Sociology Identity Control - Social Formations 2nd Edition - Resumo

SOCIOLOGY IDENTITY CONTROL HOW SOCIAL FORMATIONS EMERGE SECOND EDITION Harrison C White In this completely revised edition of one of the foundational texts of network sociology Harrison White refines and enlarges his groundbreaking theory of how social structure and culture emerge from the chaos and uncertainty of social life Incorporating new contributions from a group of young sociologists and many fascinating and novel case studies Identity and Control is the only major book of social theory that links social structure with the lived experience of individuals providing a rich perspective on the kinds of social formations that develop in the process Going beyond traditional sociological dichotomies such as agencystructure individualsociety or micromacro Identity and Control presents a toolbox of concepts that will be useful to a wide range of social scientists as well as those working in public policy management or associational life and beyond to any reader who is interested in understanding the dynamics of social life Harrison C White is the Giddings Professor of Sociology at Columbia University His books include Markets from Networks Socioeconomic Models of Production Princeton and Careers and Creativity Social Forces in the Arts Praise for the original edition In this book White has managed to cram a lifetime of singularly deep thinking about the social order that makes the best start yet on augmenting the economic understanding of manDAVID WARSH Boston Globe This work is unique in that it presents a fully formed structural theory of human behavior and organization from the ground up including seminal terms and the directions in which future research should proceed C A PRESSLER Choice This book deserves to be widely read and discussed White attempts nothing less than a comprehensive theoretical synthesis of social scientific ideas JOHN SCOTT British Journal of Sociology Cover illustration Pat Steir Kyoto Chrysanthemum 1982 Pat Steir all rights reserved Used by permission ISBN 9780691137155 90000 PRINCE TON pressprincetonedu CONTENTS DETAILED CONTENTS vii ACKNOWLEDGMENTS xv PROLOGUE Preview of Themes xvii ONE Identities Seek Control 1 Contributors Anna Mitschele and Frederic Godart TWO Networks and Stories 20 Contributors Haiko Lietz and Sabine Wuerkner THREE Three Disciplines 63 Contributors Rozlyn Redd and Don Steiny FOUR Styles 112 Contributors Frederic Godart and Larissa Buchholz FIVE Institutions and Rhetorics 171 Contributors Victor Corona and Matthias Thiemann SIX Regimes of Control 220 Contributors Matthias Thiemann and Millie Su SEVEN Getting Action 279 Contributors Larissa Buchholz and Haiko Lietz EIGHT Overview and Contexts 334 Contributors Frederic Godart and Victor Corona REFERENCES 377 INDEX 419 COPYRIGHT 2008 BY PRINCETON UNIVERSITY PRESS PUBLISHED BY PRINCETON UNIVERSITY PRESS 41 WILLIAM STREET PRINCETON NEW JERSEY 08540 IN THE UNITED KINGDOM PRINCETON UNIVERSITY PRESS 6 OXFORD STREET WOODSTOCK OXFORDSHIRE OX20 1TW ALL RIGHTS RESERVED LIBRARY OF CONGRESS CATALOGINGINPUBLICATION DATA WHITE HARRISON C IDENTITY AND CONTROL HOW SOCIAL FORMATIONS EMERGE HARRISON C WHITE2ND ED P CM INCLUDES BIBLIOGRAPHICAL REFERENCES AND INDEX ISBN 9780691137148 HARDCOVER ALK PAPER ISBN 9780691137155 PBK ALK PAPER 1 SOCIAL STRUCTURE 2 SOCIAL INTERACTION 3 SOCIAL INSTITUTIONS 4 SOCIAL NETWORKS 5 SOCIAL CONTROL I TITLE HM706W55 2008 30333DC22 2007041874 BRITISH LIBRARY CATALOGINGINPUBLICATION DATA IS AVAILABLE THIS BOOK HAS BEEN COMPOSED IN PALATINO TYPEFACE PRINTED ON ACIDFREE PAPER PRESSPRINCETONEDU PRINTED IN THE UNITED STATES OF AMERICA 1 3 5 7 9 10 8 6 4 2 454 Social Spaces 145 455 Mixture and Switchings of Disciplines 146 456 Envelope from Profiles 147 46 General Selves Actors Personages Personal Consciousness 149 461 ScaleFree Personal Styles 150 462 Personage Strategy and Intimacy 151 463 Entourages 154 464 Making History 155 465 Personal Consciousness 156 47 Communities 157 471 Overlap of Communities 159 48 Emergence and Change 160 481 Berlin and Vermont 161 482 Styles Must Mate to Change 163 483 The Story of Rock n Roll 164 49 Style as Control 165 491 Hieratic Style 165 492 Committee Styles New Guises for the Hieratic 167 493 Segmentary Style 168 494 Colonialisms Old and New 168 FIVE Institutions and Rhetorics 171 Institutions guide but need not be benign They can emerge from ongoing styles and feed into regimes with rhetorics built up for and around control by tribal elders and Roman orators alike 51 Origins and Contexts 172 511 From Status into Contract 173 512 Contexts in Natural Science 174 513 Situations Stories Networks and Pronouns 176 52 Rhetorics and Realms 177 521 Luhmanns System Theoretical Approach 177 522 Effective Rhetorics from Hierarchies of Publics 179 523 Ritual as Calculus 180 524 Disputes and StoriesBoltanski and Thévenot 182 525 Packaging Explanations 183 526 Emergence of Rhetorics 183 527 Rhetorics Disciplines and Queues 185 53 Careers 185 531 Development and Stories 186 532 StoryLines for Identities in the Fourth Sense 187 533 Positions and Plots and Events 189 534 Career System 190 535 Contingency Chains 193 536 Career and Anticareer 194 537 Projecting Reality 195 538 Demerits of Merit Systems 196 54 Stratification across Realms 197 541 Blocking Action 198 55 Production Economy as Institutional System 199 551 Prior Evolution 200 552 Origins of PuttingOut Systems 201 553 Embeddings into Production Markets 202 554 Industrial Districts 203 555 Decoupling and Phenomenology 205 556 The Reverse Side 207 557 On the Fringes 208 56 Organizations 210 561 Imitation and Fad 212 57 Evolution of Rhetorics Venality versus Corruption 213 571 Smith on Triestians versus Istrians 215 58 Disjunctions in Rhetorics of Smooth Control 215 581 Padgetts Stochastic Model 217 582 Comparing Budget Stories 218 SIX Regimes of Control 220 Regimes which embed disciplines generalize their valuation orderings Kinship roles like everyday roles evoke a rhetoric whereas a kinship system calls up a regime with narrative 61 Mobilizations around Values 222 611 Narrative around Value Contrast 223 612 Control Regime around Narrative 225 613 Types of Tie 227 614 Evolution of Control Regimes 228 62 Theories of Values 229 621 Regime Decoupling and Accounts 230 622 Values and Contexts 231 623 Packaging and Parsons 233 624 Dual Hierarchy as between Church and State 234 625 Pillarization 236 63 Functional Subsystems 237 631 Luhmanns General Formulation 237 632 Luhmanns Law 239 633 Bourdieus Art 241 ACKNOWLEDGMENTS xv PROLOGUE Preview of Themes xvii Horizons xviii Levels xviii Guidance from and to Linguistics xix Contextualizing Contexts xxi What to Do and How xxii ONE Identities Seek Control 1 Identities seek footing for control amid chaos via social support thereby generating meaning 11 Identities Out of Events in Context 2 12 Playground as Illustration 4 13 Control and Structural Equivalence 6 14 Netdoms Networks and Disciplines 7 15 Overview Identities Out of Mismatch within Contexts of Control 9 16 Meanings Come in Switchings Scientific Precursors 12 17 Culture in Play and in Emergencies 13 18 Challenging Both Extremes 14 19 Control and Social Space Scientific Precursors 15 110 Where to Go 17 TWO Networks and Stories 20 Stories mark ties within emergent networks 21 Emergence and Tracings 21 211 Political Polarization via Staccato Network 21 212 Tracings of the Small World 22 213 Network Population as Process 24 22 How Ties and Stories Mesh in Networks 27 221 Stories and Ties 28 222 Mesh Situational or Inscribed 30 223 Mesh General or Specialized 30 224 Source and Variety in Stories 30 225 Repertoires for StoryTies 31 226 Other Ways to Types of Tie 33 227 Indirect Ties and Transitivity 34 23 Networks Sort Themselves into Types of Tie 36 231 Coupling and Decoupling 36 232 Dynamics of Control 37 233 MAN Triads and Other Subnetworks 39 234 Siting through Stories into Social Times 40 24 How It Matters 40 241 Rapoports Profiles 41 242 Granovetter Ties and Medium Ties 43 243 Hanging Out in Corporates 45 244 Stratification 47 245 Ties and Selves 48 246 Modern Personhood 50 25 Modeling Emergence of New Levels 51 251 Cliques and Catnets 52 252 Structural Equivalence and Complementarity 54 253 Blockmodeling 54 254 Everyday Roles and Positions from Blockmodeling 56 26 Uncertainty TradeOffs 57 261 Ambiguity versus Ambage 57 262 Diffusion 59 THREE Three Disciplines 63 Commonsense illustrations lead in to three prototypes 31 Emergence 66 311 Valuation Order and Narrative 67 312 Tie Dynamics and Disciplines 68 313 Other Perspectives 69 314 Decoupling and Contingencies Shape Uncertainty 70 32 Embedding 72 321 Embedding with Decoupling 74 322 Embedding in Operational Environment 76 323 Involution Differentiation and Dependency 77 33 Interfaces 80 331 Supervision and Identities 81 332 Production Market and Quality Order 82 333 Embedding a Profile 84 334 Other Examples and Control Profiles 85 34 Councils 86 341 Mediation through Prestige 88 342 Factions and Autocracy 89 343 Lazeqas Law Practice 91 344 Ambiguity in Council Disciplines 93 35 Arenas 95 351 Acquaintance Dance 96 352 Gibson on TurnTaking 97 353 Arena Markets and Production Markets 98 354 Fame and Chance 100 355 Arenas as Purifiers 100 356 Ambiguity versus Slack in Arena Disciplines 103 36 Households Family and Gender Bringing It All Together 104 361 Meld of All Three Disciplines 105 37 Inventory of Disciplines 105 371 Catnet as Residual of Disciplines 106 372 In My Own Experience 107 373 Tournaments and Liminality 109 FOUR Styles 112 Style evokes from stochastic life in everyday networks a sensibility of common meaning from profile of switchings 41 Sensibility 113 411 Style as Texture of Social Dynamics 115 412 Style and Conversations 117 413 Interpretive Tone around Expertise Fashion and Warfare 118 414 NineteenthCentury American Womankindin Market Sentiments and in Protecting Soldiers and Mothers 120 42 Commerce Grows as Style 123 421 British Trade around the East Indies 124 422 Mediterranean Trade TakeoffMedieval Genoa 124 43 Person Grows as Style 126 431 Etiologies of Persons 128 432 Identities and Persons 129 433 Learned Helplessness 130 434 Mischels and Burts Persons as Identities 131 435 Persons as Styles 133 44 Rationality 135 441 Contexts for Rational Choice Theory 135 442 Professionalism and Speech Registers 137 443 Rationality as Style 138 45 Social Spaces Boundaries and Profiles 141 451 Styles around Knots and Jet Streams 141 452 Triage 142 453 Perceptions and Observers 144 634 Economy as Functional Subsystem 242 64 Corporatist 245 641 Corporatism as Blockage 246 642 Work 247 643 Consensus in CityStates 247 644 The Fronde 248 65 Clientelist 251 651 Blocking Action 252 652 Semiperiphery in World System 254 653 Nesting 255 66 Professional 256 661 Ripostes 258 67 Norman Feudalism 259 671 Kinship Gangs 259 672 Shift of Rhetoric 260 68 A Common Template for Caste and Science 262 681 Caste and Kinship across Villages 262 682 Tribal Regimes in Academia 265 683 The Template 267 684 American Academic Science 268 685 Effectiveness and Efficiency Applications 270 686 Control Applications 272 69 Template Evolution for Trading Regimes 273 691 Style and Institution Reciprocally Embed 273 692 Regime Evolution toward Capitalism through Style Feedback with Institution 275 SEVEN Getting Action 279 Breaking through the crust of common sense thrown up out of identities seeking control getting control over control 71 Mobilizing 280 711 Decoupling 280 712 Getting Action 281 713 Mobilizing for Truth 284 714 Mische on Brazil Walder on China 285 715 Intimacy and the Leifer Tie 287 72 Intervention for Control 289 721 Intervention through Disciplines 289 722 Style and Control 291 73 Agency for Control 292 731 Mechanisms 293 732 Agenda for Agency 294 74 Four General Claims and Three Angles 297 741 Reaching Through 299 742 Reaching Down 302 743 Reaching Up 305 744 Glasnost versus Career System 307 745 Suicide as Envelope 308 75 General Management 310 751 Eisenhower Style 310 752 Western Businesses 311 753 Rhetorics of Organization 313 76 Regimes in Crisis 314 761 Forms of Duality 318 762 Catholic and Communist as Structuralist 319 763 Servile Elite 321 764 Temperatures of Colonialism 323 77 Annealing from Switching 325 771 Fluctuation of Pension Fund Management in Britain 327 772 BangBang Control betwixt Firm and Market 330 773 A Lemma on Change of Style 331 EIGHT Overview and Contexts 334 Putting parts and aspects together 81 Triggers from Interlocking Contexts 334 811 A Fundamental Question and Four Answers 334 812 Context 335 813 Contextualizing 337 814 Invention of Organization 338 815 Sketch of Chapters 339 816 Other Angles 341 817 Language Thresholds 342 82 Modeling around Context 344 821 Boundary as Theory 345 822 Brass Tacks 347 823 Illustrative Models by Chapter 349 824 The Third Wave in Social Science Modeling 351 83 Modeling from Operational Environment 352 831 Embeddings with Three Dimensions 353 832 Spread within and across Cases 356 833 Other Measures and Levels for Models 357 834 Control Theory in Engineering Yields Style 358 84 Context Leached into Space 360 841 Localities 361 PhD candidates and fortunately he has also frequently given me ideas Two other recent applications by Yally Avrahampour 2007 and by Petronille Reme 2005 draw especially on my book on markets 2002 which is kin to Identity and Control I was helped by and am grateful for the insights and assessments offered in published reviews of the first editionAbbott 1994 Boudon 1993 Calhoun 1993 Meyer 1993 Stinchcombe 1993 And I benefited very much from the thorough readings and analyses in three doctoral theses that have been devoted to Identity and Control by Daniel Harrison Florida State University 2000 by Matthias Wachter Swiss Federal Institute of Technology Zurich 2001 and by Reza Azarian Stockholm University the latter of which has now been published Azarian 2006 Stephen Brint 1992 earlier published an assessment of my work leading up to the 1992 book Baecker 1997 offered what I see as a preview to this second edition Over the past year Corinne Kirchner of Columbia University made invaluable editorial contributions as well as substantive suggestions to this book project including the final copyediting And thanks to freelance copyeditor Joan Gieseke for her astute and meticulous copyediting work Of course I continue to owe all the debts that I recorded in the lengthy preface of 1992 which gives earlier background of ideas that also appear in the present book What I came to understand only when well along in this revision was the emergence of entirely new depth and power in network analysis and theorizing in American sociology and other social sciences Along with this came my recognition of major new theoretical depth in European sociology notably in Bourdieu but also in Luhmann who in 1992 was still little translated into English So the somewhat carping tone of 1992 complaining about the state of social science gives way to a celebratory tone PROLOGUE PREVIEW OF THEMES FROM STUDIES of sociocultural process of interest this book distills and integrates analytic themes Of interest To whom Use observer as placeholder for the great variety of perceivers personal or not who may singly or jointly figure in andor influence andor unobtrusively observe What is going on here What matters to whom For each ongoing sociocultural situation some implied searchlights from the different chapters of this book give us cues We work outward from situations rather than impose boundaries The datamining of Quentin Van Doosselaere 2006 will suggest how over two centuries a capitalist trade economy spun out in networks around medieval Genoa And closer to home on a smaller scale well watch Andrew Abbott 1999 tease out as analyst as observer and also as participant how a department and discipline emerged in decades of orientings and dealings and commitments as a robust cloud of common sensibilitythe style discussed in chapter 4around a scholarly journal nested in the University of Chicago Other studies and observation suggest that similar portrayals and themes can also apply for much smaller scopes in sociocultural time and space Altogether chapters 16 offer six distinct viewpoints or takes or humors or framings on sociocultural process Metaphorically these are takes on us as schools of fish in a vast river with tributaries and shoals and yet also some great depths The principal question for this book is How My colleague Charles Tilly recently published an enticing book simply entitled Why It seems to me that Why is becoming the easy question for social analysis An analyst can drown in thousands of answers sought and unsought since all studies are geared trained socialized to say why to give reasons These can just cancel out leaving the play with How which is to insist on setting context Now Tilly builds on earlier foundational labors not least by Lazarsfeld and colleagues at midcentury Columbia to tease out then probe further how folk approach causality And four years earlier Tilly gives equal billing to how and why in a book on stories to which I return in chapter 2 This is a principle of selfsimilarity as brilliantly laid out by that same Abbott in a chapter on fractal analysis in Chaos of Disciplines 2001 Geographical factors count in social process of course along with equipment and weather and myriad other factors including skills and knowhow Their impacts as settings appear only indirectly as refracted by the dynamics and topology of social process viewed anew in each chapter This book gives them little direct attention in order to instead develop deeper accounting of social process in its own terms Horizons You may already know something of social network analysis chapter 2 here a major advance in sociology and anthropology over the past half century and this is indeed bedrock for my spinning out social space through this book Identities which are the nodes trigger out of struggles for control as they seek footing with each other chapter 1 and so coevolve along with networks in one and another tangible domain of activity What is seen in searchlight focus depends on context embracing for example some degree of reflections from other networks along with their at least partly distinct identities Our metaphoric river consists in stochastic flows of events Ties and identities alike are bathed in uncertainty among crosscurrents from situations on up through births and deaths Switchings thus are endemic across combinations of network and domain Situations may be imbricated across multiple networkdomains of identities Signalings lead to utterances and thence stories that cluster for each networkdomain as a set able to account for happenings within those ties Participants too probe cohesion and connectivity among ties and may come to perceive boundaries Subsequently identities and ties may string and profile under some circumstances see chapter 4 into what you and I think of as persons Or the process of interest may go on to bloom into a style or instead mature as the institutions dissected in chapter 5 Levels Actions may instead of lolling around among network ties directly build up a disciplinary unit chapter 3for example a production team Such a team may be so robust that it figures as an identity itself on a new level nonpersonal enabling a whole new level in continuing 2 The late Niklas Luhmann powerfully theorized 1995 how identity dances with identity in a relational tie without however paying much attention to the network that emerges see his chapter 10 and also Fuchs 2001a b c process Paradoxes abound here since the new level requires and presupposes at once an embedding into and decoupling from context And a discipline may adjoin another example or another type of discipline And still further levels conjugate A higherlevel network can grow for example among nodes that are disciplines A style is itself recognizable as a new level an identity with a new sort of internal constitution The publics induced and presupposed in constructing identities and networks in chapters 1 and 2 can also be seen as a zero level The possibilities are myriad and dizzying as indeed they must be for an accounting of our vast river The idea of context its spread of reference is now seen to have a vertical depth aspect But each one of the six chapter views can be put to service over and over even with the same case so there is no precedence ordering among the six For example chapter 5 will argue that a production market discipline introduced in chapter 3 actually presupposes and must induce a style on its downstream side And of course the process in focus along with attendant context can evolve and switch such that a different view becomes appropriate The horizontal aspect of context affords a different sort of proliferation such as networks abutting a given network I pause to note that there are alternative constructions of these metaphors level and thus horizontalvertical such as laid out in Breiger 2000 Which view should capture the focus surely comes from what matters to those within the process with its entailments of glimmers from contexthorizontal and vertical Judgment is required In the final chapter 8 mathematical models are cited that can be of some assistance with one view or another in pinning down and testing the implications of framing in given context Embedding and decoupling for levels are confusing and difficult to trace Enlarging the scope as I do in the next section will offer guidance It should become evident that identity and control figure into each of the levels and processes as the core concepts It will also become clear that the concept of authority as stabilizer of control can emerge only on a higher level Guidance from and to Linguistics It is hard to doubt that language is a social construction Halliday and Hasan 1976 and yet at this banquet until now language has been the ghost All humans use speech and more important they always share some particular languages What I assert and argue in these chapters thus can also be seen as specifying language usage I foresee revealing correspondences between linguistic and sociological parsings of the 10 great social river Mine is a general frame of sociology and the match should be to linguistics Nothing in my argument so far has implied English language Mutual comprehension is the litmus test but from earliest times languages remain messy spreads across overlapping and interacting tribes And a first correspondence with linguistics for some river is between dialect or other sublanguage and social networks eg Milroy 1980 Biber and Finegan 1994 survey a hundred studies by linguists of specialized registers of speech ranging from radio sports announcers to Somali journalism Phonetics and intonations and rhythm are harnessed along with their survey of lexicon Thus linguistics surely helps to illuminate and specify the horizontal aspect of context The real insights however come from another universal facet of language namely syntax or grammar All human languages are comparable as to complexity and flexibility as well as applicability Depth along with embeddings is the subtlest and most complex aspect of context The previous horizontal correspondences work variously across one or more of my six framings Now there need not be analogues in a language to these six views of sociocultural process although I do make some suggestions in the following chapters