp. 5 1- 2. utt erances', Language, vo L 47, 197 L
69 For a recent exampl e - among very many others - see Bruce J. [:; Go ffman, Forms of Talk. p, 226.
Biddle, Role Theory (New York: Academic Press, 1979). 10 As indicated by Goffman. ibid., pp. 223 f[
70 CPST. p. 117.
7 1 Ibid.
72 A point o ft e n made in the controversy over role theory in Germany
some two decades ago. A contributio n that retains its inte rest is
F. H. Te nbruk : 'Zur deutsc he n Reze ption der Ro llenanalyse',
Kd/ller Zeilsc hrift fur Sozi% gie, vol. 3, 1962-
73 C f. Nigel Thrift, 'Flies and germs: a geography o f knowledge', in
Derek G regory and John U rry, Social Relations alld Spatial
Slnu:/ures (London: Mac millan, 1984).
74 C f. William Labov, ' Rul es fo r ri tua l insu lts', in David Sud now.
Srudies ill Social /merae/iOIl (New Yo rk: Free Press, 1972),
Time-Geography 111

established traditions of social theory, have made contributions

3 to social thought of some significance. Most such writings, I think
it would be true to say, remain unknown to the majority of those
Time, Space and working in the rest of the social sciences, although they contain
ideas of very general application. Some of these contributions are
Regionalization to be found in the work of Hi:igerstrand, but they are by no means
confin ed to his writings and those of his immediate colleagues. I"
In previous analyses of the theory of structuration I have
mentioned the significance of this approach without confronting
it directly or trying to point o ut its limitations. But in this expanded
ex position I shall do so.
Time-geography, as formulated by Hagerstrand, takes as its
Time-Geography starting-point the very phenomenon which I have mu ch stressed
1 have concentrated in the preceding chapter upon specifying - the routinized character of daily life. This is in turn connected
certain psychological qualities of the agent and upon analysing with features of the human body, its means of mobility and
interaction in situations of co-presence. The positioning of actors communication, and its path through the 'life-cycle' - and
in contexts of interaction, and the interlacing of those contexts the refore with the human being as a 'biographical project'. As I
themselves, is elemental to such concerns. But to show how these have mentioned before, Hagerstrand's approach is based mainly
matters relate to broader aspects of social systems it is necessary upon identifying sources of constraint over human activity given
to consider how social theory should confront - in a concrete by the nature of the body and the physical contexts in which
rather than an abstractly philosophical way - the 'situatedness' activity occurs. Such constraints provide the overall 'boundaries'
of interaction in time and space. limiting behaviour across time-space. Hagerstrand has formulated
Most social analysts treat time and space as mere environments these in various different ways , but his characteristic emphasis is
of action and accept unthinkingly the conception of time, as upon the following factors. 2
mensurable clock time, characteristic of modern Western culture. (1) The indivisibility of the human body, and of other living and
With the exception of the recent works of geographers - of inorganic entities in the milieux of human existence.
which more in a moment - social scientists have failed to Corporeality imposes strict limitations upon the capabilities
construct their thinking around the modes in which social systems of movement and perception of the human agent.
are constituted across time-space. As J have indicated earlier. (2) The finitude of the life span of the human agent as a 'being
investigation of this issue is one main task imposed by the 'problem towards death '. This essential element of the human
of order' as conceptualized in the theory of structuration. It is not condition gives rise to certain inescapable demographic
a specific type or 'area' of social science which can be pursued or parameters of interaction across time-space. For this reason
discarded at will. It is at the very heart of social theory , as if no other, time is a scarce resource for the individual actor.
interpreted through the notion of structuration, and should hence (3) The limited capability of human beings to participate in
also be regarded as of very considerable importance for the more than one task at once, coupled with the fact that every
conduct of empirical research in the social sciences. task has a duration. Turn-taking exemplifies the implications
Fortunately, we do not need to tackle these issues de novo. of this sort of constraint.
Over the past few years there has taken place a remarkable
convergence between geography and the other social sciences, as
"l{cfc rc ll(;cs nmy hc fO Ulid 0 11 pp. 1 5~ - 61.
a result of which geographers, drawing upon th e various
112 Time, Space and Regionaii l. arion Time-Geography 113

(4) The fact that movement in space is also movement in time. ",-- - - - --- -- -::-
(5) The limited 'packing capacity' of time-space. No two human /"
" ,
bodies can occupy the same space at the same time; physical ----:---
, - - - --(
objects have the same characteristic. Therefore any zone of
time-space can be analysed in terms of constraints over the
two types of objects which can he accommodated within it. / + - - - - -- - 7 spac<c

These five facets of 'time-geographic reality' , according to

Hi:igerstrand, express the material axes of human existence and
time L _ _ _ ==__

underlie all contexts of association in conditions of co-presence ..) Col ocatio n in lim espace'
Examined as resources (and thus , I would say, implicated in both
the generation and the distribution of power), such factors Figure 5a Figurp sb
condition the webs of interaction formed by the trajectories of
the daily, weekly, monthly and overall life paths of individuals in Figures 5a and 5b show this in its simplest guise. Two
their interactions with one another. The trajectories of agents, as individual s, say, live a mile apart in a neighbourhood: their time-
Hagerstrand puts it, 'have to accommodate themselves under the space paths across the course of the day bring them into contact
pressures and the opportunities which follow from their common in an encounter of short duration in, say, a coffee house or
existence in terrestrial space and time'.4 restaurant, following which their activities again diverge. If the
Hagerstrand's generalized conception of time-geography origina- daily activities of a specific individual are recorded , it is easy to
ted in a long-term series of studies of a local parish in Sweden. build up a gross characterization of his or her routin e activities , in
The area in question boasted comprehensive population statistics, so far as these comprise trajectories in lime and space. As a
enabling him to trace all the individuals who had lived there, and portrayal of a life path , this would involve generalized patterns of
had moved in and out of the area, for a period of something like a time-space movement within the 'life-cycle'. A person may live in
hundred years. Ordering these data as lifetime biographies, he the house of his or her parents, for example, until establishing a
sought to analyse them as composing life paths in time-space that new resid ence on marriage. This may be associated with a change
could be charted using a particular form of notation. The typical of job, such that both home and workplace , as 'stations' along the
patterns of movement of individuals, in other words, can be daily trajectory. become altered. Mobility within the housing
represented as the repetition of routine activities across days or market, marital separation or career progression, amid a host of
longer spans of time-space. Agents move in physical contexts other possible factors, may influence typical life paths.
whose properties interact with their capabilities, given the above The encounters into which individuals enter in the trajectories
constraints, at the same time as those agents interact with one of daily life are subject to constraints deriving from the list
another. Interactions of individuals moving in time-space compose indicated above. Hagerstrand acknowledges, of course, that
'bundles' (encounters or social occasions in Goffman's tenninolob'Y) agents are not merely mobile bodies but intentional beings with
meeting at 'stations' or definite time-space locations within purposes, or what he calls 'projects'. The projects which
bounded regions (e.g. homes, streets, cities, states, the outer limit individuals seek to realize, if they are to be actualized, have to
of terrestrial space being the earth as a whole - save for the odd utilize the inherently limited resources of time and space to
space traveller or two in the current age of high tec hnology). overcome constraints which they confront. 'Capability constraints'
Hagerstrand's dynamic 'time-space maps' are of definite interest are those of the sort listed above. Some affect primarily time
and provide a graphic form that has relevance to situations well distribution: for example, the need for sleep or for food at regular
beyond those for which they have bee n used so far. int erva ls e nsu res ce rtain limits to the structuration of daily
114 Time , Spa ce and RegionaJizarion Time-Ceograp hy 11.'>

activities. 'Co upling constrai nts' refer to those that condition coupling constraints in reaching the two 'stati ons' o f the nursery
activities undertaken jointly with others. The volume of time- (N I), and her place o f work (W I). Her choice of jobs is restricted
space available to an indi vidual in a day is a prism bounding th e by these constraints , and reciproca lly the fact that she has little
pursuance of projects. Pri sms of daily conduct are not just chance o f acquiring or hold ing down a well -paid occupatio n
geographical o r ph ysical boundaries but have 'time-space walls re inforces the o ther constra ints she fa ces in the trajectory o f her
o n all sides'. The size of such prisms , of course , is also very
strongly influenced by the degree of time-space convergence in
the means o f commun ica tion and transformation available to ,, -- --
The no tion of lime-space convergence was introduced by
another geograph er, Janelle, to refer to th e 'shrinking' of distan ce
in terms o f th e tim e needed to move between different locations. b
Thus the tim e taken to travel from the East Coast to the West
Coast of the United States, in tcnns of available media, can be
um e -,
calculated as follow s. On foot the journey would take more than
two years ; o n horseback e ight months; by stagecoach or wagon, W, I" -I H N~ w~

four mo nths; by rail in 1910, fo ur days; by regu lar air services

today, five ho urs; by th e fastest jet transport. just over two hours.
Time-space convergence can be plotted to describe the outer
bounds of daily prisms. However, it is obvious that there are path through the day . She has to coll ec t her chi ld in mid-
major discrepa ncies between and within social communities in afternoon, before the nursery closes, and is thus effectively
terms of the constraints on mobil ity and commun ication affecting restricted to part-time employment. Suppose sh e has a choice of
different gro ups and individuals. Seriality and turn-taking are two jobs, one better-paid and o ffering the chance to run a car
bui lt into most fo rms o f transportation. T hus , for instance, an (W 2), making it possi ble for her to take her child to a nursery (N 2)
ex press (rain may connect two cities in a time of three ho urs. But further away from he r ho me. On tak ing the mo re remunerati ve
the avai lability o f seats may be limited , even for those able a nd jo b, she finds that th e time ex pended in driving to th e nursery, to
willing to pay. Moreover, if a person misses the train , (here may and from work and th en back ho me (H) again does not allow her
be o nly local trains fo r several ho urs until the next express, givi ng time to do other necessary tasks, suc h as sho pping , cooking and
time-space convergence a 'pa lpitating' character. 1 Finally , for ho usework. She may therefo re feel herse lf 'fo rced' to leave the
those in most societies, and fo r most o f the days in an individu al's jo b for a low-paid , parHime alternative nearer home (W I).
life, mobility takes place wi thin relative ly co nstricted time-space Hagerstrand has made a particular effo rt to emp loy time-
prisms. geography to grasp the seri ality of th e life paths or 'life
Palm and Pred provide o ne example, among many that exist in biographies' of individuals. A life biography , he says, is made up
the literat ure, of an applicatio n of Hagerstrand's ideas: to the of 'internal mental exper iences and events" 're lated to the
daily prism of 'Jane', an unmarried mother. 8 Figure 6 offers a interplay between body and enviro nm ental phenomena'.9 The
representati on o f the prism of Jane's day-fo-day activities. Jan e co nduct of an individual's day-to-day life entails that he or she
cannot leave home for work before a certain hour o f th e d ay successively associates with sets o f e ntiti es emanating from the
becau se of her chi ld 's depend ence on her for feedin g and o th er scttings of interaction. These entiti es are: o ther agents , indivisible
needs. and beca use th e sole accessibl e nursery is not yet ope n. objects (solid materi al qualiti es of the milieu o f action ), divisible
Jan e has no ca r and hence is fa ced with seve re capabi lity and mat erials (air , water , min erals, foodstuffs) and domai ns. Domains
116 Time, Space and Regionalizatioll Critical Commenl5 117

refer to what I prefer to call th e regionalization of tim e-space: the the hu man being in stru ctured time-space co ntexts Hagerstra nd's
mo vement o f life paths thro ugh settings of interactio n that have ideas accord close ly with those I have so ugh; to elabo rate
vario us form s of spatial demarcation. But the properties of prev io usly. But he tends to treat 'indi vidua ls' as constituted
domains can be subjected to direct study in terms o f the coupling independ ently of the social sett ings which they co nfront in their
constraints which a given d istribUlion o f 'stations' and 'ac tivity day-to-day lives. Age nts are regarded as purposive beings in the
bundles' creates for th e overall popu lation whose act ivities are sense that the ir activiti es are gu ided by 'projects' whic h they
concentrated within those domains. T hus the nature of interacting pursue. But the nature and origin o f projects is left unexplica ted.
social patt ern s within do mains of time-space is li mited by the Second , Hagerstrand's analyses therefore tend to recapitulate the
overall organizatio n of ca pability and coupling const ra ints. There dualism o f action a nd struc ture , albeit in rathe r novel form
are 'ecological' constraints which, as Carlstein has tried to show ~eca.use ,o,f his. pr: -e min em concern with time a nd s pace.
in detail, derive from three modes of 'packing': Stations, dom~ms, erc ., are themselves take n as givens, rhe
( J) the pack ing of mat erials, artefacts, organisms and human outcome o f unmte rpreted processes of institutional fo rm ation
populations in sen lement space-lim e; and ch?n ~e . Unsurprisingly , in thi s type o f viewpo in t little
(2) the packing o f t ime-c.:ons uming aCl ivitie... in po pulat io n lime- emphasts tS placed o n the esse ntia lly transfo rmational charac ter
budgets; o f all human aCLio n, even in its most utterly ro utinized (o rms.
(3) the packing of bundles o f various sizes, numbers and durations Third, concentratio n solely upo n constraining properti es o f the
in th e popu lation system, i.e. group formatio n bC(.:8use of the body, in its movem ent throu gh time-space, is unwarrant ed. All
indivisibility and continuity constraints of individuals. to type~ o f constraim , as I have said , are also types of opportunity ,
m ~dla for the e nablement of action. The specific way in which
Hagerst~a nd tends to conceptualize 'constraint', mo reover, betrays
Critical Comments
a certa.," c ulture:bo und eleme nt in his views. For capability
The interest of time-geography to th e theory of structuration is co n s~ral.nts, couphng co nstraints and so on are typically discussed
surely evident. lL Time-geography is concern ed with the constrain ts by hIm," terms of their operation as scarce resources. It is not
that shape the routin es o f day-to-day life and shares with d iffic ult to see here o nce mo re a possible link with a versio n of
stru cturation theory an emphasis upon the significance o f the his.tc;>rical materiali~ m. There is mo re than a hint in Hagerstrand's
practical character of daily activities, in circumstances of co- writings o f the nOllon that allocatio n of scarce reso urces o f the
presence, for the constitution of social co nduct. We are abl e to body ~nd its media has some sort of determining effect upo n the
begin to flesh o ut the time-space stru cturing of the settings of orgamz.at ion of soc ial institutions in all types o f society. Such is a
interaction whic h, however impo rtant GoHman's writings may fea~i b~e p~opo~ition . I thi.nk , ~>n ly in the case o f contempo rary
be , tend to appear in those wri ti ngs as given milieliX o f soc ial life, SOCietIes , In WhlCh a premIUm IS placed upon the 'effi cie nt ' use of
Hagerstrand's concentration upon everyday social practices is reso urces. I.! Finall y, time-geography invo lves o nly a weakly
very prono un ced and clear; he wish es to use tim e-geography , he developed theory of power. Hagerstrand does talk of 'auth o rity
insists, to und erstand 'the impac t o f the ordinary day of th e constraints' , which he links to capability and coupling constraints.
ordinary person' upo n the overall organizat io n of social syste ms. 11 But these are both vaguely formulated and in voke a zero-sum
But time-geography has some ve ry distinct shortcomings, some of conceplio n of power as a source o f li mitations upon actio n. If
which, I hope, are apparent from the preced ing discussion in thi s power is conceived o f as generative, on th e o ther hand , the
book. constraints' of whi ch Hagerstra nd speaks are a[1 modalities for
The main reservatio ns o ne must have about tim e-geography lh e engende ring and sus taini ng of st ru ctures of dom ination.
are the fo llowing. First. it o perates with a naive <md d efecti ve In .ord e~ to d evelop such ideas mo re ad equate ly in respec t o f
co nce ptio n o f th e huma n age nt. In stress in g th e corpo realit y o f consl(1era tlo ns ex plored earli er in this book we have to loo k again
118 Tim e, Space and Regiona/izarion Modes of Regionalizarion 11 9

at the notion of 'place' as o rd in arily used by geographers. reasons for using the term 'locale' rather than 'place' is that
Hagerstrand's time-geography suggests a very effective critique of properties of settings are employed in a chro nic way by agents in
'place' in respect of demonstrating the significa nce, in studyi ng the constitution of e ncounte rs across space and time. An obvio us
human social conduct, of analysing th e organization of time- element of this is the physical aspect of what Hagerstrand calls
space. But his emphasis is very much upon integrating temporality 'stations' - i.e. 'stopping places', in which the physi cal mo bility
into social theory. He does not subject the notions of place or of agents' trajectories is arrested or curtailed for the duratio n of
location to a close conceptual scru tiny and uses such terms in a encounters or social occasions - as locales in which the routine
relatively unexamined fashion. The term 'place' cann ot be used activities of different individuals intersect. But the features of
in social theory simply to designate 'poi nt in space', any more settings are also used , in a routin e manner, to constitute the
than we can speak of points in time as a succession of nows'. meaningful content of interaction : demonstration of th e manifold
What this means is that the concept of presence - or. ra ther, of ways in which this occurs ranks among the major contributio ns o f
the mutuality of presence and absence - has to be explicated in Garfink el and of Goffman. Context thus connects the most
terms of its spatiality as well as its temporality. In developing th e intimate and detailed compo nents of interaction to much broader
theory of structuration I have introduced two notio ns that are of properti es of the inst itutionalization of social life.
some relevance here: the concep ts 0 1 locale and of presence
availability as involved in th e relations between social and system
integration. '4 Modes of Regionali z.ation
Locales refer to the use of space to provide the settings of
'Regionalization' should be understood not merely as localization
interaction, th e settings of interaction in turn being essentia l to
in space but as referring to the zoning of time-space in relation to
specifying its contextuality. The constitution of locales certainl y
routinized social practices. T hu s a private house is a locale which
depends upon the phenomena given pride of place by
is a 'statio n' for a large clu ster of interactio ns in the course of a
Hagerstrand : the body, its media of mo bility and communication ,
typical day. Houses in contemporary societies are regionalized
in relatio n to physical properties of the surrounding world . Locales
into floors, halls and rooms. But the various rooms of the house
provide for a good deal of the 'fi xity' underlying institutions,
are zoned differently in time as we ll as space. The rooms
although there is no clear sense in whic h they 'determin e' such
downstairs are characteristically used most in daylight ho urs ,
'fixity'. It is usually possible to designate locales in terms of their
wh ile bedrooms are where individuals 'retire to' at night. The
physical properties, eith er as features of the materi al world or,
div isio n between d ay and night in all societies used to be perhaps
more commonly , as combinations of those features and human
the most fundament al zoning demarcation between the intensity
artefacts. But it is an error to suppose that locales can be described
of social life and its relaxation - ordered also , obviously, by the
in those terms alone - th e same form of erro r made by
need of the human o rganism for regular periods of sleep, Night
behaviourism with regard to the descriptio n of human actio n. A
time was a 'fro ntier' of social activity as marked as any spatial
'house' is grasped as such o nly if the o bserver recognizes that it is
a 'dwelling' with a range of other properties speci fied by the fro ntiers have ever been. It remains a fronti er. as it were, that is
only sparsely settled. But the invention of powerful , regularized
modes of its utilization in human activity.
Locales may range from a room in a house, a stree t corner, the modes of a rtifi cial lighting has drama tically expand ed the
shop floor of a factory , tow ns and cities, to the territorially potenti alities of interaction settings in night hours, As o ne
observer has remarked :
demarca ted areas occupied by natio n-states. But locales are
typically internally regionalized, and the regions within them are The last great frontier of human immigration is occurring in time:
of critical importance in constituting contex ts of interaction. Let a spread ing of wakeful activity throughout the twenly-four hours
me develop a little furth e r the no tio n of contex t. One o f the of t he day. There is more mul liple shift factory work, more po lice
120 Tim e, Spa ce and Regiona/izatio n Modes of Regiona/izativn 121

coverage. more use of the telephone at all hours. T here are more week. In day-to-day life in the hospital the alternat io n o f 'day' and
hospitals. pharmacies. aeroplane fli ghts. hostels, always-open 'night' resembles the division o f the week into weekdays and
restaurants. car rental and gasoline and auto repair stations, bowling weekends. As the author notes , the fact that working at nights is
alleys, and radio stations, always active_ There are more emergency still considered unusual. a nd unusually demanding, is indicated
services such as auto-towing. locksm iths_bail bondsmen. drug and by the te rm used to refer to it: 'night duty'. T here is no
poiso n and suicide, gamblin g ' hot li nes- available incessantly. correspond ing term 'day duty'.'6
Although different individuals participate in these events in shifts, A useful classificatio n of modes of regionali zation might be
the organizations in volved arc continually active_j ~ offered by figure 7. By the 'form' of regionalizat ion I mea n the
Zerubavel's study of the temporal organization of a modern
hospital. where zoning is very tight ly contro lled , is relevant here. form
Most of the services of med ical care in the hospital he studied are
provided by rOLating nursing stafr. T he majority o f nurses work
fo r sel periods o n diffe re nt wards , moving around the diffe rent
sectors of the hospital. and they a lso are called upon to alternate
day and night shift work. Th e cycle of moveme nt betwee n wards dU'<I t ion
coi ncides wi th that between day an d night work, so that when
som eo ne 'goes to days' he or she also changes to anoth er sector.
Th e schedu ling of these activ iti es is complex and detailed. Whil e
nurses' work is regulated in standardized four -weekly periods, th e
rotation o f interns and residents is variable. Nurses' rotatio ns
Ch,1f3cte '
always begin o n the same day o f the week , and since they are o f
twenly-eight days, they do no t coinc ide with calendar months.
Figure 7
The activities of house staff , o n the o ther hand , are o rganized in
terms o f calendar months and he nce begi n on different days of
th e week. fo rm o f the boundaries that define the region. In most locales the
Week ly and daily zo nes are also punctiliously categorized. boundari es separating regio ns have physical or symbol ic markers.
Many rout ines occur at precise. seven-day intervals, especially In contexts of co-presence these may allow a greater o r lesser
those involving nurses. Nurses' 'tim e off' is also co unted against a number of the features of 'presencing' to permeate adjoining
weekly sched ule. Time off can be split into a number o f segments regions. As has been mentioned. in socia l gatherings the
take n separately, but each segme nt has to be a multiple o f seven regio nalization of encount ers is usually indicated o nly by body
days, and each has to begin o n Sunday and to end on Saturday to posture and positio ning, to ne of voice and so o n. In many such
co-ordinate with the rotations of work act ivities. 'Weekdays' a re gatherings , as regio na lly bou nded e pisodes, e ncou nters may be
no t ide ntical to 'weekend' days, however, because although nea rly all o f very sho rt duration. Walls between rooms, on the
operati ng upon a continuo us basis, various kinds of services are o ther hand, may demarca te regionalization in s uch a way that
restricted in the hospital during the weekend. As laboratories are lIo ne of the o rdin ary media of co-presence can penetrate. Of
closed. for example, the hosp ita l staff know that they cannot get co urse. where walls are thin various kinds of interruptio ns or
cerlain SO riS of tests carried o ut. They try to admit as few new e mharrassments to the closure o f enco unt ers can occu r. Aries,
pati ents as possible at weekends and to avoid in itiati ng new Elias and o thers have po int ed to the ways in which the internal
trea lme nt programmes fo r ex isting inmates. Saturd ays and d iffere ntiation of the houses of th e mass of th e population since
Sundays arc usually 'quiet' days: Mo nd"IY is (hc busiest da y o f the the eighteenth cenlUry has been interrelated with ch'l11ging aspec ts
122 Time, Space and Regiona li ;:ation Front Regions, Back Regions 123

of famil y life and sexuality Y Prior to the eighteenth century in co-presence. Th e 'being together' of co-presence d emands means
Western Europe the homes o f th e poor frequently had only o ne whereby actors are able to 'come together'. Hagerstrand's time-
o r two rooms , in whic h vario us communal living and sleeping geography draws o ur attention to some o f the factors typically
arrangements were found. Th e grander ho uses o f the aristocracy involved here . Communities o f high presence-availability in all
had many rooms , but these usually co nnected directly with one cultures, prior to o nly some hundred years ago, were groupings of
an oth er , without the haJlways which in modern houses permit individu als in close physical proximit y to o ne anoth er. The
types o f privacy th at were fo rmerly difficult to achieve for all corpo reality o f the agent , the limitatio ns upon the mo bility o f th e
classes of society. body in the trajecto ries o f th e du ree o f daily activity , togethe r
Regi onalization may inco rpo rate zo nes of grea t variatio n in with th e physical properties o f space, e nsured that this was so.
span or scal e. Regions of broad spans are those whi ch extend The medi a of co mmunicatio n were always identi cal to those of
widely in space and deeply in time. Of course, the intersectio n o f transpo natio n. Even with the use o f fast horses, ships, forced
'spans' o f space and time may vary , but regions o f considera ble marches, e tc., lo ng distance in space a lways meant lo ng distance
span necessarily tend to depend upo n a high degree of in time. Th e mechanization o f transport has bee n the main factor
institutio nalizatio n. All regions, as defin ed here , involve extension leadin g to the dramatic fo rm s o f tim e-space convergence no ted
in time as well as space. 'Regio n' may some times be used in previously as c haracteristi c o f the mode rn age. But th e most
geography to refer to a physicall y demarcated area on a map o f radical d isj un c ture o f relevance in mo dern histo ry (w hose
th e physical features of the material environm ent. This is not impli catio ns tod ay are very far fro m being exhausted) is the
what I mean by th e te rm , which as used here always carries the separatio n o f med ia of communicatio n, by the developm ent of
conno tatio n o f the struc turatio n o f social conduct across time- electronic signalling, from the media o f transportatio n, the latter
~ pace. Thus the re is a strong degree o f regional d ifferentiation, a lways having in volved , by so me mea ns o r ano ther, the mo bility
10 terms of class relationships and a variety of other social criteria , of th e human body. Morse's inventio n o f the electrom agn etic
between the North and th e South in Bri tain. 'The North ' is not telegraph marks as distinc tive a transition in human cultural
just a geographically delimited area but o ne with lo ng-established, develo pment as th e wheel o r any other technical innovatio n eve r
distinctive social tra its. By th e 'character' of reg io nalizatio n I did.
refer to the modes in which the time-space organization of locales The different aspects of th e regionalization of locales indi cated
is o rdered within mo re embracing social systems. Thus in many above shape th e nature of presence-availability in varying ways.
societies the 'ho me', th e dwelling, has been the physical focus o f Thus the rooms o f a dwelling may ensure thar e ncounte rs can be
family relationships and also o f produ ctio n, carried o n eithe r in sustain ed in different parts o f th e building witho ut intruding upon
parts of the dwelling itself or in closely adjoining gardens or plo ts o ne ano th er , provid ing a partic ular symmetry, perh aps, with th e
of land. The development o f modern capitalism , however , brings routin es o f the day for its in cum bents. But living in close prox imity
abo ut a differentiatio n between the home and the workplace, this within the ho use also means, o f course , high presence-availability:
differentiatio n having consid erable implication s fo r the overall co-presence is very easily sec ured and sustained. Priso ns and
organization of produ ction systems and other major institutio nal asy lums are often associated with enfo rced continuity o f co-
featu res o f contemporary societies. presence among ind ividuals who are no t ordi narily accustomed
10 s uch routines o f daily life. Prisoners who sha re the same cell
may rarely be out o f each o the r's prese nce fo r the who le of the
Fro nt Reg io ns, Back Re gion s
day and night. O n the oth er hand, th e 'disciplin a ry power' of
O ne aspect o f the c haracter o f regio nalizatio n is th e level o f prisons, asylums and other types of 'to tal in stitutio n' is based
prese nce-availability associat ed wi th specifi c fo rms o f locale. T he upon disrupting th e gearing o f presence-availability into Ih e
notio n of 'p rese nce-ava ilabilit y' is an essc mial adjun ct to th a t o f rou tin es o f d aily lnljec to rie..<; 'oulside'. T hus th e ve ry same in mates
124 Time, Space and Regionalization Fro nt Regions, Bac k Region s 125

who are forced into continuo us co-presence are denied the away' expresses the real feelings of those who enact role
avilability of easy encounters with ot her groups in the prison , performances 'up front'. While obviously this may often be the
even though those others may be physically only on the other side case, r think here we come up against the limitations of the
of the walls of the ce ll. The enforced 'sequestrat ion' o f prisoners dramaturgical mod el that Goffman employs, especiall y in his
from the 'outside world', limiting th e possibilities o f co-presence earli er writings, and we see again th e conseq uences of the lack of
to those within a single locale , is, of Cou rse , a defining feature of a a general interpretati on of the motivation of the routines o f daily
'to tal institution'. life . If agents are o nly players on a stage, hiding their {rue selves
We can furth er draw out the relevance o f regionalizatio n to the behind the masks they assume for [he occasion , the social world
structuration o f social systems by co nsidering ho w zoning is wou ld indeed be largely empty of substance. Why, in fact, sho uld
acco mplished in different settings. 'Face' and 'front ' are related th ey bot her to devote the attention they do to such performances
at all? Players in genuine theatre , after all , have a motivation to
front rt'g;on impress the audience with th e quality of their performances.
s ince they are special ists in (hose very performances as
professio nals. But this is a very parti cul ar situatio n. no t in fa ct
o ne generic to social life. To regard it as s uch is to make
something of the same mistake whic h Goffman himse lf identifies
d isc loSllrl'
in analysing talk. Th e 'faultless speech' of the newscaster is
exceptional, and bound up with the presumed expertise of one
who is a specialist in the produ ction of smooth ta lk: in most
contexts o f day-to-day life agents are no t motivated to produce
{his kind of speech.
The sustaining o f onto logical security cou ld not be ac hieved if
bad ro-'I: i on
front regions were no more than fa<;ad es . The whole o f social life
would be, in Su lli van's phrase, a desperate search to pu t on
Figure 8 'sec urity operations' to salvage a sense of self-est eem in the
staging of routines . Those who do feel this way characteristically
first of all to the positioning o f the body in encounte rs. T he display modes o f anxie ty of an ex tre me kind . It is precisely
regio nalization of the body , so important to psychoanalysis ~ heca use there is generally a deep , a ltho ugh generalized , affect ive
which, in Lacan 's phrase, exp lores 'openings on the surface' of in volvement in the routines of daily life that actors (agents) do
the body ~ has a spatia l cou nterpart in the regionaliza tion of the not o rdinarily feel themselves to be acto rs (players), whatever the
co ntexts of interaction. Regionalizatio n encloses zones of time- term in ologica l similarity between these terms. Th eatre can
space, enclosure permitting the susta ining of distinctive relations chall enge social life by its very mimicry in pantomime. This is
between 'front ' and 'back' regio ns. wh ich ac tors employ in pres umably what Artaud means in saying. 'The true theatre has
o rganising the contextuality of action and the s ustaining of always seemed to me the exercise o f a terrible and dangerous act ,
o ntological security. The term 'fa<;ad e' in some part helps to in whic h, moreover th e idea o f theatre and perfo rmance is
designate the con nections between face and front r eg ions.l~ It t.: radi cated ... .'1" Co nsider also Laing's discussion of th e hysteric:
hints. however. that frontal aspec ts of regionalization are
inh erent ly inauth enti c, and that what ever is real or substa nti al is Unless o ne is depressed. il is the others who complain o f selrs lack
hidden behind. Goffma n's discussio n o f front and back regions n f ge nuines." o r sincerity. It is regardctl as pathognomic of the
also te nds fa have (h e same implication : that whatever is 'hidden hys lcric's c harac le riSlic slralegy Ihal his o r her aClions sho uld be
126 Tim e, Space and Regio naliza lio n Disclosure and Se l f 12 7

false, that they should be histrio nic, dramatized. The hysteric, o n routines of social life. and th e ordinary propri eti es, go. Fo r these
the other hand, oft en insists th at his feelings are real and genuine. sorts of occasion do in volve fi xed perfo rman ces for audiences,
It is we who feel th ey are unreal. It is the hysteric who insists o n the though th ere is no necessary implicatio n that those in the back
seriousness o f his intentio n o f committing suicide while we speak regio ns are able to relax th e usual courtesies o f tact or 'repair'.
of a mere 'gest ure' towards suicide. The hysteric complains that he
The level o f enclosure be tween fro nt and back regions is
is going to pieces. It is just in so far as we feel that he is no t going to
neverthe less likely to be very high. since it o ft en holds that the
pieces, except in that he is pretending or making believe that he is,
mo re ritualized the occasio n, the mo re it has to be presented as
that we ca ll him an hysteric. . .. 20
an autono mous set of e vents, in which the bac kstage props are
Thus the differentiation between front and back regio ns by no ke pt entirely o ut of view o f audiences o r observers. It is worth
means coincides with a divisio n between the enclosure (covering po inting out that there is much mo re to the distinction between
up , hiding) of aspects o f th e self and their disclosure (revelatio n, 'publi c' and 'private' acti viti es th an might appear from th e
divulgence). These two axes o f regionalization operate in a seemingly mutually exclusive natu re o f these catego ries. Cere-
complicated nexus o f possible relations between meaning. norms mo nial occasions are distin cti vely. pro to typi call y publi c events,
and power. Back regions clearly often do form a significa nt o ft e n involving 'publi c fi gures'. Bu t the backstage of such
resource which both the powerful and the less powerful can occasions is not a 'private sphere' : th e chi ef figures in the drama
utilize reflexively to sustain a psychological distancing between may be able to relax even less wh en, leaving the ceremonial
their own interpretations of social processes and those enjoined arena, they move among th eir inferio rs, the individuals who are
by 'official' norms. Such circumstances are likely to approximate merely 'behind the scenes'.
most closely to those in which individuals feel themselves to be Ritual occasions seem for the most part di stin ctively different
playing parts in which th ey do nol really 'believe'. But it is fr om the range of circum stan ces in whi ch back regions are zones
important to separate out two types of situation in which this may within which agents recover fo rm s of auto nomy which are
hold, because only o ne approximates at all closely to the co mpromised or treated in frontal contexts. Th ese are often
dramaturgical metaphor. In all societies there are social occasions situations in whi ch sanctio ns are imposed upon actors whose
which involve ritual fo rms o f condu ct and utteran ce , in which the co mmitment to tbose no rms is marginal or no nexistent. The
normative sanctions reguiating 'correc t performance' are stro ng. forms of enclosure and disclosure whicb allow agents to deviate
Such episodes are usually set apart regio nally from the rest o f fro m . or flout. those norms are im portant features o f the dialectic
social life and differ from it specifically in requiring homo logy of o f control in situatio ns involving surve illance. Surveillance , as I
performance rom occasio n to occasio n. It seems especially in have pointed o ut e lsewh ere , connects two related pheno me na :
these circumstances tha t individuals are likely to feel they are t he collation of informatio n used (0 co-ordinate social activities
'playing roles' in which [he self is only marginally involved. Here o f subordinates, and the direc t supervisio n o f the conduc t of
there is lik ely to be tensio n in th e style and continuity o f th ose subordinates. In eac h respect th e adve nt of the modern
perfo rmance , and style may be accentuated much more than in state. with its capit ali st-in d ustria l infrastructure, has been
most day-to..<fay social activity. distinguished by a vast expa nsio n of surve ill ance. 21 Now 'sur-
vt: illan ce', by its very nature, invo lves disclosure, making visible.
Di sclosure a nd Se lf Th e garnering of informatio n discloses th e pattern s o f activity of
th osc to whom thaI info rmatio n refers, and direct supervision
Back regions involved in ritualized social occasions probably ope nly keeps such activit y und er o bservatio n in o rd er to control
often do quite closely resemble th e 'backstage' of a theatre or the if . Thc minimization or manipul ati o n o f conditio ns of disclosure
'off-camera' activities of filmin g and television productions. But is thus o rdinarily in Ih e int eres ts o f those wh ose behaviour is
this backstage may very well be 'on stage' so far as th e ordinary slL bj ect 10 surveillull ce - Ih e morc so accord ing 10 how far what
128 Tim e, Space and Reg iona li zation Disclosure and Self 129