regarding English My main guidance is that grammar essentially any grammar is an array of compact mechanisms for conveying fundamental meaningmechanisms that are robust across both the horizontal and the vertical aspects of context That makes sense for language as social construction Concrete meanings supplied by lexicon evidently must be at least partly geared to specific situation Functional theory of grammar Halliday 1994 shows that deeper structuring aspects of meaning come in invariant packaging starting with the fact that grammatical words are a separate breed that come in a relatively few sets each of a small number of items The truism is that each ongoing process of grammaticalization leaches meaning from that particular lexical itemfor instance the AngloSaxon word that over time turns into an English pronoun A sounder view Hopper and Traugott 1993 Levinson 1983 reveals the grammaticalization process as pragmatic adaptation Pronouns join other pointers deictic terms such as here now in supporting the easy transposition of messages from one situation to another These aspects of grammar also enable the switchings that are fundamental to building and maintaining identities and social process So they supplement registers in pragmatics of horizontal context The sentence or equivalent unit is however the core of making meaning and thus of grammar The six views or framings chapters 16 could not have evolved nor could they survive except for packing together by means of each sentence the essential distinct functions of a message This is the province of word order among classes of lexeme and of tense mood and provenance of verb They enable the six framings The very sketchy presentation of linguistics here will be expanded later especially in chapter 5 and in a projected companion monograph But this presentation supports a final main point developed further in chapter 6 with reference to the theory of Luhmann Each of the six views depends on the others so that embeddings are constitutive not optional In particular social process even thousands of years ago could develop only in coconstitution with fullfledged language As we will see only within framing six could such dominance transpire Contextualizing Contexts Sociocultural context is active not passive it gets negotiated rather than uncovered or invoked This book construes context of a process seen from one view as drawing from instances of contexts found in various other views chapters besides itself I develop this further in the conclusion chapter 8 but already recognition and use of contexts within all kinds of ongoing scenes is at the core of chapter 7 which traces how constraints in chapters 16 get shaken by knowing agents Making the Majors 1998 is a brilliant specification by Eric Leifer of how a whole new independent realm professional sports gets built It is an instantiation both of chapter 7 and of chapter 6 within a larger canvas of one hundred years of formations up and down the levels of chapters 15 Historian Lawrence Stone 1972 makes a signal contribution to social science with his Causes of the English Revolution 15291642 He has read understood cited and made use of sociological and political science analytics And Stone puts his hand on a core problematic the complex interplay across varied periods of crossing projects and perceptions and mobilizations that yield a major disjunction Yet at the end of his acute survey Stone throws up his hands and just dumps the causes into three baskets background conditions long term precipitating incidents short term and mediumterm organizing in between I hope this book provides a more helpful a more discriminating framing for a turbulent dynamics perceived in hindsight as disjunction 11 Three clocktimes cannot effectively discriminate across embedding levels and the six views So much for prelude which may be as useful and more intelligible for you as coda along with chapter 8 My hope is that you the readers will start trying out the approach in this book on your own observations whether direct and daily or drawn from extended study and I also offer quick sketches from my own experience I borrow much in ideas Altogether a hundred or so studies often qualitative and historical will be introduced The originality is in how to parse one and another by drawing on the six distinct lenses of chapters 16 plus the view in chapter 7 of disruptions What to Do and How We need to figure out how to bring this text to bear comparably on a large enough population of studies to suggest as well as test for regularities whether interpretive or demographic or cumulative Payoffs from this text can include enhancement of observation by anyone whatever the background But enhancing and adding to the few systematic claims offered here will depend on its use by and usefulness to other analysts Most of my earlier books and articles some reported on later turn out in hindsight to be pretty consistent with the views Ive reached now but I think that explicit use of this text would have improved as well as speeded them up I suspect that enrichment by combinatorial analysis Crapo and Rota 1970 Cameron 1994 will be crucial even given much further development of simulation analyses Some sort of Wiki sites or chat rooms may be helpful in these matters IDENTITY AND CONTROL ONE IDENTITIES SEEK CONTROL I DENTITIES spring up out of efforts at control in turbulent context But our everyday sense of reality then guides us Being common sense it enables communication among us and thus makes our lives work This book argues that common sense also obscures the social processes that lie behind us and our everyday perceptions An identity emerges for each of us only out of efforts at control amid contingencies and contentions in interaction These control efforts need not have anything to do with domination over other identities Before anything else control is about finding footings among other identities Such footing is a position that entails a stance which brings orientation in relation to other identities Biophysical context of course also im pacts footings most obviously as lines of visibility The control efforts by one identity are social realities for other identi ties So this identity can be perceived by others as having an unprob lematic continuity in social footing even though it is adding through its contentions with others to the contingencies they face Thus social contexts assert normality that is at odds with the impro visations and stumblings in direct experience Perceived normality is a gloss on the reality of turbulent efforts at control by identities as they seek footings Smooth social stories intrude into common sense News broadcasts imply that everyday life is not newsworthy Researchers should put on different eyeglasses that unfold the com plexities of the everyday We often work outward from observation of some tangible pattern and can disregard notions of an overarching society At all scales normality and happenstance are opposite sides of the same coin of social action Sociology has to account for chaos and normality together and this book works toward suitably flexible framings Identity achieves social footing as both a source and a destination of communications to which identities attribute meaning1 Conse quently without footing identities would jump around in a social space without meaning and thus without communication Gaining control presupposes a stable standpoint for orientation Identity be 1 Theorist Luhmann 1995 chapter 2 lays out a subtle yet precise argument for mean ing emerging in coconstitution of communication among identities 2 C H A P T E R O N E comes a point of reference from which information can be processed evaluated Footings thus must be reflexive they supply an angle of perceptions along with orientation and assessments that guide inter action with other identities to yield control So all these processes among identities in their footings can be understood only as an inex tricable intermixture of social with cultural spreads out of which meanings are constructed jointly 11 Identities Out of Events in Context A firm a community a crowd oneself on the tennis court encounters of strangers on a sidewalkeach may be identities Identity here is not restricted to our everyday notion of person of self which takes for granted consciousness and integration and presupposes personality2 Instead I generalize identity to any source of action any entity to which observers can attribute meaning not explicable from biophysical regularities Those regularities are subsidiary to social context as envi ronment and persons will appear as bundles of identities I claim that all scopes and scales of social process induce themselves in some such fashion as the following Identities trigger out of events that is to say out of switches in surroundingsseeking control over uncertainty and thus over fellow identities Identities build and articu late ties to other identities in networkdomains netdoms for short However netdoms themselves remain subject to interruption from fur ther switching with attendant netdoms Thus the world comes from identities attempting control within their relations to other identities In their search for control identities switch from netdom to netdom and each switching is at once a decoupling from somewhere and an embedding into somewhere An Internet forum as illustration can flesh out this claim There you can create an account in order to participate and use it Its not the mere subscription but the postings that create your identity in a forum while linking you by stories to others and their comments You dont exist in the forum as a whole person but as a user contributing to the specific topic of the forumeg football or sociology Since you can have ac counts in many forums you can switch between them by logging out of say the football forum so as to log on to the sociology forum We can see the forums as netdoms The important point is that although you log out your identity in that forum your account remains so 2 The work of psychologist Mischel 1990 supports this turn away from common sense see chapter 4 I D E N T I T I E S S E E K C O N T R O L 3 your postings are not deleted by the logout process In this sense your activity has left a social trace consisting of the ties to other identities in the forum But the interaction has just switched from one netdom to another The only moment in which you are less than a bundle of identities is in sleep Each mornings awakening puts together a you that had been deconstructed within social and physical protections around sleep3 You reconstruct out of various identities triggered earlier in switches among topics amid ties with others The same few general sorts of identity can be found here as in social context Many other tangible examples surround you switches in and out of committee meetings mealtime switches shopping expeditions The list is endless and subsequent chapters troll through them Communi cation remains central Human social process typically orients around meanings of events and interpretations of relations among identities Speech presupposes language and I aim for these chapters to pro vide a basis for appreciating how languages themselves emerged as byproducts of the continuing spread of dances in identity and control This communication need not be explicit speechor even extension of speech by nonverbal means For example consider how students in duct a newly arrived professor at a university into the implicit stan dards of grading and cognitive framing in curriculum for their campus eg that technical but not historical sophistication is encouraged none could articulate and most are unaware of the complex of pres sures this subtle communication brings to bear It is indeed effective control but there is no intention there It does not rely on intention to get fresh action instead smoothing the new participant into the pre viously existing flow the previously existing expectations Social organization is a byproduct of the multiplication and the cu mulation of these processes in control which inversely shape how identities result from social process The connections may be quite ob scure as in reshufflings of careers resulting from patterns of switchings in jobs Also identities and their contentions come wrapped up in and with larger contexts of many sorts cf Tilly The Contentious French In terpretations emerge in patterns weaving topics among identities and ties When contending counteractions result in some dynamic equilib rium even common sense perceives context as social structure This is for example the case with kinship or social stratification Social organization has two faces blockage and allowance of fresh action The blockage can come from the intermeshing of identities de 3 There is great variety in these protective orders from tribal fireside vigil to modern dormitory see the extensive survey in Aubert and White 1959 4 C H A P T E R O N E spite some latitude some decoupling The other face cuts open the Sar gasso Sea of social obligation and context to achieve openness suffi cient for getting fresh action Each of us has experienced how hard it is to push even the smallest social organization in a given direction By what means and when does it become possible to break through rigidity in social organization to get fresh action at large scale and small How can one effect action by intention despite social context Are there any reliable guides to getting action But then again if there are would that not generate paradox This book builds toward chapter 7 where recursive conjugations of control across levels are examined to identify ways to overcome sometimes that blockage of action that is built into social organization My central claim entails that the lives of these identities are stochas tic flows over time whose primary shapers and switchers come from the others not just in local detail but also as overall patterns and dy namicsas coconstituted context It follows that blockage and getting action provide the key contrast necessary for making sense of the com plex arguments to follow 12 Playground as Illustration In each chapter with a section marked with an asterisk I will point out how the studies there can also be seen from other perspectives This playground example will be taken up again more than casually in sections 15 and 17 and in sections 221 224 432 722 and 815 As an example both of how identities are formed and of how they help to create each other consider children interacting across a playground We can tease out some complexities from just this seemingly simple context Dynamic models can be based and tested on observation of spatial patterns in free play of young children4 Likely as not the identity for a given child on this playground was triggered from contingencies during play The childs identity links to other identities in the playground through stories in that setting eg Tom is the bad guy who always breaks the toys of other kids Strings of children may be seen rushing along some following a leading child while in other sets each child is just tagging along after a friend known from neighborhood or home or school If the children 4 Joel Cohens PhD thesis cf Cohen 1971 is a notable attempt And see the observa tions of adult freely forming groups by James 1953 as modeled by Coleman 1964 cf White 1962 for critique I D E N T I T I E S S E E K C O N T R O L 5 are older one can record some continuing networks of relations of ties between pairs of children Or a cluster of children may go about together because they are sim ilar in their own andor others eyes This recognition of similarity may be implicit as when all the members are teenagers or each child is a fan of singer X or it may be explicit as when the group are Hispan ics or are fatsoes Mostly these clusters are unnamed even unrecog nized They depend on the kinds and degree of activity going on Such clusters can come to be perceived as and act as identities if they reap pear repeatedly or in a variety of other contexts Certainly what you observe at a given moment is there only because of some underlying orderliness of process This orderliness partially comes from and is reflected in talk One can listen to the standard tales being offered across the playground in accounting for what this or that cluster does Stories go along with expressing habits and habi tus But it is conflicts and inconsistencies in which a child finds itself caught up that start generating identity With children it is not repeti tive family domestic life and not playing with the same bunch but rather clashing gangs that cause and work from identities A common set of stories as we shall see in chapter 2 is what can meld such identi ties into a network This orderliness is also affected by the physical environment How slides and swings are arrayed influences how children sort themselves into groups with geometric ordering overcoming some social disor der5 And other identities of the children come from mismatches else where between two netdoms like home and school for example when a kind of food newly enjoyed with peers at school is rejected when the child goes home Or the mismatch may occur when the clothes that classmates insist upon as their badge of belonging are disdained by a parent at home who resists purchasing them Any identity comes out of the energy for which becomes the energy from bringing together many disparate bits as when the child be comes the weird dresser in the parents eyes6 Having an identity in the common sense of that term requires continually reproducing a joint construction across distinct settings This is better described as having a bundle of identities That is the dictionary notion of the per son a placeholder term embracing identities often conflicting from different settings 5 See Alexander 1964 and the actor network theory as elaborated by Bruno Latour and collaborators 6 Garfinkel 1967 emphasized this with counterexamples odd probes such as knock ing on a restroom door to greet its occupant 6 C H A P T E R O N E Even though the playground is a casual setting one can observe con flicting identities and orderliness at the same time If the playground is observed over a long period7 certain clusters of children will emerge repeatedly This is what is meant by finding footing through control struggles Choosing up sides for games will go on This may partition children into teams almost every child going to one team or another but likely there will be a straggle of leftovers Thereby identities find positions in relation to other identities Together with the stories that tie them together structure and meaning are produced Any such crowd may partition anew into teams which make claims about spe cialization in relations and tasks Or the crowd may dissolve instead into casual chasing or gossiping Neat accounts only faintly reflect the real turbulence energized by unending searches for self and control In this sense the social never stands still Identities couple and decou ple thus continuously creating social space and time On the playing field teams may come to visit for tournaments If so grownups probably come along with the visitors and this activates local adults to come out and spend time on the playground These adults favor and slight various children patronize them according to how they themselves get caught up in the tournament A much more elaborate social organization is created or rather is shown to have been there in potential and in the perceptions of some all along 13 Control and Structural Equivalence The triggering of one identity activates control searches by other iden tities with their own impetus toward control of any and all exigencies including each others Each control effort presupposes and works in terms of realities for other identities8 Endemic efforts at control are exactly outside any given identity and are fitted into relations by drawing on the outputs of undisrupted identities Observer always is in some interaction with observed On a small scale identities in a grouping may come to be seen as structurally equivalent by themselves and by still other identities This equivalence may be because of a shared attribute or because all are 7 As has been done in a series of distinguished investigations in social science eg Opie and Opie 1969 Maynard 1985 8 In Luhmanns words An important structural consequence that invariably follows from the construction of selfreferential system is abandoning the idea of unilateral con trol There may be hierarchies asymmetries or differences in influence but no part of the system can control others without itself being subject any control must be exer cised in anticipation of countercontrol 1995 p 36 I D E N T I T I E S S E E K C O N T R O L 7 tied to each other in a clique but the basis may be more indirect and abstract To gain footings means to fashion structural equivalence Control is both anticipation of and response to eruptions in environ ing process Control projects participate in how identities array in so cial structures with social order as a possible byproduct Social pro cesses and structure are thus traces from successions of control efforts In the words of Chanowitz and Langer 1980 p 120 Control is not something that we possess It is some way that we are The exercise of control is a whole situation that cannot faithfully be fully reproduced as a number of parts or measures And further control efforts become entangled in ways that need not be visualized as proj ects of individual actors The accuracy of observing the process is enhanced through deci phering which identities are structurally equivalent with respect to context overall or partial And control can be equally real when it is fugitive since it uses disorder as material from which to evoke order9 So control efforts are responses by identities to endless stochastic con tingencies to which others control efforts add Context is crucial con text is experienced rather than designed This is why power is not the right term for these processes 14 Netdoms Networks and Disciplines Control efforts take place in demarcated social spaces Netdom is a suitable descriptor dom from domain of topics and net from net work relations Identities switch from netdom to netdom finding foot ings in different networks in differing domain contexts The dualism of network and domain is essential and make no mis take it is a radical departure from common sense We wont reach the singular person until chapter 4 And an isolated single relation or tie is accorded no reality outside the special historical and social circumstances so brilliantly portrayed by Luhmann in Love and Pas sion Netdom is not a thing it is experiential process usually transi tory but with impact so awesome that participants cannot bring it into focus Luhmann in his general theory 1995 takes a parallel road of deriving social organization with use of a single term and his com munication like netdom presupposes the mixture of relation and topic plus understanding 9 If we assume with Luhmann that all events are fugitive and that they are the elements of social systems then control becomes the attempt to constrain the possible events 8 C H A P T E R O N E Figure 11 Netdom switching is not just for smallscale and informal settings but is part of business and power life as seen in Padgett and McLean 2006 Concepts on figure are from this source not from Identity and Control Solid