they are called upo n to do in such settings is regarded as way be distinguished fro m those in which the situational
uninterest ing o r noxious. proprieties o f interaction are weak ened or allowed to lapse.
Back regions in, say, settings o f th e shop fl oor include 'odd Th ese are situations in which fro nt , th e detai ls of bodily control
corners' o f the fl oor, tea rooms, toil ets and so on, as well as the and some 'repair' procedures of care fo r o thers can all be relaxed.
intricate zonings o f displacement o f contact with supervisors At least o ne connotatio n o f ' privacy' is the regio nal isolatio n o f an
which workers can achieve through bodily movement and posture. ind ividual - or of individua ls, fo r privacy does not seem
Desc riptio ns of th e use o f such zoning in order LO contro l inevitably 10 imply solitude - fro m the o rdinary demands o f the
properties o f the setting (and there by to sustain modes o f monito ring o f ac tio n and gesture , whe re by 'infantile' types o f
autono my in power relat io nships) are legion in the literature o f conduct are permitted ex pression. T he zoning o f the body seems
industrial socio logy. For instance, here is a worker talking about in most (all ?) societies to be associated with the zoning o f activities
a characteristi c incident on th e floor of a car factory: in time-space in the trajecto ri es o f the day within locales. Thus
eating usually occurs in d efinite settings at definite times, and is
I was working on one side of th e car and the boot lid dropped. It
usuall y also 'public' in the restricted sense of involving gatherings
just grazed the head of th e fell a working opposite me. I can see it
o f family members , fri ends, colleag ues and so o n . T he dressing or
now. He stopped working, had a look round to see if anyone was
watch ing. I was pretending not to look at him - and then he held ado rnment of the body may not be universally treated as 'private'
his head. He'd had enough lik e. You could see him thinking, 'I'm b ut at least in most cultures seems to be so regard ed. In spite of
getting out of this for a bit.' He staggered, I could see him looking Elias's claims that sexual activity was carried on in an unconcealed
round. You know what it was lik e in there. Paint everywhere. He way in medieval Europe!) gen ital sexuality see ms everywhere to
wasn't go ing to fall in the paint ... so he staggered ahoullen yards be zoned as a back-region phenomenon, with many variations, of
and fell down with a moa n on some pallets. It was bloody funny. co urse, in intersecting modes of public and private behaviour.
One of the lads saw him there and stopped the line. The superv isor It seems plausible to suppose that the intersections between
came chasing across. 'Start the lin e ... start the line... .' He start ed regionalization and the expressions o f bodily care are intricately
the line and we had to work. We were working one short as well. It bo und up wi th th e sustaining of the basic security system. Back
took them ages to get him Out of there. They couldn't get the regio ns which allow the indi vidu al complete solitude from the
stretcher in. It must have been half an hour before they got hi m. presence o f o thers may be less important than those whic h allow
Him lying there, y'know. with his onc eye occasionally opening for th e ex pression o f 'regressive behavio ur' in s itu atio ns of co-
a quick look round: 'What's happe ningTn presence. Such regions may permit
Derogatio n o f those in autho rity is obvio usly extreme ly commo n profanity, open sex ual remarks. elaborate griping . . . rough
in s uch situ ations. T he inc ide nt described he re, however, informal dress. 'sloppy' sitting and standi ng post ure, use of dialect
emphasizes the fact that defamatory action o f this SO T[ is not or substandard speech, mumbling and shouting, playful aggressivity
always kept co nfined to the back region, to activities closed o ff and 'kidd ing', in considerateness for the other in minor but
fro m the prese nce o f those who are the targets. potentially symbolic acts, min or physical se lfinvolvements such as
Th e regional zo ning o f activities in many co ntexts of this sort humming, whistling, chewing, nibbling, belching and flatulence. l <
connects closely, of course , with the seriality of encounters in Far from represe ntin g a diminutio n o f trust, these types of
time-space. But again it does no t clearly converge with a divisio n hehaviour might help to reinforce th e basic tru st in the presence
between public and private act ivity. The worker makes no attempt Df intimates originally built up in relation to the parental figures.
to disguise to hi s workmate that the act of malingering is direc ted Th ey are mark ed not by th e so rt of upsurge of anxi ety brought
towards tempo rarily escaping fro m the pressures of th e assembly ahou t by critical situ atio ns bU I the reve rse - a dissi pation of
lin e. Such fro nt / back differe ntiations - ord inari ly occurring in tensio ns derivin g fro m th e dema nds of ti ght bodily and gestural
circumstan ces o f ma rked imbalan ces of power - ca n in a ge ne ral l:on tro l in o th er settings of day-Io-d ay life.
130 Time, Space and Regiona li za tion Regionafization as Generic 131

centres , and cities have th eir centres, so too do the daily

Reg ional izat ion as Gener ic trajectories of indi vidua l ac to rs. In modern socie ties, for the
majority of males at least, the ho me and workplace fo rm the two
The differentiatio ns between enclosure , disclosure, back and main centres in which the d ay's activities tend to be concentrated.
fro nt regions, apply across large spans o f time-space, no t only in Locales also tend to be centred regio nally. Some rooms in a
the contexts o f co-presence. Th ese are , o f course, unlikely to be ho use, such as spare bedrooms. for example, may be used only
as directly mo nito red refl ex ively by those whom they affect, 'peripherally' .
altho ugh such may be the case. Regio nalization within urban Centre/ periphery distin ctio ns tend frequ ently to be associated
areas in contemporary societies has been much studied since th e with e ndurance ove r time. 2sThose who occupy centres 'establish'
early work o f the Chicago sociologists Park and Burgess. In most the mselves as having control o ver resources whic h allow them to
Western soc ieti es, the zoning o f cities into neighoou rhoods with maintain diffe re ntiatio ns be tween th emselves and those in
markedl y different soci al characteristics is strongl y influ enced by peripheral regions. Th e established may employ a variety of
the operation of housing mark ets, and by separations between form s o f social c1osure2f> to sustain d istance fro m others who are
individu ally own ed ho mes and state-o perated housing sectors. effe ctively treated as inferio rs o r o utsiders.
Neighbourhoods may no t be zoned as symmetrically as so me of
the 'ecological' urban analysts suggested, but their distribution
esta bli shed
has th e consequ ence o f creating various sorts of front / bac k
contrasts. Industrial areas in northern towns and cities in England
were o nce th e most visibl e features of tbe built environm ent -
factories and mills, as ir we re, pro udly displayed. But the tendency
in urban planning in recent years has been to treat such areas as cent,,, 1 L - -- - -- I - - - -- - - - - 1 pe,iphe,,,1
unsightly, as back regions to be hidden away in enclosed enclaves, region s ["" reg ion'
or transferred to the edge o f town. Ex amples can easi ly be
multiplied. The access of those in more afOuent sec tors o f housing
markets to relatively easy transfer of property underli es the 'fli ght
to the subur bs'. changing city centres from regions o f fro ntal
display to back regio ns o f urban decay, which the 'respecta ble
classes' avo id . G heno areas may be rendered 'in visible' by the ir Figure 9
regio nal enclosure in ne ighbo urhoods having very low rates both
o f pro perty transfer and o f daily mobility in and o ut of those
neighbo urhoods. As always. vario us types of time-seri es ph eno- T he 'established ' ind ustrial natio ns o f th e Western 'core'
mena und erli e such spatial regio nalization. Illaint ain a central positio n in the world econo my o n the basis of
Reg io nali zatio n across long spans of time-space has bee n Ih e ir temporal precedence over the 'less develo ped' societies.
analysed by many writers in terms of familiar notions su ch as T he geopolitical regionaiizatio n of th e world system may be
'uneven develo pm ent' and distin ctions between 'centre' (or 'core') ch:mging - with, for example, shifts in centres o f manufacturing
and 'periphery'. Th ese noti ons, however, can be applied across [lmdu ction Lo erstwhile pe riph eral zo nes in the East - but the
the wh ole ran ge o f the setlings of locales, from large to small. 1':1(.:101" o f prio rity in tim e has so far decisively influenced pre-
Ratb er than di scussing the th eme of uneven developm ent he re, I cmill cnce in space. Within mHi o n-stat es ce nt re/periphery region-
shall deve lo p th e d ifferenti ati on of ce ntre and periph ery by ali zati o n seems eve rywh ere to be assoc iated with th e existence of
relatin g it LO emheddedness in tim e. If the world econo my has its 'cswhli shm ent s' that li e at th e co re o f th e stru c turati o n o f
132 Tim e, Space and Regionalization Time, Space, Context 133

do minant classes. n Of co urse , there are a variety o f compl ex nature o f the tim e-space paths traced out by acto rs in modern
relations involved in these phenomen a, and I offer these examples societies. The po int may, o n th e face of things, appear to be a
as purel y illust rative. banality but is actu all y very far from being so. What is al issue is
no t just differem mea ns of recko ning time, but di vergent fo rms o f
Ti me, Space, Context [he struc turatio n o f daily ac ti vities.
Consi der , fo r in stance , Bou rdie u's well-kno wn disc ussion of
Let me at this point offer a summary o f the main points in thi s time and time-reck o ning in Ka bylia. Here th e year is considered
chapter so far. Th e di sc ussion has been concerned with the to run fro m autumn towards summ er an d the day from evening
contextuality o f social life and social institutio ns. AI! social life towa rds noon. This scheme expresses , however, a conceptio n of
occ urs in , and is co nstituted by, intersectio ns o f presence and time as eternal recurrence, which is in turn part o f the basic
absence in th e 'fading away' o f time and th e 'shading o ff' of space. compositio n o f day-lo-day acti viti es. Night is symbolically a time
The physical properties o f the body a nd the milieux in whi ch it of death, mark ed by regular ta boos - aga in st bathin g, coming
moves inevitably give social life a serial characte r, and limi t into contact with stretches of water , looking in a mirror, anointing
modes of access to 'absent' o thers across space. Time-geography th e hair or lOuching ashes.29 The morning is no t just 'd ay break'
provides an impo rtant mode o f no tatio n o f the intersectio n o f bUI a triumph in th e struggle between day and night : to be 'in the
lim e-space traj ectories in day-to-day ac tivity. But it has to be mo rnin g' is to be open to the light , to th e benefice nce that is
in serted within a more adequate theorization both of the agent assoc iated with it. The 'opening' of the day is thus a time for
and of th e organization o f the settings o f interacti on. In proposing go ing o ut , when peopl e po ur from their ho uses to their work in
the ideas o f locale and of regionaJiza tion I want to fo rmulate a the fie lds. Ge lling up ea rly means pu tting oneself under
scheme o f concepts which help to categorize contextuality as fav o urabl e auspices, to 'do ho no ur to the angels'. It is no t just a
inherently in volved in the connection of socia l and system transitio n in time but a keying o f events and practices. Never-
int eg rati o n. 2~ thel ess , the creati ve potemi al of the day must be fostered by
magic or o th er mali gnant forces can int ervene , partic ularly
d ,u ly tl m "~Cl' path~
fo llow ing the zenith o f the sun 's rise. For afte r this the day goes
int o decline , signa lling the imminent return o f the decadence and
di strib utio n of "nCOun te' 5
decay of night , 'th e paradigm o f all form s of dedin e' ..lO
(c~io n al iza ti o n o f loca liO! Bearing this example in mind , let me develo p some of th e main
COntiOKl u", lily of ...-gions no tio ns consid ered in thi s cha pter, taking as an illu stration
school ing in co ntemporary societies_ There is no do ubt that
in(C. St'Clion 01 fo ca lc s
mapping the time-space pa tterns followed by pupils, teachers and
o th er staff in a sc hoo l is a useful to pological device with which to
The graphi c techniqu es d eveloped in time-geography have hegin co study th at school. Rather than using the ex act fo rms of
al read y proved their frui tfulness in several areas o f research. representation fo rmulated by Hagerstrand and his co-wo rk ers ,
There is no reason a[ all why those wo rking in a range o f fields in however, I propose to emphasize the 'reversible time' o f day-to-
the social sc ie nces sho uld no t adopt , and adapt , Hage rstrand 's day routin e conduct. Hagerstra nd usually po rtrays time-space
method o f no tation. But th e limitatio ns of time-geography , as path.'; as having a 'linear' mo ve ment through the day. Bu t a more
indicated above , must cert ainly also be borne in mind . Moreover, ;I(.:r..; ural e represe ntation of th e repetitive character of d ay-Io-day
'clock time' sho uld no t be accept ed simpl y as an unquestio ned sor..:ial life is give n if we see that most daily time-space paths
d imensio n o f the constru ctio n of to pographical models. but must in vo lve a return '. Instead o f ado pting [h e fo rm o f fi gure lOa we
be regarded as itse lf a soc ially cond itio ned influe nce upo n the might tak e as ex arnplary th at o f rigure lOb.
134 Time, Space and Regionalization Time, Space, Conted 135

'Statio ns' tend to be black boxes, as it were , in time-geography,

because the main focus is upon movement between them. As a
type of social organizati on, concentrated upon a locale having
definit e physical characteristics, the characteristics o f a school
can be understood in terms of three features: the di stribu tion of
encounters across time and space occurring within it , th e in ternal
regio nalization that it displays, and the contextuality o f the regions
l imo:
thus identified .
Mod ern schools are disciplin ary o rganizatio ns, a nd their
bureaucratic traits cl early both influence and are influ enced by
Figure 10a Figure 10b the regions they contain . Like all forms of disciplinary organization ,
th e school operates wit hin closed bound aries , its physical borders
Figure lOa is of the sort favo ured by Hagers(rand , in whi ch we being cut o ff rather clearly (ro m d ay-to-day interactio n o utside. A
look at time-space 'laterally' and th e 'time' arrow makes o ut a school is a 'container', generating d isciplinary power. The
spec ific temporal seq uence (usually equivalent to the wo rking enclosed nature o f school life makes possible a strict co-ordination
day). I pro pose not to abandon this type of notatio n but to of the serial encounters in which inmates are invo lved. Th e
supplement it - certainly conceptu all y, if not figurati vely - with segments of childre n'S time that are spent in schoo l are spatially
figure lOb , in which we are looking 'down' , as it were, rath er than and temporally sealed o ff from potenti ally intru sive enco unters
laterally. The lines mark ed with the arrows represent paths of ou tside. But this is also tru e, usually at least, of the divisions
time-space mo vement . The length of the lines refers to th e amo unt be tween different classes. Sc hools are internally partitio ned.
o f ti me, measured c hronolog ically, spent moving betwee n T he re may be some areas in a school, and some times , when
'statio ns' in the course o f a particul ar day by a partic ular or heterogeneous o r unfocused fo rms of interactio n tend to o ccur
typi cal individual ; th e d egree of elo ngatio n of the boxes indicates - e.g. at the beginning and end of classes. But for the most part
how long is spent within a specifi c locale. Thus a c hild 's day in th e distribution of encounters within a school contrasts drama-
school term looks something like the scheme indi cated in the ti cally with sectors o f social life in which the normative regulation
diagram. The child may spend th ree discrete periods in th e ho me of activity is looser. Disciplinary spacing is part o f the architectural
(H) per day - sleeping there from the middle of the evening until character of schools, both in the separatio n o f classrooms and in
the early mo rning, returning there from school (S) in the late the regulated spacing of desks that is often found inside them .
aft ernoon and coming back again aft er having bee n o ut to the T here is no doub t that spatial divisio ns of this sort fac ilitate the
cin ema (C) in the evening. Some aspects of the child's day are no ro utinized specificatio n and allocatio n o f tasks.
doubt stro ngly ro ut in ized (th e jo urn ey to school and back ), T he school timetable is fundam ental to the mo bilization of
whereas o thers (goin g o ut to th e cin ema ) may be less so. The space as co -ordin ated time-space paths. School administrators
most ro utinized types o f ac tivity ca n be represented as a profil e no rmally do no t face the same pro blems o f 'packing' as their
of time-space paths embedded in reversible time. counterparts in hospitals do. But, like aUdisciplinary organizations,
A school, in Hage rstrand 's terms, is a 'station' alo ng the sc hools o perate with a precise eco nomy o f time. It is surely right
converging paths traced by clusters o f individuals in the course of to trace th e origins of school disciplin e in some part to the
the day. He is right to point out th at th e conditions whi ch make it reg ul atio n of time and space which a ge neralized transition to
possible for ind ivid uals to come toget her within a single locale 'clock ti me' makes possi bl e. Th e point is no t that the widespread
canno t be take n fo r granted but have ( 0 be exam ined d irectl y. lise o f cloc ks ma kes fo r exact di visions o f the day; it is that time
But a locale is. o f course. more tha n a mere sto pping-po in! . cnl ers into the ca lcul'lti ve applicat io n o f administrati ve autho rity .
136 Time. Space and Regionalization Tim e, Space. Context 137

The con textual features of classrooms, as the main 'areas of A child comes in late, looking sheepish, and walks to his seat.
application' of disciplinary power, obviously vary widely. But in Other chi ldren point and laugh.
more severe forms of classroom spacing the specification of CHI LD: Hey, Duncan, what are you doing'!
bodily positioning, movement and gest ure is usually tightly TEACHER: DUlican, come here. You 're lat e again, three minutes
organized. The spa tial positioning of teacher and pupils in the late to be exact. Why'!
contex t of a class is quite different from that of most other DUNCAN : Sorry, sir.
situatio ns in which face e ngageme nts are carried on. Indeed , it T EACHER: I said, 'Why'?'
usually signals a collapse o f the teacher's control if such situatio ns DUNCAN : I slept in, sir.
come into being. The seeming minutiae of bodily posture and TEA CHER : Well, are you awake now?
mo bility [ 0 which GoHman draws attention are once more far (Other children laugh.)
from incidental here. DUNCAN: Yes, sir.
The classroom, like the school, is a 'power container'. But it is TEACHER: Well you'd better stay behind for three minutes at
not o ne that merely churns out 'docile bodies'. Contexts of co 4 o'clock and don't go to sleep again after that.
presence, as I have emphasized, can be described as settings, a nd
More laughter, Duncan sits down. Teacher finishes register. J \
settings have to be reflexively activated by authority figures in the
course of making that authority count. Discipline thro ugh What is going on here? We have to recognize, as the teacher
surveillance is a potent med ium of generating power, but it none does, that registration has a parti cular signifi cance for the ordering
the less depends upon the more or less continuous compliance of of the day's activities. It is a marker that signals the opening of
those who are its 'subjects'. The achievement of such compliance the brackets in an encounter, and it is the first salvo fired in a
is itself a fragile and co ntingent accomplishment, as every teacher battle that is joined daily between teacher and pupils. The teacher
knows. The disciplinary context of the classroom is not just a recognizes it as the first occasion to test the mood of the children ,
'backdrop' to what goes on in the school class; it is mobilized as the children do in respect of the teacher. The teacher's
within the dialectic of co nt rol. A school class is a face engagement maintenance of directive co ntrol depends upon ensuring that the
which has to be reflexively managed, like any other. children assume (he routines involved in the classroom setting.
Consider the fo llow ing st rip o f interaction , described and O n e ntry to the classroom in the mo rning the childre n are
discussed by Pollard: ex pected to sit in the ir assigned places , get o ut their reading
books and answer to their na mes whe n they are called out.
Bell for 9. 0 a.m. goes, about half class in, mostly reading books.
Poll ard interprets the teacher's jo king and teasing as a front
Teacher enters breC' ily: 'Morning - ah, that's good. getting those
IJerforman ce, which is intended to set the lo ne o f the day as one
books out.' Teacher sits al desk. tidies up, gets register our.
Meanwhile most of the other children have come into the of co-operative work, However, this strategy has its risks , as is
classroom. The later arrivals talk, swap some football cards, indicated by the response to a late arri val o f o ne of the children,
occasionally glance at the teacher. Another feels able to tease the latecomer. The teacher at once
recognizes this as the first test case of the day, in respect of which
TEACHER ~ Right. let's do the register, then, hurry up and sit down his superior authority must be demonstrated, His bantering rebuke
you football maniacs - I see that Manchester United lost 10 Duncan mixes appeal with firmn ess, a tactic shown to be
again. success ful by the laughter of the children. Thus the events of the
MANCHE.,<;;TER UNITED SUPPOHTCRS: Oh yeah, well they're still bett er day move on. If the teach er had been more overtly disciplinarian
than Liverpoo l. and had sent th e miscrea nt to the head, the response co uld have
TEACHER : (Jokey sarcasm in voice) Really? It must be all the
heen judged too seve re by the rest o f the c hildren. The result
spinac h they don't cat. Now then . .. Martin ... Doreen.
th en might have been an escalatio n o f threat and punishment
Alan ... Mark (calls register and children answer) .
138 Time, Space and Regionalization Against 'Micro' and 'Macro" Social and System Integration 139

less effective in sustaining routine than the 'effort bargain' which associations may make their power felt in influencing policies
teacher and pupils have implicitly concluded as part of a more that help to shape the life of the school. But it is intrinsic to
co-operative atmosphere. disciplinary power that what goes on in the 'power container' of
The very nature of classrooms, in which most things both the school has a significant degree of autonomy from the very
teachers and children do are visible each to the other, means that outside agencies whose ethos it expresses.
back regions usually have a strong temporal as well as spatial
definition. For children these lie in some part along the narrow Against 'M icro' and 'Macro': Social and System Integration
temporal boundaries between classes, whether or not they involve
physical movement from one classroom to another. Although the The foregoing considerations are of some importance in
weight of discipline normally bears down most on the children, it examining the relations between social and system integration. J
is sometimes felt more oppressively by teachers. Teachers usually do not employ the more familiar terms , 'micro-' and 'macro-
have a back region to which they can retreat, the staff room, socio logical' study. for two reasons. One is that these two are not
which children ordinarily do not enter. The staff room is no infrequently set off against one another, with the implication that
doubt a place for unwinding and relaxation. But it is also we have to choose between them, regarding one as in some way
somewhere in which tactics of coping with teaching tend endlessly more fundamental than the other. In Goffman's studied refusal to
to be discussed, formulated and reformulated. be concerned with issues of large-scale social organization and
It is in the nature of disciplinary organizations that the intensity history, for example, there seems to lurk the idea that in what he
of surveillance inside inhibits direct control from outside. This is sometimes calls microsociology is to be found the essential reality
a phenomenon which can be seen both in the internal of social life. On the other hand, advocates of macrosociological
regionalization of the school and in its situation as a locale within approaches are prone to regard studies of day-to-day social
other locales. Inside the school the concentration of disciplinary ac tivity as concerned with trivia - the most significant issues are
authority in separately partitioned classrooms is the condition of those of broader scope. But this sort of confrontation is surely a
the high level of control over bodily positioning and activity phoney war if ever there was one. At any rate, I do not think that
which can be achieved. But this circumstance also acts against there can be any question of either having priority over the other.
the direct supervision of the supervisor. The head is 'in authority' A second reason why the micro/macro division tends to conjure
over the teaching staff, but such authority cannot be exercised in up unfortunate associations is that, even where there is no conflict
the same way as teachers endeavour to control the conduct of betwee n the two perspectives, an unhappy division of labour
children in their classes. Schools therefore tend to have a rather le nds to come into being between them. Microsociology is taken
sharply opposed 'double line' of authority. The control which 10 be concerned with the activities of the 'free agent', which can
teachers seek to exercise over their pupils is immediate , involving safely be left to theoretical standpoints such as those of symbolic
the teacher's continuous face-to-face presence with the children. interactionism or ethnomethodology to elucidate; while the
Supervision of the activity of teachers, however, is necessarily province of macrosociology is presumed to be that of analysing
indirect and proceeds by other means. One might hazard a guess lh e structural constraints which set limits to free activity (see
that it is only in organizations in which a considerable amount of pp . 211). I have made it clear previously that such a division of
autonomy from direct supervision is given that a graduated line lahour leads to consequences that are at best highly misleading.
of authority can be achieved. The enclosed nature of the school, Why should the issue of the relation between 'micro-' and
and its clear separation in time and space from what goes on in 'm:tcrosociological' st ud y be seen as so problematic by many
surrounding locales, also inhibits supervisory control from the wrilers'! T he co nceptual division of labour just referred to is
outside, however. Thus inspectors may visit sc hools regularly to presumahly th e main reason. Reinforced by a philosophical
check upon th eir operati o n; hO<lrds o f governors and parents' dualism. it demands a more thoroughgoing reformulation of soc ial
140 Tim e, Space and Regionalizatkm Agaimt 'Micro ' Jnd 'Macro " Socia l and System IllI egration 141

theory than most authors are able or willing to contemplate. It simil ar to that advocated by Sla u (see pp . 208 - 10), Collins quite
will help to develo p this point to look briefly at one of the more rightly questions the sort of version o f 'structural soc iology' which
interesting recent discussions of the issue , that offered by Colli nsY Blau and many others propose. But in other respects . Collins's
Collins points ou t that the schism between micro- and macro- view is wanting. As I have consistently stressed , to trea t time and
soc iological approaches, as these terms are ordinarily understood , space as 'variables' is to re peat th e cha racteristic error o f most
has become accentuated over the past decade or so, While social fo rms of orthodox social science. Mo reover , why should we
theory was do minated by fun c tionalism and Marxism , or some assume that 'stru cture' is relevant o nl y to macrosocio logical
combination o f th e two , social relations in situatio ns o f co- issues? Both in th e more precise and in th e vague r senses of the
presence were typically regarded as substantially determined by term I have distinguished activi ty in microcontexts has strongly
broader , 'stru ctural ' fa ctors. However, as led especially by defined structural properties . I tak e this, in fact , to be o ne of the
ethnomethodology , mi crosociology has become a burgeoning main claims which ethnomethodological research has successfull y
field of interest and one in which the presumptions of the above sustained. Moreover , why hold that time as a 'variabl e' is relevant
approaches have been tak en to task in a fairly radical fashi on. In o nly to macrosociological concern s? Temporality is as inseparable
Collins's view , 'the newer, radical microsociology is epistemo- from a small strip of interacti on as it is from the longest of
logically and empi rically much more thorough than any previous tonglles dlln~es. Finally, why propose that stru ctural properties
method. , , . I would suggest that the effort coherently to co nsist only of three dimensions, tim e, space and number'! The
reconstitute macrosociology upon radically empirical mi cro- reason, I assume , is that Collin s still has in mind thaL 'structure'
foundations is the crucial step LOward a more successful must refer to something 'outsid e' the act ivities of social agents if
sociological sc ience,'.JJ it is to have any sense at all in social sc ience. Dispersion in time
According to Collins, the proper way forward is via a and space seems the only phenomenon left, give n that Collins
programme of the 'microtran slation' of 'structural phenomena' . accepts a good deal of the criticisms that have been levelled by
Such translation is likely to eventuate in theories which have a (h ose whom he calls 'radi cal microsoc io logists' against the
stronger emp irical basis than existing macrosociological theories. collective concepts with which their macrosocioiogical antagonists
Those who are concerned with macrosociological issues are called usually operate.
upon no t to abandon their endeavours but to recognize that their But the most important confusion in Collins's account is the
work is theoretically inco mplete. There are, in Collins's eyes, assumption that 'macroprocesses' are the 'resuhs' o f inte rac tion
only three 'pu re macrovariables': Lime, space and number. Thus a in 'microsituations'. According to Collins, the 'macrolevel' consists
concept such as 'centralizatio n of authority' can be translated o nly o f 'aggregatio ns o f mic ro-experie nces'. Now, it can be agreed
into accounts of microsituatio ns - how situated actors actually thai generalizatio ns in th e social sciences always presuppose -
exert authority in describable co ntexts. However the 'pure and make at least implicit reference to - the intentio nal activities
macrovariables' ent er in as the number o f situations of such a o f human agents. However , it does not fo llow fr om this that what
sort , in time and in space. 'H ence structural variables often turn is described as the 'macrolevel' has a rath er sham ex istence. This
out to be sheer numbers of people in various kinds of mi cro- on ly takes us back to th e phon ey war. Social in stitutions are not
situations,').! 'Social reality', then, is 'micro-experience'; it is the ~x plicable as aggregates of 'm icrosituations', no r fully describable
numerical temporal and spatial aggregations of such experien ce in terms that refer to suc h situ atio ns . if we mean by these
which make up th e macrosociological level of analysis. Th e circumstances of co-presence. On t.he other hand , institutionalized
'structural' qualiti es of soc ial systems are the 'results' Collins says, patterns of hehaviour are deeply impli cated in eve n the most
of condu ct in mi crosituations, in so far as they do not de pend fl ~cti n g and limited of 'microsi tuatio ns'.
upon numbe r, tim e and space. Let us pursue thi s thought hy indi ca ting why (he micro/ macro
Alth o ugh Collins's co nce pt of 'struct ural variables' is somew hat di stinction is not a partic ularly useful o ne. What is a 'micro-
142 Time, Space and Regionalization Agains t 'Micro' and 'Macro' ; Social and Sy . tem Integration 143

situatio n'? T he response migh t be: a situation of interaction reproduce , basi c institutional parameters of the social systems in
confined in space and tim e - seemingly Collins's view. But this is whi ch th ey are implicated . Tribal societi es (see pp. 182- 3) tend
not very helpful. For not o nly do en counters 'slide away' in time to have a heavily segmental form, the village community being
but also once we start being concerned with how encounters are overwhelmingly the most impon ant locale within which encounters
carried o n by their participating actors , it becomes clear tha t no are constituted and reconstituted in time-space. In these societies
strip o f imeracrion - even if it is plainly bracketed, temporally relatio ns of co-presence tend to do mina te influ ences o f a mo re
and spatially - can be understood on its o wn . Most aspects o f remo te kind . It makes sense to say that in them there is something
interactio n are sedimented in time , and sense can be made o f o f a fu sio n o f social and system integratio n. But obvio usly such a
them o nly by considering their ro utinized , repetitive characte r. fusio n is never co mplete: virtually all societies, no matter how
Moreover, the spatial differentiation of the micro and macro small or seemingly isolated , exist in at least loose connection with
becomes im precise o nce we start to examine it. For the forming wider 'intersocietal systems'.
and reforming of encounters necessarily occurs ac ross tracts of Since we now live in a wo rld where elec tro nic communication
space broader than those in volved in immediate co ntexts of fa ce- is taken fo r granted , it is worth emphasizing what is otherwise a
to-face interactio n. The paths traced by individuals in the course self-evident feature o f traditio nal socie ti es (of all societies, in fac t,
of the day break off some co ntacts by moving spatially to fo rm up to a little over a century ago). This is simpl y th at all contacts
others, which are then broken off and so on. between members of different communiti es or societies, no matter
What is normally talk ed about under the heading o f microl how far-flung, involve contexts of co-presence. A letter may
macro processes is th e positioning of the body in time-space, th e arrive from an absent other, but of co urse it has to be taken
nature of interactio n in situations of co-presence , and the physically from one pl ace to ano th er. Very lo ng jo urneys were
connection between these and 'absent' influences relevant to th e made by specialized categories of peopl e - sailors, the military ,
characterizatio n and expl anatio n o f social condu ct. T hese merchants, mystics and diverse adventurers - in the traditional
phenomena - th e anchoring concerns, in fact , of structuration world. Nomadic societies wou ld roam across vast tracts o f land.
theory - are better dealt with as concerning the relations between Popul atio n migrations were com mon. But no ne of these
social and system integratio n. Now, some of the questions at issue pheno mena alters th e fact that contexts of co-presence were
in the microl macro debate are conceptual problems to do with a lways the main 'carrying contexts' o f interactio n.
the long-standing controversy o ver methodological individu alism . What made possible the larger time-space 'stretch' in volved in
These I shall leave aside until the next chapter. Other aspects, what I shall call class..<fivided societies was above all the
however, do no t rest upon solely conceptual consideratio ns. They develo pment of cities. Cities establish a centralization of resources
can be resolved o nly by directly analysing partic ular types of - especially administrative resources - that makes for greate r
society. Because societies differ in their modes o f institutio nal lime-space distanciatio n than is typically the case in tribal orders.
articula tion, the modes of intersectio n of presence and absence T he regionalizatio n of c1ass-d ivided societies, howeve r compli-
that enters into their constitution can be expected to vary. I shalt cated it may be in detail , is always for med around the connections,
indicate this briefly here , introducing at the same time material to of bo th interdependence and antagonism, between city and
be expanded upon in the next chapter. cou ntryside.
Social integration has to do with interaction in contexts o f co- We tend to use th e term 'city' in an enco mpassing fashion to
presence. The connectio ns between social and system integratio n rder both to urban settlements in trad itio nal societies and to
can be traced by examining the modes of regionalization which Ihose convergent with th e fo rm atio n and spread of capitalist-
channel, and are chann elled by, the time-space paths that the imtuSlrialism. Bul this is a n obfu scating usage if it is taken to
members of a community or society foll ow in th eir day-lo-d ay impl y that in modern limes we merely have mo re of th e same -
activiti es. Such path s are strongly influ enced by, and a lso Ihal Imlay's urbani sm is o nl y a denser a nd mo re sprawling version
144 Time, Space and Regionaiization

of what we nt before, The contextualities of traditio nal citi es are

in many respects different fr om those of modern urbanism.
Rykwert, for example , points out the symbolic form that many Criti cal Notes: Foucault on Tim ing and Spaci ng
cities had , in widely removed parts of the world , prio r to modern
It is diUicull [for us today I to imagine a situation where the fo rma l
Foucault's various discussio ns o f the o rigins o f disciplinary power
order of the universe could be reduced to a diagram o f two
demo nstrate a persistent concern with temporal and spati.al
intersect ing coordinates in o ne place. Yet this is ex actly what did
distributio n. According to Foucault , d isciplin ary power has as us
happen in an tiquity: the Roman who walked along the cardu knew
that his walk was the ax is around which the sun tu rned, and that if focus the manipulation of the body , rega rded essen~i~ lI y a~ a
he followed th e d eCllm(llllls . he was follow ing the sun's course. The machine that can be fin ely tuned. The forms of admmlstratlon
whole universe and its meaning cou ld be spelled out of his civic associated with the disciplinary organizations which have
institutions - so he was at home in it.ll mushroomed from the eighteenth century onwards are different
fro m the mass mobilizatio n of labour power fou nd in large-scale
Such cities, we cou ld say , do not yet exist in commodified time projects in agrarian civi li zations. Such projects - road-building,
and space:)6 Th e buying and se lling of time , as labour tim e, is the construction of temples, public mo nume nts and so on -
surely one of th e most dist inctive features of modern capitalism. o ft en involved large numbers of peop le . But their activities were
The origin s of th e precise te mpo ral regulation of the day may co-o rdinated only in a gross fashion. The new forms of discipline
perhaps he fo und in th e chim e of the monastery bell , but it is in are tailored precisely to movements, gestu res and attitudes of the
the sp here of labour th at its influ ence became embedded in such individual body. Unlike mo nastic disciplin e, which is one of its
a way as to spread throughout society as a who le. The main historical forerunners, the new techniques of power connect
commod ificatio n of time, geared to the mechanisms of industrial discipline directly with utili ty. The con tro l o f the body ~s part of
production , breaks down [h e differentiation of city and cou nt ry- the novel 'po litical anatomy' and as such, Foucault says, mcreases
side cha racteristic o f class-d ivided societies. Modern ind ustry is the output of the body while also reducing its independence o f
accompanied by the spread o f urbanism , but its operatio n is no t orientatio n.
necessarily fi xed in any part ic ular type of area. The trad itio nal Discipline can proceed o nly via the manipulation of time .and
city, o n the o ther hand , is both the main locus o f d isciplin ary of space. h o rdinarily req uires enclosure , a sphe re of o perauons
power in class-divided societies and , as such, set off fro m the closed o rr and closed in upon itself. Foucault makes a great deal
countryside - very o ften, physically and symbolically, by the city o f the concept o f 'confinement ', the more or l es~ fo:ci ble
walls. Together with the transfo rmation of time, th e commodifi- separation o f individu als from the rest of the popu lation III the
cation o f space establ ishes a 'created environment' of a very early hospitals , in mental asylums and in prisons. However, other
distin ctive character, expressing new forms of instituti onal less embracing disciplinary o rganiz.ations also involve enclosure.
articulation. Such new forms of institutional order alter the The factors leading to the establishment of closed areas may
conditions of socia l and system integration and thereby change va ry, but th e end result is similar in all of them , .in ~o~e degree
the nature of the co nn ect ions between the proximate and remote hecause similar mod els were foll owed by the mdlvlduals and
in time and space. authoriti es responsible for setting th em up. Enclosu:e. is a
generalized basis of disciplinary power, but taken alone It IS not
enough to permit th e detail ed management o f the movements
and activit ies of th e body. This ca n be achieved o nly through
in ferna l reg iona l divisio n o r purtilio ning'. Eac h individual has his
146 Time, Spa ce and Regionalization Critical No tes : Fo ucJr on Timing and Spacing 147