lines are constitutive ties dotted lines relational social exchanges and ob longs formal organizations Dots are individuals We will repeatedly make use of studies of multiple networks in Re naissance Florence conducted by John Padgett and coauthors Already here without attempting any explanation yet of the case I will exhibit a diagramsee figure 11of theirs that you can interpret in terms of netdom switches The constructs I am introducing are not meant just for smallscale casual and current processes Networks are overview reports from the dynamics of overlapping of and transitivity in and across netdoms Each network is sustained through invocations by those identities of a common set of stories that explain away anomalies Networks lay out the space of social action A further concept to be introduced in more detail in chapter 3 is disciplines Disciplines are selfconstituting conveners of social action which each induce an identity on a new level In this book they are as im portant as networks Disciplines build around commitments that con strain constituent identities very different from networks with their flexible sets of stories Disciplines are concepts about processes rather than about structure in sociocultural life Depending on which disci pline is at work control struggles take place according to different rules and in different frames I D E N T I T I E S S E E K C O N T R O L 9 I introduce three different species of disciplinesInterfaces Arenas and Councilsaround their distinctive valuations and contingencies Much practical activitywhether the production of a frozen pizza or the dinner party in a country clubkeeps on getting done and is shaped in all these venues Disciplines can translate into normality and habit at some level But chaos and accident are the sources and bases for all identities and it is identities seeking control that fuels practical activity whatever the context 15 Overview Identities Out of Mismatch within Contexts of Control Now that I have suggested the main paths including networks and dis ciplines through which to specify social process let us look again at identities Identity is produced by the contingency to which it is a re sponse an intervention in the process to come at whatever level and in whatever realm Seeking control is not some option of choice it comes out of the way identities get triggered and keep rolling along as process So basically an identity comes along with its footing out of mismatch by drawing on both observation and reflexive selfobservation Such a mismatch can occur at many scopes and levels A position is identity triggered a level up from its occupants To illustrate Recog nition of the position of presidency is triggered by the mismatch be tween Jefferson and Washington or for that matter between Reagan and Franklin Delano Roosevelt One can see troubleshooter as an identity coming out of mismatch a further level up and so identity can take on life through imputations of others I hope for example that this book will be draped with an identity by readers and certainly mis matches will be at the root of that process mismatches with prior works and their identities as well as with observations and debates among commentators There is need for a population dynamics of identities quite distinct from current demography But within any one of these many sorts of realizations of identities there will be heterogeneity Within the same network for example identities will appear to differ in strength visibil ity and longevity Such discriminations and their inventories must be keyed to particular studies but I can make some general points Redis covery and reshaping continue for every identity10 An identity is as likely to target itself for a control effort as it is to target another identity 10 On disappearance of identities consult the discussion of case breakers and dead cases by Bearman Faris and Moody 1999 and Bearman Moody and Faris 2002 10 C H A P T E R O N E Now let us return to my initial sketch of just four general sorts of identities four senses of identity For simplicity I discuss these identi ties as tagged to individual human beings Their mismatches include the rushing and jarring of daily living along with the contingencies of ill health and of arguments I turn to four particular dynamics for individual identity Identity for a human begins as and from a primordial and continu ing urge to control which can be seen always in all contexts For exam ple a new child on a playground has an overriding need to find some sort of stable social footing so that the child can know how to act in an otherwise chaotic social world This is not necessarily a harsh strug gle over status and rank only occasionally does this lead to bullying on the playground Identity in this first sense is the expression in social context of the same urge for secure footing that in physical settings induces behavioral patterns of posture such as leaning forward when climbing stairs A grouping can also have identity in this first sense exhibited in its solidarity Seeking celebration for example can yield a label for a group All chapters of this book invoke instances of this first sense of identity Identity with a second more elaborate and quite distinct sense oc curs apart from networks This sense is akin to face It is identity achieved and expressed or operationalized as part of some distinct so cial grouping in which each member has face just because it is a social face one of a differentiated set of faces that together make up that grouping The differentiation may be uneven and the grouping may be loose A simple example is a group at a table in a dorm eating din ner chances are these students know each other and are accustomed to eating together often and so have come to tend to take certain stancesone as topic selector another as clown and so forth Here the grouping necessarily has identity as an entity on a distinct level It is also recognized by diverse other identities and observers through and as participating and communicating in social process Celebration of this identity builds narrative Around identity in this second sense each discipline builds its more complex and sophisti cated process The tension between identity and control can be seen as conformity versus creativity Identities figure in fury and fear as well as sweetness and light as aspect of identity seeking control and thereby becoming creative This creativity corresponds to an additional third sense of identity that builds on the first two This is identity from frictions and errors across different social settings This third sense of identity fig ures especially in the formation of network ties I D E N T I T I E S S E E K C O N T R O L 11 This third sense of identity arises from the central fact of social orga nization each human lives switching among netdoms Even as chil dren we mix with different groups while intermixing our living in dif ferent realms Moreover each of us continues in several different roles which cross between distinct realms such as family and village and job and secret society so that our actions and thence our selves crosscut these realms Even as adults we do not often try to include all these realms in any one narrative we call career All this transports to a higher level to description of position and the like But there need be nothing unusual or esoteric in this third sense of identity Return to the homely example of a child on a play ground The child may pick up a new way of wearing or tearing its clothes as being proper an aspect of the second sense of identity But then the child finds upon arriving home that peerproper is not fam ilyproper Such contradictionsall the screwups mistakes errors and social noisein life are just what bring about establishment of identity in this third sense It is a sense that each of us achieved when still a child and it is in the third and first senses that any identity ini tially comes into existence Identity in this third sense is urgent it thus both implodes and ex plodes with the greatest of energies These are for example the ener gies which generate and which call forth artworks along with narrative creativity This third sense of identity may be construed by an outside observer as critic assessing the outcomes through a dossier indicating some broad range of possibilities This third and crucial sense of identity has no application in utopias because identity in this sense arises precisely from contradictions across social disciplines impinging on the same actor from mismatches and social noise Literary utopias acknowledge the central fact of multiple roles for persons but what makes them utopian is imagining individu als to be in roles that are combined in consistently prescribed packages There is also a fourth sense of identity which is close to what is usually meant by identity in ordinary talk This fourth sense of identity corresponds to an ex post account after the fact about identity it is career seen from the outside Whereas change is enabled by identities in the third sense the fourth sense of identity is all about rationaliza tion and about failures of action And so the fourth sense combines with the third in network phenomenology Yet all four senses of identity attach to the same constructed reality as emanations from mismatch as it becomes observed Each sense weaves together layers of expression in myriad ways These are ways that can change A painting can reflect a second or a fourth and bor ing sense of identity just as some story or play can suggest the inter 12 C H A P T E R O N E esting third or first senses but the reverse occurs as well It would therefore be silly to reify the four senses of identity to set them up as separate personae or wholly distinct sorts of positions Narrative can and does weave them together the narrators business is to generate for the time being a larger sense of membership that embraces both auditors and author11 16 Meanings Come in Switchings Scientific Precursors Netdom shows habit as surface This is dualsided habit as one finds in Bourdieus 1996a b construct of habitus But now perception comes only with and from contrast as a process Gibson 1979 Thus fresh meaning emerges for humans only with switching as from one netdom to another Switching is central to this theory and will appear again and again at different scopes and levels Again this point is unorthodox de parting from common sense but as I noted earlier I hope to show you that it unties some knots and riddles in existing social science I make only a partial claim for originality of this theory in sociology since I think much the same root idea was found long ago in works by Garfinkel by Cicourel by Goffman and in linguistics by Halliday Recently it is again being championed by Vaughan 2002 by Powell 2002 and by Mische 2007 and see Mische and White 1998 My radical innovation is different I disallow the bracketing the setting aside of context when penetrating and following particular situations and episodes whether commonsensical or Garfinkelian Bracketing is in direct contradiction with how I conceptualize identity Instead I venture shortcircuiting proposals in order to bring contex tual reality cheek by jowl with particular situational encounters I do this rather than endlessly trace out particular situations I try to emu late playwright rather than narrator Psychological perspectives offer precursors too I have already cited James Gibson An early parallel is Personal Knowledge by Michael Po lanyi 1958 That book argues that all knowing is an essentially tacit integration of subsidiary clues from which we attend into focal wholes to which we attend Much the same was also said by Fritz Heider from whom sociologist Niklas Luhmann drew guidance to construe social process as communication Here I bring this insight still further outside the minds to dissect it into component social processes12 11 This also addresses the problem tackled by Bearman Faris and Moody in Blocking the Future see previous note 12 This extraction is supported by a recent study Arnoldi 2006 of stock market de rivatives Futures of various sorts long have been around and actively traded growing into the orgy of the 1990s that centered in sophisticated mathematical modeling The I D E N T I T I E S S E E K C O N T R O L 13 I will argue that linguistics provides the deepestrooted evidence in support of switchings among netdoms Its most direct evidence is the universality of deictics grammatical pointers like pronouns this and that here and now and so on Careful examinations as in Hanks 1990 1993 Lucy 1993 as well as in Halliday 1994 suggest that deictics have evolved exactly to support coherence of discourse across switches in netdoms by providing terms that everyone can and does use to maintain footings with others through changes in netdoms 17 Culture in Play and in Emergencies Speaking of meaning where is culture in all this Culture in the sense of museums and libraries is set aside for this discussion I think of liv ing culture as a process recognized in societal institutions and prac tices which are taken up in chapter 5 as byproducts but also cocon stitutors of social process at all levels The previous playground example could concern just some empty lot or field but I was in fact thinking of a school or city playground which would be subject to more or less explicit institution and practices even aside from coaches and teachers Left to play by themselves indoors young children often take on rolesmommy doctor nurse cowboy teacher Developmental psy chology attests to and elaborates this common knowledge And re cently sociolinguist Sawyer 1992 specified the discourse pragmatics that he observed over a year of observation One can conclude that from an early age kids are made aware of more complex forms and higher levels of social process over which they try to acquire some mastery Their play is the beginning of the sophistication in transposition that everyone needs just to participate as a normal adult Sophistication however is not the same as ana lytic awarenesssuch constant awareness indeed would induce stum bling instead of normality Accidents offer a different prism Unlike childrens play they are not pretend switchings In a city an accident often evokes an emergency team and ambulance which in chapter 3 will be modeled as a disci pline The injured person experiences a vivid switch to another net volumes became so huge and the markups so small that electronic trading from com puter terminals replaced Chicagostyle bidding auctions What Arnoldi found is that the lack of facetoface contact among a set of traders with all its back and forth signal ing through discourse and body language crippled their intuitions and thus their ac tions so much so that facetoface contact was introduced again through various sub sidiary auctions 14 C H A P T E R O N E dom and then likely a continuing succession of switches Whether in Paris or Milwaukee though the situation will unfold according to much the same script from culture inducing interlocking role behavior along the network lines presented in chapter 2 Culture is being naturalized here as the product of social process This is analogous to developments in information science such as cy bernetics early on and general systems theory especially as specified by Luhmann There is also an analogue here to dynamic control theory and to Kalman filters The latter are selflearning not just selfdirect ing programs 18 Challenging Both Extremes Within sociology and other social sciences there is a strong resurgence of an individualist mode of theorizing under the label rational choice theory Bueno de Mesquita and Lalman 1992 Coleman 1990 Cole man and Nowak 1986 Lindenberg 1989a Riker 1982 Such theory takes identity for granted and ignores the nesting of contexts and thereby tries to explain away control13 Some institutional economists themselves take exception to that theory Favereau 2005 and see the volume by Lazega and Favereau 2001 Rational choice theories build upon a myth of the person as some preexisting entity and focus on how choice is made and how choices interact once made But although one can usually impute ends from actions these ends often are despite protestations mere byproducts of previous history as adapted to current circumstance These theorists need not deny this empirical weakness because they can point to the sheer scope of prediction possible on those assumptions The push toward some rational choice theory is in itself sensible Indeed it is rational because it mimics the push in other sciences to ward what is called mean field theory14 This is an approximate theory of longrange order through calculation of selfconsistent fields At first sight of course rational choice theory might instead seem to ape mod els of shortrange order which concentrate on immediate environs But no the longrange order of a selfconsistent field is essential to the cal culations in a theory of rational choice This is because the goals and 13 But see Hechter 1987 for an attempt at institutional explanation And Pizzorno 1991 reviews exactly this difficulty in Hobbes 14 Also called the Mean Field Approximation or SelfConsistent Fields see de Gennes 1979 and Ziman 1979 and for an elegant and readable early account see Van Vleck 1932 It is discussed further in the conclusion chapter 8 I D E N T I T I E S S E E K C O N T R O L 15 ends in fact have to be read out of a pattern and only larger patterns will sustain such attempts Although any selfconsistent field approach attempts to take great care with local context it is at the cost of the subtle correlations that are central to actual process Structuralism15 by contrast to rational choice theory disdains events as when it explains the United States without the War between the States and that warthe Civil Warwithout Gettysburg and the Third French Republic without Louis Bonapartes Eighteenth Bru maire Structuralism thus takes control for granted and tries to explain away identity Structuralism builds from the myth of society as some preexisting entity Neither rationalist nor structuralist approaches can give proper account of social action Abandon structuralism including Talcott Parsonss attempt to de rive social order from values guiding individual persons and also abandon the view common in economic theory of social order emerg ing from preexisting individuals efforts to achieve their idiosyncratic wants and interests In my opinion neither of these two approaches to social theory themselves opposites take persons seriously As a result neither can treat historical trends and cultural impacts with proper sensitivity In contrast my theory aims not just to sidestep the struc ture and agency problem but to build on grounds of concepts that eliminate that problem It is silly to treat rational choice theory as the basic or general theory of social organization It is just as silly to carp at any particular approx imations it uses and then refer to the carping as an institutional theory All theory is simplification scientific theory simplifies so as to uncover new phenomena Rational choice theory has suggested new phenom ena and the present task is to determine contexts in which it is likely to be productive Chapter 4 develops theory to ground these ideas about personhood and rationality 19 Control and Social Space Scientific Precursors Now I go on to develop a more general claim I will draw on natural sciences for analogies to this claim Start with weather forecasting My first article as just a teenager published in the Tech Engineering News of MIT was about the initial introduction of radar to survey storm clouds I continued to follow the blossoming of meteorology and be came convinced that a fresh perspective was as crucial there as new technology Now I try to bring a fresh perspective to sociology and 15 Whether in Parsons 1937 or Wallerstein 1980 or later forms 16 C H A P T E R O N E encourage the reader to break out of some standard terminological frames in social science Social dynamics have peculiar features when compared with say chemical reactions There is no single unique and isotropic space for context The dynamics of control while they are playing out are also inducing and constructing their own spaces These accommodate possibilities of social action which depend on perceptions and inter pretations that must be communicated and are set only partly by the biophysical environment There are also similarities between social dynamics and chemical or other natural science reactions Extension and shape get read by the an alyst from observing mutual positionings In both realms positionings are pressured by jockeyings for control There is spread across a field Topologies of social spaces are complex varying over time and from one locale to another Insights about a topology suggest leverages for control For example the military drill is one model of control a model that subjects to caricature the preconditions and steps for control In a drill persons are induced to move in parallel within a little group which is both literally and metaphorically cut off from other social rela tions for a time Alternatively one can seek control from weaving a maze of uncoordinated and changing contexts around others Chapter 7 develops these themes My general claim makes moves analogous to three moves made by physical science in supplanting Aristotle and his insistent common sense The first key move was to divorce force from momentum so that unchanging momentum signifies no force The commonsense reality of frictions is set aside in order to achieve analytic power Coupled to that the second move was abstracting from particular objects to uni versals point masses and the like The analogous moves in sociology are to switching as to momentumforce and to identities as the actors The third key move was the later explicit development of Cartesian space completely parameterized space as the setting This allowed physics problems to become formulated analytically subject only to boundary conditions The analogue being developed for social process is networks a distinct new sort of friable multidimensional space with which a new and friable constitution of interpretive social time has to be interwoven Engineering disciplines also offer analogies Implicit in Cartesianiza tion was universal regularity of the time dimension also found in engi neering Engineering need not imply predictable control Perhaps clos est to social science is chemical engineering in which it has always seemed to me the highest art is just riding herd on enormously com plex fluid flow processes I D E N T I T I E S S E E K C O N T R O L 17 110 Where to Go An identity in a human being need not constitute person despite being mirrored in the body and in the consciousness in a mind Minds fall outside the scope of sociology as I work it here16 And all sorts of identi ties are bound up with what control is in social surroundings I ex pand on this in subsequent chapters around the following five theses Five Theses Identities emerge from turbulence seeking control from within social footings that can mitigate uncertainty Switchings are the vehicles of meaning for identity and control Switching reckons in change both of social relations and of domain of association Context gains in depth as identities embed into new levels The fifth thesis is dual context is constituted in and as patterns in dy namics across identities and control across levels for a situation I also expand on this in subsequent chapters around what become the following five senses of identity Five Senses of Identity The first sense is identity as the smallest unit of analysis Persons con sist of a bundle of these identities When this form of identity finds footing one could replace the word identity with position in a netdom The second sense is a connected bunch of the firstsenseidentities It exists only where firstsenseidentities found footings and are thus ob ject and subject of the attribution of meaning The third sense is the trace of different identities in different netdoms This identity is a report of for example a human being switching from netdom to netdom over time It is the pathway a person entity or place takes through social time If we could graphically sum up time as well as domain layers we would see this third sense of identity The fourth sense of identity is the interpretation of the third sense If a person looks back on the netdoms and identities he switched into and out of and embeds this pathway in meaning he produces the fourth sense of identity This is what a person perceives to be his or her selfa narratively embedded history of a journey through different netdoms If the third sense is for example the detailed account of the misfortunes of Oedipuss life story the fourth sense is the realization 16 Niklas Luhmanns system theory 1995 which I see as compatible with mine does treat consciousness but keeps it segregated his chapter 7 I discuss this further in chap ters 4 and 6 and then start chapter 8 on this issue 18 C H A P T E R O N E that he failed Its the fourth sense that leads a psychologist to label a certain mind disturbance the Oedipus complex I will argue that there is yet another a fifth sense of identity with very different scope It is a dynamic selfreproducing amalgam across pro files of switchings in the first four sorts of mismatch This fifth kind is on a distinct level that analytically is still more embracing than the level of discipline This fifth kind I will argue in chapter 4 is the form in which persons are realized My aim is theory that enables observation expert observation attentive to all scopes and levels Social organization is messy and refractory a shambles rather than a crystal cf Sorokin 1956 There is no tidy atom and no clearcut world only complex striations and long strings that reptate as in a polymer goo So my account challenges commonsense constructs of person and of society in order to search out selfsimilarity of social organization according to which much the same dynamic processes apply over and over again across different levels and scopes But any level and scope can be constrained and otherwise influ enced by and thus embed into as well as decouple from others Lan guage as both vehicle and outcome is central in this process17 From time to time I draw on linguistics for support that goes beyond coding of particular case studies and I intend to devote my next book to social construction of language The importance of identity and control and switchings as primitives of the theory is manifest and this has an important corollary Since they arise around irregularities and amid contentions they prove less responsive to averages than they are to dispersions that is to spreads across locale and degree of social connections and timing For example how long you wait in a queue depends as much on the dispersions of arrivals and of servicing times as on their means On a grander scale the volume of product an industrial firm ships out depends especially on the quality rank seen by buyers among competing producers who are eying each other it is dispersions across flows not averages that trigger levels in prices costs and profits that sustain a set of produc tion volumes in dynamic equilibrium18 To reach such results I first lay out network analyses in chapter 2 followed by construction of three disciplines in chapter 3 Then in 17 For example Hopper and Traugott 1993 argue this regarding grammaticalization and Halliday 1994 and see Dejoia and Stenton 1980 has long argued this for language more generally 18 McPherson and RangerMoore 1991 p 35 make a similar argument about sizes of organizations on the authority of Darwinian models of evolution there in Hardy Wein berg equilibrium the rate of change of fitness is equal to the genetic variance in fitness I D E N T I T I E S S E E K C O N T R O L 19 chapter 4 I jump to a conjecture on what sort of larger format survives as sensibility within stochastic social process Chapter 5 illustrates what practices and institutions emerge in some systems of implicit and explicit culture Then chapter 6 turns to what regimes of control are able to establish themselves with special cases of separate realms in law economy art science and even sports Chapter 7 is foil to all the prior chapters in laying out how to break through such formats sometimes to get fresh action It may well be the most relevant chapter for those readers seeking to cut their way through the Sargasso Sea of conformity which chapters 26 dissect This book draws on case studies a hundred or so diverse in scope and realm and period Ragin and Becker 1992 organized a major dis cussion on issues in the use of case studies for social inquiry I argued there that a case study concerns primarily either identity or explana tion or control Studies in the present manuscript conform to this clas sification chapters 1 and 4 concerning identity chapters 2 and 5 con cerning explanation and chapters 3 and 6 concerning control chapter 7 however crosscuts all three concerns I offer guidance about what lies under the hood of a social vehicle and I hope that sheer curiosity will bring in some readers with still others searching for guidance on practice and policy The concluding chapter will begin with an overview and you may wish to consult that as well as the prologue as you move along in the seven chapters The argument is intricate somewhat unconventional introduces some new terminology and draws some unfamiliar distinctions This is for the purpose of providing flexible tools in a supple framing to assist very diverse observation I hope to hear from you about what does and does not work for you One of my contributors see table of contents for their names sug gests that you be sure to read chapter 8 maybe even early because it gives such a good overview TWO NETWORKS AND STORIES IDENTITIES seek control Any identity may see control as slipping away and going to other identities Each control effort presupposes as well as shapes some context of particular relations across identities particularly in talk The netdom described in chapter 1 is the local and shortterm context for relations between pairs of identities These relations are called ties The present chapter canvases larger continuing contexts in patterns of ties called networks Network has entered common speech as a verb but only recently Social networks are traces from dynamics across netdom switchings As two identities come over time to focus control attention upon each other a stymied struggle can settle down into some story that marks a tie between them A story is a tie placed in context Stories structure switchings into accounts with a beginning middle and end Tilly 2002 so storymaking frames social time This chapter sketches how social process plays into from and around networks of pair relations These relations are characterized by stories told in and about them with meanings drawn from switchings between netdoms Since actual settings and the switchings among them are endlessly varied the description I give here must be kept abstract and general Existing technical and substantive prototypes eg Padgett and McLean 2006a Lazega 2001 and canvasings can guide the reader for particular implementations It is often convenient here to substitute person for the general term identity see chapter 4 for the problematics of this substitution A network can be traced as similar stories appear across a spread of dyads These ties are in an incisive phrase from Podoln y 2001 prisms for meaning as much as they are pipes for connectivity In this chapter we will look first at emergent networks and then survey some ways that ties fit together with stories into networks Next comes a discussion of how a network shakes out over time Examples are then provided of how all this matters before getting back again to the emergence of networks this time referring to a new level of network The chapter ends with a focus on the role of uncertainty NETWORKS AND STORIES 21 21 Emergence and Tracings Assessing connectivity is crucial Freeman 1979 but may be problematic for a network that has only crude coding of relations Coding itself commonly derives from a relation between observer and subject namely the instruction about the criterion for reporting a tie as in the Small World studies that follow But we begin with modeling that derives ties from behaviors evolving over time 211 Political Polarization via Staccato Network Triggerings of identities also invoke communication with others as an aspect of seeking footings Repeated communication between some pair can get recognized as a continuing relation when its frequency rises above chance expectancy in that context The pattern of such ties across identities can become seen as a network engraved in some sort of public space with an identity of its own Influences flowing through ties and their impacts are shaped by the network and in turn can reshape it Baldassarri and Bearman 2006 model how this may transpire when political issues that are already active in participants minds are the subjects of communications Repeated communication will include arguments and attempts at influencing the opinion of the other participant in a tie A threefold stochastic model is proposed for showing how the choice of what issue to discuss and with whom shift along with the existing divergence of opinion Baldassarri and Bearman ran large numbers of massive simulations of the resulting distribution and location of opinions in that public across the menu of issues The point is to see how an extreme partisan polarization can come about even within a public most of whom take moderate stands on most issues The crux of the studys findings is that even with the majority moderate on most issues this public can sift itself into rather segregated and homogeneously partisan blocs of opinion on one or a few hotbutton issues The simulations were run over hundreds and thousands of periods of discussion and the output provided the full network on each issue at each period noting the opinion level of each actor at each stage Not always but often an issue or two sorted themselves out as hotbutton without attribution of particular content This is a classic illustration of a way to use social network theory about how people interact to explain unintended outcomes that are paradoxical to common sense focused on whys When you read Baldas sarri and Bearmans full account note how perfectly the setting of the model corresponds with the vision I have laid out in chapter 1 And also note how ties and their stories are generated in an endogenous process without need for the analysts to call on attributes or ideology We will return again and again in chapter 5 in particular to the interpretation of public as referred to in this illustration 212 Tracings of the Small World Stanley Milgram 1967 and others Pool and Kochen 1978 over an extended period see Kochen 1989 developed and applied chainsearch techniques to assess connectivity across both huge and mediumsize populations Subjects actually used rather than just reported ties Milgram had each arbitrarily selected initiator aim to reach a named target persona stockbroker in Bostonwith whom they were not previously acquainted initiating a chain beginning with some acquaintance presumed to be more likely to know the target Each successive contact was to select and mail the instruction booklet on how to proceed on to a next contact reporting completion of that step by postcard to Milgram The basic finding on chains was that an arbitrary pair could connect in about a half dozen steps this within a population of one hundred million persons1 The basic finding for phenomenology is that in our society ordinary people made sense of and carried out an activity that to peasant societies2 might seem a bizarre task Coding a tie as acquaintance sums across some scope of specialized relations It also sums up implications perceived from some range of past incidents from ongoing processes And it folds in more intense relations with weaker ones It is not that the scope of acquaintanceshipthe actual number of persons known to someonenecessarily differs by society In all known times and contexts from primitive tribes to empires the scope of effective acquaintance persons known in the relevant minimal sense clusters around a median below one thousand The six steps of separation found by several researchers seem to suggest a surprising extent of overlap among acquaintance circles around distinct identities This is where intuition fails and explicit modeling is required But this type of modeling is very demanding even with smaller populations as was found in early modeling for epidemics Bailey 1957 1982 1 Dropouts from the searches terminated some chains falsely but the distribution of chain lengths can be corrected for the resulting biasing White 1970a 2 Eg the Tallensi Fortes 1945 NETWORKS AND STORIES 23 Those studies themselves come to the conclusion that more depth in the phenomenology of ties is required A generation later Duncan Watts and colleagues see Watts 1999 Watts Dodds and Newman 2002 established through extensive simulations of network formation how to characterize the evolution of chains abstractly They surveyed probability models of tie formation within sets of nodes numbering one thousand or more Watts guided his explorations with measurements of chains sampled from three widely different types of populations cocastings in movies considered as ties the power network of the western United States and the nerve network of a worm species Overlap was indeed important in particular the degree to which the nodes connected to a given node were also connected to each other Even a modest fraction of ties sent out to random targets was sufficient to generate shortcut chains that seem analogous to the six degrees of separation found with ordinary social worlds Milgram however was studying actual search behavior by subjects rather than measuring each of a large inventory of chains formed according to a probability model The analogy is problematic although people do search out the shortest cocasting chains in the Kevin Bacon Game Watts and his collaborators subsequently adapted Milgrams chainsearch technique to messaging on the Internet Far more chains were initiated across far more targets across the whole world again with somewhat arbitrary recruitment of the initiators One study Dodds Muhamad and Watts 2005 closed a glaring loophole in the technique by having each contact verify that he or she was indeed an acquaintance of that sender Another study Kossinets 2006 assessed chain reliability more generally Later I describe classic studies by Rapoport and others and by Granovetter that differentiate acquaintanceship ties by coding them according to their strength Recent studies are also able to explore variations according to attributes of the messagesenders on each of whom a dossier was gathered Watts Dodds and Newman 2002 propose an explicit model for how search is conducted by respondents the first explicit explanation of searchability3 Grossetti 2005 studied networks in Toulouse by interview and survey and also pointed to the importance of common attributes and memberships in forming network ties 3 Because of Milgrams arbitrary selection of the initiators and because of a very low success rate Kleinfeld 2002 concluded that the Small World phenomenon is not empirically reliable In a simulation Watts 2003 chap 5 came up with six as the median number of steps in a chain search rehabilitating Milgram And a whole new domain of data on mobile phone calls has been probed Eagle Pentland and Lazer 2007 Presumably that dataset when combined with contentanalysis programs from linguistics could supply an empirical base for the Baldassarri and Bearman modeling described earlier Milgram appreciated bizarreness in our world his Small World of acquaintanceships abuts the phenomenological world of Goffman and Simmel who attuned us to sensibilities of life in city streets constantly amid strangers on errands unknown to you and me But to understand any world and its origins requires more than modeling Small World linkages we should also be thinking in terms of for example vacancy chains among jobs and kinds of exogamyendogamy in kinship It may be vacancies not participants whose moves reflect the underlying dynamics across organizational systems White 1970b Stewman and Konda 1983 And generalized cycles of exchange may dominate local titration in ties of direct exchange as Bearman 1997 and D White and Johansen 2006 have shown for some kinship systems The Small World can be seen as an artifact using social networks more than it is an architecture in social networks It should evoke in us queries about how the overall context constructs itself amid the diversity of mode and multiplicity of level extent and incidence for which this book is trying to offer handles Further discussion to respond to such queries is called for and is provided by the discussion of contextualization that is in the final chapter 213 Network Population as Process Identities of actors and events come out of mismatches and they embed among the ties spun in seeking control These identities array in networks or when there are many balancing condensations give way to the vaguer catnets a concept that I will introduce in the last part of this chapter and to identities on new levels The resulting ensemble can be called network as population Euphemismsworld school society and so onare often used for population which is possibly the most deceptive term in the social sciences just because it seems so obvious this set of people here But a population of identities each seeking control is through these struggles coming to specify its own social space rather than boundaries being imposed arbitrarily as an observer is tempted to do Identities are embedding via some stories with respect to various other identities in a network population evolving during the course of continuing struggles for control Ongoing demographic flows births and NETWORKS AND STORIES 25 deaths of course are also refracted into these social networks as are clusterings in neighborhoods and skills When a person strikes up a pleasant chat with a stranger at a bus stop this does not constitute a network tie What counts is that each identity is and knows it is committed to some entailment to still other ties Take an opposite example Even in present society although you may not like or seek out your cousin this person remains known socially as your cousin Although you dont perceive a tie to this cousin that person is embraced by cousinhood in social reference4 The requisites are a domain context and also coordinate ties which is to say a network context An apparently simple pairtie can be seen to be a considerable social accomplishment Research often works outward from some tangible behavior patterns or topics such that it need not call for boundaries implicit or explicit Think of the qualitative work by Erving Goffman 1971 or even scan the quantitative studies in our journals Networks could be used as method metaphor and form for such research Knox Savage and Harvey 2006 In most actual projects sociologists need not trouble themselves with reifications such as society Nonetheless theorists insist that observers use some explicit framing which need not conform to framings by those being observed Niklas Luhmann 1995 developed an especially penetrating formulation in the idiom of systems theory around communication that is wary of environment A new version presented in Against Essentialism by Stephan Fuchs 2001a takes the key step of regrounding social construction in social networks A specification of tie as an overall general pair relation also called a multiplex tie sums up implications for ongoing process that are perceived from some range of past incidents Or alternatively each multiplex connection in a general network sums across some scope of specialized relations it incorporates all types of tie Can observation of discourse suffice to identify ties and their network Social networks are rooted in the reflexive nature of language in talk and as enhanced by the three gs of semiotics glance gesture and grunt5 Choices in networks reflect the representations that people have about those to whom they tie as well as assessments of sacrifice opportunity and time 4 For further background on roles see for example Nadel 1957 on roles in general and Boyd 1991 and White 1963a on kinship role networks and Berkowitz 1988 on phenomenology and Pattison 1993 on models 5 From this base can grow sophisticated realizations of solidarity from what Doreian and Fararo 1998 formulate as ideational and relational aspects also see White 2006 A relation in a dyad can be expressed without stories Subtle realtime interactions in a pairtie a dyadasprocess have many facets but these do not necessarily require verbal expression by way of a story Handholding is a nonverbal way of expressing a relation It is simultaneously very personal and yet also manifestly public seeable by anyone around There are whole classes of other nonverbal ways such as glances and grunts to express relations and thus to constitute a tie in the given population And such ties have different meaning and occur more or less commonly depending on the history of that particular network population Moreover the cast of characters should be expanded to include objects Relations of various youths to a snappy roadster are indispensable to capturing the network dynamics in the movie Saturday Night Fever So were the relations of the hero in the same movie played by John Travolta to a routine job and to the tailoring of his new suit French sociologists have developed the insight about objects in a call for recasting theory of social networks Callon 1998 Latour 1999 How can a tie capture the ambivalence and complexity of interaction What is being coded as a tie is dynamics from control attempts around a dyad Pair balancings of control efforts can become generalized as a set of stories held in common And indirect ties can gain standing in some strings of ties and stories Expectations grow up as to both content and participants in ties To be in one relation is to be enmeshed in further relations to some of those tied to you and your alters to know of further warranties and entailments thus generating new ties with other identities So a tie is as much a projection as a record The result across all identities is a network as more than a set of identities and their ties And networks and ties are also shaped by storied shadows from identities that have vanished or did not come into being That notion of potential according to the wide choice social life leaves you figures large in the accounts I will give later of styles chapter 4 and careers chapter 5 Even if the initial claim