or her 'proper place' at any particular time of the day. The largest of the buildings, which was 110 m etres long, with three
partitioning o f disciplinary tim e-space has at least two con- storeys. On the ground fl oor block printing was carried out.
sequences. It helps to avoid the format ion of large groups which There were 132 tabl es, set up in two rows running the le ngth of
might be a source of independent will formation or o f opposition. the workshop ; two employees worked at each table. Supervisors
and it allows for the direct manipulatio n of individual activities. would walk up and down the central aisle , being there by able to
avoiding [he flux and indeterminacy which casual enCOUnlers supervise the labour process in gene ral and th e activities of each
tend 1O have. Wha t is in volved here, according to Foucault , is an individual worker in particular. Work ers could be compared for
'analytical space', in whic h indi vid uals can be watched and their speed and productivity and the ir activities correlated with
assessed , their qualities measured. Th e partitioning of disciplinary o ne another. By assorting workers according to strict principles
space may have been infl ue nced by the example of the mo nastic o f classific ation , each elemenr o f th e labour task could be
cell , but oft e n o riginat ed also in architectu ral forms th at were c haracterized and related to discrete mo tio ns o f the body. The
established fo r pu rely practi cal purposes. In France the naval doctrines of Frederick Taylo r are no t much more than a late
hospital at Rochefort served as a model. It was set up as part of for mulatio n o f the disciplinary power that accompanied the rise
an attempt to cope with th e contagious disorders rife in a po rt o f large-scale industry ove r a century earli er.
teeming with num erous disparate groupings of people engaged in T he character of disciplinary space, according to FOllcault,
war or trade. Co ntrOlling th e sp read of disease involved other derives primarily not from the associati o n of an o rga nization with
kinds of supervisory reg ulatio n o f transient populations - that of a specific piece of territory but from th e farming of space. Lines,
the military over deserters and of th e local administration over columns, measured walled intervals are its distin guishing features.
the flow of goods, rations and raw materials. This led to pressure It is not any particular part o f th e bui lding that matters, but its
for the rigorous control of space, which first involved caring for o verall relational form. T he classroom exemplifi es this pheno-
valuable commodities rather than organizing human beings. But menon. In the eighteenth ce ntu ry , in France and elsewhere ,
the practi ce of tagging goods, categorizing and controlling th eir classes come to be divid ed in tervally into clearly delimited rows,
distributio n was later a pplied to patie nts. Case records began to externally separated by a connecting system o f corridors. These
be ke pt. The overall number o f patie nts was carefull y regu lated; are curricular as well as spatial divisio ns. Ind ividua ls move through
restrictions were placed o n their movemenl and the tim es at such partitions not o nly in the course of the day but also during
which they were visited. T he emergence o f 'therapeutic spac ing' their educational careers.
thus was develo ped fro m 'administrative and po litical ' spac ing.' In organizing 'cells', ' places' and 'ran ks', t he disciplines create
The partitio ning o f space came about in rat her differe nt complex spaces that are al o nce architectural. fun ctional a nd
circ umstances in factories in the late eighteenth centu ry . He re hierarchical. It is spaces that prov ide fi xed posit io ns and permit
the te nde ncy was also to distribute individuals in demarcated circulatio n; they ca rve out individua l segmen ts a nd establ is h
space, but this distribution had to be directed towards the co- operational links: they mark places and indicate va lues; they
ordination o f machin ery. Thus the arrangement of bodies in guarantee the obedience of individ uals, but a lso a better economy
space had to co rrespond to th e tedmical demands of productio n. of time and gesture.!
But this 'articul atio n o f productio n space' can al so be shown to
Disc ipline depends upo n th e calcu lative divisio n of time as well
have been infused with disc iplinary power. Foucault quotes the
as space. The monastery, after <t il , was o ne of the first places in
Oberkampf manufactory at Jo uy as an example. The manufactory
which the day was temporall y regulated in a precise and ordered
was co nst ru cted o f a seri es o f workshops identified according to
fa shi on . The religiou s orders were th e m asters of the methodical
the type of productio n o peration. T oussaint Barre design ed th e
co ntrol of time , and their influ e nce , diffu se o r more direct, was
R c rc rc n cc~ 1I1ay be f''',II,] ' 01' pp. IhO- I. k it everyw here. As in most aspects o f di sciplin ary pow er , the
148 Time, Space and Regionalization "
Critical Notes: Fouc;ylt on Timing and Spacing 149

army provides an apt illustratio n. Soldiers had long been trained T iming also stretches across the progress io n o f careers.
to march in regu lar formatio ns. T he Dutch were the early pio neers Fo ucault comp ares two phases in th e developme nt of th e
of the precise timi ng of mi litary manoeuvres.3 By the end o f the ma nufactory schoo l of the Gobe li ns. T he manu factory was
sixteenth century a method had been develo ped in the Dutch established by royal ed ic t in 1667; a sc hool fo r app rentices was
army whereby troops were trained programmatically to manoeuvre planned as part of the scheme. T he superintendent o f royal
in an ordered way while main taining a steady and continuo us rate bu ildings was to select six ty scho larshi p child ren fo r participation
o f fir e. This was accomplished by timing the vario us movemen ts in the school. the educatio nal process being organized alo ng the
of the body. The method was later applied to the gestures involved ty pical lines o f guild apprenticeship. T he pupils were first of all
in loading, firin g and reloading weapons a nd to many o tber the respo nsibility o f a master , later serving six years' apprentice-
aspects of mili tary o rga niza tio n . It was in re lation to such ship . Fo llowing fu rther service lasting fo ur years and [he successful
develo pm ents, in fact. that the term 'disciplin e' und erwent a passing of an ex am inatio n, they were able to set up their own
change in meaning. In its original sense it referred to a learning worksho ps. Here there was a diffu se process of transmission of
process and was regarded as a tra it of th e 'di sciplin ed '. However, knowledge. in volving an exchange of services between masters
in th e armed forces it came to be applied as ordinaril y it is tod ay, and apprentices. The tempo ral orga nization of the apprentices'
as to do with an overall mode of regul ation rather th an wi th the lives - by the stand ard s of what was to fo llow - was lax. Som e
process of in struction itse lf. 4 seventy years after the sc hoo l was set up, a new type of training
Th e timing of activiti es is mo re th an their subordinatio n to was initiated for th e apprentices; it was fir st of all complementary
measured temporal in tervals. It is perhaps the most bas ic to the existing mod es o f proced ure. Unlike those modes of
condition o f the 'co-o rdin atio n of the body and th e gest ure'. proced ure, it was based on th e carefu l serial arra nge ment o f time .
Disciplin ary powe r does not consist only in the impositio n of T he children attended the school for two ho urs a day. Classes
control over specific gest ures, bu t is max imized where gestures were d ivided acco rdin g to abi lity a nd prev io us ex peri ence.
are related to the positioning of the body as a whole . T he efficient Allo tt ed tasks were carried ou t in a regu lar fas hio n, appraised by
use of the body means that nothing remains idle o r unused; the teacher and the most able rewarded. Progression between
attentio n must be foc used wholly upon the act with which the classes was governed by the results of tests ad ministered to all
indi vidual is concerned. A disciplined body is a trained body: in pupils. Day-to-day behavio ur was recorded in a book kept by
this, o ne might say, th e traditio nal sense o f 'discipline' pe rsists. teachers and th eir assistants; it was pe riod ically looked at by an
Th e positioning o f the body is the main mediating factor between inspector.
two temporally arti culated sequences. O ne is the disaggregatio n T he Go belins school was o ne insta nce o f a gene ral trend in
o f the gestu re in to a timed series of movements, specifying the eighteenth -century educatio n, in Fouca ult's words an expressio n
parts o f the body to be used . T hus Maurice o f O range broke of a 'new technique for taking c harge of the lime o f individu al
dow n the hand ling o f the musket in to a seri es o f forty -three I.:xi stences'. Disciplin es 'which analyse space , break up and
separate movements , t hat o f the pi ke into twenty-three, co rearrange activities' have 10 be concentrated also in ways which
ordi nated within a formation of soldiers in a battle uni t. S However, makc possibl e 'adding up and capitalizi ng time'.ft Four metho ds
th e parts of the objects handled are also specified and in tegrated can be used to effec t thi s.
with the gesture. Precise t iming is essential fo r this, SlO ce
weaponry and machin ery have in creasingly become designed 10 j I ) T he di vision o f lives chrono logicall y, such th at ph ases of
o perate in a sequenti al way, each step in its operation bein g a developm ent are specifi ca ll y ti med. T hus the period of
prerequi sit e to what is do ne nex t. Disciplinary power d epends tra ining can be sepa rated o ut in a dear fas hi o n fro m a career
upon no t just th e explo itat io n o f pre-given materi als but also th e proper. Within th e traini ng period steps in attai n ment can
establi shment of a 'coercive link with the apparatus o f prod uc tion'. be de marcated, and all those receiving instructio n can be

Critical Notes' Foucaft on Timing and Spacing 151
150 Time, Space and Regiona/ization

made to move sequentially through all of them. 'exercise'. Exercise is the imposition of regular and graduated
(2) The separate phases of training and the subsequent 'career' physical training of the body, with an end state of fitness in view
- a word which thereby attains only its modern sense - - 'fitness' referring to the preparedness of the body but also to a
can be organized according to an overall plan. Education generalized capacity to carry out designated tasks. The idea and
has to be freed from the personalized dependence entailed practice of exercise had religious origins but became a secular
in the relation between master and apprentice. The theme of most of the disciplinary organizations. Exercise demands
educational plan has to be set out in impersonal terms, regular participation over time and works on specific parts of the
wherever possible dismembered into their most elementary body. It expresses in a direct fashion the significance of control of
operations, which are then readily learned by anyone the body, in relation to other bodies, which is essential to
undergoing instruction. discipline as a whole. The body is treated as a moving part in a
(3) Each of the temporal segments has to be concluded with an larger composite. Discipline, in sum, demonstrates the following
examination, which not only guarantees that every individual main characteristics. It is 'cellular' (in terms of spatial distribution);
will undergo the same process of instruction but also it is 'organic' (coding activities according to programmed
differentiates each in terms of his or her relative capabilities. procedures); it is 'genetic' (in respect of serial phases); and it is
The various examinations involved in the pursuit of a career 'combinatory' (uniting human activities as the paths of a social
are graded so that they each have to be successfully machine). Foucault quotes Guibert:
undertaken before the novitiate can move on to another.
(4) Different forms or levels of training can be designated for The state that I depict will have a single, reliable, ea~ily controlled
administration. It will resemble those large machines, which by
the achievement of ranked offices. At the conclusion of
quite uncomplicated means produce great effects; the strength of
each series some individuals can be hired off and allocated
this state will spring from its own strength, its prosperity from its
to a particular grade, while others continue to higher grades. own prosperity. It will disprove that vulgar prejudice by which we
Every individual is involved in a temporal series by means of are made to imagine that empires are subjected to an imperious
which his or her office or rank is defined. law of decline and ruin.
The 'seriation' of successive activities makes possible a whole There is an obvious similarity between Foucault's discussion of
investment of duration by power: the possibility of a detailed disciplinary power and Max Weber's analysis of modern bureau-
control and a regular intervention (of differentiation, correction, cracy. To be sure, the focus of their respective writings is different.
punishment, elimination) in each moment of time; the possibility Weber concentrates upon the 'heartland' of bureaucracy - the
of characterizing, and therefore of using individuals according to state and its administrative offices. In Foucault's work, on the
the level in the series that they are moving through; the possibility other hand, the mechanisms of the state are rarely analysed
of accumulating time and activity, of rediscovering them, totalized directly; the state is examined 'symptomatically', via seemingly
and usable in a final result, which is the ultimate capacity of an more marginal forms of organization, hospitals, asylums and
individual. Temporal dispersal is brought together La produce a prisons. However, in each author there is a stress upon the
profit, thus mastering a duration that would otherwise elude one's
emergence of novel types of administrative power, generated by
grasp. Power is articulated directly on to time: if assures its control
th e concentrated organization of human activities through their
and guarantees its use.'
precise specification and co-ordination. At first sight the theme
Thus disciplinary methods reflect a specific understanding of of the transformation of time and space seems lacking in Weber's
time, one which is an equal-interval scale. In the seriation of time, writings, and it is worth indicating how Weber's ideas can be
Foucault proposes, there is a procedure corresponding to the shown to incorporate such a theme. Admittedly, it is latent rather
mapping of partitioned space on to bodily activities: this is than manifest. Consider first Weber's treatment of the nature of
152 Time. Space and Regionaliza fi on Critical No tes. Foucf/r on Timing and SpaCing 153

modern capitalist enterprise. Wh at differentiates 'rational capital- and gives a measure o f auto no my to those within them . and also
ism' fro m preceding fo rms? Above all , it is its stable, regular serves as a powerfu l marker of hierarchy.
c haracter. Pre-existing types o f capitalistic enterprise take place Weber also stresses th e impo rtance of the separatio n of the
in sporadic, stuttering fashio n across time and space. Rational o rfice fr o m the dom ic ile o f the wor k e r. ~ One o f the main
capitalism involves the forging of regu larized market relationships characteristics of bureaucracy is that the vocational life of the
across space, something that can o nly become well-develo ped o fficial is segregated from home and family life, Impersonal
with th e formation of a burea ucratic state, which guarantees not fo rmulae of bureaucrati c diSCipline can be much more effectively
o nly pro perty rights but also other esse ntial institu tio ns, most appli ed when corporate moni es and eq uipm ent can be kept
no tably a regularized form of paper money exchange. separate from the privat e possessions of officials, when personal
Bu t control of time is equally necessary. Th e rational capita list o r kin ties are not the basis upon which decisio ns are concluded
enterprise is one that is a ble to o perate in a stable. orderly o r appo intments made and when matters concerning the house-
fashio n. Weber's emphasis upo n the significance of do uble-entry ho ld are distinguished fro m business affairs. Th e tho ro ughgoing
book-keeping for the develo pment o f modern capitalism is readily separation of the home fro m the workplace, We ber mak es clear ,
understandable in these terms. Double-entry book-keeping makes is fo und only in the modern West. But we might also note the
possibl e co ntinuous capital accounting over long periods o f time. importance of differenriation o f locales in distinguiShing between
Cap ital accounting is the valuation and verificatio n o f pro fit - th e sph eres of operat io n o f varying types o f bureaucratic
makin g opportunities. This mea ns making a valuation o f tota l organization. Anyone who doubts the influence o f th e differen -
assets at th e beginning of a transaction o r venture and comparing ti atio n of space and setting in shaping and refl ecting social
it with assets at a later date. Profitability depends, among o th er patterns should po nder the position of the 'C ity' in Britain. Its
fa cto rs, upon being abl e to predict future events and subject spa tial districtiveness fro m centres of 'indu stry', and its sheer
them to calculation. Double-entry book-keeping is a kind o f time- concentration in o ne area, ex press major institutio nal cha racter-
mach ine , because it both expresses and allows the quantification istics o f ttie society o f which it is a part (see pp. 319- 26).
o f units by reference to which the perfo rmance of an ente rprise Here we might return to Foucault. In this brie f excursus I am
can be judged in 'orde red tim e'. ~ no t interested in assessing the historical rights and wrongs of his
Co ntro l o f time is characteristic o f bureaucracy in ge neral. no t cxpositio n, or in probing the th eoretical sho rtcomin gs which
just of cap ital enterprises, Doubl e-entry book-keeping is a device 11' ight be discerned in th e general views upon which it draws, I
which 'stacks' past events as well as anticipating future o nes. want o nly to add a po int o r two to his interpretation of the
Bureaucratic rules are also a way of doing this. Mode rn rela tion of disciplinary power to modalities of tim e and space. Let
bureaucracies, Weber asserts, could not exist without the collatio n ll1e begin with the discussion given in reference to We ber in the
of documents which are both records o f the past and prescriptio ns preceding paragraph. Foucault treats disciplinary orga nizations
for the future - the 'fil es'. The fil es are not o nl y documents o f as epito mized by the prison and th e asylum - 'to tal institutions'
bureaucratic proced ure; they exemplify that procedure and make in Goffman 's phrase, 'comple te and austere institutio ns' in the
possible the continuo us and regular operatio n upo n which c hamcterizatio n Foucault adopts fro m Beltard. 'The prison', as
bureaucratic discipline depends, Files are usually o rganized within Fouca ult remarks, 'has ne ither exterior nor gap; it cannot be
definite o ffi ces and are part of what gives eac h office in a il1t crrupted , ex cept when the task is totally completed; its action
bureaucracy its distinctiveness. An 'office' is a physical se lling as 1 \11 th e individual must be uninterrupted: an increasing discipline

well as a level in an administrative hi erarchy. Although Weber . . . it gives almost to tal power over the prisoners; it has its
bare ly to uches upon the point, the physical distribution of o ffi ces illh.: rnal mechanisms o f rep ress ion and punishm ent: a despotic
in bureaucracies is a distin ctive feature o f such organiza tio ns, d isl: ipli nc.'10 Fa ctories. o ffi ces, sc hools , barracks and other
The physical sep<lratio n o f o ffi ces insulat es each from th e o ther l'onl cxls whe re surveill ance and disciplinary power are brought
154 Time, Space and Regionaliza tion Critical Notes : Fouca ult on Timing and Spacing 155

into play are mostly not like this, as Fo ucauh admits, without embracing character, impose a totalizing discipline upon those
developing the point. It is an o bservation of so me sign ificance, who are placed within them. 'Adjustment' to these circumstances
however, because 'complete and austere institutio ns' are the implies, and usually directly leads to , a process of degradatio n o f
exception rather than the rule within the main institutio nal sectors self, by which the inmate is stripped of tokens of self-identity at
of modern societi es. It does not fo llow that becau se prisons and the same time as the ordinary components of autonomy o f action
asylums maximize di sciplinary power, they express its nature are heavily constricted . 'To tal institutions', it may be said , both
mo re clearly than the o ther, less all-embracing o rganizatio ns. express aspects o f surveillance and diSCipline found in other
The journey to work (or school) pro bably indicat es as much co ntex ts in modern societi es and yet a lso stand out in relief
a bo ut the instituti o nal c haracter o f modern socie ti es as do against those other contexts. 'Total institutions' ordin arily involve
carceral organizatio ns. The tim e-space separatio n o f different what Go ffman calls 'civil d eath' - the loss of the right to vote
sectors of social life may indeed be th e condition o f the large- and to engage in oth er forms of political participation, of the
scale operation of disciplinary power. Most children attend right to will money, writ e cheques, contest divorce or adopt
schools o nly for part o f the day and at certain periods o f th e year. children . But in additio n inmates simply do not have separate
Mo reover , within the school day discipline is often o bserved in ilS spheres o f activity whe re rewards de nied in one secto r can be
stricter form s o nly within the definite tim ed periods that cOUO[ as pursued in another. G o Hman's comment on such matte rs is very
'lessons'. There is no doubt tha t disciplinary power can be relevant :
systematically ge nerated only by th e 'packing' of human beings
into specific ph ysically demarcat ed settings. But We ber is surely Th ere is an incompatibility, then, between total institutions and
right to say th at administrative di scipline is most eff ective Ih e basic work-payment stru cture of our society. Total institu tions
precisely when other aspects o f individuals' lives are separated are also incompatible with another crucial element of our society,
o ut fro m it. Fo r it involves the regularized applicatio n o f criteria the fami ly. Family life is sometimes contrasted with solitary li ving,
but in fact the more pertinent contrast is with batch living, for
o f conduct th at do no t accord with th e enactment o f activities in
(hose who eat or sleep al work, with a group of fellow work ers, can
o ther spheres o f life . T his is not solely because of th e factors that hardly sustain a meaningful domestic existence. I!
We ber mention s but also because of th e 'machine-lik e' nature of
di scipline. Fo uca ult is led into diffic ulti es in this regard. The Fo ucault treats th e investigative procedures of crimin al Jaw,
po int is not just that human be ings resist be ing treated as psychi atry and m edici ne as illustrating th e nature of disciplinary
a uto mata , something which Fo ucauh aCCepL'i ; the prison is a site power in general, especial1y as these are applied within carceral
o f struggle and resistance. Rather , it is that Foucault's 'bodies' are organizations. But again 'to tal institutions' stand out in this respect
not agents. Eve n th e most rigo ro us fo rm s of discipline presume as d ifferent from the d aily life paths of those outside. What
that those subj ect to them are 'capa ble' human agent s, which is Go ffman calls the 'territo ries of the self' are violated there in
why they have to be 'educated' , whereas machin es are merely ways whi ch do not apply to those not within their walls. Four
designed. But, unl ess subjected to th e most extrem e deprivation dislin c.tive features o f 'total institutions' can be mentioned in this
o f resources, capa ble agents are li kely to s ubmit to discipline only respect .
fo r parts of the day - usually as a trad e-off for rewards that I I ) Interrogati ve procedures frequ ently transgress what for most
deri ve from bei ng freed from such discipline at o ther times. o f [he populatio n are regarded as legitimate 'info rmation
In this respect reading G offm an o n 'total in stitutio ns' can be preserves' about th e self and about th e body. In other wo rds,
more instructive than reading Fo ucault. For G o ffman stresses d ata about inmates' charact eristics and past co nduct -
th at e ntry to prisons or asylums is demonstratively different from whi ch would oft en be regarded as discreditable by th em and
moving between o th er setlings in whi ch indi viduals may spend by o lh e rs and pro tected by s uppressio n or tact - are
pa rt o f the ir day. 'To ta l ins titutio ns', by virtu e o f the ir .. 11- co ll ected in doss iers availabl e to slaff.
156 Time , Space and Regiona l izat ion Critica l Nol es' Fo ucaul t on Timing and Spaci ng 157

(2) Th ere is a dissolution of the boundaries between enclosure conversion , colonization , and loyalt y to the inmate gro up .... 'lJ
and disclosure that ord inarily serve to protect a sense of Th ere is no doubt , as many sociological studies have demon-
o ntological securit y. Thus it may be the case that excretion, strat ed , that such inmate gro ups can exert co nsid erable control
the maintenance o f hygie ne and appearan ce not o nly have over day-to-day activities even in the most stringentl y disciplin ed
to be carried o ut publicly but are subjected to regimentation carceral organizations. But the modes o f control exerted by
by o thers. subordinates in o ther contexts, s uc h as that o f work, is likely to
(3) T he re are o ft en fo rced and continual relations with o thers. be greater beca use of a furth er way in which these contexts
Hence just as there are no back regio ns for to ilet acti vities. co ntrast with carceral o nes. T his is that supero rdinates have an
th ere are no back regions in which sectors of social life can inte rest in harnessing the acti vities o f those subject to their
be kept free (ro m the disciplinary demands made elsewhere, authority to the enactment o f designated tasks, In prisons or
Lik e Bette lheim , GoHman notes that in 'total inst itutio ns' asylums the 'd isciplining o f bod ies' co mes close to describing
human beings are redu ced to states of childlike dependence, t1 what goes on; the administrative staff are no t concerned with
(4) The temporal se riatio n o f activities , in the short and lo ng producing a collaborative endeavour at prod uctive activity. In
term , is specified and contro lled. Inmates do not have 'free workplaces and schools, o n the othe r hand, they are. Managers
tim e' or 'th eir own time', as workers do. Moreover, th ose have to coax a certain level of perfo rmance fro m workers. T hey
who undert ake se ri al examinations or pass throu gh serial are concern ed not only with th e time-space differentiation and
stages o f a career in th e o utside world are normally also able positioning of bodies but also with t he co-o rdination of the
to co unterpose these to other temporal units which have a co nduct of agents , whose behavio ur has to be channelled in
different pattern. The temporal distribution of marriage and defin ite ways to produ ce collaborative o ut comes. Foucault's
raising children. for example, is initiat ed separately from bodies do not have faces. In circumstances of surveillance in the
those pertaining in other spheres of life. workplace - where surveillance means direct supervision, at any
rate - discipline involves a grea t deal o f 'fa ce work' and the
In carceraJ organiza tions the significance of th e dialectic of exercise of strat egies of con trol that have in some part to be
contro l is st ill considerabl e. There are contex ts in which that daborated by agents o n the spot. The time-space 'packing' of
autono my specifically charac teristic o f the human agent - the groupings o f individu als in confin ed locales , where continuous
capability to 'have acted o therwise' - is severely red uced. T he s upervision in circumsta nces o f co-presence can be carried o n, is
fo rms o f co ntro l whi ch inmates seek to exert over their day-to - o hvio usly highly impo rtant to th e generatio n o f disciplinary
day lives tend to be concentrated above all upo n protec tion power. But the demand that agen ts work togeth er to effect some
against degradatio n o f th e self. Resistance is certa inly o ne o f so rt o f producti ve o utcome gives those agents a basis of contro l
these and no doubt is an impo rtant consideration that in some Ilver the day-to-day ope ration o f the workplace which can blunt
degree imposes itsel f, whatever policies the administrative staff .o,; upc rvisory efficacy. Supe rvisors and managers are as aware o f
might fo llow in th e implementation of disciplinary proced ures. thi s as anyone, and often bu ild that aware ness into the type of
But vario us other forms of reaction can be readily identified. disci plinary policies th ey fo llow. '4So me o f the forms of control
T hese include what Goffman calls 'colonization', the constructio n u pl.! n to workers in a tightl y integrated discipl inary space (e .g .
of a tolerable world within the interstices of managed tim e and th e possibi lity of disrupting or bringing to a halt an entire
space , and 'situational withdrawal', refusing, as it were, any lo nger produ ction process) do not ex ist wh ere a workfo rce is dis-
to behave as a capabl e agent is expected to do. But probably the OI).:gregated in tim e and space.
most commo n amo ng prisoners, as among the 'mentall y ill '. is Ll.!t me offer o ne final comm ent o n Foucau lt and Goffman.
simpl y 'play ing it cool'. This Goffman aptly descr ibes as 'a But h writ ers hav e as o ne o f th e leadin g themes in th eir work the
somew hat o pportunist ic combinatio n of secondary adju st me nts. posi1 ioning and discip lining o f th e body. Lik e Fo ucault , Goffman
158 Time, Space and Regionaliza tion References 159

has also pursued at some length questions of th e nature of comments on Hagerstrand's t ime-geography', Economic Geography.
'madness'. Their common concern with carceral organizations vol. 53, 1977; Don Parkes and Nigel Thrift , Times, Space.~ alld
might lead o ne to o verlook the differences in their respective Places (Chichest er: Wi ley, 1980); Nigel Thrift , 'On the determina-
views o f madness. Goffman's perspective actually places that of t ion of social action in s pace and time" SocielY and Space. vol. 1.
Foucault radically in quest io n in respect o f the relatio ns between 1982.
2 Hagerstrand: ' Space, time and huma n condit ions' . d. also Parke...
'insanity' and 'reason'. Fo ucault argues that what we call 'madness' and T hrift. Times. Splices alld PI(lce.\, pp. 247-8.
- o r, following the triumph o f the medical metapho r, 'mental 3 Alan Pred, "The impact o f technological and institutio nal inno-
illness' - has been created in relatively recent times. Madness is vat io ns of life cont ent : some time-geograp hic o bserva tions',
the suppressed , sequestered , dark side of human awareness a nd Geographical Allalysis. vol. 10, 1978.
passion, which Enlightenment and modern thought is unable to 4 Hagerstrand, IIIIIOVa/ioll as (I Spallal Proces.\" (Chicago: Chicago
conceive of in any o ther way save as 'unreason'. In traditio nal University Press. 1967), p. 332. cr. also Amos H. Hawley, HI/man
cultures , or at least in medieval Europe, foll y/fo lie encapsulated EcoloRJI (New York : Ronald Press, 1950), chapters \3-1 5: E.
its own reason, permitting something of a direct access to God. Gordon Ericksen, The Territorial Experience ( Austin: University
But by the middle o f the seventeenth century and thereafter, of Texas Press, 1980).
'Madness has ceased to be, at the margins of the world, of man o r 5 After Parkes and Thrift, Tim es, Space.l alld Places. p. 245.
death, an eschatological figure; the darkness on which the eyes of 6 O. G. Janelle, 'S patial reorganisat io n: a model and concept', Annals
of Ihe Associa/ioll 0/ Americal1 Geographers, vo l. 58, 1909, and
madness were trained , out o f which the forms of the impossible
ot her articles by the same author.
were born, has evaporated ... ,' 15 But perhaps this view invests 7 P. Forer, in Cariste in et al., Timifl g Space (lnd SpacillR Time.
madness with a grandeur whi ch it does not have and has never R R. Palm and A. Pred, 'A tim e-geogra phic perspective on problems
had? In seeing madness as the oth er face of reason it may express of inequality for women', in D. A. Lanegran and R. Palm, All
just those Enlightenment claims it affects to disparage. It may In vitation to Geography (New York: McGraw-Hili, 1978).
very well be that the clues to the character of madness or, in its 9 Hagerstrand : 'Survival and arena: o n th e life-history o f individuals
modern guise, 'mental illness' are to be found not in the in relation to their geographical environment', in Carlstein el a/.,
extravagance o f d elusions, visio ns of o th er worlds, but in mu ch T iming Space alld Spacing Time, vol. 2. p. 123.
more mundane features o f bodily and gestural impropriety. Social 10 Carlste in, ' Innovatio n, time-allocatio n and time-space pack ing', ibid.,
disability , no t a mysterio us access to a lost contine nt o f unreason , p. 159: Carlstein. Time Resources. Sociely and EculoRJI.
may express its real nature. II Cf. T. Carlstein, 'The sociology of struc turation in time and space:
a tim t..-geographic assessme nt o f G iddens's theory', Swedish
Geographical Yearbook (Lu nd: Lund Un iversity Press, 198 1).
References 12 T. Hagerstrand, 'W hat about peo ple in regio na l science?', Papers of
the Regional Science A ssociatioll. vol. 24, 1970, p. 8.
Time, Space and RegionaLization 1.1 CCHM. chapter S.
14 Ibid., pp. 161 ff.; CPST, pp. 206- 10.
See T. Hiigerslrand, 'Space, time and human conditions', in A. t.'i M. Melbin, 'The colo nisat io n of time', in Carlstein et al.. Timillg
Karlqvist, Dynamic Allocation of Urban Space (Farnborough: Sf)aC e alld Spacill}!, Time, voL 2, p. 100.
Saxon Ho use, 1975); Derek G regory, Ideology, Science and Hum af! th Evitar Ze rubavel. Pa((ems of Tim e ill Hospiwl Li/e (Ch icago:
Geography (London : Hutchinson, 1978), and 'Solid geometry: notes University of Chicago Press, 1979), p. 22; d. also P. A. Clark, ' A
on the recovery o f spatial structure', in T , Carlstein et al. , Timing review o f the theor ies o f lim e and Slructure for organisational
Space and Spacing T ime (London: Arnold, 1978); T. Carlstein, sociology', Ulliv ersity 0/ Astoll Malla gement Cemre Working
TIm e Resources, Sociely and Eco logy (Lund: Department o f I'llfl er.\'. no. 248, 1982: Zcrubave l, Hiddell RYlhms (C hicago:
Geography, 1980); Alan Pred, 'The choreography of existence: Universi ty of C hi cago Press, IqS t ). One might poi nt ou t that while
160 Time , Space and Region aiiz<J !ioll References 161

the 'year', 'mo nth' and 'day' have links with nalUral events, the 2 Ibid., p. 148,
'week' does nOI; cr. F. H. Colson, The Week (Ca mbridge: 3 Cf. Maury D. Feld, Th e StmclUre of Violellce (Beverly Hills: Sage.
Cambridge University Press. 1926). 1977), pp. 7 ft
17 P. A ries, Centllrie.rofChildhood( Harmondsworth: Penguin, 1973); 4 Ibid., p. 7.
Norbert Elias, The Civihl'ill/J Proce.u (Oxrord: Blac kwell, 1978). 5 Jacques van Doorn, The Soldier alld Social Change (Beverly Hills:
IS Edward T. Hall, The Hidden Dimellsion (London: Bod ley Head, Sage, 1975). p. 11.
1966), p. 98. 6 Foucault, Di.I'CljJlille al/d P!lIlI:~h. p. 157.
19 Antonin Artaud, Le thealre ella Jc ience (Paris: Sc uil, 1947), p. 98. 7 Ibid. , p. 160.
20 R. D. Laing, Self and Others (Harmondsworth: Pe nguin . 197 1), p. 52. 8 Max Weber, Economy alld Society (Berkel ey: Un iversity of
21 CCHM. p. 169. California Press, 1978), pp, 86-94.
22 Huw Benyon, Workillgfor Ford(Lo ndon: Allen Lane, 1973) , p. 76. 9 Ibid., p. 957.
23 Elias. vol. I. 10 Foucault, Discipli/le atld Punish, pp. 235 - 6.
24 Erving GoHman. The PreselltaliOIl of Self ill Evelyday fAye (New II Erving Goffman, A rylums (Harmo ndsworth : Pengu in, 1961 ), p.12.
Yo rk : Doubleday, 1959), p. 128. 12 Ibid., p. 33.
25 Cf. N. El ias a nd J. Scorson, The Ertablished alld the Olltsider.r 13 Ibid. , p. 64.
(Leicester: University of Le icester Press, 1%5). 14 Cf. Andrew L Friedman, IlIdu.ftry (I/ld Labollr(London: Macmillan,
26 Max Weber. Economy alld Society (Berkeley: University of 1977).
California Press, 1978), vol. I , pp. 341 - 4. 15 Foucault, Foiie et derai.\OIl (Paris: Plan, 196 1). p. 5 1. Foucau lt's
27 CSAS, c hapter 9. preoccupation with exclusio n, sequestration. e tc., is not accom-
28 CCHM, chapter 5, and passim. panied by a concern with the excluded th e mselves. who appear
29 Pierre Bourdieu, Outline of a TheolY of Practice (Cambridge: only as shadowy figur es. Thus in his analysis o f the case of the
Cambridge U niversity Press, 1977), pp. 143 - 52. murderer Pierre Riviere the c haracter himself bare ly e me rges from
30 Ibid., p. 153. the testimo ny discussed, which is treated o nly as a 'discu rsive
31 Andrew Po llard, 'Teac he r interests and changing s itua tio ns of e pisode', Ca rlo Gi nzburg's descriptio n o f the cosmo logy of
su rvival threat in prima ry school classrooms', in Pete r Woods, Mennochio, a s ixteenth-century heretic, o ffe rs a teUing comparison
Teacher Strategies (London: Croom Helm, 1980). in this respect. See Foucault e( 01.. Moi. Pierre Riviere . .. (Paris:
32 Randall Coll ins. 'Micro-trans lation as a theory-building strategy', in PIo n, 1973); Carlo Gi nzburg, The CheeJe and fhe Worms (London:
K. Knorr-Cetina and A, V. Cicourel, Advan ces in Social Theory Routledge, 1980), pp. xvi i- xviii. and passim.
alld M ethodology (Londo n: Rou tledge, 1981). See also idem, 'On
the micro-foundations of mac ro-sociology'. American j Olll'l1ul of
Sociology, vo1. 86,1981. For Goffman's thoughts on the matter -
given in a lecture which, sadly, he did not live to deliver - see' The
interaction order', America" Sociological Re well', vol. 48, 1973.
3J I bid .. p. 82.
34 Ibid .. p. 99.
35 Joseph Rykwert, The Idea of a Ta wil (London: Faber & Faber,
1976), p. 202.
36 CCHM, c hapter 5.