about network analysis is true just how much leverage is demonstrated A series of papers out of the Add Health study recently took on this question across ties of all sorts Bearman Moody and Stovel 2004 p 46 Moody and White 2003 table 1 Also Barry Wellman stands out as having devoted an entire career to exploring and documenting natural social worlds in network terms 22 How Ties and Stories Mesh in Networks Meanings that come from switchings fold into stories which thus come from and also become a medium for control efforts in ties Since social situations include stories nonverbal relations and instantaneous ties I conclude that social networks emerge only as ties mesh with stories Particular ties and stories get spun off as byproducts of some particular history but I can offer some general guidance for specification and analysis I associate a single overall story such as acquaintance with a general network in multiplex tie whereas stories specialized to types of tie may call up multiple networks For participants stories are the key and they may suffice to discriminate among types of tie resulting in multiple networks as for kinship relations A set of dyads may each exhibit several qualities of relation that may be discriminated and explicitly coded only by an analyst He or she can factor the set into multiple networks Yet it may be that the whole set of stories proves necessary to sustain the metabolism of a single general network such as of acquaintance Participants may induce and call on a set of excuses and disclaimers and allowances that legitimate and keep viable a network of acquaintanceship Walter Johnson 1999 chap 1 was drawn to this conclusion concerning a chilling special case relations in the slaveholding antebellum South Concerning the stories from slaves Some incidents appear so often that it seems certain they are stock figures But these stock figures have a truth of their own to tell they gesture at the way the world looked to people whose access to information and technology was limited Whether or not every one of these stories was true and we know some were collectively they tell a truth p 11 And concerning court stories by others about slaves from Louisiana docket transcriptions But this is the conclusion of a researcher an observer and netdom analysis perhaps should be closer to lived experiences foggy and fuzzy and elusive and stochastic Such may be going on with the current reifications of social networks among business organizers and marketers within the military in everyday talk and among social analysts 35 Captured in the neat script of a law clerk are conversations a century and a half old I have generally read the docket records as if they contained only lies And yet these lies describe the circumstances of a specific sale in terms of a shared account of what was likely to happen in the slave market A few stock stories supported much of the testimony p 12 Nothing is simple and clearcut in process across social situations but we can lay some bases for possible guidelines for analysts The Internet forum illustration in chapter 1 opens out into different types of ties For example strength of tie derives from the number of direct responses to another users postings andor from the intimacy of the content of communication There is asymmetry if one user never responds to your comments And remember that incidences of types of tie are not some extraneous analytic matter They are part of the armaments of manipulation for control With general networks one looked for effects according to the absolute or relative efforts and resources devoted in ties rather than to their specialized domain Generally multiplex ties also play into selfhood as we will see later After canvasing stories first consider two pairs of contrasting framings for the mesh of ties and stories note variants within each and then use the framings to crosstabulate example networks Next we turn from meshes to sources and varieties of their stories and ties The talking that underlies storyties requires constant use of pronouns and other deictics which are prominent in every language no doubt because of their utility 221 Stories and Ties Each tie that persists encapsulates struggles for control Each tie is a metastable equilibrium of contending control attempts and as such it induces chronic reports Ties portray connections but these need not be onceandforall interconnections among fixed identities Ties always reflect but also are implicated in activity as seen by observers as well as participants As the reports accumulate invoked also in other ties they fall into patterns that tend to be accommodated as stories A whole set of stories can go with or come from a type of tie A convention Lazega and Favereau 2002 is such a set of stories Conventions emerge over time with networks of ties as their context This process goes on right under our eyes again and again The playground will have its neighborhood argot The occasion and arena are there for a primitive language to emerge as a vehicle for contending accountings 36 Rules of thumb which often appear in packages Simon 1945 are one form of conventions for a network Rules of thumb are widely transposable across concrete social contexts and across frames of interpretation Rules of thumb applied here affect the application of rules of thumb there or their application here at other times They are transmitted and vouched for along strings of interconnection in a network A language makes them available in idioms and formulae Rules of thumb can supply the story set for a network Regular life is shot full of contradictions They are less obtrusive to adults than they are to children The contradictions may even become invisible Everyday life has trained us and supplies us with nice packages of stories At any given time we have learned to apply just some one of the set and suppress memories of the switchbacks and changes that at other times we use and embroider to get along Much of social science has been an auxiliary to this provision of packages of stories sufficient to account for most anything we findbut only by suitable ex post selection of one rather than another story This explains how it is that stories have become so universal how they communicate so effectively across diverse hearers and audiences including social science No one has made this sections case as well as Charles Tilly I quote at length from his recent masterful reweaving of a generations worth of sustained analysis Effective explanations require the peculiar combination of skepticism about the stories told with close attention to how stories work Most of social life consists of interpersonal transactions whose consequences the participants can neither foresee nor control Yet after the fact participants in complex social transactions seal them with stories Identities are social arrangements reinforced by socially constructed and continuously renegotiated stories we can contextualize stories which means placing crucial stories in their nonstory contexts and seeing what social work they do Tilly 2002 pp xxiv And from further on Consider the place of standard stories in social construction For reasons that lie deep in childhood learning cultural immersion or perhaps even the structure of human brains people usually recount analyze judge remember and reorganize social experiences as standard stories in which a small number of selfmotivated entities interact within a constricted contiguous time and space Some social and economic theorists are working to adapt preferences and goal maximization to the realities of perception accommodated by stories One rubric in this awakening is framing effects eg Kreps 1988 chap 14 Lindenberg 1989a b Even if the individuals involved harbor other ideas the embedding of stories in social networks seriously constrains interactions hence collective actions of which people in those networks are capable They recast events after the fact in standard story form Tilly 2002 pp 8 9 222 Mesh Situational or Inscribed Political polarization from an emergent network as in the Baldassarri and Bearman 2006 modeling example that I gave at this chapters beginning is a pure example of situational mesh Ties are observed to congeal out of a soup of discussions to emerge from crosscutting situations By contrast kinship relations exhibit a mesh of significations that are inscribed that can be transposed in setting or time There are of course ranges of inbetween meshes For example Small World traces chain together situations but according to a criterion inscribed by the searcher 223 Mesh General or Specialized This contrast can also be expressed as inclusive versus delimiting Small World traces here too are in between since they mesh ties with all sorts of content but always under the rubric of recognition But taking multiplex versus uniplex to be the relevant contrast instead obscures the importance of strength of tie measured as a continuous cline see the Rapoport example that I give later 224 Source and Variety in Stories A relationship gets interpreted in stories both by its members and by onlookers Amy Shuman has traced this process in depth with a group of city schoolgirls She records how stories over many months emerge through switchings back and forth between oral account and diary entries Considered more generally how does this process come about Identities perceive and invoke the likelihood of impacts from other identities which are seen to do the same These relations get coded from raw reports into various shorthands of discourse and deportment Then sets of signals communications on topics get transposed from one situation to another Eventually these sets can settle down into stories These stories are fresh in any particular application but they are also familiar from before and elsewhere so that relational ties can indeed be recognized by stories Indispensable to all of this are resources in language first of course discursive Silverstein 1998 but also grammatical The sentence is a marvelous mechanism for packing three strands of meaning into brief utterances that interact to sustain talk Halliday 1994 p 34 A sentence can carry a story which can also draw on lexical distribution and on other cohesive resources of text Halliday and Hasan 1976 Stories include everything from the simplest line heard on the playgroundErnie loves Sue true truethrough artful excuses and basic daily accounts and on through recondite nuggets of professional gossip Stories are invoked without hesitation endlessly But a story in itself does not suppose or require identities and relations Gossip can be about the collapse of a skating rink roof under the weight of snow or a shout that the surf is mounting a call to the beach A story is at root an authority a transfer of identity which explains its binding to network This holds as much for respondents answering a survey as for civil servants issuing reports therefore social science must attend to this truth Anything about which you tell a story can get reflected in a relation Everyday time spent with stories building and hearing them in gossip or whatever suggests that they are crucial in social process And imbibing a formal story or film is so similar to imbibing real life that their authors and directors also like gossipers in ordinary life must have found effective shorthands for expressing identities and control in social relationships Stories can and do conceal projects of control Failures too require accompanying stories Even setting aside chicanery concealment would still remain in social space Every identity continually seeks control to maintain itself and in that struggle breaks as well as establishes relations with other such identities Both the tensions and their overcoming induce stories and may require sets of stories to characterize relations within a network 225 Repertoires for StoryTies Differentiation of ties is not a passive detached affair Types of storytie evolve as a byproduct from endless trading off among different This idea is due to Pizzorno 1991 Some sort of social network may be uncovered for other social species besides man for wolves and monkeys at least One finds pecking orders and ties and certainly control struggles there eg Wilson 1979 WynneEdwards 1985 These involve communication but at a simple level that need not rise above the pheromone level of an ant society Wilson 1970 This suggests that meaning and stories are what set human social action apart Without stories social action would have a monotone quality there would not be all the colors that humans observe and use in social settings those two networks yet some independence in sequences of action continues across situations Accuracy of perception within a complex social formationperhaps measurable by the density of sociologists or of mothersinlawincreases the need for couplersdecouplers This is because wide and widespread misperceptions including ignorance are principal conditions for several coupler mechanisms to work effectively For example many regularities of size distribution such as will be discussed in chapter 4 are largely the cumulative results of random processes yet their results are commonly interpreted as coherent signals of ability and leadership Coupling and decoupling do not deal with levels with embeddings into new levels of identity which I will take up later in this chapter and in chapter 3 Coupling can be coded as the provision of channels between parts and aspects of networks as with chaining of ties as well as multiple networks Decoupling can be seen then as the primary process the exchange through these channels of different types of uncertainty Inhomogeneity of networks remains a challenge to measurement and perception despite attempts to sidestep it Wellman 1981 Howell 1969 yet it may be a necessary aspect of rhetoric Rhetoric must accommodate to inhomogeneity not only of networks but also of the rest of the context of that institutional system as we will see in chapter 5 So pervasive decoupling must underlie any rhetoric which therefore calls for a calculus of tradeoffs among ambage ambiguity and contingency as will be introduced at this chapters end and in chapter 3 232 Dynamics of Control Stories serve to describe the ties in networks These are ties of contention as well as of cooperation and of complementarity There will be many distinct perceptions many stories about particular ties and interconnections of ties Stories serve to soothe identities irreducible searches for control which can be captured in stasis as stories representing ties New and additional control can be achieved by some actors when a network of multiplex ties becomes factored into distinct subnetworks of types of tie One illustration is the trenchant analysis by Padgett and Ansell 1993 of the rise to supreme power of the Medici faction within the Florentine polity of the 1400s They detail how particular stories control efforts across identities Multiplex networks initially report how the various identities have spread their presence in the course of these struggles As struggles for control continue the ties themselves which report chronic states of struggle are subject to splitting into distinct types of tie This factoring process can be illuminated in the abstract by an extensive literature based on observation of experimental discussion groups Much of it is distilled by Bales 1970 through modeling Before sketching two simple ways to characterize a tie or a type of tie abstractly I turn to some of what we know empirically about repertoires Start with a focus on just those types of relations that are intense enough to persist indefinitely once established for a given ego There seem to be around sixteen as a modal average In early societies these relations were reckoned in a kinship frame In those societies above all you have to deal with your inlaws for one or more spouses at a time The inlaws relate differently with your immediate kin as well as the grandchildren in ways you wish to monitor The relative age of your siblings will circumscribe some of these choices A definitive study of huntergatherer demographics Howell 1979 198812 finds about sixteen relatives recognized by a particular ego Every kind of relation practical and emotional is construed in kinship terms in this mode of human social life13 Thus sixteen is a good bet for the upper limit of distinct relations sustainable by human beings In current society but at much smaller scope Sampson14 in his meticulous and finegrained study of a monastery with its entering novitiate differentiates eight types of tie but he imposes them as a grammar of affections which do not in fact all produce distinct configurations This collapse to very few distinct types confirms extensive experience with sociometric testing on small populations compare Bjerstedt 1956 on classrooms and Newcomb 1961 on fraternities Regarding large populations see Burt 1987 Fischer 1982 and D White and Johansen 2006 In our own society key ties are dispersed over peer kin work neighborhood and so on Thus many distinct sets of stories can earn recognition for different types of tie But you still can set up in the emergence of identity as is illustrated in the next chapter Again the grammatical features of language are crucial resources Types of tie then can be explained and labeled and their number can be estimated in terms of specializations Specialization describes how the ecological is patched into the social for example in work Udy 1970 where it has a technical or engineering cast You are living in multiple networks each limning a distinct realm Some network of allpurpose multiplex ties of low intensity is all that may be perceived by the actors involved as well as uncovered by most research studies on a population But upon further scrutiny the ties may be seen to devolve into special networks each a network of ties with a focus from a particular set of stereotyped stories An identity can be very differently perceived in these different networks In the playground domain with the peer type of tie for example a child may be perceived as on the offensive a identity that may be compensating for its defensive role at home where a different set of stories characterizes the kinship type of tie The analytic task is to sort out types of tie in a particular concrete population using methods that are transposable Discriminating among types of tie is a hallmark of expertness in the sociocultural milieu but there are formal techniques to help an observer sort out the discriminations cf Burt 1987 1990 Techniques are needed that go beyond both tribal kinship lore and early sociometric studies described later we will return to this in section 25 the Modeling section 227 Indirect Ties and Transitivity Another universal basis for separate types of tie becoming recognized is the institutionalization of indirect ties that is ties compounded from adjoining ties Participants and observers alike persuade themselves it is cogent to single out some such tie for a network of its own A relationship grown beyond acquaintanceship implies sufficient familiarity with the other party to know to whom there is a further tie For example a friends friend is a reality gives orientation to action yet the tie to a friends friend need not be considered as a friend relation itself Thus a crude distance can be measured in steps away given some calibration of intensity of relation required to code the presence of a tie Some sense of distance and cumulation of social space is a byproduct of and motivation for network thinking Burt 1990 Kinship is the premier instance eg grandparents cousins uncles in our Western parsing or mothers brothers daughter fathers sisters daughter or elder siblings descendants or mothers mothers brothers daughters daughter and the like in various other kinship systems Schneider 1968 In Riggss Thailand 1966 two clientship ties if both to the same party generate by observable processes on the ground a nexus of behavior between those indirectly connected that is sufficiently distinct to be recognized as such Thus is encouraged recognition in that society of other sets of parallel ties as a separable type Riggs 1964 It seems from the wide gamut of case studies available that the indirect tie will tend to be more homogeneous in intensity and in concrete attributes than the direct ties that occasion its phenomenological construction One example from the Norman feudal regime treated in chapter 6 is how magnates conscious efforts to enhance recognition of indirect ties of fealty tended to generate a much more uniform quality of relation than held across the direct ties which however still seem more uniform than in the preceding AngloSaxon regime And what makes the ambiance of Thai society so distinctive is not the patronage tie in itself but rather the universality of recognition of indirect ties carried to the bounds of total population it is the limiting case of network population as universe A base type of asymmetric tie may change so as to no longer generate indirect relations This was true in the Norman development of the mere household knight see chapter 6 Indirect ties generated from symmetric ties are more likely to gain separate recognition yet while indirecttie formation can generate recognition of distinctiveness for new types of tie at the same time the relevant scope of population tends to be enlarged which in itself would tend to weaken recognition of distinctness When indirect ties get lumped in with the base type of tie that network will fill in of course And ties of that type will then exhibit more transitivity Such a partial tendency to transitive filling in is surely common It has been noted in Small World studies although those studies emphasize how sheer connectivity accommodates with local clustering This point triggers attention to another way to arrive at a single network For this purpose I introduce the notion of a public as an overarching realm in which only an overall bland set of stories is invoked in relations Think of Goffmans strangers on the stage of our city streets and subways Or think of Peter Bearmans 2005 Manhattan doormen talking with apartment house visitors The relations are muted and thereby transient and so widespread as to wash out distinctiveness in network incidence and in impact on other networks that those identities invoke Triads are of course a special case Many problems of network analysis and not of course only social networks require locating and factoring out subnetworks in a range of sizes 234 Siting through Stories into Social Times Social structures are often made to seem the antipodes to or at least unrelated to details and nuances of sequencing in timing This is in part because of the influence of structuralism eg LéviStrauss 1969 Social times should instead be accounted as much part of structure as are network spaces In the words of network theorist Granovetter It is also important to avoid what might be called temporal reductionism treating relations and structures of relations as if they had no history that shapes the present situation Structures of relations also result from processes over time and without such an account analysts slip into cultural or functionalist explanations both of which usually make their appearance when historical dynamics have been neglected In Breiger 1990 chap 2 p 8 Social process creates and defines distances for time just as it does for network