Critical NOles: Foucallit on Timing alld Spacing

Mic he l Foucau lt, DI:ITiplill f! (lnd Pllllish (Harmondsw{)rth: Pe ng uin .
1979), pp. 143 - 4.
Societies, Social Systems 163

developed wholly on a conceptuallevcl, ho wever. Just as I gave

4 some substance to the d iscussion o f age ncy and sclf in the sha pe
of an account of mo tivatio n . so I shaH introduce a classifi cation
Structure, System, and interpretation of soc ietal types to give fl esh to the analysis of
struc tural properties. This will in turn lead back again [0 questions
Social Reproduction o f 'history'. which will pre pare the way fo r a consideration of
problems o f analysing social c hange in the fo llowing chapter.
A book has a seque ntial fo rm. which can be overcome to some
degree by 'circulating in and o ut' o f a range o f connected issues
but which in evitably has its o wn presentational spacing. In the
light of my discussion in c hapte r I I take it that , while the sections
o n the agent and upon co-presence precede in the text those on
Let me at this point try to e nsure that the main threads of th e large r social systems, it will no t be presumed that I am
di scllssion do not become 100 d isaggregated in th e read er 's mind conceptually 'starting with th e individu al' , or that I hold that
by summarizing th e overall thrust of the preceding sections of the individuals are real in some way in which societies are not. ] do
book. In stru cturati on th eory a range of dualisms or oppositions not accept any such views. as the C riti cal Notes appended to this
fundam ental to other sc hools o f social thought are reconcep- chapter sho uld make clear.
tualized as dualiti es. In partic ular. the dualism of th e 'indi vidu al'
and 'society' is reco nce ptualized as the duality of agency a nd Soci etie s, Social Systems
structure. Thus far I have concentrated mainly upon developing
a series o f concepts which serve to elu cidate what the 'individual' It is easy to see that in o rd in ary usage th e term 'society' has two
is as a reflex ive age nt . co nn ecting refl exivity wi th positioning and main senses (among oth ers, suc h as 'soc iety' in the sense o f 'high
co-presence. The disc ussion of regionalization , however, begins society'). One is the generalized conno tation of 'social association'
to point the way towa rds show ing how these concerns intersect or interac tio n ; the o ther is the sense in whi ch 'a society' is a unity ,
with the stud y o f social systcms stretched across large spa ns o f ha ving boundaries whic h mark it o ff from o lher, su rrounding
time-space. T he next ste p , therefo re. is to look in mo re de tail at soc ie ties. The ambiguity o f the term in respect o f these two
the conce pt of society. take n by many [0 be the main unit o f senses is 1e..'iS unfortunate than it looks. Fo r societal totalities by
analysis in th e social scie nces. The term needs [ 0 be exam ined 11 0 mea ns always have clearl y demarcated boundaries , although
carefu ll y, and I shall pro pose that some usages are best avo ided Ih ey are typically a.ssociated with definite fo rms o f locale. The
altogether. lende ncy to suppose that societies, as social who les, are easily
In certain traditions o f social th eory th e concept o f society is definable units of study has bee n influenced by several noxious
cha racteristica ll y link ed in a direct way with that o f constraint. presumptions in the social sciences. On e is the tendency to
The advocat es o f stru ctural sociology have , in fact, tendcd to IIlId ersland 'social system s' in cl ose co nceptual relation to
regard constraint as in some way the defining characteristic of hiol ogical systems, the bodies o f bio logical o rganisms. There are
social ph enomena. In rej ectin g such a view, I shall try to clarify re \\l loday who, as Durkheim , Spe ncer and many others in
the conte ntion that. th e stru ctural properties of social systems arc lIill elee nth -ce ntury social tho ught were pro ne to do, use direct
both e nabling and constraining , and shall specify how 'stru ctural nrg<lni c analogies in describin g soc ial sys tems. But implicit
constraint' sho uld be understood. This in turn involves indicating pOira lkls remain very co mmon, eve n among th ose, for instance,
how a number o f conce pt s associated with that of 'stru c ture' \\I ll! ' I,ilk of soc ie li es as 'open sys lems'. A seco nd fa ctor is the
mi ght best be fo rmu lat ed. Such a fo rmulati o n ca nn o t be prcvakn ce o f wha l I call 'c nd ogc no us' o r 'unfo lding models' in
164 Structure, System, Social Reproduction Societies, Social Systems 16j

the social sciences. 1* Such models presume that the main (2) The existence of normative elements that involve laying
structural features of a society, governing both stability and claim to the legitimate occupation of the locale. The modes
change, are internal to that society. It is fairly evident why this is and styles of such claims to legitimacy, of course, may be of
frequemly connected to the first type of view: societies are many kinds and may be contested to greater or lesser degree.
imagined to have properties analogous to those which control the (J) The prevalence, among the members of the society, of
form and development of an organism. Finally one should mention feelings that they have some sort of common identity,
the widespread proclivity to generalize to all forms of societal however that might be expressed or revealed. Such fe-elings
totality features that are in fact specific to modern societies as may be manifest in both practical and discursive conscious-
nation-states. Nation-states have clearly and precisely delimited ness and do not presume a 'value consensus'. Individuals
territorial boundaries, but other types of society, by far the more may be aware of belonging to a definite collectivity without
numerous in history, do noLl agreeing that this is necessarily right and proper.
Resisting these presumptions can be facilitated if we recognize It is important here to re-emphasize that the term 'social system'
that societal totalities are found only within the context of should not be understood to designate only clusters of social
inlersocieta! system.l distributed along time-space edges (see relations whose boundaries are clearly set off from others. Th e
pp. 244 - 6). All societies both are social systems and at the same degree of 'system ness' is very variable. 'Social system' has tended
time are constituted by the intersection of multiple social systems. 10 be a favoured term of functionalists, who have rarely
Such multiple systems may be wholly 'internal' to societies, or abandoned organic analogies altogether, and of 'system theorists',
they may cross-cut the 'inside' and the 'outside', forming a diversity who have had in mind either physical systems or, once more,
of possible modes of connection between societal totalities and snme kinds of biological formation. I take it to be one of the main
intersocietal systems. Intersocietal systems are not cut of whole features of structuration theory that the extension and 'closure' of
cloth and characteristically involve forms of relation between societies across space and time is regarded as problematic.
societies of differing types. All these can be studied as systems of The tendency to take nation-states as 'typical' forms of society,
domination in terms of relations of autonomy and dependence hy reference to which others can be assessed, is so strong in the
which pertain between them. 'Time-space edges' refer to inter- lilerature of social tbeory that it is worth developing the point.
connections, and differentials of power, found between different The three criteria mentioned above apply differentially in varying
societal types comprising intersocietal systems. s(leietal contexts. Consider, for instance, traditional China at a
'Societies' then, in sum, are social systems which 'stand out' in relatively late date, about AD 1700. It is common amongst Sino-
bas-relief from a background of a range of other systemic l<lgisls to speak of 'Chinese society' at tbis period. Under this
relationships in which they are embedded. They stand out because . 1:lhel scholars discuss such phenomena as state institutions, the
definite structural principles serve to produce a specifiable overall ' ge nlry, economic units, family patterns and so on, regarding
'clustering of institutions' across time and space. Such a clustering Ih ese as convergent with a specifiable overall social system.
is the first and most basic identifying feature of a society, but China'. But 'China' as designated in this way refers to only a
others also have to be noted.' These include: sillall segment of the territory that a government official would
(1) An association between the social system and a specific ha ve regarded as the land of the Chinese. According to his
locale or territory. The locales occupied by societies are not pc rspective, only one society existed on earth, centred upon
necessarily fixed areas. Nomadic societies roam across time- .( 'hina' as the capital of cultural and political life but stretching
space paths of varying types. aW:lY 10 include a diversity of barbarians on the outer edges.
AlllHHlgh the latt er acted as though they were social groupings
Rcfcrcn<;cs may he fOUlld 011 pp. 12 1-4.
distilll.: l from the Chin ese , th ey were regarded in the official view
166 Structure, System, Social Reproduc tio n Societies, Social Systems 167

as belo nging to China. The C hin ese o f 1700 included T ibet, groups living on the o ther side o f the border. in the southern
Burma and Ko rea within their concept o f 'China', as these were slates. A farmer who did not have such contacts wou ld none the
in ce rtain ways co nnected with the ce ntre. There is so me basis for less have treated someone from beyond the border as a member
th e more restricted notion of 'C hin a' espoused by Western o f his own people rather than as a foreigner from another state.
hi storians and social scientists. But even acceptance that there Suppose, however, he encountered someone from Kan-su province,
was a distinct 'C hinese society' in 1700 , separat e from Tibet , etc. , in the north-west of the T'o-pa state. Such a person would have
usua lly means includ ing under that designation several millio n bee n treated as a complete stranger, even if tha t individual were
ethnically dist inct groups in Sout h China. These tribes regarded working alo ngside him in the fields. T he stranger would have
themselves as independe nt and as having their own organs o f spoke n a different language (probab ly a Mongo lian or Tibetan
go vernment. Th ey we re, however, continuously mo lested by d ialect), dressed differently and practised different customs.
represen tatives of Chinese officialdom , who treated th em as Ne ither the farm er nor the visitor may have been aware that they
bel onging to the central state. were both 'citizens' of the To-pa empire.
Modern Western nat ion-states a re highly intern ally co-ordinated T he Buddhist priests o f the time were a different matter again .
administrative unities compared with larger-scale agrarian societies. But with the exceptio n of a small mino rity who we re directly
Let us shift the example somewhat furth er back, to fifth- ce ntury appointed by To-pa ge ntry to serve in their official temples , they
Chin a, and ask what social ties might exist between a Chinese also had little contact with the dom in ant class. Their locale, in
peasant farm er in Ho-nan prov in ce and the T'o-pa ruling class. whi ch their lives were concentrated, was the mon astery, but they
From th e point of view of th e members o f th e dominant class, the had networks of social relationshi ps whi ch ranged fro m Central
farm er was at th e lowest level of the hierarchica l ord er. But the Asia to the south of China and Korea. The monasteries contained
socia l relations o f the farmer were qu ite discrete from the social people o f quite different ethnic and linguistic o rigin , brought
world o f the To-pa . Most of the farmer 's contacts wou ld be with toget her by the ir common religious pursuits. Their scho larship
o th ers in the nuclear a nd extended fami ly: many vill ages were distingu ished th em from other social groupings. They travelled
composed only o f lin eage membe rs. T he fields were usually so across state fro ntiers without restrictio n, regardless of those to
arra nged that members of lineage groups rarely met anyone other who m they were no minally 'subj ect'. They were not , however ,
than kin in the course o f th e working day. The farm er wou ld have regarded as 'outside' Chinese society, as was th e Arab community
visited neighbo uring villages o nly o n two or three occasio ns in in Can Io n of the Tang period . The state administratio n treated
the year , and perhaps a local town as infrequently. In the market- Ihal community in some ways as belo nging within its jurisdiction ,
place o f a nearby village o r town he would have encountered fl: qui ring taxes from them and setting up special o ffi ces to deal
o the r classes o r ranks o f people - craftsmen , artisans, traders, with them. But it was also recogn ized that they belo nged to a
and a low-ranking o fficial of the state ad ministration , to whom he se parate social o rder and therefo re were not on a par with others
wou ld pay taxes. Over his lifetime he would in all probability wi thin the realm of the state. On e final example:
never see a To-pa. Local officials who visited the vill age would
In the nineteenth century we fin d in Yun-nan province a political
have to be given deliveries o f grain or cloth. But the villager rule o f a bu reaucracy which was controlled by Peking and
would probably avoid any o ther contacts with higher o fficialdom represented the 'Chin ese' government; there were villages and
if they were ever imm inent. Fo r they could potenti ally mean ci ties in the plains, inhabited by oth er Chi nese who interacted with
brushes with the cou rts, im prisonment or enforced military service. Ihe government representatives and to some degree identi fi ed with
T he borders recognized by th e To-pa administration would Ih at government. But on the slopes of the mountains th ere were
not have coincided with the span o f activiti es of th e farmer if he llih er groups, in theory also subjects of China. yet livi ng th eir own
were in certa in areas in Ho-nan . T hro ugho ut the To-pa pe riod li fe. as far as Ihey were allowed, and having their own va lu es and
many farm ers had susta in ed contac ts with me mbers o f thei r clan institulions. even Iheir own economic system. Interaction wi th the
168 Structure. System, Social Reproduction Structure and COllsfraint : Durkheim and Others 169
valley-living Chine.'ie was minim al and restricted to the sale of fire-
wood and buying of salt or textiles. Finally, there was often a third
group on the top of the mountains, again with its own institutions, Stru ctu re and Constra int: Durkh eim and Others
language, values, religion. We can, if we like, bypass such
cond itions by calling these people minorities. Yet the earlier the Most forms of structural sociology , from Durkheim o nwards,
periods we study, the more such apparent minorities were truly have been inspired by the idea that struc tural properties o f society
self-contained societies. link ed sometimes loosely by economic fo rm constraining influences over action_In contrast to this view ,
ties. and by occasional interact ion; the relationship of such a structuratio n theory is based o n th e propositio n that structure is
society to Ih e ruling power was typica lly that of subject to always both enabling and constraining , in virtue of the inh erent
conqueror at the end of a war, wilh contacts held to a minim um relation between structure and agency (and agency and power).
from bot h sides" All well and good, a critic may say - and some indeed have said?
In thinking of units larger than imperial states, we have to - but does not this conceptio n in fact sacrifice anything akin to
avo id the tumbl e into ethn oce ntrism which it is so easy to make. structu ral 'co nstraint' in Du rkhei m's sense? Does not speaking of
We are pro ne today to speak readily of 'Europe' as a d ist in ct structure as both constraining and enab ling pay o nly lip se rvice to
sociopo li tical entity, fo r example, but this is often a result of the former? For in structuration th eo ry 'structure' is defined as
reading histo ry backwards. As many historians interested in ru les and resources. It is perhaps easy to see how struc ture in this
perspectives wider than those conce ntrated within nations o r sense is implicated in the ge ne ration of action but not so apparent
even 'continents' have pointed out, if the complex of societies where constraint enters in . For there see ms to be no way in which
stretchi ng across Afro-Eurasia were to be divided into two , a th e 'externality' of social ph e nomena to individual activity is
cleavage betwee n Europe as o ne portion (the 'West') and th e rest sustained . Such a no tion must be defend ed , it might be suggested.
as th e 'East' wou ld not make much sense. The Mediterranean what ever the flaws in the writings o f those mainly respo nsible for
Basin , for instance, was an historical unity both before the Roman advocating it. Thus Carlstein remark s:
Empire and for hundreds o f years subseq ue ntly. Indi a marked a a major drawback in Gidderi=s'paradigm is that the enabling aspects
greater cu ltural disjunctio n . travelling eastwards, than did the of structure are not sufficien tly balanced by constrailling ones.
various Mid-Eastern lands with those bordering in 'Europe'; a nd There are too few principles of limi tation, and by this I do not
there was yet greater discontinuity with China. As o ne histo rian simply mean the moraHegal-normative social const raints empha-
has laconically ex pressed it, 'The Himalayas were more effective sized by Durkheim and Parsons, i.e. structures of legit imation. I
even than th e H i nd u Ku s h .'~ The differe nces between majo r am re ferrin g 10 basic constraillls of mediation and resollrce
'c ulture areas' were o ft e n no t mu ch less marked than those /im italion rooted in certain biotic-cu m-physical realities of
between the units we would o rdinarily recognize as 'societies'. ex istence. Surely, structure must also imply limits to variation and
Regio nalizat ion o f wide scope shou ld not be treated as composed to con tingency in social systems (socio-env ironmental systems) . Of
simply of aggregate relations between 'societies'. Such a view has L'ourse there is room ror variation and human creativ ity. History
some validity when appli ed to the modern world of interna lly has proven over and over agai n how the application of ideas and
centralized nation-states bu t not when speaking of previo us eras. inventions in all realms of practice alt ers the received struct ure.
Thus, fo r so me purposes, the who le Afro-Eurasian zone ca n be But the latter is heav ily biased towa rds the past, and imposes hard
treated as a unity . 'Civi lizatio n', from 6000 Be onwards, did no t scr..:c ning on things that are produced and reproduced .. . .~
develop just as the creatio n of divergent centres; iI, was in so me 1 shall argue here, however, that th e th eory o f stru cturation in
ways a co ntinuous expansion 'o utward s' of the Afro-Eurasian 110 way minimizes th e significa nce o f th e constraining aspec ts of
zone as a whol e. ~ slnll,;[lIre. But 'co nstmint' as d isc ussed in structural sociology
[l; n(\ s to have several se nses (DlIrkh eim's termino logy , fo r what it
170 Struc ture, System, Soc ial Reproduction Struc ture and ConS fraint Durkneim and Otners 171

is worth, actually oscillated between the terms 'conlrainte' and point can be put as follows. Human societi es, or social systems ,
'eoereilion'); and 'constrai nt' canno t be taken as a uniqu ely wou ld plainly not exist witho ut human agency. Bu t it is not the
defining quality o f 'stru ctu re'. case that actors create social systems: th ey reprodu ce o r transform
In struclUralio n theory st ructure has always to be conceived of th em, remaking what is already made in the continuity o f praxis. ~
as a properry o f social systems, 'carried' in reprodu ced practices T he span of time-space dista nciatio n is relevant here. In general
embedded in time and space. Social systems are o rganized (altho ugh certainly not uni versally) it is true that the greater the
hierarchically and latera lly within societal to talities, [h e insti- t im e-space distanc iatio n o f social systems - the mo re their
tutio ns o f which for m 'artic ulated e nsembles'. If this po int is institu tio ns bite into time and space - the mo re resistant they are
igno red , [he nOlio n o f 'structure' in the theory of stru cturatio n to manipulatio n o r change by any indi vidu al agent. This meaning
appears more idiosyncratic th an it really is. One o f the circ um- o f constraint is also coupled to enable menl. Time-space distan-
stances which Durkhei m usually associates with constraint (also dation closes off some possib ilities o f human ex peri ence at the
hinted at in the quotation from Carlste in ) depends upo n the same tim e as it opens up others.
observatio n that the longll l:! duree of institutions both pre-exists Durkh eim's own formul atio n o f this issue, however, is wanting,
and out lasts the lives of indi vidu als bo rn into a particular soc iety. because it is couched in the termi no logy o f what has come to be
This is not only wholly compatible with structuration theo ry but cal led by many writers 'emerge nt pro pert ies' . Thus Durkheim
is also inherent in its very formulation - although th e remarks:
'socialization' of th e individual into society should be understood
as involving mutu al time process , connecting the 'life-cycles' o f The hardn ess of bronze lies neither in th e copper. nor in the tin,
bo th infant and parental fi gures. In his earlier writings Durkh eim nor in the lead which have been used to form it, which are all soft
heav ily emphasized the constraining elements of socialization , and malleable bodies. T he hardn ess arises from the mixing of
but later he in fact ca me to see more and more clearly that th em. The liquidity of water, its sustainin g and other properties,
socializati on fuses constraint and ena bl em ent. This is easily an; not in th e two gases of which it is composed, but in the
demo nstrat ed in th e in stance o f lea rning a first langu age. No o ne complex substance which th ey fo rm by coming together. Let us
'chooses' his o r her nat ive language. although learning to speak it apply this principle to sociolob'Y' If, as is granted to us, this
:\ynthesis sui gCllcris, which constitutes every society, gives rise to
involves definit e ele me nts o f compliance . Since an y language
new phenomena, differen t from those which occur in conscious-
constrains tho ught (a nd actio n) in the sense thaL it presumes a
nesses in isolation, one is forced to ad mit that these specific fac ls
range of fram ed , rule-governed properties , the process o f language res ide in the socicty itself that produces them and not in its parls -
learning sets certain limits to cognitio n and activity. But by the na mely its members. In this sense therefore they lie outside the
very same to ke n the learning o f a language greatly expands the I.:on:\ciousness of individuals a~ such, in the same way as the
cogni ti ve and practica l capac iti es o f the indi vidual. distinctive features of life lie outside the chemical substances that
A second context in whic h Durkheim tends to spea k o f ma ke up a living organis m. In
constraint also offers no log ical difficulties for structu ratio n
theory. Howeve r, we have to be careful to avoid some of the have quoted this passage at some le ngth just beca use it is so
dilemmas to which Durk heim's own analyses at this po int give we ll-known and has bee n re ferred to so o ft en as a particularly
rise. Societal totalities, Durk heim points out, not only pre-ex ist 1H.:rsuasive fo rmul ati o n. Soc ial systems do have structural
and post-date th e lives o f th e individuals who reproduce th e m in (llOpe rties that cannot be descri bed in terms o f co ncepts referring
their activiti es; th ey also stretc h across space and time away fro m ", lhe consciousn ess o f age nls. But human ac to rs, as recognizable
any partic ular age nt co nsidered singly. In this sense th e stru ctural 'l"( Hnpelcnt agents', do not ex ist in sepanltio n from o ne another
properti es of social systems are certa inly ex terio r to the activiti es :IS co pper, tin a ncI lead do. T hey do no t co me togt: lher ex Ilihilo
o f 'the individ ual'. In strll ctunttion th eory the essentials of this !I I form a new e ntit y by th ei r fu sio n or ilssociatio n. Durkheim
172 Structure, Sys tem, Social Reproduction Structure and Constrainr : Durkheim and Others 173

here confu ses a hypo th etical conception o f individuals in a state coupling constraints, within definite mate ri al settings, do indeed
of nature (umainted by association with others) and real processes 'screen' (as he puts it) the possible forms of activity in which
of social re production . human beings engage. But these ph e nomena are also at th e same
A third circumstance in which 'constraint' appears in Durkheim's lime enabling feature s of ac tio n . Mo reove r, as I have pointed out ,
writings is in juxtapositio n to the scope o f action of the age nt. the re are majo r shortco min gs in the us ual formulatio ns of time-
Durkheim gives the fo llowing amo ng o ther examples: geography_
Th e above aspects o f constraint/ enable ment are not the same
When I perform my duties as brot her, husband, or citizen, and
as, and are not to be reduced to, the opera tio ns o f power in social
carry out the commitments I have entered into. I fulfil obligations
which are defined in law and custom which are external to myself life. Durkheim's soc io logy, in fact , may be seen as irremediably
and my actions. Even if they conform to my own sentiments and I fl awed in respect o f the absence o f a conceptio n o f power
fe el their reality within me, that reality does not cease to be distinguished from th e generalized constraining properties o f
objective, for it is not I who have prescribed those duties .... " 'social facts'. Consider one final celebrated passage from Durkheim.
Constraint. he says, is
The point here is that 'social fa cts' have properties that confro nt
each single indi vidual as 'objective' features which limit that intrinsically a characteristic of [social I fa cts .... th e proof of this is
individual's sco pe o f action. T he y are not just external but also that it asserts itself as soon as I try to resist. If I attempt lO violate
externally defin ed , in corporated in what others do or in what the rules of law, they react against me so as to for estall my act ion,
they consider right and pro per to do. if there is still time. Alternatively, th ey annul it or make my action
There is sure ly something co rrect about this claim , but conform to the norm if it is already accomp lished but capable of
b~illg reversed; or they cause me to pay th e penalty for it if it is
Durkheim was prevented from spelling it out satisfactorily because
irreparable. . . . In other cases th e constraint is less violent;
of ambiguiti es a bo ut the notion of externality. In linking
n~vcrth e l e""<i, it does not eease to ex ist. If 1 do not con form to
externality and constraint , especially in his earlier writin gs , he ordinary conventions, if in my mod e of dress I pay no heed to what
wanted to reinforce a naturalistic conception of social science. In is customary in my country and in my social class, the laughter I
other wo rds, he wa nted to find support for the idea that the re are provoke, the social distance at which J am kept, produce. although
discernible aspec ts o f social life governed by forces akin to (hose in iJ lUore mitigated form, the same results as any rCiJl penalty.ll
operative in the material world . Of cou rse, 'society' is manifestly
not external to indi vidual acto rs in exactly the same sense as the Cu nstraint here refers to the stru cturatio n o f social systems as
surrou nding e nvironment is ex tern al to them. The parallel thus r(IrlllS o f asymmetri cal power, in conjunc tio n with which a range
turns om to be at best a loose o ne , and a concern with it rests ( I f normati ve sanctions may be de ployed against those whose

uneasily in Durkheim's later work alongside a recognitio n th at ~ o ndu ct is conde mned , o r disapproved o f , by others. As
the 'facticity' o f th e soc ial wo rld is in certain basic respects a very Durkheim's statement indicates, the constraints ge nerated by
differe nt ph e nomeno n fro m the 'giveness' of nature . di He re nt types of reso urce may range from naked physical
Durkheim co ncentrated mostly upon social constraints in his l'()c rcion to much mo re subtle ways of produc ing compliance.
various disc ussio ns o f the nature of sociology. However , as lltl! it does no good at all to collapse this meaning of constraint
Carlstein quite rightly points out - and as I have accentuated illtl) th e others. Moreover , as 1 have stro ngly underlined, power is
earlier, drawing upo n the time-geography of which he himself is l1l;v~ r merely a cons traint but is at the ve ry origi n of the
an exposito r - fundamental cons traints upon action are ~ :lpabilities of agents to bring about int e nd ed o utcomes of action.
associated with th e ca usal influences of the body and th e material Eadl of the various forms o f constraint are thus also, in varying
world. I have already indicated that these are regard ed as of ways, fo rm s o f enablemenl. Th ey se rv e to open up certain
essent ial importance in struc turation theory. Capab ilit y and pll.<.;s ihiliti cs o f act ion at th e same tim e as th ey restrict or deny
174 Structure , System , Social Reproduction Three Senses of 'Constra int' 175

others. It is important to emphasize this point because it shows derive from the impact which the activ ities or social ties of actors
that those, (incl udin g Durkheim and many others) who have have upon those of other actors. Physical capa bility and coupling
hoped to find a distin ctive id entity for 'sociology' in th e constraints are limits to the feasible social lives that people can
identifi cation of structural constraint are embarked on a vain lead.
enterprise. Explicitly o r otherwise , such authors have tended to T he time-geographic approach of beginning social analysis
see in structural constraint a source of causatio n more or less from identifying physical constraints is surely useful if certain
equivalem to the o peration of impersonal causal forces in nature. q ualificatio ns are born e in mind. One , as I have said , is that the
The range o f 'free actio n' which agents have is restricted, as it physical properties of the body and its material milieux of action
were , by external forces that set strict limits to what they can are enabling as well as constraining. and these two aspects have
achieve. The more that structural constraint is associated with a to be studied togeth er. A no ther is that the identification of
natural science model, paradoxically, the freer the agent appears physical constrai nts provid es no particular fuel to defend a
- within whatever scope for individual action is left by the mate rialist interpretatio n of social life. Al l human be ings have to
ope ration of co nstraint. The structural properties of social cope with the constraints of the body, its media of mobility and
systems, in other words, are like the walls o f a room from which comm unicat ion. But it does not follow that the modes of coping
an individu al cannot escape but inside which he or she is able to with such constraints have so meho w a mo re fund a mental
move around at whim. Stru cturation theory replaces this view influence o ver social activity than do other types of constraint.
with one which ho lds that structure is implicated in that very Turning to power as a sou rce of constraint , again it needs to
'freedom of action' which is treated as a residual and unexplicated be stressed that power is the means of getting things done,
category in the various forms of 'structural sociology'. very definitely enablemcnt as well as constraint. T he constraining
aspects of power are experienced as sanctions of various kinds,
Three Senses of 'Constrai nt' ra nging from the direct application of force or violence, or the
threat of such application . to the mild expression o f disapproval.
Let me first of all consider the meaning of constrain t in respect of San ctions only very rarely take the shape of compUlsion which
materi al constraint and constrain t associated with sanc tio ns, then those who experience them are who lly in capable of resisting, and
move to structural constraint. What is constraint when we speak e ve n this can happen o nly for a brief mo ment, as when o ne
of the constraining aspects o f the body and ils location in contexts person is physically rendered helpless by another o r others. All
o f th e material world ? It evidently refers here to limits which the o th er sanctions , no matter how o ppressive and comprehensive
physical capacities of the human body , plus relevant features o f they may be , demand some kind o f acq uiesce nce from those
the physical environment , place upon the feasible options open .. ubjecL to them - whi ch is the reason for the more o r less
to agents . Th e indivisibility of the body , finitud e of the life span uni versal purview o f th e dialec tic o f contro l. This is familiar
and 'packing' difficulties in time-space emphasized by Hagerstrand cnough grou nd . Even the threat o f death carries no weight unless
are all exam ples of such limits. The sensory and communicative it is the case that the ind ividual so threate ned in some way values
capabilities of th e human body are others. We are so used to life. To say that an individ ual 'had no choice bu t to act in such
treating these as e nabling qualities that it is necessary to make alld such a way', in a situation o f th is sort evidently means 'G iven
something of a conceptual swi tch to stress that they are his/her d esire not to die, th e o nl y alternative open was to act in
constraining also. Of course, these constraints are not who lly th c wuy he or she did.' Of course, where the threat offered by a
'given', once and for all ; the invention of electronic communica- s~ lIl c ti o n is not as le thal, comp liance may depend more on
tion, for example, has- altered the pre-existing relation between I1I cc han isms of conscience th an o n fear of any sanction ~
presence and the sensory media of th e body, Alone among the som ething, in fa c t, upon whic h Durkhei m laid considerable
categories mentioned above, constraint in this sense does no t e mphasis in talking of 'mo ral san ctio ns'. In th e case of sanctions
176 Structure, Sys tem , Social Reproduction Three Sen ses of 'Constra int' 177

there are obviously major asymme lri es in the constrain t/e nable- placing limits upon the range 0/ oplion.5 open (0 an aClor. or
ment re latio n. One person's constra int is anothe r's enabling. plurality 0/ actors, in a given circums tance or type 0/ circumstance.
However, as critiques of zerosum theories of power have shown , Take (he example given by Durkheim, that of the enactment of
such asymmet ries by no means ex haust the scope o f th e concept con tractual obligations, or one particular type o f conrract, the
of power. labour contract. Co ntract , of course, involves strongly defined
We sho uld bear in mind bo th th e rather vagu e se nse wh ich legal sanctions, but let us conceptually filter th em o ut. The
terms lik e 'acquiescence' or 'compli ance' tend to have, and the contractual relations of modern industry face th e individual with
fact that by no means all 'acqui esce nce' in a given set o f power a set of circumstances whi ch limit available options of action.
relatio ns is directly motivated. T o acquiesce in a partic ular course Marx says that work ers 'must sell themselves' - o r, more
of actio n might be tho ught to s uggest co nscious acceptance of accurately, their labour powe r - to employers. The 'must' in the
th at course o f action and even ;voluntary' acceptan ce o f the phrase expresses a constraint which derives from the institutional
broader power relatio ns in which it is enmeshed . Understood in order o f modern capitalist enterprise that the worker faces. There
s uch a fashio n. acquiescence wo uld cover on ly a sma ll and is o nly o ne course of act io n o pen to the worke r who has been
relat ively marginal proportio n of instances in whic h th e cond uct rendered propertyl ess - to sell his o r her labour power to the
of o ne actor or aggregate of actors conforms to what oth ers want , capitalist. That is to say, th ere is o nly one feasible o ptio n, given
o r what is in their interests. Sa nctions are usuall y very 'visible' that the worker has th e motivation to wish to survive. The
o nl y wh ere some sort of designated transgression actually occ urs 'option' in question cou ld be treated as a sin gle o ne or as a
or is perceived as likely to occur. Power relations are o ft en most multiple set of possibi lities. That is to say, a work er may have a
pro fo undly embedded in modes of conduct which are taken for cho ice of more than o ne job opening in the labo ur market.
granted by those who follow them, most especially in ro utinized , Marx's point, however, is that these options effect ively are of a
behaviour , whic h is only diHusely mo tivated. \ single (ype. In respect o f the rewards they o ffer to the worker ,
and o f o ther features o f the worker-employer re latio nship , all
''''<llm ;,1/ <'OIlSlf;J;m (NCRaliw) ~llnCI;on StruOur,,/ COflSffdffll
wage labour is effectively th e same - and supposedly becomes
ConSlrdint de ri vi ng Constrilil11 de fivi ng Cunstrai nt dO!fivi ng
from Ihe c haraCler frolll pun itivt! f. om the co nlC'x!\w li ty
c ve n more so wit h the furth er de velopment of capitalism.
of tll (- mat erial wor ld ft!spon St!s on thtt of action. i.e., frum All stru ctural pro pert ies o f social systems have a sim ilar
il nd from th e physical !hl rt of SO rl1t' i1gc rlt s th e 'g ivl:! n' charac \t! r ohjectivity' vis-d-vis the individual agent. How far th ese are
qUi1 l it i ~s of Ih l:! towMd s ot h,)rs of structural prop(: rt i c~
body visJvi., sit U.11C(i (lc to rs constraining qualities varies according to the conlext and nature
Il f any given sequence o f action or strip of interaction. ln other
What , the n, of structural constraint '! Once constraint deriving words. the feasible o ptio ns open to agents may be greater than in
from sanc tions is separated o ff, Durkheim's other points colla pse Ih e case of the labour contract example. Let me reaffirm on ce
into one if scrutinized at all closely. To say that society pre-ex ists mo re (he theorem (hat all struc tural pro perties o f social systems
the lives of eac h of its individual members at any given mo ment is are enabling as well as constraining. The cond itio ns o f the
o nly to identify a source of constraint in so far as its preexistence G II)italist labo ur contrac t may heavily favour employers as
in some way limits possibilities open to them. To emp hasize that co mpared with workers. But once they have become propertyless,
individua ls are contextually situat ed within social relations of wo rk ers are d ependent upon th e resources that employers
grea ter or lesser span is sim il arly o nl y to identify a source of prov ide. Both sides derive (h eir livelihood from th e cap ital / wage-
constraint if it is shown how this limits their capabiliti es. In each 1,1I1(Iur relation , heavily asymme trical though it may be,
case constraim ste ms from th e 'object ive' ex isten ce o f stru ctural T his analysis does not invalidate the sorts o f claim that social
properti es that the indi vidual agent is unable to chan ge. As with scie ntists or historians mak e wh en they talk o f ;soc ial forces'
the co nstraining qual ities of sa nctio ns, it is best d escribed as witiltlut refere nce to agen ts' reasons or intentions. In institutional
178 Str{lctw e, System, Socia l Reproduction Cons traint and Reification 179