Social times are woven together with meanings through switchings so times go with stories as well as with relations Stories cite behavior Behavior guides stories But to quote the dictionary definition Behavior is action on specific occasions involving essentially external and sometimes superficial relationships Story goes beyond behavior to weave interpretation into and around relationships as they then interweave over time into network forms Chapter 7 will look at problems of timing in simple exchange reciprocity as discussed in Leifer 199018 The systems of tenses in language are of course key They enable continuing updating of stories as social process continues Meanings get written on network as palimpsestthat is as tracings of what went before 24 How It Matters I will explore how all this matters through examining a few of the classic studies that gave teeth to social network analysis The first two studies do not have the vividness that comes from ethnographic study of particular scenes as I present subsequently but conversely the quantitative ones can come to more reliable measures 233 MAN Triads and Other Subnetworks There can be subnetwork population as process too A host of external impacts can break off some fragment of a network as the focus of everyone there I return to this in the last sections of this chapter Any control regime that survives and can be observed must also encompass strategic moves by participants and these must involve a switch in type of a given triad And ties centered around affect and emotions can impact one another heavily in an immediate locality Such triad analyses were pioneered by Davis Holland and Leinhardt DHL in various combinations Davis 1979 They emphasized an interpretation of triads as triangles in which there is only one type of tie which is either symmetric Mutual or asymmetric A or absent Null Even in a single triad thirteen distinct patterns are possible These authors provide extensive statistical framing in which to evaluate for particular populations hypotheses about which of the thirteen patterns occur at greater than the likelihood of chance and which occur at less than chance likelihood Across a larger population of nodes there obviously will be heavy overlaps among triads which can obscure the picture The analysts explorations show that results are heavily influenced by the totals of M A and N edges across the population hence the designation as MAN they control for this in evaluations The observed distribution of triads by type in a network can suggest ideas about control there based on this profile of the outcomes of struggle The focus of DHL is on the degree of transitivity predicted among the ties in a network Similarly a later colleague Jensen 1985 derived algorithms for laying out likely conformations to transitivity of the whole network as treelike structures partial hierarchies Theory can then seek interpretations for diverse substantive settings I expand on this in chapters 3 5 and 6 Much the same setup can be given a very different interpretation in which M now represents likingcall it the positive tie p A represents ties of disliking and N is the label for each pair without a tie One can try to test the familiar conjecture from Heider balance theory that in any triangle without null ties there will be either two ties of dislike or none This has been extended in various ways around a tendency for each closed loop of like and dislike ties across the network to have only an even number of dislike ties One can then show a tendency for the set of all nodes to split into two sets with dislike ties only between not within the two sides That version of a MAN approach can be accommodated in the blockmodel approach taken up at chapters end rion 20 is asked and answered within this school whose principal can ensure that each child is recording choices according to instructions The size of this school approximately one thousand actors turns out to be apt big enough to exhibit nontrivial connectivity and yet small enough to be manageable21 What are the significant parameters what is the social shape of this network representation this indigestible mass of claims and verifications of who knows whom As observers we do not for example care which Suzy is most popular although we may be interested in how divergent individuals are in popularity The single most interesting question to Rapoport and predecessors was how interconnected the school of youths was and thus also how fragmented It is a question with limits Social action always arises from accidents and speculations and gamings that become aspects of more farreaching and crisscrossing projects of control Later we report incidental observations of neighborhood bars and acquaintance dances But Rapoports traces are important tools of measurement despite being divorced from particular incidents Rapoport exploited the combination of two ideas First conceive connectivity as how many people can be reached from some person taken as a representative location within the population and according to the remove to the number of steps through intervening acquaintances Second distinguish ties from one another according to the perceived intensity of the relation Each idea requires many subsidiary notions to become operational The conception must be based on sampling from a statistical ensemble of possibilities Try out a number of randomly selected actors each as the center from which to trace out connections useful parameters will then derive from averaging the resulting traces Trace out a given sequence as far as it can go Realize that from some Sam at the center one may reach some given Suzy by any number of distinct chains sometimes through completely distinct sets of intermediaries According to the first idea Rapoport presents a nesting of cumulative curves Each reports the averaged percentage of the school that had been reached at the jth remove from the child making the direct choices These are the children reached indirectly and often repeatedly through any and all chains of ties through whatever intermediaries traced from the trial center who made the initial direct choice One curve traces just through the first choices another traces just for the second choices and so on So there is a separate curve for each order of closeness of acquaintanceship choice The profiles of rising connectivity constructed in this way are reported in the diagrams used by Rapoport and his associates The second idea was operationalized by having respondents list acquaintances in order of closeness or some such criterion Rapoport then pretended as if the fifth choices for example really made up a world of their own identifying a certain intermediate level of intensity Thus traces are given separately for each successive intensity level of ties The main message of the work is exactly in the neat nesting of successive intensity profiles in Rapoports graphs The monotonic rise in the number connected is definitional and the eventual height of the asymptotic proportion of the total school reached may be rather erratic as a measure even after averaging What is fascinating is that the indicator of intensity which necessarily is crude and may not be valid distinguishes whole trace profiles so neatly Quality of life among students in the school may correlate more closely with the sheer relative incidence of the different strengths The relative absence of mediumstrength ties may go with higher incidences of bullying and marginalizing And of course density of the absence of ties which here was set by study design will contribute information too as I note later in the section on diffusion Chapter 4 will further pursue this argument of quality as derived from profiles in order to operationalize styles as a construct 242 Granovetter Ties and Medium Ties Ask yourself this simple question Should the profile for best friend lie above or instead below the profile for a choice of a weaker relation Granovetter brought Rapoports work to the attention of and use by the social science community through demonstrating that this was indeed the question Moreover Granovetter showed that it was most fruitful to simplify the question into weak ties versus strong ties label this union of ties based on intensity Granovetter ties Granovetter 1973 1982 derived from Rapoports results the conclusion that ties and network were intertwined in a manner that was at first sight paradoxical Ties that were intrinsically weaker more casual yielded higher connectivity across the network weak ties are strong That is the way in which weak ties spread themselves around is such that they connect a larger fraction of a world together than do the same number of strong ties spread out in their way Strong ties ties given precedence by the issuers are weak in the broader context because they do not bind as large a fraction of a world into a corporate whole in connectivity Granovetter elaborated all the nuances implied Strong ties did fit into strong if tiny corporates so inwardly turned as only to choose each of the few intimate others again and again without attention to the larger context of persons Sum it up abstractly Closeknitness of a network is highly correlated with involuteness Granovetter 1974 then showed that access to jobs was dependent upon the implications of these Rappoport traces as no doubt could be shown for sexual access also Not only are network perceptions shown to be intertwined with concrete networks but also the accumulative impact of social structure is demonstrated Granovetter captured so much attention because his results were not obvious they could become plausible with thought but they were not accessible to uninformed intuition We may discover that strong ties can subsist only between like actors whereas weak ties of everyday networks may be incident between any pair of identities But Granovetter ties both weak and strong are also multiplex connections between identities Multiplex ties maintain themselves through narrative stories able to account for uncertainties both physical and social But a blackandwhite portrayal as reflected in Granovetter ties cant accommodate cultural nuance and intermediate intensity Granovetter ties instead deal with connectivity and clustering as shown earlier Note that social times are also articulated by networks of ties A strong tie as seen from inside it constitutes a continuing struggle for control between two identities and this struggle defines the phenomenological presentwhich in terms of biophysical spacetime is a fuzzy set Zadeh et al 1975 rather than an instant For that very reason it can sustain itself over long periods years Granovetters weak ties are casual as seen either by others or by selves and may have only fleeting existence The focus of interest now becomes ties of intermediate strength Any tie is defined by induces and responds to stories but only as communicated in that dyad between identities Control pressures within an identity also furnish a base of comparison of the strength of ties Except when control pressures between identities either are smaller or are greater than the pressures inside them stability can be anticipated for such networks The inference that is relevant now is that Granovetter ties weak and strong are excluded All the other types of tiethose that are qualitatively differentiatedcorrespond to the whole intermediate range of social times nei 44 ther short nor long Then is when multiple networks by type of tie can be recognized Types of tie are factored as sets of stories Any tie of intermediate intensity can be diffracted into one or more types of tie where each type goes with a particular story from some set of stories a menu So qualitative distinctions among ties hold just for intermediate periods as well as for ties of intermediate strength One can confirm this by turning to how identities themselves distinguish types of tie especially in complex and differentiated contexts These are contexts that invoke unbundling overall relations into the types of tie One class of examples comes from working out connotations of the terms favoritism nepotism and venality A variety of relevant studies eg Kelsall 1955 Namier 1961 Mousnier 1971 1984 Swart 1949 establish story sets and a time frame Job placements for example commonly involve such unbundlings of preexisting ties that are neither weak nor strong22 This is developed further in chapter 6 with regard to professional regimes 243 Hanging Out in Corporates Networks contextualize identities but connectivity is not the only basis for discriminating context A clique at one extreme of network topology exhibits full interconnection of all pairs This may come from or become marked by belonging in a group of comparable actors Designate as corporate any such group in which membership is recognized But this need not imply intimacy since membership in a corporate is and is perceived as a state that is exactly to be taken for granted Membership presumes as the norm lack of questioning Character is a suitable term for what membership is to reflect As stories are seen to characterize identity formation in general so gossip in particular tends to cluster around character as a member Members of a corporate may be deemed structurally equivalent to one another on grounds of affiliations held in common Comparability is achieved and presupposed as character in a corporate group Stability is to be guaranteed through ritual as it does for clans in tribes And a clique might bind into a distinct identity at a higher level a development I will sketch in the next section Then in 45 22 But that depends on overall context For example under clientelistic spoils systems see chapter 6 no differentiation of skills may be recognized and hence there is no unbundling of tie in awarding spoils Note the institutional context is different here than in the job searches that Granovetter studied 46 the next chapter arena disciplines will come into the picture as centered around a gradient superposed on comparability The important point is that corporate in this sense is very different from corporation in its ordinary current meanings although it is true that such corporation may but need not contain various corporates Corporation can be defined as formal organization an institutional system in the sense described in chapter 5 and recognized in a rhetoric within a legal system Take as a contrasting example a body of professors in a longestablished university say the Faculty of Arts and Sciences or of the Law School at any given time no doubt each will have some formal organizational and legal definitions but these definitions will not account for nor predict to collegiality and commitment Those qualities are governed by informal relations in networks and thus by corporates Nationalism presents somewhat the same distinction on a larger and hazier scale Deutsch 1953 James Coleman 1961 reports on corporates with equal richness from a smaller and more specialized canvas the American high school In Colemans account it is initial networks among youngsters feeding in from diverse elementary schools and family clusters which are overtaken by corporates that emerge among the children in straggly fashion Colemans substantive theme is the preoccupations and machinations of identities situated in these networks to become assimilated to the right sorts of corporates These are the in crowds on a social level specializing variously around clothes and clubs and hangouts and sports and so on Of course there can be more or less bullying too correlating with differences in architectures across corporates Colemans central vision is the importance of corporateness above mere categorical identification at the basic level of phenomenology Adolescents above all are seeking a sense of belonging through emotionally grounded inclusion in the right sets Coleman argues that school sports especially in competitive leagues are the single most important venue Scholastic achievements were not broadly appealing not foci around which youngsters wished to identify with a particular school And yet abstraction is what high schools engender regarding belonging it seems from Colemans comparative study of eight schools near Chicago Younger kids have corporate identifications too but these not only are usually smaller they also are specific and concrete High school induces one to perceive and to structure ones actions toward corporates that simultaneously are real and are abstract These are precursor to larger arrays of attributes that become the building blocks of adult working life Learning about belonging is not an easy task expectations gyrate and one can expect intense emotions to be 47 generated Rituals come into being within corporates offering some privacy separate from the broader conventions Not just one sort of dynamic supports memberships in corporates Go beyond Colemans high school context Think of neighborhood bars which intermix network and corporate aspects They cannot be reduced to just groups or to just networks Neighborhood can be vague and an amorphous context for distinguishing actors and their ties So a bar can be a significant influence in shaping perceptions that there in fact exists a neighborhood and who in what pairs are in that neighborhood as a penumbra to the bar But what determines an establishments going down this road of being identified with its locality When instead is it identified as a gay bar or an ice hockey hangout or whateverand when may several of these identifications all roll into one consistent compendium A tie to another who is habituated to the bar is a major avenue of initial attendance there You may have heard some other way that your kind of person hangs out there To become a neighborhood bar is to grow a particular kind of corporateness Networks are pressured Over time persons in the paths by the bar and the area around it become very much more likely than others to get caught up in a tie or identification with the bar Just as real others who visit the bar simply by chance may not have ties and identifications proffered to them but rather find signs of lack of welcome and even exclusion A neighborhood bar helps establish some corporate membership that operates largely from networks the literal geographical locale being an amorphous field of possibility that is consistent with endless alternatives see Wellman Carrington and Hall 1988 for more extensive development The corporateness is fuzzy and in reality it is never inclusive of any complete local population Wishful speculations and gaming calculations of social advantage underlie what we find the supposed sociableness of drinking is less a truth than a stipulation conveniently shared among the speculations in the gamings 244 Stratification Intermixing of networks with corporate groups is endemic These can be Small World or staccato networks and may involve affective ties or work ties Such intermixings can be seen to engender and shape the experience of social stratification Psychiatrist Elizabeth Bott 1957 in a landmark monograph contrasted levels of wellbeing in their social lives across a sample of married couples in England This was a pioneering analysis in which Bott let the research findings guide her to the astonishing realization that social network structure dominated other more obvious attribute and 48 situational factors She identified two sorts of network context In one the husband and wife had much the same network of friends In the other each gender had a largely separate set of friends and thus were segregated in their sites of social interaction There was a strong correlation to social class stratification such that the couples with segregated networks tended to be workingclass whereas professional and middleclass couples tended to exhibit the joint network Now turn to a totally different setting and another classic study of stratification effects from networkcorporates Antals study of Renaissance Florence 1965 is a prototype of how accurately past corporate formations can be reconstructed even in medieval European cities His study was focused on artwork production and it shows how this cultural activity is very much enmeshed with the other aspects of and stratification in the round of social life Thrupp 1948 had done much the same for all London guilds in a somewhat later time Baxandall 1980 follows both prior authors in making brilliant use of art data for cities shaped around guilds He builds from detailed studies of individual sculptors in their guild settings among a system of cities Baxandall goes into detail about the cognitive bases and economic context of the particular jurisdictional joustings among separate but interdependent guilds One can derive a sense of the stratification profile As Emirbayer and Goodwin 1994 argue with their anticategorical imperative network analysis can supplant the need for conventional categorical attributions of causation 245 Ties and Selves Real selves cannot be disentangled from intimate ties in modern social contexts The modern concept of friendship is wholly aside from tit for tat from favor for help Silver 1989 esp p 274 1990 provides a convincing argument and evidence for this evolution of friendship he fixes its beginning as the Scot Enlightenment Equate Silvers tie of friendship with the multiplex tie the overall tie that we take for granted as the basis for civil and civic life This overall tie presupposes a lack of concern with the detailed balancing of obligations found in traditional formations There is no tangible concrete mechanism perceived as concerned with enforcement of obligation This multiplex tie itself comes to seem a recent modern innovation along with a great desire to create privacy So it is ironic that its intimacy comes with pulling out of a particular realm of a type of tie and shifting over into a public with a multiplex tie probably without conscious awareness 49 A tribal or feudal society does not know the luxury of a single overall tie that is not built into a concrete economy of obligation Such social formations make use of relations each part of which is lucidly factored out by audience and occasionas kinship as village whatever A multiplex tie is a late and sophisticated construct in this interpretation of evolution What the overall tie can bring about is the person By this argument which will be elaborated in chapter 4 the person is a late construct in perception Self becomes the creation as well as the creator of person with multiplex tie as the occasion for perception of a self The tie of marriage is an example The key is that the multiplex tie requires and contributes to disparate populations Bott 1957 emphasized this point She showed how to predict apparent changes in marital goals from underlying divergences in the interlocking between goals and concrete social relations Marital style was read by Bott from the immediate context of social networks which changed from couple to couple in London A joint form of marriage relation was distinguished by Bott from segregated roles of domestic wife and pub husband Only the joint form required each person to straddle different populations The world of romantic love is a precursor and correlate of the jointform marriage Romantic love is induced by the same modern contexts in which attitudes are reified Love can be an induction of identity but only as a metastable state Romantic love is so important because development