analysis it is pe rmissibl e to establish regularized connections I take it as one of the main implicat io ns of th e for egoing points
whi ch are set o ut in .m 'impersonal' manner. Suppose , by way of that there is no such entity as a distin ctive type of 'structural
illustration , we isolate a relatio n between technological c hange e xp lanation ' in the social scien ces; all explanations will involve at
and patt erns o f managerial o rganization in business firm s. The least implicit reference both to the purposive, reasoning behaviour
expanding use o f microchip technology, le t us sa y, might be o f agents and to its intersection with constraining and e nabling
shown to be associated with a partial dissolution o f more rigid features o f the social and mate rial contex ts o f that behaviour.
fonns of hi e ra rc hical aut ho rity. The 'social force' in volved here is Two qualifications require to be added to this obse rva tion , one to
not like a fo rce o f na ture. Causal gene ralizations in the social do with the historically shifting c haracte r o f constraint , the other
sc ie nces always presum e a typical 'mix' of inte nded and associated with the pheno me no n o f reificatio n.
uninrended conseque nce o f act ion, o n the basis of the rationaliza
tion of conduc t , whe the r 'carried ' on the level of discursive or o f
prac tical conscio usness. T echnological change is not something Co nstraint and Re ificat ion
that occ urs in de pe nd e ntly of th e uses to which agents put
technology, rh e c ha rac te ristic modes of innovation, etc. It is odd The nature of constraint is historically variable, as are the enabling
that man y stru c tural soc io logists who are perfectly able to accept ljualities generated by the contex tu alit ies o f huma n action. It is
this - that technology does not change in and of itself (how variable in relation to the material a nd institutional circu mstances
could it'!) - do not seem to see that exactly the same applies to of activity, but also in relation to th e forms of knowledgeability
the social for ces linking technological change with such a that agents possess about those circ um stan ces. To have under-
ph enomenon as manageria l hie rarchies. Somehow, wh e th e r stood this is one of the main achievem e nts of Marxist thought
mainly as a res ult of conscio us planning or in a fashion more or where it has not relapsed into obj ectivism . Wh e n it has done so, it
less comp letely un intended by any of those involved, actors has become methodol ogica lly just anoth er version of a structural
modify their condu ct and that of others in such a way as to socio logy, insensitive to th e multipl e m eanings whi c h co nstraint
res hape modes o f au thorit y relations - presuming that the mu st be recognized as having in social analysis. Why should such
connect io n is indeed a genuinely causal one. insensitivity exist? Th e answe r , I think , is fairl y clear. It is usually
Wh y is it that some social forces have an appare ntly 'in e virable' associated with those types o f social tho ught whic h suppose that
look to the m ? It is because in suc h instances the re a re few thc aim of the social sciences is to un cover laws o f social activity
options ope n to the actors in quest io n , given thal they be have whic h have a status similar to that o f natural scie ntific laws. To
rationally - 'rationally' in this case meaning effectively aligning luo k for sources of 'structural constraint' is presumed to be more
mo tives with the e ndresult o f whatever cond uct is involved . T hat , ,( less the same as looking fo r th e lawgoverned conditions that
is to say, the actors ha ve 'good reasons' fo r what they do , reasons pUI limits on the bounds of free ac tio n. This, fo r many writers, is
whic h th e stru c tural sociologist is likely to assume implicitly ..:x ac tl y where 'sociology' finds its role as a distinc tive endeavour
rather [han ex pli citly attributing to those actors. Sin ce suc h good among the ot her social sciences. But according to the view
reasons involve a c ho ice from very limited feasible alternatives, suggested here , it produces a form o f reified d iscourse not true to
their conduct may appear to be drive n by some implaca bl e fo rce I II ..: real c haracteristi cs of hum a n age nt s.
similar to a physica l force . There are many social forces that ' Rcifi cation' has been understood in a variety of different ways
actors, in a mea ningful se nse of that phrase, are 'unable to resist'. ill litera ture of social theory. Amo ng those diverge nt uses three
That is to say, they canno t do anything about the m. But 'cannot ' c hara c te ristic senses can be most commo nly disce rned. One is an
here means that th ey arc unahl e to do anything o th e r th an ,lIlimi sti c sense, where socia l relatio ns become attributed with
conform to whate ver th e tre nd s in qu est ion are, given th e mo tives pc rsnnified ch arac teristics. A ve rsio n o f this is to be found in
or goa ls whi c h und e rli e lh ei r aclio n. M,lrli'S ce le brat ed disc ussio n o f th e 'fe tishi sm o f commodities' , in
180 Structure, System. Social Reoroduction The Concept of Structural Princioles 181

which he compares commodity relatio ns to the 'mistenve loped terms of the implacable causal forms which struct ural sociologists
regio ns o f the religious world ', Just as in religion 'the productions have in mind when th ey e mphasize so strongly the associatio n of
of the human brain appear as independem beings endowed wilh 'struct ure' with 'constraint' . Structural constra in LS do no t o perate
life, and entering into relation bot h with one another and (h e indepe nde ntly of the mo tives and reasons tha t agents have for
human race', so it is in the 'world of commoditi es' with the what they do, They ca nn ot be compared with the effect of, say,
'p roducts of men's hands'Y Ano ther sense in which the term an eart hquak e which destroys a town and its inh abita nts without
reification is often employed is to refer to circumstan ces in which th eir in any way being able to do anything abo ut it. The only
social phenomena become endowed with thing-like properties moving objects in human social relations are individu al agents,
whi ch they do not in fact have, Again there is a reputable who employ resources to mak e things happen , intentio nally or
ancestry fo r this coin age in Marx: 'In exchange value, th e social o therwise, T he stru ctural properties of social syste ms do no t ac t,
con nect io n between perso ns is transfo rmed into a relatio n o r 'act o n'. anyone lik e fo rces o f nature to 'compel' him o r her to
between things, , , ,I. Finally , 'reification' is sometimes used to behave in any parti cular way. (Fo r furth er disc ussio n in relatio n
designate characterist ics o f social theories which trea t concepts to problems of empirical research , see pp, 304 - 10.)
as thou gh they were the objec LS to which they refe rred , as However. th ere is a range of furth er notio ns relevant t0 speaking
attributing properties to th ose concepts, o f 'stru cture' in socia l analysis, and th ese req uire special
The second of these senses is th e one J shall adopt , but it is not co nsideration. J shall di scuss them in the foll owing order. First,
acceptable as it stands because it implies that the quality of being how sho uld the concept o f 'struc tural principl e' be d eveloped?
'thing-like' does not need further explication and becau se it does Second, what levels of abstraction can be distinguished in studying
not make it clear that reifica tion is a discursive notio n. The (he structural properties of social systems'! Third , how are diverse
concept sho uld not be understood si mply to refer to pro perties of social systems articulated within societal totaliti es'!
social systems which are 'objectively given' so far as specific , In identifyi ng struc tural principles the discussio n has to move
situated actors are conce rned. Rathe r, it sho uld be see n as back from the fo rmal to the rather mo re substantive. Let me
refe rring to forms of discourse which treat such pro perties as reca ll . to begin with , a main strand o f slru cturatio n theory ,
'object ively given' in th e sam e way as are natural pheno mena. introdu ced in th e first chapter. The 'problem o f ord er' in the
Th at is to say, reified disco urse refers to the 'facticity' with which ' (heory o f structurati o n is the problem of how it comes about that
social ph enomena confront individual actors in such a way as to s(}cial systems 'bind ' time and space , incorporating and integrating
igno re how they are produ ced an d reproduced thro ugh human presence and absence. T hi s in turn is closely bo und up with the
agencyY Reification thus sho uld not be interpreted to mean pro blematic of tim e-space d istan ciation: the 'stretching' of social
't hing-lik e' in such a conno ta tio n: it concerns, rather, the sys tems across tim espace. Stru ctural princ iples can thus be
co nsequences o f thinking in this kind of fashi on, whe ther such unde rstood as the principles of organizat io n wh ich allow
thinking is done by those who would call themselves social n::cognizably consiste nt fo rms o f time-space distanciation o n the
scientists or by lay members of society. The 'reified mode' sho uld hasis o f definite mechan isms o f societal integration. Drawing

be considered a form or style of discourse. in which the properties
o f soci al systems are regarded as having the same fixit y as that
presum ed in laws of nature,
upon a range o f compara tive and historical s tudies .'~ I propose a
three fo ld classifica tio n o f types of society as below:

IIWIAl SOC IETY Tril ditionl(commUlla l (Fusion of soc i.}1

I( )" ,( c ultures ) prilC I,CCS) and ~ystem
Th e Co nce pt of Structural Prin ciples { Kin~ hir> inlr:g ra lion)
Gro u p ~il n ct l O!l S
Th e implications of the fo rego ing sec tio ns of this chapt er can be
desc ri bed as follows. Struc tural constraint is not exprcs..o;ed in
182 Structure, System, Socia l Rep roduction The Concept of Structural Principles 183

CLASS-DIVIDED 5CX: IETY Tradi ti on (communal (Diffe rentiation found along an axis relating urban areas to their rural hinterlands.
pra cti ces) of ~ial and
Kin ship system integra ti on)
The city is far more than a mere physical milieu. It is a 'storage
Politics - m ilitary power container' of administrative resources around which agrarian
Eco nomic interdependence states are built. The differentiation o f city and countryside is the
STATE { (l ow lateral and vertical
integratio n) means of the separation of social and system integration, although
the two are not necessarily coincident, for th e symbiotic relation
D ominant locale organ ization Sym b iosis o f ci ty and o f city and countryside may tak e various fo rms. IS In class-divided
country~ide societies traditional practices and kinship relations , even tribal
identifications, remain very prominent. The state is unable to
ClASS 5CX:IETY Rout inilation (Different iat ion of penetrate deeply into localized customs, and sheer military power
(Capitalism) Kinship (fam ily) social and sys tem
Surveilla nce intcgrm ion) is one of the principal foundations upon which government
STATE Poli tic s- m ilitary power officialdom is able to 'contain ' o utlying regions where direct
Economic interde pendence administrative control is particularly weak. Class-divided society
(high lateral and vert ical
Integration) is marked, however, by some disentangling of the four institutional
spheres distinguished above (p. 33). The po lity, with its officials,
Dominant loca le orga nization The ' created environment' is separated in some part from the procedures of economic
activity; formal codes of law and punishment exist; and modes of
This scheme is described in some detail in A Contemporary symbolic co-ordination, based in written texts , make their
Critique of Historical Materialism, and I shall gloss it rather appearance.
rapidly here. " In tribal societies or small oral cultures th e Modern capitalism is not o ne type of 'civilization' among others,
dominant structural principle operates along an axis relating and it does not mark an evolutionary development 'out of' c1ass-
tradition and kinship, embedd ing themselves in time and space. divided societies. The first gen uin ely global type of societal
In these societies the media of social and system integration are organization in history, it has its origins in a double discontinuity
the same, depending overwhelmingly upon interactio n in the in the development of the West. There are long-term divergencies
settings of locales of high presence availability. Of course, a in the formation of the West, as compared with that of the other
variety of diffe rent sUb-types of society can be distinguished major 'civilizations', over a period of some two mille nnia; Europe
within this general category. I should emphasize that I do no t remained a 'state system' , and no dominant imperial centre was
inte nd to present this classification as a surreptitious evolutionary re-established in its midst after the disintegration o f the Roman
scheme. Oral cultures sho uld not be understood as societies in Empire . Within this broad divergence, however, a range of
which system integra tion has 'not yet' become disentangled from massive discontinuities from other types of society was introduced
social integration , As Levi-Strauss has done more than anyone by the intertwining of political and industrial revolutions from the
else to make clear, tribal societies - in which humankind has e ighteenth century onwards. The distinctive structural principle
lived out all but a small fraction of its history - are substa ntially o f the class societies of modern capi tal ism is to be found in the
divergent from 'civilizations' , of whatever type. The inventio n of d isembedding, yet intercon necting , of state and economic
writing, so closely involved with the formation of states and institutions. The tremendous economic power generated by the
classes, alters the character of time as lived experience, by the harnessing of allocative resources to a generic tendency towards
very means wh ereby it permits an expanding of time-space tec hnical improvement is match ed by an enormous expansion in
distanciation. Ih e administrative 'reach' of the state. Surveillance - the coding
The dominant structu ral principle of c1ass-divided society - o f information relevant to the administration of subject popula-
which obvio usly also includes a range of sub-types - is 10 be lions , plus their direct supe rvision by offi cia ls and administrators
184 Structure , System, Social Reproduction Struc tu res, Struc tural Propertie _s 185

of all sorts - becomes a key mechanism furthering a breaking during that phase a greater variety of types of society existed in
away of system from social integration. Traqitional practices are relation with one another than at any other period before or
dispersed (without, of course, disappearing altogether) under the afterwards. For since that time the increasing ascendancy of
impact of the penetration of day-to-day lify by codified admini- Western capitalist societies, challenged only by the state socialist
strative 'pr2ce~ures. -The locales which provide the settings for societies 20 in terms of their industrial and military power, has
interaction in situations of co-presence undergo a major set of implacably destroyed or corroded tribal and class-divided
transmutations. The old city-countryside relation is replaced by societies, which perhaps are forever disappearing from the face
a sprawling expansion of a manufactured or 'created environment'. of the earth. The contemporary world system is, for the first time
A categorization of intersocietal systems can be formulated - in human history, one in which absence in space no longer
in a broad way at least - in terms of the above classification of hinders system co-ordination. Is it necessary to stress again that
society types as follows: the development of the world nation-state system is not coeval
with the expansion of cohesion or consensus? For the same
Triba l soc iet ie s ] 'Prehi ,torica l' and frag mentary syl lt'ml developments which have created at once that distinctively
Class-d ivided societies modern form of society, the nation -state and its involvement in a
] Im pt' rial world syste m,
Triba l societ ie s global system of a new type, have at the same time brought into
Capitali st soci eties being schisms which, in the nuclear age, threaten the very survival
Cl ass-d ivided >ocieti. ,> ] Ea rly ca pita li st world ec on o my of humanity as a whole. 21
Tribal soc ieties
'Surer-iCapita li st societies
6f;:~~ LS tate SOCi dl ist societi e s Structures, Structural Properties
Contemporary ca pita list wo rld economy
'deve l<J ping cou ntries ' (wo rld nation-state wstem ) As I have previously emphasized, the concept of structure may
,-------- ----,
ICl ass divided socieli.,> I be used in a technical and in a more general way. Understood as
:.!~b! '---s::'c..'e~i ",-:' _ _ _. _ : rules and resources, structure is recursively implicated in the
reproduction of social systems and is wholly fundamental to
This categorization, it should be pointed out, is not at all structuration theory. Used in a looser fashion, structure can be
symmetrical in respect of historical chronology. The smallest spoken of as referring to the institutionalized features (structural
category figuratively - systems of tribal societies - is by far the properties) of societies. In both usages 'structure' is a generic
largest in terms of span of time. Intersocietal systems involving category involved in each of the structural concepts given below:
tribal societies have always been relatively fragmentary, however, ( I) structural principles: Principles of organization of societal
in the sense that they have been confined in respect of their
configurations across time-space. They have dominated the world (2 ) stru ctures: Rule-resource sets, involved in the institutional
for most of human history, but they have not formed 'world articulation of social systems;
systems' in Wallerstein's senseY That is to say, 'civilizations' have (:\ ) Sfruclural properties: Institutionalized features of social
developed centres of power which have influenced large segments
systems, stretching across time and space.
of the globe, and they have fired the 'heat' of rapid social change.
Imperial world systems, however, have existed only in uneasy The identification of structural principles, and their conjunctures
relation to a diversity of forms of tribal societies and have ill int ersocietal systems, represents the most comprehensive level
frequently succumbed to attacks or pressures from such societies. \ ) f in stituti o nal analysis. That is to say, the analysis of structural

The phase of the early capitalist world economy was a transitory prin c iples refers to modes of differentiation and articulation of
on e in history , lasting no lon ger than two centuries or so. Yet in slilLlli o ns across th e 'dee pest' reach es of tim e-space . The study
186 Structure, System, Social Reproduction Structures, Structural Properties 187

of structural sets, or structures, involves the isolating of distinct C-M-C, the simplest form of the circulation of commodities, is
'clusterings' of transformation/mediation relations implied in the the beginning point of capital. As contrasted with the landed
designation of structural principles. Structural sets are formed by property of feudalism, capital first takes the form of money - the
the mutual convertibility of the rules and resources implicated in capital of the merchant and the usurer. The first distinction
social reproduction. Structures can be analytically distinguished between money and capital is simply a difference in the relation
within each of the three dimensions of structuration, signification, of transformation involved, expressed as M-C-M. This formula
legitimation and domination, or across these. I have offered expresses the transformation of money into commodities and of
elsewhere an illustration,22 on which I shall comment at rather commodities back into money - in other words, buying in order
greater length here. This is the example of private property in to sell. Money which has undergone this transformation has
Marx's analysis of modern capitalism. become capital. Like the other relation, M-C-M involves two
Consider what is involved in the following structural set: linked phases of transmutation, In the first money is changed into
a commodity; in the second the commodity is changed back
private propert\' rn o n., y ca pita l : la bour contra ct : prof it again into money. But the combination of these phases, Marx
argues, 'constitutes a single moment' whereby a commodity is
The structural relations indicated here mark out one of the most bought in order to be sold. It might seem as though money has
fundamental transmutations involved in the emergence of simply been exchanged for money - more or less, according to
capitalism and hence contribute in a significant way to the overall the success or otherwise of the transaction, But where money has
structuration of the system. In feudalism (in my terminology, one been transformed into capital it has gone through a 'characteristic
among other types of class-divided society) private property in and original movement' quite distinct in type from that of, say, a
the means of production was based predominantly on ownership peasant who sells corn and uses the money thus acquired to buy
of land, and such ownership was hedged about with numerous clothes, The transformations involved in M-C-M, as compared
qualifications upon alienability. In so far as these conversion with C-M-C, differ more than in the mere difference in the
relations pertained at all, they were confined to marginal sectors 'direction' of change.
of the economy. In capitalism, by contrast, private ownership of The difference is that in the C-M-C relation the money is
the means of production takes on a different form - land co nverted into a use value, which is then 'consumed'. In the
becoming only one type among other resources mobilized within \)pposite form, M-C-M, the money is not spent; it is 'advanced' -
production - and a diversity of goods becomes freely alienable. I hc secret of the transformation of money into capital. In the

Bssential to this process, Marx demonstrates, is the universalizing C-M-C form the same element of money changes its place twice,
of commodity form. The condition of such universalization is the completing the transaction. But the contrary is the case in the
development of a full-blown money economy. Money, Marx says, M-C- M relation: in this connection it is not the money that
is 'the metamorphosed shape of all other commodities, the result d lan ges hands twice but the commodity. The transmutation of
of their general alienation'. 2J Money (M) represents, on the one mo ney into capital depends upon the renewal of the operation, its
side, a sold commodity (C) and, on the other, a commodity to be 'reflu x', which only the M-C-M relation makes possible. M-C-M
bought. M-C is a purchase but is at the same time C-M, a sale: ~ II(1ltld thus more accurately be written as M-C-Mt, as an
'the concluding metamorphosis of one commodity is the first \'x pansionary process. The circulation of commodities has here
metamorphosis of another' or, as Quesnay expressed the same heco mc separated off from a direct relation Lo use value. Capital
thing in his Maximes generales, 'vendre est acheter .' The I fad e." 11 0t in use values but in exchange values.

differentiation of commodities into commodities and money does M-C- M\, howcver. can re present mercantile capital as well as
not dissolve the material differences between commodities: it i l1(lll ~ trial capital. It is th erefore only the 'general formula for
develops, Marx says, a modu s vivendi, 'a form in which they can l ,tpilal' . A further stru c tural relation is impli cated in the
exist sid e by sid e.!4 developm cnt o f indu strial o r manufacturing capital, one which,
188 SUucture, Sys tem, Social Reproduction Structures, Stnlctural Properties 189

like the altered nature of private property, presu mes a major T he re is no definite cut-off po int between the three levels of
process of social change. This furth er relation is the possibi lity of abstractio n distinguished in the above diagram . T he speci fi catio n
the transfo rmation of capital imo labou r and vice versa, something of structural sets , as indicated previo usly. is of basic importance to
which pres upposes a massive expropriation of work e rs fro m th e elaboration of overall stru ctural principles, bu t the o ne task
control of th eir means of prod uct ion , such that they have to o ffer nhvio usly merges in to the o th er. The same holds for the lowest
their labour power for sale on th e market in order to attain a leve l of abstraction, th e iso lating of element s or axes of
livelihood. Labour power is a commodity that has the peculiar structuration. Distin guishing ele ments of stru cturation preserves
feature, among others, o f bein g a so urce of the creation of value. [h e epoche of institutio nal analysis, but brings the level of study
Th e capitalist labour contract is inhere ntly involved with the dose r to the direct exami nation of relations of co-presence, In
transfo rmation of money into an equivalent of labou r power. order to preserve continuity wi th the preceding discussion, let me
'This relation has no na tura l basis, neither is its soc ial basis o ne fo llow th rough Marx's disc ussio n in respect of a major feature o f
tha t is commo n to all historical periods. It is clearly the result o f a capitalist productio n, the d ivisio n of labour. It is a n analysis with
past histo rical developm ent , the product of many econo mic which 1 am largely in accord , a ltho ugh my mai n purpose here is
revolutions, o f the extinc tion of a whole series of o lder forms o f an illustrative o ne. 2b
social production. '!5 Thus the isolation of such a connection helps The division of labour , Marx seeks to show, is closely bound up
to diagnose one of the key st ructural features of the novel wi th the nature of manu facture and therefore with the structural
institutional form constituted by capitalism. That labour power is re lations portrayed in the forego ing paragraphs o f this chapter.
a commodity is not given in th e 'general formula for capital'. T he division of labour links th e broader structural c haracteristics
Th e capitalist labour contract presumes that emp loyer and or capitalism , as identifi ed previously, with th e more proximate
wo rker 'meet in the mark et' in circ umstances in wh ich each is Ilrganiza lion of the industrial enterp ri se. Manufacture , a pre-
'fo rmally fr ee'. This is a basic aspect of the class relatio ns o f eminent feature of capitalism that has advanced beyond commerce,
capitalism. O ne is a buyer of labour power , the other a seller. The is associated with two modes o f th e emergence o f workshops.
'owner' of labour power sells it o nly fo r a definite period , as does ()lte is the assembling, under the contro l of a partic ular employer ,
the employer who 'takes o n' labour. Slavery, in which some IIf workers with different craft skills in a specifi c locale. These are
persons are owned by others, does not permit the commodification co-ord inated in th e making o f a single product. But such co-
o f labou r power. Th e value of labour power, in common with that 1!I'dination tends also progressively to strip away aspects of the
of oth er commodities, is governed by the labour time invo lved in sk ill s o riginally possessed by workers, leading to the splitting up
its production and therefore by what is demanded to ensure the Ill' tasks into 'detailed ' processes, 'each of which crystallizes into
physical survival of those who supply labour. The transfo rm ation th e e xclusive fun ction o f a parti cular workman, the manufacture,
o f the hire of labour power into profit , of course, is dependen t as a who le, being carried on by th e men in conjunctio n'.n A
upo n (he generation of surplus value. 'Necessary labour lime' is sccond way in which manufac ture arises is something o f the
that given over to the sustai ning o f the source of labour power, reverse o f this. It is the assembling within o ne locale o f a number
the worker; surplus labour is the source o f profit. Il f wo rke rs who all do th e same task, each worker mak ing the
l' lltire commodity. However. 'extcrnal circumstances', Marx says,
structur al IJri nciplt:s ic,lt! !o c hanges in mu ch the same direction as th ose occurring in
Ilt c first type of setting. La bo ur is therefore redi stributed; instead
level of struc tu ra l sct ~ ( ~tru c ture s )
suc i<lI/SYS I(:m 1>1' w(lfkers all occupi ed in th e sa me way sid e by side, operations
ab~t racti()n inll'Il,ati o n
hecomc brok e n down int o detail ed tasks, organized in a co-
npcr;([i ve fashion. The final form is thu s the same in both cases:
';. pl'\~du c ti ve mechanism whose parts are human hei n gs'. 2~
190 Structure, SySfem, Social Reproduction Structures, Structural Properties 191

The detailed division of labour is of major importance to the identifying the sources of social stability alone. They serve indeed
organization of the capitalist enterprise in several ways. It to indicate some of the main form s of cha nge involved in the
enhances the opportunities for direct surveillance of the work- transition from one type of societal totality to another. What
force and the consolidation of labour discipline. But it also both 'must happen ' for certai n conditio ns of system reproduction to
expresses and makes possible the connection o f labo ur , as labour occur is posed as a counterfactual questio n, not as a covert
power, with the technology of machine prod uctio n. For the versio n of functionalism.
'detail labourer' carries o ut a c ircumscribed num ber of repetitive A reproduction circuit can be sketched in d iagrammatic form
operatio ns that can be co-ordinated with the movements o f (see figure II ):
mechanized production processes. Division of labour within the
en terp rise is not simply an aspect or extension o f the division of
labour outside, the 'divisio n of labour in society', but these none
re/l e~ivcmOflilorin g str uctura l pronerti~s struetlJ.,,1 p. incipl ,,~ :
the less react upon o ne another. The 'division of labour in society' "f act ion med iat ion/transformati on inst itutiona l doma ins
depends upo n the purchase and sale of products of different
secto rs of ind ustry; the division of labour within the en terprise
derives from the sale o f the labour power of a plurality of wo rkers
dUJl ity "I struct ure
to an employer who appl ies it in a co-ordinated fashi on.
Division of labour within the workshop implies the undisputed Figure 11
authority of th e capitalist over men, that are but parts of a
mechanism that belongs to him. The division of labour within the The reintroduction of the duality of stru cture means leaving
society brings into contact independent commodity-producers, 1he virtual time-space of inst itut ional analysis, thereby re-entering
who acknowledge no other authority but that of competition.... It history'. All structural properties of social systems , to repeat a
is very characteristic IMarx adds caustically I that the en thusiastic
Icad ing theme of strucLUration theory, are the medi um and
apologists of the factory system have nothing more damning to
ou tcome o f the contingent ly accomplished activities of situated
urge against a general organization of the la bour of society than
[hal il would turn all society into one immense faclory.19 a<.:tors. The reflexive monitoring of actio n in situatio ns of co
presence is the main anchoring feature of social integration , but
To analyse the d ivision of labour in this way is to eluc idate an hot h the conditio ns and the o utcomes of situated interactio n
axis of structuration connecting the internal form of the enterprise stretc h far beyond those situatio ns as such. The mechanisms of
with broader aspects of the societal totality, indicating at the 'stre tching' are vari able but in modern societ ies tend to involve
same time contrasts with the 'division of labou r in society'. Of reflex ive monitoring itself. That is to say, understanding the
course, these relations could be spelled out in very much greater cunditions of system reproduc tion becomes part of those
detail. In institutional analysis this involves detailing the cond itions of system reproduction as such.
transformation/ mediatio n relations implicated in the 'clustering' We can trace these observations thro ugh more concretely by
of institutionalized practices across space and time. However, retu rning to th e struct ural set discussed previously. T he two
once we abandon the epoche o f institutional analysis, all the upposed but complementary transformations C-M and M-C occur,
structural relations indicated above, at whatever level, have to be of co urse, o nly through the activiti es of buyers and sellers acting
examined as condi tions of system reproduction. They help to ill a range o f divergent sett ings. According to Marx, the C-M-C
pick out basic features.of the circuits o[ reproductio n impli cated re lat io n brings into co-relation three 'dramalt:f personae'. The
in the 'stretching' of institutions across space and time. Analysing ow ner of a commod ity comes into contact with a possessor of
circu its of reprodu ction, it sho uld be cl ear, is not eq ui valent to mOlu:y, th e mon ey becoming , Marx's words , 'its transient
192 Structure, System, Social Reproduction Cont radiction 193

equivalent-form'. Money, the 'final term of the first transmutation' , electronic communication dating from the invention of the Morse
is the origin of the third , the buying of another commodity.JO But Code.
as Marx expresses it, this is unsatisfactory. For structural relations
are not isomorphic to the acts of corresponding individuals who Contrad ict io n
personify them. It is in just such tendencies of Marx's argument
that one can see where Althusser derives textual corroboration It is commonly remarked that the concept of contradiction should
for the view that human agents are no more than 'supports' for remain a logical one rather than being applied to social analysis.
modes of production. Moreover, it is also easy to see how such a One can indeed see considerable justification for such a
style of analysis slips over into functionalism. For if the relations judgement because the term is often employed so vaguely that it
between structural properties, once isolated, are treated as having has no particular connection with contradiction in logic. Given
their own 'inner dynamics' , as functional necessities rather than that it is used with some care, however, I think the concept to be
as continually reproduced conditions, the activities of historically an indispensable one in social theory. I propose to use it in two
situated individuals do indeed seem rather redundant. The overall senses: that of 'existential contradiction' and that of 'structural
conditions of system reproduction are in no way 'guaranteed' by contradiction'. Each preserves some continuity with the logical
the structural relations upon which (counter-factually) they usage of the term while not being a direct extension of that usage .
depend. Nor does analysing those relations in virtual time-space By existential contradiction T refer to an elemental aspect of
explain in any way why they came about. This means that it is human existence in relation to nature or the material world.
highly important to shift conceptual gears when moving from There is, one might say, an antagonism of opposites at the very
such analysis to the study of the conditions of system reproduction. heart of the human condition, in the sense that life is predicated
By circuits of reproduction I mean fairly clearly defined 'tracks' upon nature, yet is not of nature and is set off against it. Human
of processes which feed back to their source, whether or not such beings emerge from the 'nothingness' of inorganic nature and
feedback is reflexively monitored by agents in specific social disappear back into that alien state of the inorganic. This might
positions. When Marx uses the term 'circuits of capital' he seems see m to be an unabashedly religious theme and as such to be the
to have something of this sort in mind; however, Twant to refer to proper province of theology rather than social science. But I
actual conditions of social reproduction, while Marx sometimes think it to be in fact of great analytical interest , although I shall
uses the term in reference to what I have called structural sets. not attempt to develop that contention here.
Reproduction circuits can always usefully be examined in terms Structural contradiction refers to the constitutive features of
of the regionalization of locales. There is no harm in thinking o f human socreties.. T suggest that structural principles operate in
such circuits as having something in common with electronic contradiction. What I mean by this is that structural principles
circuits, which can be traced out in a visual display - the graphic o perate in terms of one another but yet also contravene each
techniques of time-geography , in fact , might be relevant here . ol heL J ! 'Contradiction' in this sense can be further divided into
The reproduction circuits associated with the M-C-M! set - as Iwoo By primary contradictions I refer to those which enter into
Marx himself makes clear - actually depend upon vast processes I he constitution of societal totalities; by secondary contradictions
of change not just within societies but on an international scale. I mean those which are dependent upon, or are brought into
The concentration of th e population in newly expanding (and heing by , primary contradictions. I do not intend by these simply
internally transformed) urban areas is one of these processes of an abstract series of distinctions; they have to be related to the
change. Others concern the nature of the workplace. But as sl udy of the societal types described above. The concept of
important as any of these is the mechanization of transportation , struc tural contradiction has reference to a specific characteriza-
the treme ndous expansion of means o f communi cat ion from th e li on o f the stat e. Excep t in th e case o f tribal society. the state is
late eighteen th century onwards and the development of regard ed as th e foc Lis (a lth o ugh n OI as such also the origin) of
194 Struc ture, System , Socia l Reproduction Contradiction 195

primary struct ural contradi ctio n. of individual existence is interpolated within a dimension of
Of the th ree types of society I have distinguished , tribal societies moral timelessness. Th ere is no need to po rtray suc h circum-
exist in closest relatio n to nature. By this I do not refer to th eir s ta nces o f social life as a Rousseauia n idyll; the poin t is that
technological develo pment , o r at least no t to that alone. In tribal whe th er pastoral and bucolic life 'red in [Ooth and claw', or in
societies huma n beings li ve closely with each o ther in cond itio ns oral cultu res expresses directly the prox imity o f huma nity and
of co-presence a nd with in the rhythms o f nature in their day-to- nature.
day condu ct. but they also integrate the natural world cogn it ively
flW3AL SOCIETY PU"-eminence o f existential contr.ldiClion
with their ac tivities. Fro m the point of view of civilizatio ns - (Ora l cultures ) Absence of state
especially that of th e modern West - this is something to be seen
only negati vely, a failure to rise to a higher level on a cognitive CLASSDIV IDED SOCIETY Structural contlad ic tion/exi stential contrad ict ion
scale. Levi-Stra uss expresses this very well when he comments: State form : d ty/ co untrysi de relation
'Anthropology, we are apt to say ... is concern ed with societ ies CLASS SOCIETY Prc--eminc nce of struc tural contradic tion
that are floll-civi lized. withoul a system of writing, and pre- o r (Capit.1Ii sm ) Stat~ form : nation-s tate
lion-industri al in type.' In some respects , however, it is 'modern '
socie ti es that should be defi ned in negative terms. Our relations Tribal cultures are segmented in character. That is to say, t hey
with one anot her <Ire now only occasionally and fr agmenta ril y consist of multiple centres of high presence availability, in which
based upon 'generic ex perience', the 'concrete "apprehensio n" of the boundaries between different 'societ ies' are usually no t clearly
one person by ano th er'.J2 The mythic 'world view' and the modes marked. In such decentred systems structural contradiction is
of represe ntation that it employs serve to establish ho mo logies nonexistent. Existential contradictio n traces o ut th e conto urs of
betwee n natural and social conditions or. more accurately, make th e natural world. Struc tural contrad ic tio n is signalled by the rise
it possible to eq uate significant contrasts found on differe nt o f the state , which is in turn associated above all with the,
planes: the 'geographical. meteorological , zoological, botanicul , formation of cities. I do no t mean to say that the state is simply'
technical. econom ic, social. ritual , religio us and philosophical'.n hased 'in' the city. Rather, cities are power co ntainers which , in
Myt hs mediate ex istential contradictio n cogniti vely. That is to L:onjunc tio n with their re la tio ns to the countryside , generate th e
say. in myth themes o f incest. o f sexuality, o f life and death are s tructural nexus of the slate fo rm . Ex is[emiai contrad ictio n is
ex plored and 'explicated ' fo r those who tell them and those who weakened by the introduc tio n o f structural contradictio n but no t
listen to them. If tribal societies are cold cultures - c ultures dis.solved altogether. The city is a milieu alie n to that o f nature
whic h are no t caught up in a flu x of change to whic h their and the refore helps to foster attitudes and symbolic systems
institutio ns are geared - it is not because th ey are poorly discre pant from those that aJly themselves with natural elements
'adapted' to nat ure. as evolutionary theories would have it. On and events. The c ity wall may sym bolically and materially seal off
the contrary, it is beca use those institutions in termingle with the urban mOieli from the ou tside. But traditio nal cities could
nature in an imm ediate and embracing fash ion. Ex isten ti al I;xist only through th eir transactions with th eir agrarian hinter-
contradiction is d irectly expressed, as it were , in th ose in stituti ons lands . T heir internal layout and archit ecture still maintained
by virtue of th e key role of kinship and of tr~diti o n. Kinshi p c lose co nn ections with the natura l environment , usually in
relation s are the main format around which that 'co ncre te co nju nc tio n with traditionall y established sym bols. In traditional
"apprehensio n'" of indi vidua ls of which Levi-Strauss speaks is citi es . as has been mentio ned prev io usly. the di stribution of areas
buill. Th ey are also th e mea ns whereby life is produced - or, in <Inc! th e alignment of buildin gs o ft en ex pressed sacred c~)smo
th e o rigin al sense of the lerm , reproduced. Traditio n , o n the II )gica l di stinctio ns.
other hand , is th e source o f th e injectio n of mo ral meaning into [ do no t propose to o ff er a disc ussio n o f th e state or the origin s
th e reversible time o f day-to-day life; immersed in it , the finitud e 41 1' sta te powe r h e re :~' Suffi ce it to say Ihat I ho ld th e 'ea rl y state'
196 Structure. System. Socia l Reproduction Con tradiction 197

to be a co ntrad icto ry formation in the following sense. The state, elsewhere,J5 is not me rely a fo rtuito us one. Nation-states, to
ex pressing the c ity- countryside relation , represents a new type ex press the matter in a n oversimplified way, are the new power
of struc tural prin ciple that is coun ter to the old while still containers that replace cities. Th e transfo rma tio n of the city-
depend ing o n it. The symbiotic/ antagonistic relation of city and counrryside relation thro ugh the emergence o f 'c rea ted environ-
countryside is the specific form of this structural contradic tion. ments' - exemplifi ed by , but not limited to , the 'built
As power containers, cities generate potential dynamism of a e nvironment' of mode rn urbanism - is part and parcel o f the
novel type in 'hislOry'. That is to say, they break with the fo rmation of the nalio n-state. The transmuted c haracter of space
'ahistori car characte r o f cold cultures. In class-divided socie ties a nd of rime is essential to bot h the po litical formation o f the slate
'economy' is typically no t clearly distinct from 'polity ', and the and the differentiated 'economy'. Such a process of transmutation
sense in which tbe state lodges claims to represent the society as a severs structural from existential contradictio n , and the former
whole is minimal. State power has not lost its connection with now becomes pre-eminent over the latt er. Put in less wordy
existential contradiction and is symbolized in persistently religious fa shion, this means that human social organization no longer has
form. The state may have escaped from tradition in the sense of any symmetry with nature; nature becomes a means to th e
being able to innovate through the use of consolidated power. ex pansion of production. Th e suppression of ex istential questions
But it must none the less continually yield to tradition in another and problems is not , and cannot be , wh olly complete. Indeed ,
way, because traditional beliefs and practices retain th eir ho ld they are fundamental to the structural contradictions introduced
everywhere outside the main centres of concentration of state by capitalism and are part of what gives them their peculiarly
agencies. In so far as the power of the state depends upon ex plosive potential. J6
surveillance , this is ce ntred primarily in the physical locales o f the T he primary contradiction of th e ca pitalist (natio n-)state is to
agencies of state: palace, temples and administrative buildings. be found in the mode in which a 'private' sph e re of 'civil society'
The emergence o f state-based societies also alters the scope is created by , but is separate from and in tension with, the 'public'
and pace o f 'histo ry' by stimulating secondary contradictio ns. sphere of the state. It is a mistake to suppose that civil society is
States bring into being , or at least greatly accentuate , socia l everything that lies outside the sco pe of the state, if that is taken
relations across considerable reaches o f time and space. That is 10 mean institutions which precede, and are not incorporated
to say , at th e same time as they generate and consolidate with in , the realm o f stale power. T he o rigins of th e modern state
centralized power, 'drawing in' various aspects of social activity are also the origins of th e sphe re of c ivil soc ie ty - so I wish t.o
within their scope, states stimulate the developme nt of o ther ties claim a l any rate , ahhough I shall leave it here as a bald assertion.
and in terconnectio ns which cu t across the social and territorial Civil society is the sector withi n wh ich ca pital accumulation
realms over wh ich they claim sovereignty. Structural contradiction tlcc urs, fu e lled by the mechanisms of price , pro fit and investment
in this contex t concerns the sovereignty of the state over a given in labour and commodity mark ets. I therefo re take the contra-
territorial area, wh ich is an tagonistic to and yet depends upon dic tion between civil socieLY and state to be at least roughly
processes that cu t across that sphere of jurisdiction and involve p.tra ll el to the classical formu lation o f th e capitalist contradiction
different mechanisms. These include external relation s with other hetween 'private appropriation' and 'socia li zed production'. The
states but also the existence of cross-cutting trading enterprises, eapitalist state , as a 'socia lizing' centre represe nting the power of
religious groups, intell ectual commu nities and so on. th e co mmunity at large, is depend ent upon mechanisms of
Th e secondary contradictions associated with the formati on of production and reproduction whi ch it helps to bring into being
modern nation-states. whose development is intertw in ed with hut which are set off from and antago nisti c to it.
that of industrial capitalism as a mode of economic enterprise, Seco ndary contradiction in the nove l glo bal order ushered in
are s ubstantia ll y d iffere nt from those o f previous eras. Th e hy Ih e advent o f modern capitalism is conce ntrated upon the
con nection between cap italism and the nation-state, I have argued (elision between th e int ern ationalizing o f capital (and of capital-
198 Structure, System , Soc ial Reproduc tion Making History 199

istic mechanisms as a whole) and the internal consolidatio n of the conditions under which actors not only are aware of their
nati on-stat es. It is probably because these push in different interests but are both able and motivated to act on them are
direct ions that most schools of social theory have seen the widely variable. It is right to say, for exampl e , that the existence
conn ections between capi talism and the nation-state as no more of class division presu mes o ppositio n of inte rest (as well as
than an accident of history. The dominant trend in social thought. common interests). But the conditions under which class conflict
in fa ct, has tended to see nation-states as little more than occurs are certainly no t to be inferred directly from this
epiphe no mena o f, o r as me re impediments to, the natural observation . Thus in agrarian states o r c1ass-divided societies
propensity o f capitalist productio n to dissolve pol itical and conflict between domina nt and subordinate classes is relatively
cultural differences. It is no t hard to detect the origins of this type rare; this is mainly because there is very little contact between
of view in nineteenth-centu ry social thought. They lie in classical them which wou ld supply the contexts in which conflict could
political economy and in its main o pponent, Marxism. For both, actually occu r.l7
in spite of their maj or divergencies in other respects , economic According to the conceptio ns I have ou tlined above, the pre-
relationships disclose th e true origin of political formati ons, and eminence of existential contradictio n is characteristic of those
it is economic change which is the leading sou rce of transforming societies immersed in traditio nally sanctioned reversible time -
the modern wo rld , Th is view fails to see that the separation of the societies which 'have no history'. The emergence of structural
'economic', as a sph ere of continued and rapid change, has as its contradiction (the origins of which 1 am no t concerned to try to
necessary cond itio n the power of the modern state. The modern ex plain here) 'heats up' processes of social change. But it is only
state is intrinsically, not just co ntingently , a nation-state , existing with the development of modern capitalism that such processes
in a world of o th er nation-states. become 'white-hot', Compared with the modern world, with its
What is the relation. analyti cally expressed, between contra- extraordinary rates of prolonged social transformation, traditional
diction and conflict, si nce the two terms are often used in the empires and other types of state appear to be marked by an
same breath" absence of change rather than the reverse. What Marx took to be
c haracteristic of the 'Asiatic mode of production' , and rather
ConfHct Strugg le between ac tors or co lle(:tivities co ntemptuously referred to as social and economic stagnation , is
expressed a s definite social prac tices
in fa ct a distinguishing feature o f alliarge-scale agrarian societies
(Structural) Disjunction of structu ral pri nciples of whatever kind . As o ne o bserver has rema rked , it is the
contradiction of syste m o rganizat ion 're latively overwhelming absence of major social and economic
change' that characterizes the variant forms of society that existed
By c_o nnict I mean actual struggle between actors o r groups, across the fa ce of world histo ry until some two o r three centuries
however such struggle may be carried on or through whatever ago. J8
sources it may be mobilized. Whereas contradiction is a structural
concep t, con fli ct is not. Conflict and contradiction tend to Making History
cOIncide because contradictio n expresses the main 'fau lt lines' in
the structural co nstitution of societal systems. The reason for this I shall distinguish two main types of collectivity according to the
coincidence is that contradi ctions tend to involve divisions of form of the relations that enter into their reproduction. I shall
interest betwee n d ifferent gro upings or categories of people c(lll these associations and organizations, and I shall separate
(including classes but no t limited to them), Contradictions express th em from social movements. In associations. as in all social '
divergent modes of life and distributions of life chances in relation systems. social reprodu ction occurs in and through the regularized
to possible worlds wh ich th e actual world discloses as immanent. co nduct of knowledgeabl e agen ts. The settings of interaction in
If contradiction does not inevitably breed conflict, it is beca use which ro utine encounters occ ur are reflex ively monitored by
200 Structure, System, Social Reproduction Making His tor y 201

their constituent actors in the reprodu ction of mutually linked in history. This is true no t o nly because the fo rms o f storage and
role relationships. Bu t while such mo nitoring is th e conditio n of retri eval o f inform ation generated by wri ting allow an expansio n
their reproducti on, it does not take the form o f an active att empt o f time-space distanci atio n but also beca use th e nature o f
to control o r to alter the circumstances o f reproductio n. T here 'traditio n' becomes altered , c hanging the sense in which human
tends (Q be a close conn ectio n between traditional modes o f be ings live 'in ' histo ry. C lass-divided societies have always
legitimacy and the preva le nce o f associatio ns. Traditio n is mo re maintained a strongly traditio nal fo unda tio n , especially o utside
thali a partic ular fo rm o f the experience o f temporality; it the relatively restricted sphere o f c ities. T he works of philosophers
represents the mo ral command o f 'what went befo re' o ver the o f pre-Ch'in China conceived o f the intersectio n o f past and
co ntinuity o f day-to-day life . It is a mistake to suppose tha t present as a mobile relatio n, in whic h no t o nly does 'present '
traditio n, even in the coldest o f cold cultures, is who lly refrac tory penetrate into 'past' and vice ve rsa but also histo ry is 'fl at' rather
to change or to d iversificatio n o f conduct. Shils's characterization th an linear. That is to say , it ru ns laterally rather than 'back.wards'
o f traditio n is probably very apr. Tradition is like 'the movement into time. Life was represented as governed by th e Ii, o r traditio nal
o f raind ro ps on a windowpane. , . . A wavering stream o f water rituals , continuo usly transmitted. According to Hsun Tzu , ;Past
slipping down ward at a n angle, co mes into contact with ano th er and present are the same. T hings that are th e sa me in kind,
stream mo ving at a d iffe rent angle. Th ey fuse into a single stream though extended over a long period , con tinu e to have the self-
for a bri ef mo ment , which th en breaks into two streams, each of sam e principle .'40 None the less, the introduc tio n o f writing means
which might brea k apart again , if the window pane is large that tradition becomes visibl e as 'traditi on' , a specific way , among
enough and th e rain heavy eno ugh.'J'I What the metaphor does o th ers, of doing things. T raditio n' whi ch is kno wn as such is no
not carry, however, is the very aspect of tradition which grounds Jonger a time-honoured basis o f custom but a discursive
the routin e in 'traditio nal societies'. In this respect Levi-Stra uss is pheno menon open to in terrogati on,
surely right to emphasize that tradition is the medium of the So far as 'history' is conce rn ed , it is worth whil e at this point to
reversible time linking the duree o f dail y life with that of the re lurn to Marx 's dictum that h uman beings 'mak e hi story' . It was
lo ngue duree o f institutions, nol merely whimsical to ask earlier what it is that is 'made' here,
Th e disti nctio n between associations o n the o ne side a nd as th e debate between Sartre and Levi-Strauss shows. All human
organizations and social movements on the Olher coinc ides with a beings li ve in histo ry in the sense that their li ves unfo ld in time ,
distinctio n in modes o f reproductio n that I drew in the first but this they share in commo n with all things ex tant. As reflexively
chapter. O rgan izatio ns and social movements are collecti vities in fo unded prac tice , human society is distinc t from that o f the
whic h the refl ex ive regulatio n o f th e conditio ns o f syste m animals, but this in a nd o f itself hard ly explains what 'histo ry' is or
reproduction looms large in the continuity o f day-to-day practices. what there is that is specific to huma n histo ry . T o suggest that a
Orga nizatio ns and social movements are characteristically fou nd respo nse to these issues has to be h isto rical conta ins no paradox
in segments o f class-clivided societies - and , indeed , in some hecause , o f course, 'history' trades o n two meanings : the
degree mark their separatio n fro m tribal societies. For refl exi ve occ urrence o f events in the elapsing of time and the chro nict ing
self-regulatio n, as a pro perty o f collec tivities , depends upo n the or explicatio n o f those evenrs, T he fact that today we tend to
collation of information which can be controlled so as to influ ence d ide th e two senses is expressive o f some key features o f the
the circumstances of social reproduction. Information contro l. in co ntemporary era and again indi ca tes what extrao rdinary
turn , depends upon informatio n storage of a kind distin ct fro m complexities und erlie the innocent pro positio n that human beings
that availabl e in indi vidual recollection, in myths o r story-telling 'mak e history'. For its eluc id a tio n pres um es a philosophical
or in the practical conscio usness o f 'Jived tradition', The invention account of time . We return here to some o f th e matters upon
o f writing, the prim e mode o f the colla tion a nd sto rage o f whi ch I tou ched in th e ve ry first sec tio ns of this book in relation
inform ation in ctass-d ivided societ ies, mark s a radi cal disjunc ture ( ll str ucturatio n th eo ry.
202 Structure, System, Social Reproduction Making History 203

Uvi-Strauss's analysis of 'savage thought' perceptively identifies Thompson is no doubt correct to see in this an anticipation of
some of the relevant questions. Tn Totemism he shows a parallel Marx, as many others have done. But regarding Vi co as a direct
between Bergson's concept of duree and ideas 'common to all the forerunner of Marx means ignoring features of his thought which
Sioux, from the Osage in the south to the Dakota in the north, preserve a divergent view of time and of 'experi.ence'. Thompson
according to which things and beings are nothing but materialized casually dismisses what he calls 'Vico's own attempt to attribute
forms of creative continuity'.41 Bergson's attempt to formulate a to process a cyclical intelligibility', concentrating instead upon
philosophy of time, like the more powerful notions of Heidegger, 'his superb expression of process' arguing , 'this is the point from
can be seen as an endeavour to escape from the 'linear' or which all sustained historical thought must start.'45 But 'cyclical
'unitary' view of time expressed in the world view of modern intelligibility' is fundamental to Vico's views, and it is only
Western culture. Bergson wants to apprehend duree as fusing the relatively recent 'historical thought' that has taken as its point of
continuous and the discontinuous, the order of differences that departure 'history as process'.
actually constitutes 'reality'. Similarly, in the cosmology of the Modern organizations and social movements operate in a social
Sioux, as one song describes it: world in which the retreat of the gods and the dissolving of
Everything as it moves, now and then, here and there, makes stops. tradition create the conditions in which reflexive self-regulation is
The bird as it flies stops in one place to make its nest and in manifested as history - and as sociology. The modern era,
another to rest in its flight. A man when he goes forth stops when dominated by the rise of capitalism in the West over a slim period
he wills. So the god has stopped. The sun, which is so bright and of a few centuries, is one marked by the prevalence of hi,\,toricity,
beautiful, is one place where he has stopped. The moon, the stars, awareness of the 'progressive movement' of society shaped by
the minds, he has been with. The trees, the animals, are all where that very awareness, the 'feeling for world history' of which
he has stopped .... 4l Spengler wrote. The collation, analysis and retrieval of information
that both stimulates and expresses historicity is made possible,
In this version of 'history', as the explication of events, time is
first, by the development of printing and mass literacy and,
associated not with social change but with repetition, not with the
second, by the invention of electronic media of communication.
capability of human beings to transform the world or themselves
Each expands time-space distanciation by an 'alienation' of
but with their involvement in nature.
communication in circumstances of co-presence. Any written
If 'history' , in the phrase human beings 'make history' means
text becomes distanciated from its author; printing is for the most
the conjunction of a linear view of time with the idea that,
part a quantitative extension of such distanciation . Electronic
through expanding the knowledge of their past, agents can change
media separate presence in time from presence in space , a
their future, it is a conception that begins no earlier than Vico.
phenomenon of decisive significance for contemporary forms of
Indeed, Vico's writings might be seen as bridging an older
understanding of time and continuity and a newer, emergent one.
Organizations and social movements are what Touraine calls
Thus in a celebrated passage - quoted and endorsed by
'decision-making units',46 utilizing certain typical forms of resources
Thompson 43 - Vico asserts:
(authoritative and allocative) within discursively mobilized forms
It is true that men have themselves made this world of nations, o f information flow. The study of social movements has been
although not in full cognizance of the outcomes of their activities, distinctly under-represented within the social sciences as compared
for this world without doubt has issued from a mind often diverse, with the vast literature given over to the numerous vying
at times quite contrary, and always superior to the particular ends daborCitions of 'organization theory'. There seems little justifi-
that men had proposed to themselves.. .. That which did all this o,:,lti on for thi s in a century in which revolutions and the clash of
was mind, for men did it with intelligence; it was nol fat e, for they ri val doctrin es ori ented towards radicCiI social change have been
did it by choice; not chance, for the results of their always so so promin e nt, and o ne mu st agree that Touraine and oth ers are
acting are perpetually the sam e.'" ri ght to cl aim Ihat th e noti o ns o f o rganizati o n and soc ial
204 Structure, Sys tem, Social Reproduction Making Historv 20S

movement are o f equ ivalent importance in the mode rn era. Social prototypical of contemporary social moveme nts. In the circuit of
movements can be conceptually differentiated from populatio n capitalist reprOduc tion discussed earTier 'labour powe r' ap pears
movements, migratio ns, etc., precisely because they suppose a as a commodi ty. to be 'translated' into ot her commodities.
high degree o f rerlexive self-regulation. Social moveme nts ca n be However, labour power , o f course, is no t a commodity like any
cogently defin ed as 'collective enterprises to establish a new o th er. Labour moveme nts have the ir o rigin in the fo rms o f
o rder o f life'.47 Unlike organizations, social moveme nts do not 'defensive contro l' whe reby work ers seek to achieve some
characteristica lly ope rate within fixed locales, and positio ning measure o f mastery over circumstances in which they are denied
within them does not have the clarity o f definition associated rights o f participation in d ecisio ns tha t affect them. In so far as
with 'ro les'. labour movements have been infused with socialism, and mo re
Cohn 's characterizatio n o f millennial movements in medieval partic ularly with Marxism , they incorporate historicity in a direct
Europe helps to indicate some of the distinctive elements o f fashio n into the scope of their activities. La bour movements have
social movements in the modern period . As described by Cohn . been animated by mu ch the same nexu s of ideas as th e capitalist
millenarian movements are inspired by the phantasy of a salvatio n organizations aga inst which they have bee n pitted. Whether
which is to be reformist or revolutionary , such movements have been concerned
to foster, although in an egalitari an fashi on , those very forces of
(a) coll ective, in the sense that it is to be enjoyed by the faithful production which their oppo nents seek to develop through the
as a group; acc umulation of capital. Here , however, is where the labour
(b) terrestria l, in th e sense that it is to be realized on this eart h movement ceases to exemplify modern soc ial move ments in
and not in some other- worldly heaven;
general. For Marx it was to carry th e burde n o f a ge neral
(c) imminent, in the sense that it is to come both soon and
reformatio n of the societal totali ty, acting in the general interest
sudden ly;
(d) total, in the sense that it is utterly to transform life on eart h, so against the sectional interests expressed by class divisions. The
that the new dispensation will be no mere improvement on limits of this visio n have become increasingly ap parent , no t o nly
the presen t but perfec tion it.~elf; because of the failure of th e pro letariat to make the r evo lut io n ,~
(e) accomplished by agencies which are consciously regarded as not even because of th e tendency to reduce all sec tional interests
s u pernalu ra l. ~ 10 class interests. but exactly because o f a n un covering o f th e
hislOrical roots o f historicity itself. Our era is o ne which entertains
Cohn's work has been cited so o ft en that some cautio n is rad ical doubts about th e accomplishments of enlightenment
necessary against over-generalization o n the basis o f it. Not all guided by science and by techno logical innovation , o ne in which
medieval social movements can readily be described in terms o f histo ric ity loses its e rstwhile unq uestio ned pre-eminence.
th e above fea tures, and, o f course, mille nn ialis m does not In a similar way the capitalist enterprise is in some respects
disappear with the closing o f the Middle Ages. But we can say bot h typical o f modern o rganizations and one o f the main sou rces
with some confidence that most latt er-day social movements o f in novation ge nerating the circumstances in which they have
differ from all these characteristics of millennialism, with th e arisen. As analysed by Marx, capitali sm is a mode o f productio n
exception of the second and , not infrequently, the third. 49 Modern in which reflexive self-regul ation within th e enterprise - a
social movem ents are almost exclusively this-world ly and are phenomenon clarified by Weber's demonstration of the signi-
invariabl y oppositional in character. They are situated in the fi cance of double-e ntry book -keep in g to th e capitalist firm - is
same 'fi elds of historicity' as the organizations and associations no! matched by re fl ex ive contro l over econo mi c life as a whole.
they confro nt. However, as Weher has again done ma rc than anyo ne else to
The labour move ment may not provide the solution to 'the mak e dear, re fl exive se lf-regu la ti o n gains mo mentum in many
riddle o f histo ry', as Marx fo resaw . But it is in certa in ways sec lo rs of social li fe . Here in li es o ne of the most profo und iss ues
206 Structure, System, Social Reproduction

which faces us today. Is the expansion of a diversity of different

fonns of organization - in which th e conditions of reprod uction
are reflexively monitored - a medium of emancipation from pre-
established modes of exploitative domination? There is no doubt Critical Notes : 'Stru ctural Sociology' and
that Marx believed such to be the case in the context of his Methodological Indi vidualism
anticipation o f the revolutionary overthrow of capitalism by
socialism. But Marx's critics and adversaries, from Weber to
Foucault, have provided mo re than good cause to treat this basic Blau: a Version of Struc tura l Socio logy
tenet of Marxism with caution, if not with outright scepticism.
There are strong connections between a n emphasis upon a
'structural approach', as used by those writing ou tside traditions
of structuralism, and objectivism in the social sciences. Some
mo tifs continually cro p up in the works of those who regard
themselves as taking such an approach. These include particularly
the Durkheimian ideas that 'societies are more than the sum of
their constituent individuals' and (a conceptio n I have already
criticized) that structural properties are qualities of social systems
which are to be defined solely in terms of their constraining
influence over actors. 'Structural approaches' also tend to stress
endurance in time and extension in space. Structures are 'supra-
individual' in the sense that they outlive the individual agent and
spread well beyond the scope of the activity o f individual agents. I .
These considerations obviously overlap considerably with themes
o f my preceding discussion in this book. But something of an
epistemological element is o ften involved too. Fo r it is frequently
held . or assumed , th at to examine structural features of social
activity is to demonstrate causal influences over human conduct
akin to those which o perate in nature .
Thus Wallace identifies the 'crucial difference' between what
he calls 'social structuralist theory' and 'social actionist theory' in
the following way: 'social structuralist theory treats purposiveness
and other subjective o rientational factors as at least secondary
ilnd at most [? ] irrelevant in explaining social phenomena ... .'2
The bluntness with which this view is expressed is not at all
unusual. Consider a recent acco unt o n these lines set out by
Mayhew. Mayhew identifi es the proper concerns of sociology as
structural'. Structures refe r to networks of relations, and such
ne two rks can and should be a nalysed without any allusion to the
d wracte ristics of individual s: 'in stru ctural socio logy th e unit of
analysis', he says , 'is always th e soc ial netwo rk , never the
208 Structure. System. Social Reproduction Critica l Notcs : -Structu ral Sociology' 209

individual.') A 's tructural approac h' is here linked , as is very often scale, and include , for example, wealth , income and education .
the case , with an e nd orsement o f a rather naive form of One o f the main objects of structural study is to examine the
beh aviourism. Mayh ew argues that 'structuralists do no t employ relation between these parameters. in so far as they are associated
subjectivist concepts such as purpose o r goals in th eir analysis.'4 with clusters of interaction. Where the re is considerabl e differen
81au has developed a more sophisticated version o f ideas such tiation along eith er parameter , there wi ll be less chance of such
as these in a number o f recent publications, and his views no interaction eiusters being formed . Parameters can thus be
do ubt are representative of a subs tantial segment o f socio logical analysed so as to explain the fo rm s and degrees o f social
o pini o n . ~ Like most Anglo-Saxon soc io logical write rs, he will differentiation and integration. 81au writes as a 'structural
have nothing to do with Lhi-Slrauss's co nception of structuralism d eterminist', 'who believes that the structures of o bjective social
or with kindred standpoints. He also , however, carefully separates pos itio ns among which people are distributed exe rt more
himself from fun ctio nalism, proposin g a notion o f structure fund amental influ ences on social life th an do cultural va lu es and
'stripped of its broader cultural and fun ctional connotations 10 its n o rms '.~ His aim is to explain variatio ns in the structural features
core properties. 6 Accepting that 'structure' has been used variably of societies , not fac tors relevant to individual attitudes , be liefs or
by different autho rs, he po ints out that it is usually agreed that , in mo tives. Structural analysis in his sense o f the term , he no tes, can
its most elementa l sense, it refers somehow to social position_s a nd be carried on without inves tigating overall charac te ristics o f
relations between soc ial positions. As specified by 8lau , structural societi es.
socia l science is concerned with parameters of po pulation However, he does make claims relevant to those characteristics.
distributions , not with <lctors as such. A 'structural parameter' is Thus he remarks, fo r example, that in small oral cu ltures kinship
any cri terion of ca tegorizing aggregates o f individ uals relevant to is the main co-ordinating stru ctural axis of differentiation and
social positions which individuals migta occupy. He ex plains this integratio n. Industrialized societies. by contrast , are characte rized
as follows: by 'multifo rm heterogeneity', the comple x intersection of
struct ural paramete rs, producing diverse forms of association and
Thus, we speak of th e age structure of a popu lation. th e kinship
clusters of interactio n, In the currenl era, he adds. a co nsiderable
stru cture of a tribe, th e authority structure of an organizalion. the
power structure of a co mmunity, and the class structure of a stru ctural consolidation is going on in Western societies - Blau's
society. Th ese arc not types of social structure but analytical ow n version. in fact , o f th e looming threat of a 'onedimensional'
elements of it dist ingu ishing social positions in one dimension only. social o rde r.9
T he different positions gen erat ed by a single parameter are On the basis of th ese concepts, Blau attempts to fo rmulate
necessarily occupied by different persons - an individu al is either what he calls a dedu ct ive theory o f social structure. T he theory
a man or a woman. old or young, rich or poor - but the case begi ns from propositio ns involving ve ry simple analytical terms
differs for positions generated by several parameters, becau se the (for example, the size of aggregates or groups) and builds up
same person simultaneously occupi es posit ions on different more complex generalizations on this foundation. Some of the
parameters.... Social stru ctures are reflected in diverse for ms of assumptions involved, 8lau says, depend upon 'essentially psycho,
differentiation, which must be kept ana lytically distinct. 1 logical principles'; he quotes as an example the genera lization
t hat people prefer to associate with o the rs who have traits similar
The task of studying stru ctural pa rameters, according to Blau , to their own. However, the structural properties an alysed can not
delimits the distinctive concern of soc iology. he derived directly from such psychOlogical theorems. Blau's
Two types of stru ctural parameter can be di stinguished. d edu ctive theory is a complicated affair, involving several dozen
'Nominal paramet ers' are lateral. separating a given population ge ne ralizations about 'structural effects' , ranging fr o m the
into categories , suc h as gende r , religion o r race: 'gradua ted marvellously trit e ('peop le associate nOI o nly with membe rs of
panlmeters' are hierarc hical. diffe rentiating individuals <llo ng a (heir own groups but also with members o f other groups'). through
210 Stf{Jctwe, System, Soc ial Reproduct ion Critica l No tes. 'S tructura l Socio logy' 211

the mildly interesting altho ugh quite debatable (,decentralization parameters can be defined ind epend ently of 'valu es', 'norms' or
o f authorit y in an association in creases informal association 'cultural traditions' on the other. Hi s progra mme o f discovering
among admin istrati ve ranks') to th e provocative but perhaps the 'independent influen ce the stru cture of social positio ns in a
su bstantially mistaken ('high rates o f mobility promo te stru ctu ral society or community exerts o n social relatio ns' is supposed to be
change'). 'T he theory is sociological', according to Blau , 'in the accomplished 'independently o f cultu ral values and psycho logical
specific sense that it eKplains patlerns o f social re la tio ns- in terms mo tives'. II But reductio n to psycho logical generalizatio ns is not
o f properties o f social stru'c t ure , no t in terms of the assumptio ns the same as formulatio n in terms o f c ultu ral valu es o r meanings.
made , whether o r no t these are derivable from psycho logical The latter have refere nce to the inevita bly he rmeneutic task of
principles. T he nature o f the logical fo rmulatio ns employed makes the generatio n o f social d escriptio ns, paraSitic upon agents'
the explana tio ns struc lU ral." (I concepts that help constitut e the m. A characteristic mistake of
Blau's views are in some ways idiosyncratic but for th e most th e advocates o f structural sociology is to confuse two d ifferent
part exemplify the ambitio ns o f 'stru ctural sociology' in a genera l senses in which the 'objective' nature o f structural properties can
way. He eKpresses in a cogent fashion the persistent feeling be counterposed to 'subjectivity' . Stru ctural parameters, as Slau
shared by many th at SOCio logy can and must be clearly separated de fin es them , are 'no n-subj ecti ve' in the se nse that they cannot be
from other neigh bouring diSciplines, particularly psychology, T he described in terms of individu al pred icates. But th ey canllot be
point is made forcibly th at what gives sociology its distinctiveness 'non-s ubjective' in the sense th at th ey canno t be c haracteri zed at
is its specific concern no t onl y with social structure but also with all independently of 'cultural traditions' , wh ere thi s term refers to
the ways in whi ch th e constraining properties of structure mak e agents' meanings. Thus Blau regards kin ship ca tego ries as
themselves felt in rega rd o f th e co ndu ct of individuals. Acco rding 'structural'. But kinship categori es palemly depend upon concepts
to Blau , neith e r the fo rmulatio n of stru ctural analysis nor and discriminations employed by actors. Th e very term 'position',
stru ctural explan ation needs to make reference to 'values or so basic to Blau's notion o f stru cture, clearly invol ves agents'
norms'. In th is latt er respect he seems to d iverge from Durkheim , co ncepts. Social positio ns, lik.e all ot her aspects of 'structural
but in other ways what he has to say could be taken as a latter-day parameters', exist onl y in so far as actors make discriminatio ns in
version of a Durkheimian manifesto, Discussio n of the s ho rt - the ir conduct based upon the attri butio n o f certain ide ntit ies to
comings of his s tandpo int will both serve to reiterate features o f ot hers.
s[ruc turatio n theory mentio ned earlier and help to point up T he notio n [hat the study o f structural parameters is convergent
aspects o f 'stru cture' and 'struc tu ra l propertie.s' as I want to with the distinctive character o f socio logy might be plausible if
understand those te rms. th ere were some definit e causal properties associated with the m,
Th ere are some in teresting and instru ct ive features o f Slau 's thus mak ing 'sociological eKplanation' convergent with 'structural
ideas. He skirts fun ctionalism, and he avoids id entifyi ng structural e xplanatio n'. But the causal relatio ns s upposedly at work are
analysis with some unex pli cated influence which society 'as a o bscure - although eviden tly supposed to operate in some way
whole' has over its ind ividual members. He recognizes that ou tsid e the scope o f the reasons that agents might have for what
societies are not all o f a pi ece - that is to say, that one of the they do. Thus Bla u pro poses the genera lizatio n that an organi-
aims o f structural study sho uld be to show exactly what levels o f zation \ increasing size prod uces greater internal differentiation
integration can be discovered within and between social gro upings. and hence raises th e proportio n o f administrative personnel it
None the less, (h e limitatio ns of such a conception of 'stru ctural contains. According to him , thi s rel<ltionship can be grasped
sociol ogy' are mark ed. 'without investigatin g the mo tives o f indi vidua ls in o rganizations'.11
Bl au's app roach confuses th ~ demand to dist ingu ish th e But , as Blau develops it , thi s propositio n is false. If the implication
influ ence of stru ctural properties from psychological explanat io ns we re that certain typical mo ti ves ca n be ass umed by the th eorist ,
of conduct o n th e o ne hand wi th th e assertion that structural and co uld be spe lled o ut if n\.!ccssary, the sta tement could be
212 Structure. Sys tem, Socia l Reproduc tion Crit ical Notes : 'Stru ctu ra l Soc io logy' 213

defended. But this is no t what Blau has in mind. He makes it clear repeat: there is 110 such thinK as a distinctive category of :5tntClllral
th at he thinks that the specificatio n of motives (and reasons or explanation ', only an interpretatio n of the modes in which varying
intentions) is actually irrelevant to th e factors in vo lved in the form s o f constraint influ ence human action. There is nothing
generalization. And this is not so at all. It is , on th e contrary, myste rio us about what 'influen ce' means here. Take the
necessary precise ly to its causal ex plication. The increased gen erali za tion that hi gh rates of mo bility promot e structural
pro portion of administrators will tend to come a bout as actors change. We can probably assum e that high mobilit y rates are
respond to what they see to be new problems and issues which largely unintended and that the resultant changes they indu ce are
increased organizatio nal size presents,lJ also , altho ugh it may be [he case , fo r example, that edu catio nal
T he .'struct~ral' generalizatio ns given by Blau may in fact on policies have been set up in order to enh ance mo bility and hence
closer mspectlon turn out to be formulae which actors use to that what is going o n is part of a reflex ively monito red process.
produce the results indicated. If we know nothing about what the Suppose , however , that the mobility in question is unintended, is
~ge nt s t~em~elves beli eve they are doing - becau se this type of that of women , is upward occupatio nal mobility , and that th e
mformatlo n IS thought to be distinct from the analysis of struc tural 'struc tural change' it promo tes is highe r (or lower) divorce rates.
effects - we cannot assess the likelihood that such may be the We can pro be what th e causal innuences might be , but o nly by
case. Those who administer organizatio ns have their ow n theori es- know ing about the motives and reasons of those invo lved -
in-use about them a nd may , indeed. be well aware of the academic wives, husbands and o thers. It could be that women who become
literature on the subj ect. Consider th e proposition that d ecentra- successful in occupatio nal careers spend less time at home than
lization of authority in organizations increases informal associations they would otherwise, leading to th e (unintended) result of placing
between administrati ve ranks. As with th e generalizatio n about a strain on the mariral relationship ; that they see marriage as
size and internal differentiation , this may presume intended unimportant compared with success at work ; that their husbands
conseq u.ences which agents have reasons fo r bringing about , or , resent their success , etc.; o r a combina tio n of all of these fo r
alternatIvely, the o utcome may be largely uninte nded. It is differe m individ uals.
essential for the social observer to know which is th e case in
order to be able to elu cidate what is going on. At least some of An Alte rn at ive? Methodolog ical In d ividu a lism
the ag~n ts . involv.ed coul~ be ,acting in the light of the very
gen.erahzatlOn which Blau Identifies. It might very well be that a Conceptio ns o f distinctively 'structural ex planatio n' in socio logy
policy o f decentralizatio n is followed specifically in o rder to have lo ng had a natural e nemy in methodological individualism.
increase certain sorts o f informal association among different The d ebate betwee n th e two posi tio ns is in some part the
ranks of administrato rs. methodological cou nterpart to the dualism of subj ect and social
What these comm ents demonstra te is that a 'str uctural object that has characterized the ont o logy of the social sciences.
~ppro ach' to th e social sciences cannot be severed from an Although Max Weber has frequently been adopted as a 'stru ctural
examin ation of th e mechanisms o f social reprodu ction. It is soci ologist' , he mad e his own preferences cl ear enough. In a
perfe~tly c~rr~c~ . of course , to emph asize that society is no t a kite r whic h he wro te no t lo ng befo re his d eath he observed: 'if I
c re~ [l on o f mdlvldu al acto rs and that the Structural prope rties of have beco me a sociologist ... it is mainly it:! o rder to exorcise the
SOCIal systems endure beyond the lifetimes of individua ls. But s pectre of collec tive conceptions wh ich still lingers among us. In
stru c ~ure, or structural properties, or 'structural parame ters', exist oth er wo rds, socio logy itse lf can only proceed from th e act io ns of
~nly In so far as th ere is continuity in social reprodu cti o n across o ne o r mo re separate individuals and must therefore adopt strictly
tIme and space. And such continuity in turn exists o nly in and indi vidu alist ic m e th od s:l~ Human action. as Weber says in
thro ugh the reflexively mo nitored ac tivities of siluated ac to rs ECOIIOIIIY lIlId Sociely, 'ex ists on ly as the beha viour o f o ne or
having a range of intended and unintended consequ ences. Let m~ mo re illdhidulI! human be in gs. I~ The dcba te over what clai ms
21 4 Structure, System, Socia! Reproduction Critical Notes: 'StructuraI Socio!ogy' 21 5