of this sort of subtle ambiguous gamed bond of some duration is a main path to unique personal identity in Goffmanesque social contexts Here is where the esoteric game theory that economists misplace onto rational decision might yield empirical insight23 Unique identities as persons are difficult to build they are achieved in only some social contexts they are not pregiven analytic foci Yet the modern multiplex tie need not be fraught with fearsome potency from balanced duress in gaming Romantic love is merely representative of extreme examples not of medians in multiplexity Some extreme examples notably the feud can also be found in traditional formations 23 It could be argued that the crippling of game theory is the worst effect of rational choice theory for which see chapter 4 At its introduction by von Neumann game theory had the potential of refounding the theory of social action Unfortunately it devolved into the hands of economic theorists The results for many years were increasingly arid exercises except in the work of Schelling 1978 who eschews systematic theory or modeling and of Shubik 1984a b New developments are afoot but effective game theory has to concern the induction of identities and disciplines of social organization 246 Modern Personhood The modern vision of social milieus is exemplified by Goffmans 1963 1967 1971 and Simmels 1955 strangers These are persons who are so little reinforced by siting in specific social locations as to have only shadowy existence to be creatures mostly bracketed to be abstract actors triggered into concreteness only through encounters Yet encounters can yield ties This modern type of vision fits as well in the province of romantic love which at first sight seems so surely and purely a matter of one preexisting person intensely attuned to another Perhaps But in fact the two persons are meeting in the Small World This Small World has been emerging since around the time a phenomenology of romantic love was spreading beyond troubadour circles Bloch 197724 It is just when persons are shadowy in their social sitings that intensely personal attractions are generated as Swidler 1986 and Leifer 1982 have argued But this modern vision also reminds us that an identity may have just one or two facets or personae out of its corporate self exposed to observation at any given time These facets are cobbled together by the identity from different frames perhaps spun together in stories and yet decoupled as well Romances are outcomes of mutual searches among networks and groups that are keyed to occasions to specific times marked by assemblages of persons as at acquaintance dances Attention can be given in social as well as physical perception only to limited numbers of contacts at a time Romantic acquaintanceship can be pursued only one date at a time Romantic personages are a construct out of a byproduct of distinctive eddies and enclaves that emerge as by chance in a Small World in which eligibility itself perceived active participation is induced by the states of queues in what can be visualized as a system of stochastic social servers Romantic love is a colorful topic The same truths can be argued however about more everyday stuff of our life in Small World contexts Attitudes are perhaps the most distinctive invention for modern contexts of social life Substantial portions of time and cognitive attention are given to attitudes as the adaptations and replenishment of hosts of notions on topics remarkably far removed from any tangible aspect of the actual daily living of the persons concerned 24 There are no person pheromones that signal one person as special to another ie there is no doing ant societies Wilson 1970 one better A queuing system of stochastic servers Kleinrock 1964 Riordan 1962 is a formalmodeling way to situate romantic love Persons become unique in identities yet must be validated as stereotypes So today to some extent the person is the attitudes Each set of attitudes can be seen as distinguishing an identity25 Each such set can then find it comfortable to recognize and define as persons other such combinations variously alike or compatible or neither Chapter 4 develops these matters at length Here just note that modern democratic politics and its incessant stream of news events made meaningful by episodic campaigns and elections can be seen in part as contributing to the sustenance of personhood to the creation of persons as distinctive combinations of attitudes that can be perceived as having meaningful continuity over time Modern politics is a very inexpensive way to create the voters as civil persons whatever the significance of the ostensible processes of decision and governance Modern politics can be seen as a phenomenological byproduct of the Small World Milgrams 1977 use of persons whose name is not previously known to be the targets of search do not seem bizarre to us as civil beings in part because we have come to attend to and know politicians as names remote from our direct experience Politics is just an example Sports is another or rather each is a cascading modern family of examples Persons are defined and define themselves most easily and inexpensively within the Small World that also yields romantic love by what particular families of attitudes they penetrate into as much as by particular attitudes upon topics apparently portentous enough that every person should indulge in a holding Trivial Pursuits is a parlor game emblematic of an age 25 Modeling Emergence of New Levels Now look further into how to discriminate distinct networks with differentiated ties in order to deepen insight into how sociocultural formations build But corporates also deserve further examination and we will also reprise and generalize the phenomenon of hanging out in groups which do supplant networks in many evolutionary accounts of social organization Transitivity can be worked out in terms of a single network the union of underlying differentiated networks being recognized in the overall type referred to as multiplex tie But the last example from the playground showed how a multiplex tie can be factored into particular types of tie that differ as particular flavors of the relation The Rapo 25 Combinatoric calculations show the number of possible combinations or sets to be astronomically larger than any human population see Cameron 1994 port model developed an intermediate approach in which each of the many degrees of friendship being reported by students was treated quite artificially as a separate network for multiplex tie and we didnt examine the mutual impact of these networks just the differences in involution 251 Cliques and Catnets A new way to differentiate type of tie brings us back to the notion of corporates This way discriminates a type of tie on the basis of the pattern of overlaps observed among identities and corporate groups Especially in tribal contexts eg Hart and Pilling 1960 Rose 1960 age and gender corporates spring into relief as intersections such as old men in councils But Breiger 1974 proposed a general duality between person networks and corporate ones for example the earlier networks from cocasting invoked by Watts in his Small World study A generalization to groupascommonattribute is in the literature eg Feld 1981 and is being further developed and applied by the Watts team eg Kossinets and Watts 2006 and see McPherson SmithLovin and Cook 2001 And absolutist France provides a rich tapestry of enacted and selfconscious corporate forms based on underlying identities and offers an array of overlaps Mousnier 1984 magisterially surveys the inclusions and exclusions among those corporate forms that served to underline and define the importance of various attributes This patterning provided the basis for that regime as we shall see in chapter 6 The anthropologist Smith has attempted 1975 to raise an entire general theory of social action just on corporates and their interlocks Classic field studies of communities in our own day eg Hollingshead 1949 Gans 1962 portray overlaps and exclusions and nestings that are less tangible than corporates I suggest the term catnet as a general designation for such less tangible outcomes generating meanings and interpenetrations of lives and ties The following paragraphs flesh out this concept and the basis for the term Start with a limiting form to catnet As the density of ties among a subset of persons reaches some threshold value the subset may come to regard itself as having an identity Most of the pairs in the subset may not be actively connected at any given time by network relation but because of the perception of an identity all relations will be regarded as present in a latent way In other words any member in a clique will feel free to mobilize a particular individual relation with another member in the clique A clique can continue to grow by the process of foldingin eg Coleman 1957 1961 Burt 1982 If one or more clique members have a friend in common in the network outside the clique other clique members will tend to assimilate the additional person When several types of network tie occur in a population cliques can form in each Foldingin processes now work across the different types of relations and tend to yield cliques with common membership across several netdoms Similarity in attributes will generate groups out of categories Yet network ties will continue to be recognized intertwining across categories without recognition of the network as such Label this generalization and loosening of the corporate concept a categorynetwork abbreviated as catnet Given the tendencies in a catnet toward the focusing and alignment of relations it becomes easier and more common to perceive indirect relations with a wider segment of the surrounding population In the bar setting you may feel that you have a relation to a stranger because last week you saw him there laughing with a good friend The clique consisting of you your good friend and your other good friends can then grow by folding in the stranger the similarity of attributes being the bar you all visit frequently Now we can say much the same thing in more general abstract terms one reckons relations through clique memberships and the likethat is through the latent relationships rather than tracing out some of the usually long chains of concrete ties that would be necessary to reach most other persons in the population The network comes to be projected in perception into a network among nodes that are clusters or cliques with persons in a clique treated as equivalent unless there is some short actual path to a given one Each person in the catnet system thereby secures a less fragile place in a social formation that is less definite but as ineluctable as any given identity When particular concrete ties are disrupted there are clearly acknowledged sets of other persons with whom new ties of equivalent sort can be acknowledged and mobilized quickly Yet at the same time the catnet system is more decoupled from random disturbance Being surrounded by a wider range of reliable ties a person can be less attentive to and concerned about every rumor and disturbance that passes along the concrete nets Catnets are thus important controls over ecology In still larger scopes around it a catnet tends to become perceived in a broader view as an entity that itself grows links and evolves into membership in a still larger catnet system We each have had the experience of leaving one earlylife stage and moving on to another in fixed patterns of ties no ties to the identities in a zeroblock and ties to those in a oneblock which give footings to the actors And after all until other role frames are activated there is no phenomenological basis for discriminating roles in a given frame One will not be seen as a parent until one is also a workmate or scholar or tribesman you will not be seen as a playmate until you are also a schoolboy and so on In the playground example the child may be braggart bully docile follower and the like sorted out by some familiar complementarities held in tension that may become recognized amount to an institution what Bourdieu calls a habitus to which we return in chapter 5 Separately in the home there is some menu of roles visavis parents and siblings Separately again among gangs there may be task leader social leader follower and so on Whyte 1943 A position then appears in each of several network populations of identities The position may have some tag associated with it such as for the child teachers pet or butt of ridicule or big shot A particular position may bring together a set of distinct identities from distinct networks They are brought together into a moreorless integrated whole This whole can be tied to one account of social personality Mischel 1968 as is done in chapter 4 after bringing in the dynamics of switching This vision seems compatible with the structuration perspective advanced by the sociologist Giddens 1979 Each actual implementation requires relational data from across some population such as for example Gross Mason and McEachern 1958 collected long ago in studies of school superintendency positions and on a much larger scale recently collected by Bearman Moody and Stovel 2004 in the Add Health study of high school social life 254 Everyday Roles and Positions from Blockmodeling Types of tie confound differences between relations with differences in their framing that is they confound distinctions in structural context for with distinctions by asymmetries and strengths and qualities in the pair ties Even so types of tie have face validity and enable effective prediction of changes in both relations and their structure Theoretical interpretation leads to rediscovery of role frames as one of the most effective forms of constraint and control Blockmodels derive various possible implications for how multiple networks may imply aggregation of actors and aggregation of types of tie simultaneously into some articulated larger structure Any such structure is one of an array of such structures latent in the context dependent for activation upon impetuses of chance and control Incidences of types of tie are to repeat not some extraneous analytic matter they are part of the armaments of manipulation for control Blockmodels identify different possible stable balancings of control projects within that population separated from production pressures that stem from the biophysical ecology The outcomes are possible partitions of the original population and each partition defines a set of possible identities called blocks which are candidates to become disciplines chapter 3 if and when that partition emerges from control struggles And relations between the blocks are candidates to outline a role frame as discussed later Substantive interpretation as in the previous section invokes roles and positions One can argue that a type of tie already by itself describes a role Discourse in all the dyads for that type is being framed in one distinctive set of stories I have avoided that claim since the substance of role involves a pattern across different domains of discourse and thus implies the complementary notion of position There are still other ways to assess and visualize overlaps and orderly ecology across networks as we saw in the first section of this chapter And this is equally true in formal and business settings see Stuart 1998 Further advances along those lines will be cited in chapter 8 26 Uncertainty TradeOffs Modeling tends to suppress the noisiness of reality At this point in the discussion confusion is to be treated as a lawful aspect of process an aspect to be assessed I refer to confusion from and across switchings through publics around networks Refined distinctions are needed as tools for this analysis We can assess phenomenological confusion in terms of tradeoffs Make these tradeoffs be between specifically social slack and specifically cultural ambiguity 261 Ambiguity versus Ambage Slack has too many extraneous overtones of intentional carelessness and the like to be effective as an analytic term a tool of discrimination Instead contrast ambiguity with the term ambage Ambage designates slack in the sense of uncertainty in a purely social context Ambiguity designates uncertainty in a purely cultural context From its origins the word ambage signifies winding or indirect and roundabout ways27 27 In previous centuries ambage also carried meanings of concealment and deceit but as the Oxford English Dictionary 1933 vol 1 makes clear that usage has become archaic The established meaning is circuitous indirect or roundabout ways or proceedings delaying which our previous highly refined and detailed perception of network among the young peers we leave is quickly distilled into a lumped representation of that whole stage as an entity a catnet Such is one genesis of more complex social organization studied in later chapters 252 Structural Equivalence and Complementarity Structural equivalence is a more general concept than membership or network It includes as a special case but may be contrasted with the cohesion of corporate interconnection cf Burt 1992b It concerns mutual positioning what partition into sets of identities would signal what partition of types of tie Note the duality Blocks of structurally equivalent identities are built according to tie profiles For an explicit definition consult Breiger Boorman and Arabie 1975 There may be no ties at all between structural equivalents Two lonely kids alike isolated on the fringes of a playground illustrate the pervasiveness of marginality in networks Romo 1991 analyzes this as the Omega Phenomenon Also structurally equivalent are two stars who each reach out to gather the other kids into their respective orbits but have little to do with each other Or structural equivalence can be abstracted from the particular others so that two quarterbacks are equivalent even though there is no overlap between the kids in their orbits Winship and Mandel 1984 The result is positions The central point is to look for a partition of a population such that the nodes in each set tend to relate to the rest of the sets in much the same way in the pure case they have the same incidence of the same sort of ties into each other set According to this principle just call it streq those in a set see the rest of the world the same way but need not even be aware of each other much less be tied as a clique There is a dual aspect Look at the types of tie and separately at each type of compounding of the sorts we discussed earlier around MAN triads This huge array of separate patterns found for compounding may itself be partitionable into sets so that there is streq on the ties too dual to the streq on the nodes 253 Blockmodeling Computer implementations of intricate search algorithms called blockmodels are required to get usable results from structural equivalence even with small populations Structural equivalence is entirely relativist for further discussion see D White and Reitz 1983 It requires bootstrapping That is equivalence among identities assigned to a given compound actor is determined only with respect to other compound actorsbut each of them in turn is defined only with respect to the presumed existence of the others including the initially given one And this relativism is simultaneously dual with respect to the sorts of ties distinguished from among the endless array of compounds as being distinct in implications for social action So searches have to be openended For early expositions see Burt 1978a and White Boorman and Breiger 1976 A blockmodel interprets multiple networks one from each type of tie in terms of a particular population26 Blockmodels parse distinct relational aspects among actors into feasible clusterings This is the obverse of our focus earlier on establishing some continuity in character for multiplex networks Each of these two foci has distinctive cultural accoutrements and quite different procedures are appropriate for modeling and measuring structural equivalence groupings on the one hand and on the other hand continuity Observer accounts are but sketches offering just hints and clues But this seems the promising direction for assessing how multiple networks impact social process and organization Now start again from the beginning in building up a substantive context for explicit modeling concerning structural equivalence Coding of a tie is at minimum two names in brackets to indicate juncture A name makes an identity transferable at the same time as unique so that a name is the primitive of position But the child is more than the playmate represented by a named node in the playground network of chapter 1 Children also come from several elsewheres from families from schools and likely also from neighbohood gangs Each of these network populations of identities in which a given child is found may contain few or none of its other playmates from that playground To represent the child as a whole takes a bundle of such nodesincontexts from beyond any one physical locale and any one network Maybe these bundles can be fitted together into roles but the sustained abstract analysis of role theory by Nadel 1957 suggests we should be skeptical about that even for an American playground So do traditional statistical analyses Qualitative accounts in textbooks are unpersuasive Social positions are presupposed by earlier role theory but their provenance has been a continuing puzzle Biddle 1986 One needs to look for a set of positions within network data as a bridging structure among distinct role frames A role frame has to do with more or less 26 I use the term blockmodel generically to refer to several distinct lines of development that are surveyed by Freeman White and Romney 1989 The one fulllength monograph treatment is Boyd 1991 Ambiguity can begin to be measured in spread across a set of stories We can call that set a convention used in that network as we already discussed in the earlier section on mesh section 221 These conventions are not mere matters of perception Exactly because a convention can be fit loosely to any situation it is not subject to refutation by ongoing observation Pressures for change of conventions will come as byproducts of efforts at control The conventions actually used which are not determined by the social mechanics going on can be expected to be very resistant to change Lazega and Favereau 2002 Ambage on the other hand concerns the concrete world of social ties in networks of ties and corporates among nodes Ambage can be operationalized in several ways28 A test of reliability of measuring ambage is needed One test is correct prediction of the appearance of a tie of specified sort in some action context This test applies separately for each participant and for observer Thus ambage is dual to ambiguity fuzz in the concrete embodiment as opposed to fuzz in the rules of perception and interpretation One can see there should be some sort of tradeoff between ambage and ambiguity Consider how ties bundle and unbundle into types of tie Blockmodeling treats reduction of ambage measured through zeroblocks and increase of ambiguity from adjoining larger numbers of initial identities into the oneblocks the corporate nodes of a partition Here ambage decreases and ambiguity increases for an involved observer The particular blockmodel predicted is one that has some intermediate level on both Other levels emerge The