We ber and other 'methodological individualists' might in fact be

making has stretched far and wide , but there is no doubt a According to this principle, the ultimate constituents of t~e soci~1
genuine difference of opinion between them and the 'structural world are individual people who act more or less appropriately In
socio logists'. The details may be complex, but the ou tline is the light of their dispositions and understanding of their situation.
relatively simple. The methodological individualists agree with Every complex social situation, institution or event is the res ult o f
the view I have stated above: the search for 'structu ral a particular configuration of individuals, th ~ ir disposit ions,
explanation ' is futile and perh aps even harmful. situations, beliefs, and physical resources and environment. There
Let me follow through one of the more influential acco unts o f may be unfinish ed or half'way explanations of large-sca le social
the issues raised by various versions of methodological individu al- phenom ena (say, inflation) in terms of other large-scal.e phenomena
ism. Luk es discusses and seeks to 'render harml ess' each of what (say, full employment); but we shall not have arnv~ at rock-
he takes to be the main expressio ns o f methodo logical individual- bottom ex planat ions of such large-scale phenomena until we ha~e
ism." The doctrines that advocate methodological individualism dedu ced an account of them from statements about the diS-
involve o ne or more of the foHowing theses. positions, beliefs, resources and inter-relations of ~ndi v i~u als: ~T he
individuals may remain anonymous and only typical dispOSitions,
(1) 'Truistic social atomism'. T his is the view which ho lds that it etc., may be attributed to Ihem.)19
is se lf-evident that social phenomena can be explain ed only
in terms of the analysis of the conduct of individuals. Thus Lukes's bomb-disposal sq uad o f arguments designed to defust
Hayek says: 'There is no oth er way toward an understanding methodological individualism advances on t.w0, fr,onts. None ~f
of social phenomena but through our understand in g of th e claims mention ed under the fo ur categones IS In the least bit
individual actions directed towards other people and gu ided plausible when examined closely . Since the fi rst is truistic (that is
by their expected be haviour' l7 (a formulatio n close to to say, trivially true) , it is neither here nor there . That 'soc i ~t y
We ber's definition of 'socia l act io n', in fact). consists of people' is a 'banal proposition about the world ' which
(2) The idea that all statements about social pheno mena is 'analytically true , that is, in virtue of the meaning o f words'.20
such as Blau's exposition o f structural parameters - can be The second , third and fo urth points are demo nstrably false. The
reduced, without loss of meaning , to descriptio ns of the fa ct that the description or analysis of kinsh ip relations, such as
qualities of individuals. This view would d eny that Blau's those designated as 'cross-cousi n marriage' , cannot be accom-
talk o f 'structure' makes any sense; he is merely aggregating plished without reference to the knowledgeability of,human agen.'s
properties of individuals. does not entail that such relations can be deSCribed solely 10
(3) The assertion that only individuals are real. Thus it seems to terms of predicates of individuals. If point (3) implies somehow
be held by some writers that any concepts which refer to that only individuals are di rectly observable, it is mi stak~n. -
properties of collectivities o r social systems (one might again although there is no reason in any case to suppor~ th ~ propoSItion ,
instance 'structural parameters') are abstract models, con- associa ted with behaviourism , that only that which IS observable
structions of the theorist , in some way that th e notio n o f is real. We may not be able to o bserve the elements Bla u ha:> in
'individu al' is not. mind when he speaks of structural parameters, but we certamly
(4) Th e allegation that there cannot be laws in th e social can observe social phenomena in circumstances o f co-p resence ,
sciences, save in so far as there are laws about the such as the formation a nd enactment of en counters. Finally,
psychological dispositions o f individuals. ' ~ po int (4) is covered by what I have s~id p~evi o u s ly : there is no
shortage of generalizations in the SOCial s~l e n ces, alth? ugh they
All of th ese four elements wou ld appear to be found in th e mu ch- do not have the sam e logical fo rm as uDlversal laws In natural
qu oted statement that Watkins gives of what he calls the 'principle sc ience.
of methodological individu alism ': Th ese arguments, Luk e concedes, do not yet render me thodo-
216 Structure, System, Social Reproduction Critical No tes : 'Structural Sociology' 217

logical individu alism harmless. Th ey do not even attack its main exclude the possibility o f more struc tural analyses , o r these
strength , which is concerned wi th explanation . The most characteristics are covered by the re hultal o f (3) above and do , in
important assertion in the quotatio n from Watkins, and perhaps fact, involve social (structural) characterizatio ns anyway. Hence
also that from Haye k, is to be found in th e declaratio n that ;rock methodological individualism has been neutralized. Those who
bott om' explanations of social phenomena have to involve the advocate a reduction ism involving phYSio logical characteristics
'disposi tions , belie fs, resources and inler-relations o f individ uals'. o f the organ ism cannot make their claim count for anything as
It is here thal Lukes feels the potentially explosive power of regards the actual pract ice o f the social sciences, but o thers
methodological indiv idualism to li e. and from where the fu se has cannot find any properties of indi vidu als that are no t irreducibly
to be delicately removed. Wha t are the 'dispositio ns, etc.' of 'contaminated' by the social.
individuals? And what , in any case, is 'explanati o n'? As regards There Lukes leaves th e matter. I do not think that this will do;
th e latter, Lukes is able to demo nstrate rather easily that many we have to formul ate th e issues rather differently . However,
proponents of methodOlogical individualism have in mind an befo re picking up some of the threads left dangling by Lukes's
overly restricted no tio n of what explanation is (this is equally true discussio n , it will be instru cti ve to refer to some quite similar
of Blau and most of the structural sociologists). T o ex plain is (0 pro bl ems raised fro m a differe nt quarter - in int erc hanges
answer a why question, and often this involves making a particular between Thompso n and Anderso n about the character of
social phenomenon intelligible sim ply in the sense o f providing Marxi sm,u Thompso n has long regarded structural concepts as
an accurate characteri zation of it. 21 Explanation here o perates, as suspect, without rej ecting them altogether, and has consistently
it were, on or aro und the baselin e of the necessarily hermeneutic emphasized the signifi cance of studying the texture and variety of
natu re of the social sc iences. It is undeniably important to human agency. Thu s in describing the views infonning his analysis
emphasize that 'explanation' partakes o f the contex tu ality of all o f class de velopme nt in England in the eightee nth and nin eteenth
social ac tivity , whether this be in respect of the inquiries of lay centuries , he comments, 'class is d efin ed by men as they live thei r
actors or those of soc iological obse rvers. However, let us own history, and , in th e end, this is its only definiti o n: u In the
co ncentrate upon the more confined meaning of 'exp lanat ion' as co urse of a sustained po lemic against Althusser and those
having to do with the formulation not just of generalizations but influ enced by him - prompting a b oo k ~length reply from
of ca usal generalizations - in o ther words, generalizations which And erso n - Thompson spells out the implications of his
do not simply assert that a rela lion o f an abstract kind holds standpo int in some d etail. I shall mak e no attempt to characterize
between two categories o r classes o f social pheno me na but also the de bate as a who le but shall mentio n o nly a few aspects of it
id entify the causal connections involved. whic h are relevant here.
In what sense do these causal connections necessarily relate to Althusser is taken to task by Thompso n - rightly, in my
individuals? According to Lukes, in some versions o f methodo o pinion14 - for offering a deficient account of human agency and
logical individualism the qualiti es o f individual s invoked in a d eterministic conceptio n of structure. Human beings are
explanations are physiological traits of the organism or organically regarded not as kn owledgeable agents but only as the 'supports'
given needs. But th ese e xplanatio ns turn out to be quite fo r modes of production . This 'derogatio n of the lay actor', as I
implausible. No o ne has been able to produce any accounts have called it , Tho mpson ex presses in blunter fashio n. Althusser,
which reduce social ph eno mena to organic properties. So these and most others associated with either structuralism or runctional-
forms of meth odological individualism are at best hypot hetical ism , 'proceed from th e same "latent anthropology" , the same
cl aims: they have no direct bearing upon the materi als of study ult erio r assumption abo ut "Man" - that all men and women
with which social scientists operate. In other interpretations of (exce pt th emse lves) are bloody s il~)' . 2 s Social life, or human
methodo logical individualism. however. either th e characteristics histo ry, Thompson says, sho uld be und erstood as 'unm astered
aHributed to indi viduals and incorporated in explana tions do nOl human practice'. T hai is to say, human beings act purposively
218 Structure, Sys fem, Social Reproduction Crit ical Notes : 'Structural Sociology' 219

and knowledgeably bur without being able either to foresee or to not rise like the sun at the appointed time , It was present at its
cont rol the consequ ences of what they do. To understand how own making' and its formation 'owes as much to agency as to
this happen s we need a term which , Thompson says, goes missing conditioning',25 But in spite of the plaudits which the work has
in Althusser: it is what Thompson simply calls 'human expericnce'.26 justly rece ived , And erson points out, it does not really resolve the
Experience is the connection between 'structure' and 'process' , issues thus raised,
th e real material of social or historical analysis. Thompson stresses
that such a view does not bring him close to methodological For if the claim for the co-determination of agency and necess ity
individualism. In fact , he finds a certain affinity between were to be substantiated, we would need to have at a minimum a
me thodologica l individualism and Althusser's Marxism. For conjoint exploration of the objective assemblage and transformation
Althusser believes th at 'structu res' exist only wit hin theoretical of a labour force by the Industrial Revolution, and of the subjective
domains, not in reality itself; hence this stance resembles the germination of a class cu lture in response to it. ... [But ] the advent
nominalism of the met hodological individualists. But yet in the of industrial capitalism in England is a dreadful backcloth ro the
end it is not easy to see just how di sti nct Thompson's ideas are book rather than a direct object of analysis in its own right. ... The
jagged temporal rhythms and breaks, and the uneven spatial
from methodo logical individualism. Many of the passages in his
distributions and displacements.. of capital accumu lation between
work where he characterizes his overall views resemble con- 1790 and 1830 inev itably marked the composition and character of
ceptions suc h as that of Watkins quoted above, Thus, talking the nascent English proletariat. Yet they find no place in this
agai n of the co ncept of class, he insists: 'When we speak of a class account of its formari on.2'J
we are thinking of a very loosely defined body of people who
share the same categories of interests, social experie nces, tradition The interchange between Thompson and Anderson is not at all
and value-system, who have a disposition to behave as a class, to conclusive , but it is useful to place it alongside the more abstract
define themse lves in their actions and in their co nsciousness in debate about methodological individualism, The latter debate
relation to other gro ups of people in class ways,'2'7 seems large ly played o ut, but th e livelin ess of the polemics
There is much that is attractive about Thompso n's views, but between Thompson and Anderson graphically demonstrates that
Anderson does no t find it difficult 10 find some sho rtcomings in the issues are not dead. There is one very important sense in
them. When Thompson writes of 'people' and of the primacy of which they cannot be, Every research invest igatio n in the social
'ex perience', how are these seem ingly transparent terms actually sciences o r history is involved in relating action to structure, in
to be understood? In emphasizing th em Thompso n clearly means tracing, explicitly or otherwise, the conjunction or disjunctions of
to accentuate the significance of human agency in making history, intended and unintend ed consequences of activity and how these
But what 'agency' is remains unexplica ted . in spite of the profusion affect the fate of individuals. No amount of juggling with abstract
of historical ex amples Thompson offers in the course of his concepts could substitu te for the direct study o f such problems in
o riginal works an d by way of criticizing Althusser. 'Ex perien ce' the actual contexts of interactio n. For the permutations of
- as we kn ow from Dilthey's attempts to grapple with Erleb nis influences are endless , an d there is no sense in which structure
- is a notoriously ambiguous term. One use of the word , for 'determines' action or vice versa. The nature of the constraints to
example, connects direc tly with empi ricism, in which ex perie nce which individuals are subject, the uses to which they put the
is a passive registration o f events in the world, something very far capacilies they have and the forms of knowledgeability they
from the active connotations of the term which Thom pson wishes display are all themse lves manifestly historically variable.
to accentuate, Moreover, Thompson nowhere effect ively teases Conceptual clarifi cation can at least help in suggesting how
out the relation between action and st ru cture. This is even tru e of lhese matters are best approached, What connects the arguments
his major book. The Making of/he English Working Class. The o f Tho mpson with those of Watk ins el al. is that both rest their
book opens with a celebrated paragraph: 'The working class did cases too much upon an intuitive . untheorized conceptio n of the
220 Structure, System, Social Reproduction References 221

'individual' or agent. They are quite justified in being suspicious though they were, as in the above examples? We tend to do so
of the aspirations of 'structural sociology', whether it takes the when there is a significant degree of reflexive monitoring of the
form offered by Blau or that elaborated by Althusser. Methodo- conditions of social reproduction, of the sort associated especially
logical individualism is not, as Lukes suggests, harmless in respect with organizations, although not exclusive to them. The govern-
of the objectives of 'structural sociologists'. The methodological ment decided to pursue policy X' is a shorthand description of
individualists are wrong in so far as they claim that social decisions taken hy individuals, but normally in some kind of
categories can be reduced to descriptions in terms of individual consultation with one another, or where a resulting policy is
predicates. But they are right to suspect that 'structural sociology' normatively binding. Decisions that are taken by governments or
blots out, or at least radically underestimates, the knowledge- other organizations may not represent the desired outcome of all,
ability of human agents, and they are right to insist that 'social or the most desired outcome of any, of those who participate in
forces' are always nothing more and nothing less than mixes of making them. In such circumstances it makes sense to say that
intended or unintended consequences of action undertaken in participants 'decide' (individually) 'to decide' (corporately) upon
specifiable contexts. a given course of action. That is to say, individual members of a
'Structural sociology' and methodolo,6ical individualism are Cabinet may agree to he bound by the outcome of a meeting with
not alternatives, such that to reject one is to accept the other. In which they disagree or a proposal which they voted against yet
some respects, as Luke says, the debate between the two sides is which found majority support. It is important to understand that
an empty one. The point is to discard some of the terms of the 'The government decided .. .' or 'The government acted .. .' are
debate while elaborating others further than any of its contributors shorthand statements because in some situations it may matter a
have done. What the 'individual' is cannot be taken as obvious. great deal which individuals were the main initiators or executors
The question here is not that of comparing predicates but of of whatever decisions were taken (or not taken) and whatever
specifying what human agents are like - something I have tried policies followed.
to do in respect of the basic concepts of structuration theory.
This presumes abandoning the equation of structure with References
constraint. The relation between enablement and constraint can
be fairly easily set out on a logical level, given a beginning point Structure, System, Social Reproduction
in the notion of the duality of structure. History is not 'unmastered
human practices'. It is the temporality of human practices, CPST, pp. 222-5.
fashioning and fashioned by structural properties, within which 2 CCHM chapter 8.
diverse forms of power are incorporated - not by any means as 3 Ibid., pp. 45-6. My discussion here modifies my earlier version of
neat a turn of phrase, but I think it is more accurately put. this problem only slighlly. For other sections drawn upon here, see
A further question raised by the debate over methodological pp. 157 - 64 and 166-9.
individualism is: are collectivities actors'? What does it mean to 4 I have followed Eberhard's discussion closely in the preceding few
paragraphs: Wolfram Eberhard, Conquerors and Rulers (Leiden:
say, for example, 'The government decided to pursue policy X"?
Brill, 1965), p. 9, and passim.
or 'The government acted quickly in the face of the threat of S Marshall G. S. Hodgson, 'The interrelations of societies in history',
rebellion'? Various distinctions need to be disentangled here. Campara/ive Studies in Sacie/y and History. vol. 5,1962-3, p. 233.
Action descriptions, as I have mentioned in a previous chapter, 6 H. A. Gailey, A History of Africa, 1800/0 the Present (New York:
should not be confused with the designation of agency as such. Ho ughto n-Mifflin, 1970-2,2 vols.; Rene Grousset, The Empire of
Neither descriptions of action nor accounts of interaction can be the Steppes (New Brunswick: Rutgers University Press, 1970).
given purely in terms of individual predicates. But only individuals, 7 T. Carls( ein, 'Th e sociology of s(ructuration in time and space: a
beings which have a corporeal existence, are agents. If collectivi- tinK'-geographie assessment of Giddens's theory', Swedish Geo-
ties or groups are not agents, why do we sometimes speak as ~mf)hi(: {/I Ye,,/"/look (Lu nd: Lund University Press, 1981); Derek
222 Structure, System, Soc ial Reproduction References 223

Layder, Structure, Interaction and Social Theory (London: 23 Marx, Capital, p. 110.
Routledge, 1981); 1. B. Thompson, Critical Hermeneutics 24 Ibid., pp. 110 and 103.
(Cambridge: Cambridge University Press, 1981 ); Margaret S. 25 Ibid., p. 168.
Archer, 'Morphogenesis versus structuration: o n combining struc- 26 for an earlier version of some of these points, see CSAS, chapter 6.
lUre and action', British Journal of Sociology, vol. 33, 1982. 27 Marx, Capital, vol. I , p. 337.
8 Carlstein, 'The sociology o f struc turation in time and space', pp. 28 Ibid., p. 338.
52- 3. See also John Thompson, Cn'tical Hermeneutics (Cambridge: 29 Ibid., p. 356.
Cambridge University Press, 1981), pp. 143-4. 30 Marx, Capital, vol. I, p. III .
9 Roy Bhaskar, The Possibility of Naturalism (Brighton: Harvester, 31 CPST. pp. 141ff.
1979), p. 42. 32 Claude Levi-Strauss, Structural Anthropology (Londo n: Allen Lane.
10 Emile Durkheim, Th e Rules of Sociological Method (London: 1%8), pp. 365-6.
Macmillan, 1982), pp. 39 - 40. 33 Claude Levi-Strauss, Th e Savage Mind (London: Weidenfeld &
11 Ibid .. pp. 50 and 52. Nicolson, 1966), p. 93.
12 Ibid. , pp. 2-3. 34 This is a major preoccupation of Between Capilal,:9m and Socials:9m.
13 Karl Marx, Capital (London: Lawrence & Wishart, 1970), p. 72. An 35 CCHM, chapters 7, 8 and 9. I also leave o ut of consideration here
instructive discussion of this issue appears in Gillian Rose, The the very important question (a lso ana lysed in CCHM) of the
Melan choly Science (London: Macmillan, 1978), chapter 3. relations between capitalism, the state and class divisions.
14 Karl Marx, Gnmdrisse (Harmondsworth: Penguin, 1976), p. 157. 36 A theme more fully developed in Betweell Capitalism and Socialism.
15 See CPST, chapter 5. 37 See John H. Kautsky, The Politics of Aristocratic Empires (Chapel
16 Prepared for the writing of CeHM but not in the final version Hill: University of North Carolina Press, 1982): 'If a class is
included ~h erein. conceived of as a grouping in conflict with a nother class, then,
17 The classfication also leaves open the possibility of other types - indeed, aristocracies and peasantries are not classer at all' (p. 75) .
for example, state socialist society as distinct from capitalism, as 38 Ibid., pp. 5-6. See also Henri J. M. Claessen and Peter Skalnik,
well, of course, as o ther forms of societal organization that might The Early State (The Hague : Mouton, 1978).
conceivably develop in the future. 39 Edward Shils, Tradition (Londo n: Faber & Faber, 1981), p. 280.
18 The view expressed in CCHM, p. 164, 'The city is the locus of the 40 Arthur WaJey, Three Ways of Tho ught in Ancient Chilla (London:
mechanisms which produce system integration', is rather inade- Allen & Unwin, 1939), p. 38. For an exte nded discussion , see J. G.
quately formulated. Moreover, I do not want to convey that the A. Pocock , "The origins of the study of the past', Comparative
city- cou ntryside relation is a unitary or s ingle one; it is he tero- Studies in Society and History. vol. 4, 1% 1- 2.
geneous a nd complex whe n considered across the generality of 41 C laude Uvi-Strauss, Totemism (London : Merlin. 1964), p. 98.
societies. 42 Ibid. Levi-Strauss also re marks, 'The Dakota language possesses no
19 Immanuel Wallerstein, The Modern World-System (New York: word to designate time, but it ca n express in a number of ways
Academic Press, 1974); cf. Spengler: 'Is it not ridiculous to oppose modes of being in duration. For Dakota thought, in fact, time
a "modern" history of a few centuries, and that history to all intents constitutes a duration in which measu rement does not intervene: it
localized in West Europe, to an "ancient" history which covers as is a limitless "free good'" (p. 99). Inte resting observatio ns relevant
many millennia - incidentally dumping into that "ancient history" to these issues are made in Birgit Schintlholze r, Die Auflosung des
the whole mass o f the pre-Hellenic cultures, unprobed and Ge~'chichtbegriff\' im Stnlkl1lralismus, doctoral dissertation
unordered, as mere appendix matter?' Oswald Spengler, Th e (Hamburg, 1973).
Decline of the West (London: Allen & Unwin, 1961), p. 38. 43 E. P. Thompson, Th e POI'erly of Th eory (London: Merlin, 1978),
20 Cf. note 2 above. pp. 86 and 291.
21 Cf. my essay, 'The nation-state and violence'. tl4 G. Vieo, Th e New Science (Ithaca: Cornell University Press, 1968),
22 CPSTpp.104 - 5. p. 3R2. para. 1108.
tiS Thompson. Th e Po verty of Th eory, p. SIl.
224 Structure, System. Social Reproduction Reference s 225

46 Alain Tourain e, The Self ProeillctiOIJ of Society (Chicago: Un iversiry 12 Peter M. Blau, ' A for mal t heory o f diffe rentiatio n in organizations',
of Chicago Press, 1977), p. 238. American Sociology Review. voL 35, 1970. p. 203.
47 Herbert Blumer, 'Collective behaviour', in Alfred M. Lee, Principles 13 This po inl is made in Ste phen p, Tu rner. ' Blau's t heory of
of Sociology (New Yor k: Barnes & Noble, 195 1), p. 199. differe ntiation : is it explanatory?" Sociological Quarlerly, vol. 18,
48 Norman Cohn, ' Mediaeval millenarian ism: its bearing upon lhe 1977. Some of these issues are ai red agai n in Blau: ' Co mments on
comparative slU dy o f millenarian movemen ts', in Silvia L Thru pp, the prospects fo r a no mo thetic theory o f social structure', }oumal
Millenial Dreams ill A ctio" (The Hague: Mo uton, 1% 2), p, 3 1. for the Theory of Social Behaviour. vol. 13, 1983. See also .an
49 C f. J. A. Banks. T he Sociology of Social M ovemellts (Londo n: extrao rdinary piece by Mayhew, in the sa me volume o n ' Causahty,
Macmillan, 1972). pp. 20 - 1. and passim. histo rica l particularism and o ther errors in socio logical discourse'.
50 Andre Can. FarclI'e ff 10 Ihe Worki" g Class (London: Plu to, 1982). Blau's contribut io n co nti nues to display the sho rtcomings I have
ind icated. ( 1) Herme ne utic elements in the fo rm ulation of conce pts
o f social analysis are suppressed in favour o f the view (hat 't ~ e
objective of sociology is to study the influence o f the "social
Crilical NOleJ: 'Sll"Uctural Sociology ' environment' on "peo ple's observable tende ncies'" (p. 268). (2)
and Mel hodological l ndividllahml Rererence to agents' motives, reasons and intentio ns is persisten tly
equated with ' psycho logy', relegat ed to a realm separate from the
C f. Ray mond Bo udon, Th e Uses of St ructuralism (Lond on : concerns of 'sociology'. (3) A version of a d iscredited philosophy of
Heinemann, 197 1). Boudon categorizes a number o f divergent uses natural science. in wh ich 'explanatio n' is regarded as necessarily to
of th e concept. For a rather different set of approaches, see Peter do with ' nomo th etic-deductive theorizing' (p. 265), is accepted
M. Blau, Approaches ( 0 the Study of So cial Slmclure (Londo n: unqu estioningly. (4 ) No consideration is given to the poss ibility that
Collier-Macmi llan, 1975). even if th e philosophy o f natural sc ience th us impli ed were
2 Walter L. Wall ace, 'S tructure and act ion in the theories of Cole ma n acceptabl e, the character o f ' laws' in soc ia l science might be
and Parsons'. in Blau, A pp roaches to the Swdy of S ocial Slrucru re. fu ndamen tally differe nt fro m laws of nature. (5) T he whole
p. 121. standpoin t is wrapped up in the familia r but erroneous claim t hat
3 Bruce H, May hew, 'S tru cturalis m versus ind ividualism', Parts 1 and social science, as compared with natural sc ience, is in t he early
2, Social Forces. vol. 59. 1980, p. 349. phases of irs d evelopme nt. Blau accepts that there are, 'at least so
4 Ibid., p. 348. fa r" ' no deterministic la ws in socio logy' (p, 2(6). But he ex presses
5 Petcr M. Blau, Ineqllality and Heterogeneity (New York : Free fa ith that t hese will o ne day be fo und - we certainly canno t write
Press, 1977); 'SIrUClU ral effects', A merican Sociological R eviell'. o f( the possibility because ' nomot hetic theory o f social structure is
vo l. 25, 1960; ' Paramete rs o f social structure', in Blau, A pp roaches undo ubtedly still in a most r udimentary stage' (p. 269).
10 f he Sludy of Social Sf l1J.CW re; ' A macrosocio logical t heo ry o f 14 Q uoted in Wolfgang Mo mmsen, ' Max Weber's political sociology
social structure', A merican Jou rnal of Sociology. vo l. 83, 1977, a nd his philosophy o r wo rld history', I lllem (l(iollll l Socia l Science
6 l ll equafify and H eferogell eify. p. ix. Jou rnal, vol. 17, 1965, p. 25. Of course, it is a moot po int how far
7 ' Parameters o f social structure', p. 22 1. We ber's substantive writings were guid ed by Ihis principle.
8 Illequality and Helerop,eneify. p. x. 15 Max W eber, Ecoll omy and Society ( Ber keley: University of
9 ' Parameters o f social stru cture', pp. 252- 3. 'W hat poses this threat Cal iforn ia Pres..., 1978 ), vo l. I, p. 13.
is the domin ant position of powerful o rganizations in contemporary I(l Sleven Lukes, 'Methodological ind ividualis m reconsidered', in
society, such as th e Pentagon, the White House, and huge Essays in Social Th eory (London : Macmillan, 1977).
conglomerates. T he trend has been toward increasing concentrat ion 17 F. A. Haye k, II/divid ualism all d Economic Order (Chicago :
of economic and manpower resources and powers derived fro m University o f C hicago Press. 1949). p. 6.
t hem in gianr organizations and their to p executives, whic h im p li es IH Lu kes also id en tifi es a furth er co nno tatio n o f met hodological
a growing consolidat ion of majo r resources and forms o f power. . ind ivid ualism. a doctri ne o f 'socia l individua lism' which '(ambigu-
10 In equality Gil d Heterogelleity, p. 246. ollsly) aserts th at society has as its end the g()od o f ind ividuals'.
II 'A macrosocio logical theory o f social struct ure', p. 28. Lukes, ' Methodolog ica l ind ividua lis m reconsidered'. pp. 18 1- 2.
226 Structure, System, Social Reproduction

19 1. W. N. Watkins, 'Historical ex planation in th e social sciences', in

P. Gardiner, Th eories 0/ History (G lencoe: Free Press, 1959).
20 Lukes, 'Methodological individualism reconsidered', p. 178.
21 Cf. N RSM. chapter 4.
22 E. P. Thompson. Th e Poverty of Theory (London: Medin, 1978); Change, Evolution and Power
Perry Anderson, Arguments within En[;lish Marxism (London:
Verso, 1980).
23 E. P. Thompson, Th e Making 0/ the English W orking Class
(Harmondsworth: Penguin, 1968), p. 40.
24 CPST, chapter 1, and passim.
25 Thompson, The Poverty 0/ Theory, p. 148.
26 Ibid., p. 30. I wan t to argue in this chapter for a deconstruction of a whole
27 Ibid., p. 295. Italics in the original. range o f theories of social change, particularly those of an
~8 T hompson, Th e Making 0/ the English W orking Class. p. 9. evolutionary type , and for a reconstruction of the nature of
:).9 Anderson, Ar[;uments within Engh~h Mar:xl:ml. pp. 32-4. power as inherent in the constitution of social life. To deconstruct
theories of social change means to deny that some of the most
cherished ambitions of social theory - including those of
'historical materialism' - can be realized. This does not imply
making the relatively weak claim that such theories cannot be
sUpt'orted by the available evidence. It involves a much stronger
and'more controversial contention: that they are mistaken about
the types of account of social change that are possible. A
deconstruction of theori es of social change can proceed through
three sets of considerations of progressively diminishing generality,
as below:

level of genera lity


, 0c i,,1 e vo lution

h ist o rica l materia li,m

A great d eal of social science, in academic sociology as well as

Marxism, has been based upon the presumption that it is possible
10 formulate th eorems of structural causation which will explain
Ihe determination o f social action in general. ' Most versions of
sl ruclural determination are linked to the thesis that the social
sciences can uncover universal laws, these laws identifying the
e ffec ts o f stru ctural constraints. A given occurrence or type of
hdHlviour would be show n to be an instance o f a general law,

Udcrc n",-" ,n ay he [<lund "11 pp. 27 .. \ - 9.

228 Change. EvolUfion and Power Evolutionism and Social Theory 229

certain boundary conditio ns for the o peration o f the law having called 'endoge nous' or ;unfo lding' models o f change, which I have
been specified. 'Determin ation' here equals a particular form of criticized earlier. Th ese so rts o f evo lutio nary theory have in fact
determinism. The so-called 'covering law' debate has explo red often been close ly connected with fun ctionalism - th e works of
these sorlS o f issues at some length , and without entering into it CornI e being a notable instance - and th e separation between
directly it is eno ugh to say here th at suc h a view is quite fun ctio nalism and evolutio nism introduced by Malinowski and
inco nsistent with the character o f ge neralizatio ns in the social o thers should perhaps be regarded as something of an abe rration
sciences as r have portrayed it previo usly (see also pp. 343- 7).1 rath er than a natural state o f affairs. Organi c metapho rs have
To deny th at a general account of structural determination can oft en provided the relatio n between the two. A pl ant o r an
be achieved is to take a stance to which a great deal o f this boo k organism contains within itself a trajectory of growth , an unfolding
is relevant. l o f latent potentia lities. Change here is understood as governed by
So me theo ri es of social change are link ed to the abo ve the mec hanisms in volved in such unfolding, with societies being
conception s. It has som etimes been th ought, for ex ample, th at regarded as clearly bounded unities. Extern al conditio ns are held
there are universal laws governing social change an d that a th eory to accentuate or hold back processes of growth , but th ey are
o f social change should be o rganized around such laws. But there really a background against which the mecha nisms o f change
are many attempts to ex plain change which, while they do no t o perate, Some evolu tio nary mod els have treat ed change as
postulate laws, specify certain limited principles of determin atio n inhere ntly slow and cumulative. T hus Durkheim regarded poli tical
o f change whi ch are held to apply in something like a universal revo lu tion as agitatio n on the surface of social life, in capable of
fashio n. Among these, evolutio nary conceptions have been by far givin g ri se to major transformations o f soc iety because th e
the most prominent. evolu tio n o f basic social institutio ns is always necessarily slow.'
'Evolutionism', of course, canno t be easily categorized, since a But unfOlding conceptions o f cha nge are certainly no t alien to
variety of different sta nd po ints have been associated with the theo ries which propose that evolution proceeds thro ugh processes
term, and th e po pularity of evolutionary conceptions has waxed of rev~utionary t.ransition. Marx's vi ews represent a case in
and wan ed in the social scien ces. Th e second half of the po int . T he chi ef motor of social change, in the scheme Marx
nineteenth century was certainly the high po int o f evolutio nism po rtrays in the 'Prefa ce ' to A Contribution to the Crilique of
in soc ial theory, inspired in some considerable degree by the Politica/ Economy. is the expa nsio n o f [he fo rces of production
ac hievements of Darwin in biology.4 Evolution ary notions sub- within a given ty pe of society. At some po int such an ex pansio n
seque ntly tended to dro p o ut of fashi o n, especially amo ng (;un no lo nger be co ntained within the existing institutio ns of the
anth ropologists, who for the most part became strongly influ enced society, leading to a process of revolutio n, fo llowing which the
by o ne or other interpretatio n of 'cultural relativism'. But s uch same process occurs all over again.1 The sources o f c hange are 10
no tio ns reta ined some defe nders in anthropology, and in he fo und in the tendential pro perties o f class soc ieties, which
archaeology evo lutionism has consistently remained do minant. conta in th e 'seeds o f their own transform ation'.
In the Anglo-Saxon wo rld th e rise o f functionalism, as led by Ho w should th e term 'evolutio n' be unde rstood? The word
Malinowski and Radcliffe-Brown in anthropology and subsequently it self comes from th e Latin evo/Illia, deri ved from e- ('o ut of') and
in sociology by Merto n and Parsons, was in some degree m/llfllS ('ro lled '). It was useQLO refer t<i:the unro iling o f parchment
respo nsible fo r the eclipse of evolutio nary thinking, altho ugh a hoo~ks. Th ~~ n c~p t W~ no t applied in anything like its modern
revival of evolutionary th eory was later initi ated by Parsons se nse until the la te seve nteenth century , when it came to mean an
himself. ~ ord erl y process o f change , pass ing thro ugh d iscernible stages.
Co mt c was o ne o f fhe first leading social think ers to make
Evol utio ni sm a nd Soc ial Theory cx t-ensive use o f the not io n, and his fo rmulat io n is not very
d iffere nt rrom those pro posed by many subseq uently (i ncluding
Muny th eori es o f evolu tio n fo rm prim e ex amples o f what 1 have Pa rst ll1 s, sec PI'. 2():\ - 74). T he va ria ti o n of societal types, th eir
230 Change, Evolution and Power Evolutionism and Social Theory 231

differentiation and synthesis, promoting 'order with continuity' from old. On the other side, evolution generates progress: higher
- these were Comte's themes. 'Aucun ordre reel ne peut plus forms arise from, and surpass, lower. The first of these directions is
s'etablir, ni surtout durer, s'il n'est pleinement compatible avec Ie Specific Evolution, and the second, General Evolution . a
progres; aucun grand progres ne saurait effectivement s'accomplir, different taxonomy is required in examining these two aspects of
s'il ne tend finalement a l'evidente consolidation de l'ordre. 'R evolution. Concerned with lines of descent, the study of specific
Let me mention some latter-day definitions of social or cultural evolution employs phylogenetic classification. In the general
evolution, culled more or less at random: evolutionary outlook emphasis shifts to the character of progress
itself, and forms are classed in stages or levels of development
Whether the adjective 'biological' be used or not, the principle of without reference to phylogeny. (Sahlinsl])
evolution is firmly established as applying to the world of living
things .... Such basic concepts of organic evolution or variation, There are significant points of variation between these
selection, adaptation, differentiation, and integration belong at the fonnulations. What Sahlins calls 'specific evolution', for example,
centre of our concern, when appropriately adjusted to a social and is the only sense of evolution recognized by Steward, who actually
cultural subject-matter. (Parsonsp directly rejects 'general evolution'. But the definitions do tend to
have certain common traits, stated or implied, and they can be
Evolution can be considered as an interest in determining recurrent utilized to characterize what a theory or approach has to be like
forms, processes and functions .... Cultural evolution may be to be worth calling 'evolutionary'. I shall take it for granted that
regarded as either a special type of historical reconstruction or a 'evolution' is to be more than a casually applied term (to which
particular methodology or approach. (Stewardl'o there can be no objection), synonymous with 'development' or
'change'. For 'evolutionary theory' in the social sciences to have a
Evolution (both natural and social) is a self-maintaining, self- distinctive m;;:anTng,rsllalr say";~ it" shoulO. show the following
transforming and self-transcending process, directional in time and characteristics. (I do not think these are arbitrary or overly
therefore irreversible, which in its course generates every fresh strong.) /.
novelty, greater variety, more complex organization, higher levels Fi ~ there must be at least some presumed conceptual
of awareness, and increasingly conscious mental activity. (Huxley)" continuity with biologicat !=<v_olution. As the above definitions
malUn:tear-,-thlsis acriterion which many but not all of those who
Evolution may be defined as a temporal sequence of forms: one regard themselves as evolutionary theorists are prone to
form grows out of another; culture advances from one stage to emphasize. It is a claim that makes sense, for even if it originated
another. In this process time is as integral a factor as change of primarily within social thought rather than in biology, it is the
form. The evolutionist process is irreversible and non-repetitive .... latter which has given 'evolution' a fairly precise designation and
The evolutionist process is like the historical, or diffusionist, process elaborated an account of evolutionary transformations - one
in that both are tempora~ and therefore irreversible and non- which illuminates evolution without using any teleological notions
repetitive. But they differ in that the former is nomothetic in
at all. To use the term 'evolution' in the social sciences is rather
character, whereas the latter is idiographic.... To be sure, the
gratuitous if it does not have at least some connections with the
evolutionist process always takes place somewhere and in a
temporal continuum, but the particular time and the particular conceptual vocabulary which has become established in biology.
place are not significant. It is the temporal sequence of forms that It does not follow from this that a complete conceptual
counts. (White)12 correspondence is either necessary or desirable. Evolutionism, or
at any rate Darwinianism, has recently come under strong attacks
In both its biological and cultural spheres evolution moves within natural science, and it is conceivable, if not at all likely,
simultaneously in two directions. On the one side, it creates that it may be discarded there while being sustained in the realm
diversity through adaptive modification: new forms differentiatt: o f social sci e nce.
232 Change, Evolutio n and Power Adaptation 233

S~co nd ,M>..cial. evolutionism mu st specify something mo re than because o f e thnocentri c assumptions whi ch , while probably not
ju~t a progression of cha nge in respect of certain designated ' logically implied in evo lutio nism, are very difficult in prac tice to
~~it eria , that something being a mechanism o f change. Tbis point avoid.
needs looking aC in a certain amo unt o f detail , because it is Fo urth.....Jdentifying a mec hanism of soc ial change means
impo rtant. Some evolutionists te nd to believe that to de fend the exp1il'ining change in .some way which applies ac ross the whole
co ncept o f evolution in soc ial th eory, it is enough to sho w th at spectrum -of human histo ry, not as an exclusive mechanism of
progressio n has occurred, in respect of a certain social it em or ch~tnge but as th e dominant o ne. There is no do ubt about the
items, over th e course of histo ry fro m the earliest period of which prime candidate here, sin ce it figures somewhere in virtually all
we have evidence of human soci ety up to the modern era, Thus , evolutionary theories, ho wever much they may differ in other
f0t-example, White has constructed an index of evolutio n o n t he respects. T~_is is 'adaptatio n' - usually meaning adaptati on to the
basi~ o f e nergy productio n, Socie ties, or in White's termino logy , material environment.
'cultural systems', vary as means of harnessing e nergy. Some are Not all accounts o f social change whic h depend upo n the
mo re effective in this regard than o th ers. Different c ultural concept of adaptatio n a re evo lutionary, since they may not
systems may therefore be ra nk ed along a scale by co mpa ring confo rm to th e first three criteria. But the no tio n o f ad apta tion is
coeffi cients derived from relating th e amo unt of energy harnessed so ~ rtant in evolutio nary theo ries that wi tho ut it th ey lose
and expended to the number o f huma n beings involved in those mos t o f their cogen cyY It makes sense, therefo re, to hold that if
syste ms,14 From Comte and Spence r onwards , evo luti onary in the explication o f social change the concept of adaptation
thinkers have referred to increasing complexity, differentiati on turn s out to be without valu e (as I shall claim ), evolutionism is
and so on, Of course , 'evo lutio n' cou ld be used just to refer to stripped of much of its appea\. But I shall also pursue two further
such progression , abstracted fro m tim e and space. It may be critical avenues of auack upo n evolutionary theories: th ey forc e
justifia ble to say, for example , that small , o ral cultures are at o ne human history into a mo uld which it does no t fit descriptively,
end o f a continuum of ene rgy cons umptio n and distributio n (or a nd they tend to be associa ted , although no t inevitably, with a
time-space distanciation), with the modern , industrialized societies number of unfortun ate corollaries.
at the o ther, There is no diffic ulty, either , in sustaining th e cl aim
th at certain technical develo pme nts, o r forms of social o rganiza Adaptation
tion , are prerequisites to o th e rs. 'Evolution' in thi s se nse is
unco ntentious as a concept. But to use 'evolution' in this way is T he concept of adaptatio n, used in a social co ntext, can be
not to ex pl ain anything abo ut social change and does no t meet shown characteristi cally to be either (1) vacuo us , i.e " so wide and
the criterion of having a reasonably close affinity to bio logical vague in its meaning as to be mo re confusing than illuminating, or
evo lutio n. (2) impli cated in a specio us and lo gically d efi cient claim to
Thl.!"d , a sequence of stages o f social development must be func tio nalist explanatio n, o r (3) involved in the predicatio n o f
specified , in which the mechanism o f change is linked to the dynamic te ndencies in human societies that are d emo nstrably
dls(5lacement o f certain types o r aspects of social organizatio n by false.
others. These stages may be arranged in the form eithe r of T o address the first po int : th e notion of adaptatio n can be used
specific o r o f general evolutio n, o r some kind of combin atio n of in a fairl y precise way in bio logy, whence it derives,' 6 where its
th e two. No presumption must be smuggled in suc h that usual meaning is to refe r to mod es in whi ch th e gene pool of
progression up such an evoluti o nary scale means progress as organisms is influ enced by interaction with th e environment as a
judged in terms of moral criteria , save in so far as thi s is explic itly f\!SlIlt o f selective survival traits, 'Adaptatio n' ca n perhaps be
justifi ed in some way , As I shall emphasize below, evoluti o nary formulat ed in a coge nt way in so cial scien ce if it is taken as a
theories are highly prone to me rge 'progress ion ' with progress' gene ral labe l referring 10 the gamul o f processes whe re by human
234 Change, Evolution and Power Adaptation 235

beings respond to and modify features of their physical environ- including other societies (i.e" the 'social environment') within the
ments. Thus Rappaport defines the term as 'the process by which term 'environment' and/or by including as 'adaptation' more or
organisms or groups of organisms, through responsive changes in less any major social process which seems to further the changes
their states, structures, or compositions, maintain homeostasis in of maintaining a society in something like a stable form. Once this
and among themselves in the face of both short-term environ- has been done, however, the concept becomes so vague that it is
mental fluctuations and long-term changes in the composition or useless as a means of explaining anything at all.
structure of their environments'.17 It is characteristic of evolu- Second, it is often because of its vacuous character, as
tionary social thought, however, to extend this usage so much expressed in such formulations, that the notion of adaptation
that the term becomes irremediably amorphous. For instance, features so widely in spurious 'explanations'. It is if little value
Harding begins his discussion of adaptation by defining the indeed to claim that those societies or types of society which have
concept as 'the securing and conserving of control over the survived for a given period of time, because they survived, must
environment', which is unobjectionable enough. But he then goes have survived. But that is exactly what explanations which involve
on to say that in evolutionary theory adaptation concerns not just 'adaptation' frequently amount to. Thus it is common to propose
the relation between societies and nature but 'the mutual that the survival of a social item can be explained in terms of its
adjustment of societies'. superior adaptive capacity. But how is adaptive capacity under-
Adaptation to nature will shape a culture's technology and stood? In terms comparable with those above - all the elements
derivatively its social and ideological components. Yet adaptation which need to be invoked if that item is to endure while another
to other cultures may shape society and ideology, which in turn act does not. Where 'adaptation' is understood in a more limited way,
upon technology and determine its future course. The total result however, proffered explanations tend to be equally defective,
of the adaptive process is the production of an organized cultural embodying versions of functionalism. ,Q An example which is
whole, an integrated technology, society, and ideology, which typical of much of the relevant literature and has had a great deal
copes with the dual selective influence of nature on the one hand of substantive influence is the following, from O. H. Childe, who
and the impact of outside cultures on the other.'~ starts from the obvious fact that man cannOllive without eating. So
Adaptation here has simply become such a diffuse notion as to a society cannot exist unless its members can secure enough food
include all possible sources of influence upon social organization to keep alive and reproduce. In any society approved beliefs or
and transformation! institutions that cut off the food supply altogether (if for instance
This sort of usage is entirely typical of evolutionary theories in all Egyptian peasants had felt obliged to work all year round
the social sciences (compare, for example, Parsons's usage of the building a superpyramid), or stopped reproduction (as a universal
concept, discussed on pp. 270-1). The reasons for this are plain and fanatical conviction of the virtue of celibacy would do), the
enough. Where 'adaptation' is specified with some degree of society in question would soon come to an end. In this limiting
precision - as in the formulation by Rappaport - and where case it is quite obvious that the food supply must exercise a final
control in determining even beliefs and ideals. Presumably. then,
what is adapted to is also clearly delimited, the notion is manifestly
methods of getting a living in the end exercise a similar control
inadequate as a general mechanism of social change. If environ-
more concretely. The way people get their living should be
ment means 'natural environment', and if 'adapting' to it means expected in the long run to 'determine' their beliefs and
responding to distinguishable changes in that environment in institutions.'H
ways which have this effect of modifying existing organic or
social traits, 'adaptation' simply is much too narrow to be a However, what is obvious to Childe does not follow at all from his
credible candidate for such a mechanism. It can be made plausible premise . To identify a functional exigency of a society or social
only by expanding one or both aspects of its meaning - by it em carri es no implication at all, in and of itself, about its actual
23& Change, Evolu tion and Power Evolution and History 237

influence upon the shaping o f the institutions which meet it. logical time and the progressio n o f the species are integrated - is
Turning to the last of the three charges, adaptatio n wo uld be an inappro priate meta pho r by whic h to analyse human socie ty .
given expl anatory force if a dynamic were found which success Human beings make their histo ry in cogniza nce of th at histo ry,
fully interpreted the diversity and the succession o f the major that is, as re nexi ve bei ngs cognitively appropria ting time rather
types o f human society in histo ry. Here evolutionary theo ri es than metely 'living' it. The po int is a hackneyed e no ugh one, but
show th emselves to be empirically wanting. If it were th e case us ually figures in the di sc ussions of evo lutio nists o nly in relation
that th ere were some sort of generali zed motivational impulse fo r to th e question of wh eth er o r not there is a d istinctive break
human beings progressively to 'adapt' more effectively to th eir be tween proto-humans and Homo sapiem. Tha t is to say, they
material environments . there would be a basis for sustainin g regard it simply as some thing new added to exi stin g evolutionary
evolutionary th eory. But th ere is no t any such compulsio n,2l processes - ano ther fac tor complicating natura l selection. Th e)
Alternatively, it might be supposed that some sort of equivalent nub o f the matte r, howeve r, is that the relexive na ture o f human
to natural selectio n could be fo und in respect o f human societies. social life sub"y~rt~ the explicatio n o f soCia l c hange in terms of
T his is certainly what many nineteenth-century evolutio nists any simple-and sovereign set o f qa us~ mechanisms. Getting to
supposed. Spencer preferred his term , 'survival o f the fitt est', to know what goes o n 'in' his tory bec6--nles no t o nly a n inhe re nt part
'natura l se lection' , but th e idea is the same. He interpreted o f what 'history' is bu t a lso a means o f t@".s(o rming 'hi story'.
'survival' less as a result of meeting the material requirements o f a Evolutio nary th eory in bio logy depends upo n postula tes of the
give n environment than of besting other societies militarily. The ind ependence of the o rigin o f species and th e un cha ngeability of
fo rmatio n of larger and larger societies through war, Spen cer speci es save through mutatio n. Th ese conditi o ns do not apply in
says , 'is an inevitable process thro ugh which the varieti es o f men human history. 'Societi es' simply do not have th e degree of
most adapted for social life suppl ant the less adapted vari eties'.22 'closure' that species do. Biologists can fa irly easil y answer the
But if this sort of view has become largely discarded today, eve n questio n: what evo lves? But the re is no readil y ava ilabl e 'unit of
amo ng evo lutio nists , it is fo r sound empirical reasons. Th e evolutio n' in th e sphe re o f the social sciences.ll I have already
influence o f war upon social change is real eno ugh. But military made this po int (pp. 163 - 8), but it needs to be repea ted here_
strength simply does not have the o verall explan ato ry valu e Evol utio nists usua ll y speak o f th e evolutio n of ei ther 'societies' or
necessary to turn 'adaptatio n' into a viable evolutio nary mecha- 'c ultural systems', with the presumptio n tha t those whi ch are
nism . O nce we start adding in o th er factors, howeve r, we are most highly advan ced are simply differentiated versions of the
back to th e situation where the concept explains everything and less advanced. But what constitutes a 'society' or 'culture' varies
nothing. with the very traits upo n which evolutionary thinkers tend to
co ncentrate. The debate between evo lutionists and 'diffusionists'
Evolution and H istory helped only to concea l thi s problem because both tended to treat
socie ties or c ultures as discrete entities, differing primarily in
Human history does no t have an evolutio nary 'shape', and positive respect o f their divergent appraisa ls of the so urces o f cha nge that
harm can be do ne by atte mpting to compress it into o ne. Here I a ffected th em.
shall list three reasons why human histo ry does no t resemble an Human history is not, to use Gelln er's lerm . a 'wo rld-growth
evolutionary model of the speci es and four dangers to which story'. As G ellner re ma rks, for two ce nturies it has bee n difficult
evolutio nary thought in the social sci ences is pron e. Most o f the for anyone from the West to
ground has been well-traversed by critics of evolutio ni sm from
th e nineteenth-century onwards, but it is perhaps wo rth while think about human a ffairs without the image . . . o f an all-embracing
spelling th ese items o ut. An evo luti onary 'shape' - a trunk wi th upwa rd growth.. . It seemed a natu,al conclusion from the
branches , o r a climbing vin e, in which the elapsing o f chro no - paU crn o f Western hislory. which was generally treat ed as (iIe
238 Change, Evolution and Power Evolution and History 239

history of humanity. Western history seems to have a certain ascent of civilization; it conforms more to Toynbee's picture Of\
continuity and a certain persistent upward swing - or at any rate, the rise and fall of civilizatio ns and th eir conflictual relatio ns with
so it seemed, and so it came to be taught. Emerging from the river tribal chiefdoms. T his pattern is ended by the rise to global pre-
valleys of the Middle East, the story o f civilization seems one of eminence o f the West , a phe no meno n which gives to 'history'
continuous and in the main upward growth, only occasio nally q uite a different stamp from anything that has gone before , 1
interrupted by plateaus or even retrogressions: history seemed to truncated into a tiny pe riod o f some two o r three ce nturies. !
creep gently around the shores o f the Mediterranean and then up Rather than seeing the modern world as a furth e r accentua tio n of
the Atlantic coast, things getting better and better. Oriental conditio ns that existed in class-divided societies, it is much more
empires, the Greeks, the Ro mans, C hristianity, the Dark Ages, the illuminating to see it as placing a caesura upo n the traditional
Renaissance, the Reformation, industrialization and struggle fo r world , which it seems irretrievably to corrode and destroy. The
social justice ... the familiar story, with variants especially in the ~ld is born out of disco ntinuity with what wen..!. bef~re
later details, stresses and anticipation; all this is extremely familiar rather t!go COntinuity with it. It is'the nature o f this discontinuity
and still forms t he background image of history for most of us.... the specificity o f the world ushered in by the advent of
The picture of course dovetailed with biological evolutionism, and industrial capitalism, originally located and fo unded in the West
the victory of Darwinism seemed to clinch the matter. Two quite - which it is the business of sociology to exp lain as best it can.
independent disciplines, history and biology, provided, it seemed, Let me conclud e by briefly listing four dangers which
different parts of the same continuous curve. 24
evolutionary thought courts - dangers which are best avoided by
The voyage of the Beagle symbolized, as it were , the journeys breaking with it in a radi cal way. They are those of what I shall
that brought Europeans into contact with diverse and exotic call (1) unilineal compressio n, (2) homological compression, (3)
cultures, subsumed and categorized within an embracing scheme normative illusion and (4) temporal distortion.
in which the West naturally stood at the top. There is no sign that T he first danger, unilineal compression, means the tendency of
evolutio nary schemes today are free from this sort of ethno-_ evolutio nary thinkers to compress general into specific evolution.
~nsm , Where can o ne find such a scheme in Western social T hus feudalism precedes capitalism in Europe and is the social
science which holds that traditional India is at the head of the nexus fro m which capita lism develo ps. It is therefore, in o ne
scale? Or ancient China? Or , for that matter, modern India or sense at least, the necessary forerunner o f capitalism. Is feudalism ,
China ?~ th en, a general 'stage' in the evolutio n o f capitalism?26 Surely no t,
However, there is no need to pose such questions - which are altho ugh there are versio ns o f Marxism , and other schools of
obviously not logically waterproof in terms of their damaging social though also , that would have it thus.
implications for evolutionary theories - to show that history is By homological compressio n, the second danger , I refer to the
not a 'world-growth story'. The history of Homo sapiens is more te ndency of some writers to imagine that there is a homo logy
accurately portrayed as follows. No one can be sure when Homo between th e stages of social evolution and the development of
sapiens first appeared, but what is certain is that for the vast bulk the individual personality. It is worth discussing this in at least
of the period during which human beings have existed they have moderate detail because although it does not directly depend
lived in small hunting-and-gathering societies. Over most of this upon the postulates of evolu tion ism discussed thus far , it is none
period there is little discernible progression in respect of either the less quite often associated with evolutio nary thought. Baldly
social or technological change: a 'stable state' would be a more stated. it is supposed that smail, o ral cultures are distinguished by
accurate description. For reasons that remain highly controversial, forms of cognition, affectivity or condu ct found only at the
at a certain point class-divided 'civilizations' come into being. first re latively early stages o f the developm ent of the individual in
of all in Mesopotamia, then elsewhere . But the relatively sho rt more evo lved societies. The level of complexity of societal
period of history since th en is not one marked by the continuing organization, fo r instance , may be supposed to be mirrored by
240 Change, Evolurion and Power Evolution and History 241

that of personality developme nt. A correlate of this view is that people on whom the individual depends in all his actions. , , . As
increased complexity o f soci ety implies a heightened degree of more and more people must attun e their conduct to that of others,
repression of affect. Freud's Civilizal io n and its Discontents is the the web of actions must be organ ized more and more strictly and
IOCliS cln.uicu.'i o f such a standpoint. Freud uses th e term accurately. , , , The individual is compelled to regulate his condu ct
'civilization' to refer to 'the who le su m of the achievements and in an increasingly differemiated, more even and stable manner. ...
the regulations which distinguish o ur lives from those of our The web of actions grows so complex and so extensive, the effort
animal ancesto rs and which serve two purposes - namely to to behave 'correctly' within it becomes so great, that beside the
indiv idual's conscious self-control an absolute, blindly fun ctioning
protect men against nature , and to adjusllheir mutual relations'.17
apparatus of self-control is firml y es t abl is h ed . ~
In strongly emphasizing progressive control over the material
world , Freud's disc ussion o f 'civilization' - a notion abom which Elias does stress certain specific cha racteristics of the modern
much more could be said - shares some strong affinities with West , but these are largely subm erged in a generalized evolution-
historical materialism. Perhaps it is not as surprising as may ism. In the 'less complex societies' th ere is lower individual self-
appear at first sight , lh en, that some Marxists have picked up control, greater spontaneo us expressio n o f emotion, etc. People
other aspects of Freud's conception of social development. in such societies are rather like children , spo ntaneous and
Marcuse's attempt to harness Freud's interpretation of 'civili- volatile.
zation' to a critique o f the capita list mod e of production acce pts If this view is wrong, as 1 beli eve it to be, the re is a variety of
the fundam e ntals of Freud's view. The transmutation of 'animal implications that can be drawn as regards both the nature of
man ' into th e 'human being' represents a movement from primitive modern capitalism and th e liberating potential that it might
barbarism to civilization: contain.J() But why is it wrong, and what type of perspective
shou ld replace it? In some part we have to look to the findings of
From To modern anthropology , which surely dispel the id ea that 'primitive
immcdiille satis(lIction d elllyed satisfaction societies' are primitive in anything other than their material
pleasolt! restrain t or pleasure teChnology. The study of language perh a ps pro vides something
joy (p la y) to il (work)
receptive ness produc ti veness o f a baseline here , There simply is no disce rnible correlation
absence of repression sccurily18 between linguistic complexity and the level of material 'advance-
ment' o[ different societies. This [act in itself would indicate that
there is unlikely (0 be any general differences of psychic
Marcuse differs from Freud o nly in supposing thai the 'struggle
o rganization between oral c ultures o n the one hand and
with nature' that is the basis of human mate rial existence can be
'civilizations' on the other. We have to be careful even with the
alleviated by the productive forces generated by, but no t capable
supposition that civilizations are more complex than oral cultures.
of humane expression withi n, the economic order of cap italism.
Civilizations _ . but , above all , that specific fo rm of global order
A comparable utilizatio n of Freud, although stripped of the
ushered in by the ascendancy o f the West over the past two
vision of a radical reconstitutio n of society, is to be found in the
centuries - involve greater timespace distanciation than do oral
writings of Elias. Elias builds his theory very directly around the
c ultures. They bracket more ex tensive segments of time
theorem that increasing complexity of social life necessarily entails
(probably) and space (certainly). However, some features of social
increased psychological rep ression:
activity found in oral cultures, such as those associated with
From the earliest period of the history of the Occident to the kinship institutions, are exceptionally complex. Of course, it
prescnt. social fun ctions have become more and more differen- might be pointed out that Freud's view, and that of others who
tiated under th e pressure of competition . The more differentiated have adopted a similar position, is centred upon the repression of
th ey become, the larger grows th e number of fun ctions and thus of affec t, or relative lack of it , in ora l cultures . But the evidence
242 Change, Evolution and Power Evolution and History 243

simply does not support the proposition that such cultures are The views established in th ese so urces conform to all the main
universally associated with spontaneity of emotional expression. criteria by which I have identified evo lutionism and also carry
Some oral c ultures (as the ego psychologists , among others, have some of its noxious secondary implicatio ns. It is true that Marx
sought to demonstrate) have very strong moral prohibitions that sometimes wrote as tho ugh he were do ing no more than producing
cover a range o f daily conduc t. and the repressions inculcated in a history of Western Europe. But he was surely no t just writing an
child training may be very severe. J1 interpretative account of one corner o f the world . His scheme o f
By the tendency o f evolutio nary theo ry to normative illusio n, the development , invo lving tribal society, the ancient world , feudalism ,
third danger, I mean th e inclin at io n to identify superio r powe r, capitalism , plus the Asiatic mode of production , is an evolutionary
economic, po liti cal o r military, with moral superiority o n an framework in which adaptation , in th e guise o f the expansion o f
evolutio nary scale. Such an inclination is no doubt closely related the forces o f, production , plays th e leading ro le. Why is the
to the ethnoce ntri c conno tatio ns of evolutionism, but it is not Asiatic form of society 'stagnant' co mpared with the West?
exactly the sa me thing. The concept of adaptation is again a Because it does not allow for the deve lo pm ent of the forces of
hazardous one in this connectio n. It has an ethically neutral sound, production beyond a certain po int. It would, of course, be a
as if superior 'adaptive capacity' were ipso jac(o superiority in mistake to bracket Marx too closely with o ther versions of
respect of normatively superi o r social traits. When applied to nineteenth-century evolutionism , his admiration for Darwin
human societies, however, the term is more often than not a notwithstanding. His preoccupatio n with th e in creasing mastery
synonym for sheer might. If tbe adage that might does not confer o f nature which human beings achi eve expresses a version of the
right is an old o ne , it is frequ ently forgotten by evoluti onary notion of adaptation not essentially different from many other
theorists as a conseq uence o f their very evolutionism. Jl uses of the idea. But in Marx there is an in verted Hegelian
Finally, by temporal d islOrtion, the fourth danger, I mean th e dialectic , tortured into a partic ular developmental shape , that has
proclivity of evo lutiona ry thinkers to presume that 'h istory' ca n no direct analogue in more o rthodox evolutio nary theories.
be written only as social change, that the elapsing of time is th e Marx's evolutionism is a 'world-grow th story' and displays the
same thing as change, the confusion of 'history' with 'histori city'. shortcomings of unilin eal compression and temporal distortion.
Is historical materialism a fo rm o f evolutionism '? With certain But one must object to it primarily in terms of the role it accords
reservatio ns, we may say that it is, if the term is understood in a to adaptive mechanisms. Childe's versio n o f historical materialism
certain way. Suppose 'historical materialism' is understood in a may be in some respects a peculiarly c rude one , but it does have
very general sense. The term thus appro priated can refer to th e Ihe vi rtue of bringing into the open assumptio ns that are often
idea. stated in the quo tatio n 'human beings make history', that mo re surreptitio usly made. The fac t that human beings must
human social life is formed and reformed in p raxis - in the survive in the material environments in which they live tells us
practical ac tivities carried o ut in the enactmen t of everyday life. nothing about whether wha t they do in order to survive plays a
This is exactly th e kind of view I have tried to argue for in setting do minant role in social transformation.
o ut [he basic te nets o f structuration theory. But 'histori ca l I do not think it possible to repair the sho rtcomings of either
materialism' is mo re commo nly used, especially among those who evolutionary theory in general o r historical materialism in
designate themselves Marxists, in a much more definite sense and particular. u That is why 1 spea k o f deconstructing them. We
one which certainly has a great deal of textual support in Marx. cannot replace them, in other wo rds, with a th eo ry of a similar
This is 'histori cal mat erialism' based on the scheme of soc ietal fo rm. In explainin g socia l change no sin gle and sovereign
development which Marx and Engels sketch out in the first few m\!chanism can be specified; th ere are no keys that will unlock
pages o f The GermalJ I deo l op,y and in th e Comnlllflis( Malli/esfo 'h e myst eries of human socia l development, reducing them to a
and which Marx stat es succ in ctly and brilliantly in the 'Preface' to unilary formula, o r that wi ll account for th e major transitions
A CO llfriblllio/l fo fh e Criliqlle of Polifical Ecollumy. hclwee n socielall ypes in suc h a way eith er.
244 Change, Evolution and Power Analysing Socia l Change 245

o f s urplus produc tio n o n the part o f spatially pro ximate village

communities in areas o f high pote ntial ferti lity may be o ne type
Ana lysing Social Change
o f pattern leading to the eme rgence o f a slate combining those
The foregoing considerations do not mean (hat we cannot communi ties under a single o rde r o f administration . But it is only
generalize about social change and do not imply that we sho uld o ne among others. In many cases the co--ordinatio n of military
relinquish all general concepts in terms of which change might be power used coercively to establish a rudimentary state apparatus
analysed. Five concepts are particularly relevant in this respect. I is the most important fa cto r. Agrarian states always exist along
have mentioned three - structural principles, time-space edges tim e-space edges in un easy relations of symbiosis and conflict
and intersocietal systems - in the previous chapter. To these I with , and partial domination over , surro unding tribal societies, as
want to add the notions of episodic characterization (or, more well , of co urse, with other states wh ich may struggle for hegemony
brieny. episodes) and world lime.14 over a give n area. To insist that social change be studied in 'world
time' is to emphasize the influence o f varying fo rms o f inter-
Struc tural princ i p le~ Ana lysis of modes o f institutional ar ticulatio n societal system upon episodic transitions. If all social life is
contingent , all soc ia l change is conjunctural. That is to say, it
Episodic cha ra cte riza tions Delineation of modes of institutio nal change d epends upon conjun ctio ns of c ircumstances and events that may
of comparahle form
diffe r in nature according to variations of context. where co ntext
Inlcrsoc ieta l systems Spccifi cnlion of re lations between soc ie tll (as always) involves the refl exive monitorin g by th e agents
tota li ties
involved of the conditions in which they 'make histo ry'.
Time-Space edges Indication of connections belween .;.ocieties of We can categorize modes of social change in terms o f the
differing struc tu ral type dimensions represent ed belo w. these being combin ed in the
World time Examina tion of coniunc lu re~ in the light of assessme nt o f the nature o f specific forms of episode. In analysing
ref lexively mo nitored ' history' the origins of an episode, o r series of episodes studied in a
comparative fashion , vario us sorts of consideration are o rdinarily
All social life is episodic , and I intend the notion o f episode, re levant. In the modern world the expansion in the time-space
like most of the concepts of structuration theory, to app ly to th e distanciatio n of social systems, the intertwining of different modes
whole range of social activity. T o characterize an aspect o f social o f regionalization involved in processes of uneven development,
life as an episode is to regard it as a number of acts or events th e prominence of con tradict ions as structural features of
having a specifiable beginning and end , thus involving a particular societi es,J5 the prevalen ce of hi storicity as a mo bilizing force of
sequence. In speaking of large-scale episodes I mean identifiable social o rganization and transmutation - all th ese fa ctors and
seq uences of change affec ting the main institutio ns within a mo re supply a backdrop to assessing the particular o rigins of an
societal totality , or involving transitio ns between types of socie tal ~ pi sod e.
totality. Let us take as an exampl e the emerge nce of agrarian
states. To treat the formation o f a state as an episode means o rigin

analytically cutting into 'histo ry', that is, identifying certain

elements as marking the opening of a seq uence of change and
tracing through that sequence as a process of institutional
transmutation. State formati on has to be studied in the context of
I mo ""'''tum I
the involvement of a pre-existing society in broader intersocietal
relatio ns (without, of course. neglecting endogenous forms of
change), ex amined in (h e cont ex t o f the structural principles
implicated in the relevant socielallolalilies. Thus the accumu latio n
246 Change, Evolution and Power Analysing Social Change 247

In referring to the type of social change involved in an episode definite modes of training, recruitment and status attributesY
I mean to indicate both how intensive and how extensive it is - Such a formulation, or one very much like it, has been adopted
that is to say, how profoundly a series of changes disrupts or by many prominent contributors to the field, notably in the case
reshapes an existing alignment of institutions and how wide- of classic discussion of Fortes and Evans-Pritchard ..}!! What is the
ranging such changes are. One idea that is relevant here, which I obverse, the type of social situation from which states develop?
have outlined in some detail in other sources,J6 is that there may The answer might be thought to be self-evident - societies which
be 'critical thresholds' of change characteristic of transitions do not possess state institutions as defined above. But matters are
between overall societal types. A set of relatively rapid changes not, in fact, so obvious , or they are so only if we unthinkingly
may generate a long-term momentum of development, that apply an endogenous model. For it is probably not usually the
development being possible only if certain key institutional case that state institutions develop within an already constituted
transfonnations are accomplished initially. 'Momentum' refers to 'society' that remains more or less unchanged. On the contrary,
the rapidity with which change occurs in relation to specific the development of states very often fuses previously unarticulated
forms of episodic characterization, while 'trajectory' concerns the social entities and may at the same time break up others that have
direction of change, as mentioned earlier. existed hitherto.
Let us look briefly at the problem of the emergence of agrarian We have to bear this point in mind when distinguishing states
states in order to illustrate the concepts just introduced. How far from chiefdoms. It may be that the latter are usually the
can the development of such states be regarded as a single type of antecedents of the fonner (and remain when states are destroyed
episode? Even such an apparently innocuous question turns out or collapse), but the one rarely derives simply from the 'expansion'
to be much harder to answer than is suggested by the relative or 'internal differentiation' of the other. The distinction between
simplicity of most theories which have been put forward about states and chiefdoms is by no means as easy to make as is
such states - for example, that they have their origins in warfare, seemingly often presumed in the anthropological literature. The
in irrigation schemes, in the rapid accumulation of surplus normal basis of the distinction has to do with centralization. In
production and so on. To make an episodic characterization, as I contrast to states, chiefdoms have a range of equivalent office
have mentioned, means making a number of conceptual decisions: holders, under the chief; these offices entail more or less the
about what social form is the 'starting point' of a presumed same power and status. There is no doubt that this distinction
sequence of change, about what the typical trajectory of does help to order the relevant empirical materials. None the less,
development is and about where the 'end point' is said to be. the dividing line can be variously placed. Consider, for example,
First of all, we might register that the term 'state' is an the case of Tahiti. 39 Here there were three endogenous descent
ambiguous one. It can refer either to the overall fonn of a 'state- groups, stratified in some degree by status and political respon-
based society' or to governmental institutions of a definite type sibility. Chiefs, presided over by a paramount chief, were drawn
within such a society. To simplify the issue, 1 shall take 'state' for from the upper of these groups within different parts of the
this purpose to mean the second of these alternatives. The initial island. But are these groups worth calling 'states'? Claessen says
characterization problem, then, becomes one of deciding the yes,-4() but the author who has devoted most energy to studying
main contrasts being looked for in juxtaposing circumstances in ancient Tahitian society, Oliver, says no.41
which certain political institutions exist to those in which they do The difference of opinion is not so much empirical as
not. This question does seem to admit of an answer, although not co nceptual. It is important because it is symptomatic of the
an uncontroversial one. Following Nadel, we may suppose that a difficulties involved in specifying classes of social objects. My
state exists when the following conditions are found: (a) centralized view is that Claessen makes the criteria for the existence of states
organs of government, associated with (b) claims to legitimate too lax . Of course, it is apparent enough that criteria of
territorial control and (c) a distinct dominant elite or class, having classification and the predication of definite mechanisms of
248 Change, Evu/ution and Powe r Analysing Social Change 249

institu tiona