nodes of a network for example may be identities or may be corporate blocks such as derive from a structural equivalence analysis such as blockmodeling And you will see that the next chapters embrace regimes of disciplines crossed with styles that cross network populations It is difficult to transpose the measures of ambiguity and ambage proposed earlier to the other levels and the concept of publics needs reconstruing But transposition of measures across such levels is essential to establishing selfsimilar theory that can hope to deal with the scrappy mess that is social organization We will return to this in chapter 7 around discussion of agency Let us return to the discussion of coupling and decoupling in section 232 Note that going to parties is an analogue and prototype for the communion ceremony which decouples ambage in the past from ambage in the future Age grading and other seniority systems when seen as a succession of cohorts moving through the same set of positions eg generations are a most obvious form of couplerdecoupler What pattern and process of coupling and decoupling would prove robust for example among university departments each hunting for funds and prestige In complex societies the law court is the institution most specifically concerned with couplingdecoupling A judicial sentence on an individual is a coupling point which from that individuals point of view transforms ambiguity in his past role to ambage in his future role whereas a duel or a killing in a feud does the opposite A change in substantive laws operates in a similar way for social groups 262 Diffusion In each chapter with a section marked with an asterisk I point out how the studies there can also be seen from other perspectives Diffusion examples are taken up more than casually in the following sections 241 footnote 7 of 32 422 456 footnote 4 in 622 821 823 and 84 Diffusion of information significant enough to trigger consequential actions observable across a population provides a basis for assessing the evolution of types of tie in network patterns Burt 1987 1990 2000 has built upon and summed up a tradition of study and modeling for diffusion of innovation The tradition rests on a few definitive field studies In particular Coleman Katz and Menzel 1966 generated a reliable record of actual prescriptions of a new drug in a city by doctors whose network patterns they investigated the strength of the study was its anchoring in specific events Bothner 2003 extends the paradigm to diffusion for new computer technology Salganik Dodds and Watts 2006 bring the 1966 paradigm into the Small World framing of this chapters beginning with a huge population and experimental design they are able to penetrate further into the extent of social suasion on adoption preferences Morris 1993 brings diffusion via social networks to epidemiology And Strang and Tuma 1993 take on modeling effects of temporal as well as spatial heterogeneity on diffusion Innovations generate ambivalence High standing can come from adoption but so can scorn depending on what comes to be accepted as the worth of the innovation That worth in turn is no isolated technical truth but rather is negotiated by the interaction of numbers of actors who adopt and are pleased and it is also assessed with haught tier and more specialized verdicts likelier to shape cultural traces To be crude early but not too early seems an apt stance for those already of high standing such as physicians as a group and within them the betterplaced ones The original study has much information and insight on corporate aspects of these issues There are ways to crudely allow for corporate effects as Burt shows But there are no data for discriminating networks by intensity of tie29 As often in social science a good probe becomes unusable as the population being examined shifts from one that is inexperienced and low status like junior high school students to loftier adult persons of affairs But on the other hand for much the same reasons in the latter type of study population timing becomes more visible in actions and these tend to be actions that are consequential Adoptions of the new drug among this towns physicians controlling for corporate statuses and effects generate a mosaic in time that Burt 1987 shows to be interleaved with the mosaic of structural equivalence in social space Network ties are identified from a single crude measure with whom ego discusses professional medical matters among colleagues in the town Social space is mushed into an artificial Cartesian representation but that is done on the basis of the quintessential social relation of structural equivalence Structural equivalence asserts how important it is that two actors see and relate to especially on similar topics rather much the same set of other actors if similarity in the views and acts of the given two is to emerge through a myriad of instant transactions Close location in the Cartesian space conjured to the observers eye boils down an average of a great deal of structural equivalence Timing of major acts such as first prescriptions of the new drug should interlock with this array of locations in social space It does A few clusters of doctors are so completely interconnected with each other in the cluster that stringing becomes cohesion and is indistinguishable from structural equivalence And just here a particular tie of advicegiving between a particular pair within the cluster does not stand out and predict closer timing Such a cluster is so saturated with closeness that more elaborated cultural and hierarchical influences intervene in decision making Most physicians as in any actual population of size have little connection and thus not the raw material for much structural equivalence with most of the others A crucial set of physician dyads are intermediate in that they are rather close in the Cartesian space they share and are structurally equivalent with respect to many neighbors in the space yet by no means do they cohere into a cluster of nearcomplete connectivity Timing is significantly correlated just in these circumstances A reported tie of professional discussion activates similar adoption dates just when the physician pairs are acquaintances so to speak Time has different social meanings and results according to its interpenetration with locale and topology of social space Boundary is seen to be a problematic concept for social phenomena in time or space Boundaries are both matters of perception and of construction and thus subject to speculation and to gaming The physicians are demarcated only fuzzily on their own account as is the community in which they are operating In fact it seems natural to think of several subpopulations of doctors which overlap only partially and differ on locale and specialty and standing and age Further layers of embedding and of control eg of certification and of residence must be invoked and recognized to yield an edge for analysis and there is little sign of that in this particular naturalistic observation of diffusion of prescribing a new drug Networks and corporates were uncovered as explanatory pointers in fieldwork and analysis among the physicians In contrast boundaries either were impositions of convenience to demarcate the field work or pointed entirely outside the scenes accessible to this fieldwork like regulations on issuing and recording prescriptions Our axiom is that ties and identities alike are socially constructed not just imposed by observers And multiple types of tie are generated not just the overall multiplex tie Thereby identities come to be perceived as embedded in more commodious network spaces while at the same time being constrained by ecological space Note that similar insights follow from the studies of tracing for the Small World The principal conclusion is that processes even of diffusion much less of manipulations for control cannot be treated properly by a network stripped down to sheer connectivities For a more recent overview of diffusion models see Edling 1998 The claim is that social network analysis adequately captures some essential aspects of the lived social experience toward which chapter 1 was directed That claim originally rested on little evidence It was bold and provocative but two generations of research have now provided a menu of results and tools Arriving at positive results despite noise from imperfect implementation of messy constructs brings me extra conviction Results have reached beyond connectivity and clustering for general networks on to aspects of roles across specialized networks using structural equivalence and ambiguity assessments Control is the driving energy of identities spinning out social networks and coping with ecology Stories emerge from interacting control projects as these build networks Stories come from these energies to embody their spinning out they give color to human social life they shake it up Ceremony and ritual emerge from social goingson to smooth their junctions with ecological spacetime they along with stories derive from and build valuations So network analysis provides only one window on social process and structure Yes the lead insight regarding some situation in its moving context may come from this chapter 2 Parts of later chapters will point out especially close groundings in particular parts of this chapter But more often this book will instead call primarily on one of the other four framings offered in subsequent chapters It is a mistake to cram other framings in under the rubric of network analysis which is demanding and complex enough by itself Networks supply some sense of social space to us as observers but only partial insights into social processes which involve more complex interactions such as the disciplines taken up in the next chapter THREE THREE DISCIPLINES DISCIPLINES offer rules of the games that yield coordination in tasks in an otherwise messy world Joint tasks of many sorts get done and keep getting done Disciplines order ties between identities enabling joint accomplishment of tasks To persist and reproduce itself any joint accomplishment must root in and emerge from some focusing some disciplining of the ties and talk as presented in chapter 2 among the identities presented in chapter 1 In chapter 1 we saw that a public was produced jointly as a forum for the fleeting netdoms that are the phenomenological base of networks and disciplines can emerge in such a forum as well Consider how that can be Durkheim has sold us on the primal necessity of deviance of at least some sort and degree Such deviations can regroup coming to constitute an invidious ordering around a valuation order that unfolds in talk around netdom switchings The results are task groups as status systems made up from socially patterned judgments around networks The ties in these networks reflect interactions among flows of action along with the ways they are judged by the participant identities This chapter focuses on processes whereas the descriptions and models in chapter 2 were about patterns as outcomes Those patterns did correlate with participants interpretations but little attention was given to the question What is going on here Which identities are doing what with whom and on what basis By introducing the concept of discipline this chapter shows how struggles for control which in stalemate constitute ties between identities can also evoke a whole new identity on a different level over the existing set This chapter proposes and elaborates three genres of joint disciplines Varieties of public ranging say from Goffmans strangers on a street to Habermass citizens in forum here give way to three disciplines proven able to accomplish joint tasks For example if Goffman ambled on down to William Foote Whytes particular Street Corner 1943 he would have seen his anodyne public replaced as arena or as council discipline I begin with some everyday examples before developing the concepts further Then the subsequent section traces the emergence of discipline out of network form it sketches how struggles for control coalesce with and around some valuation order to yield a discipline Each discipline is a local status system with different bases of valuations and different systems of ties The succeeding section argues that each of the disciplines necessarily embeds a task into a context that embraces not only outside relations with recipients observers clients plaintiffs etc but also the physical and engineered environment Disciplines embed into their operational environment on three dimensions dependence with respect to the operational environment differentiation among identities and involution Later sections lay out each of the three disciplines separately These presentations are laced with case studies as well as theoretical elaboration Tradeoffs in uncertainties are discussed throughout Then at chapters end I begin to show how disciplines switch and hitch and stack as larger social process builds Disciplines can be seen as status systems that are made up simultaneously of evaluative judgments and of network patterns created by interaction of those judgments with task flows For initial orientation I turn to familiar university settings to provide a commonsense take on each genre along with a label For an interface discipline task flows are production which may be of material goods but could be lectures or briefs oral or written in a university or a law firm Evaluative judgments tend to be from the impacts on the flow of these products eg scientific papers out of that status system to a larger context and the usefulness seen from returns back eg citations fame flow of graduate students This is the valuation of quality In turn particular instances eg law firms as wholes could get swept up into some discipline on a higher level hierarchy of fees among Chicago law firms practicing real estate law chapter 6 portrays such further embeddings which influence as well as build on functioning of the local discipline In contrast the Greek letter fraternities and sororities or houses in American universities are organized around judgments involving inclusion and exclusion the affirmation of group boundaries and thus the purity of those inside contrasted with the danger of those outside These judgments induce an arena discipline One might be willing to work in a laboratory with a nonGreek independent is the euphemism but one would not want to dance with or marry one These houses constitute a university status system of selecting future friends and lovers The flows across university boundaries relate this system to larger systems of inclusion and exclusion eg by generational connections to adult country clubs Rotary clubs comingout parties etc TABLE 31 Genre Process Valuation Order interface commit quality council mediate prestige arena select purity Indeed members identify it as a marriage market the flows in being largely unmarried and the flows out either engaged or eligible and thus it also becomes an arena on a higher level In university councils people such as deans and research entrepreneurs compete to shape issues as well as for dominance over resources Within these council disciplines judgments center on prestige in the sense of perceived capacity to influence corporate action Judgments involve bets about others power as well as about their soundness in corporate political matters Such councils relate to the outside through regulating the flow of people and resources into the control of the local system The basic notion then is that these three different kinds of local status systems are disciplines with different bases of valuation and different systems of ties that coordinate what is getting done Networks around councils tend to reach out to control resources and people whereas networks in arenas tend to break off at boundaries to avoid introducing impurity into the group And the interface pumps flows from outside upstream to downstream outside Each genre of discipline points to its prototype process participants commit to producing flows in interface whereas in council they mediate among proposals and in arena they select from candidates Table 31 correlates these with the label and valuation order for each genre of discipline I build pictures of disciplines using constructs that are realizable for different scopes histories and levels Although some explicit and detailed models are proposed accounts of disciplines perhaps serve best as heuristic guides to observation including those folded in among participants own discourse Here flexibility enhances scope of use The following detailed discussions of each genre will for example offer variants on the prototype process and valuation order in table 31 as did the introductory three examples Neat and precise discriminations are at odds with the stochastic messiness of social life 31 Emergence Valuations are the idiom in which irregularity gets replaced by discipline Each discipline is a mechanism of social action that configures an identity but only as adapted to some context It is a discipline of process toward accomplishment from in and for that operational environment The earliest example is a work group such as one for hunting or gathering Networks themselves grow around and with work and other material production as these open out into ties and stories Disciplines also emerge in contexts more purely sociable divorced from work such as the childrens playground of chapter 1 How a discipline emerges also presages the forms in which it continues and reproduces itself and it points toward a process distinctive for the continuing discipline of that genre To begin consider some homely examples of discipline process namely meals as social processes A sitdown urban dinner party among professional couples is an example of an arena discipline It is concerned with establishing yet bounding some sort of identity of the evening with a set of stories that index professionalism So status ordering on purity is at play A church supper by contrast can be considered a council discipline ordered by prestige valuation There is unending concern with balancing and disciplining conflicts among factions At the same time the integrity of the congregation as a church is being reaffirmed A cafeteria at mealtime can be seen as an interface discipline This is the production discipline effectively delivering foods into people arrayed to receive them An interface mediates between social worlds In many places in the chain of food production and service there are interfaces interface is generalizable to multiple levels For instance the checkout counter at a fastfood restaurant doles out hamburgers and such in return for money They do this with minimum contact the interaction being mostly stereotypical The stereotypes hide the complexity of the networks on either side in the worlds of food manufacturing and services and in the worlds of the customers I return to this in chapters 5 and 6 On the one hand control is attempted over specific occasions in social context And one the other hand control induces efforts to verify or regulate by comparison with some standard if only implicit or relative or historical This is how discipline emerges out of contention And discipline extends over multiple occasions For example Ochs and Taylor 1992 show from a hundred case studies in America how a familys dinnertime typically engenders narrative that reinforces a political order what I have called a council discipline Valuation Order and Narrative Switchings among netdoms from chapter 1 underlie the general mechanism in which discipline emerges from network context Valuation order is also key One can try to infer and induce the valuation by looking at storysets active in the local discourse Standard stories for networks can settle out across situations and then transmute into narrative There are common elements across sets of stories in the networks being drawn into an emerging discipline Stories vary but some framing by valuation is taken for granted As a discipline emerges the takenforgranted valuation order becomes hegemonic It is this hegemonic quality that can afford the discipline independent standing as an entity as an actor on its own For example The Right Stuff which is both title and theme of Tom Wolfes novel 1980 exactly captures the inarticulate hegemony of a valuation as prestige among a group of test pilots Focus first on the origins of a council contentions for control can trigger each other across netdoms within the initially blend setting of a public Some of the players are sociometric stars with many ties and thus potentially wide influence toward setting an agenda As they recruit each other into joint discussion on those topics that they argue are central to the situation in that public some consensus emerges among them as to their efficacy in setting agenda and settling disputes through such discussion Such consensus can crystallize into a valuation ordering among them by prestige Such ordering of them as council prioritizes their own interventions and as this prestige order gains legitimacy in others perceptions it inhibits efforts by outsiders to gain a direct say Dispersion then is based on internal social prominence attempts at influence by outsiders are eventually directed along ties to council insiders This is a messy affair of repeated trial runs as to topics and ordering before any council discipline coalesces but thereafter feedback is biased against deviation from council guidance Deviance and variation are the original raw material from which discipline emerges It gets crafted across control struggles that trigger net dom switches Dispersions are key rather than averages These are dispersions in time switchings as well as within network standing Durkheim long ago noted We have only to notice what happens particularly in a small town when some moral scandal has just been committed They stop each other on the street they visit each other they seek to come together to talk of the event and wax indignant in common From all the similar impressions which are exchanged for all the temper that gets itself expressed there emerges a unique temperwhich is everybodys without being anybodys in particular That is the public temper Durkheim 1947 p 102 Dispersion is of course also key for emergence of an interface genre This is dispersion not on internal social prominence but rather in access through ties outside both those offering supply to and those seeking delivery of production Discussions are around perceptions of how the members can utilize these external relations and the valuation order has to conform to perceptions from the external contacts which is to say order by quality of product The repeated trial runs before successful establishment if any can be anticipated by an external observer who looks for structural equivalence streq maps The quality order disciplines commitments members make by some tradeoff between quality and volume of delivery with quality rank Arena discipline can come out of dynamics between hostilities and likings in netdoms Some are perceived as gatekeepers and come into array according to how stringently they exclude as they select for successive degrees of purity the valuation order Any number of particular arenas can emerge from trial runs out of an initial configuration of netdoms No order or membership is predetermined rather they are emergent 312 Tie Dynamics and Disciplines Now stand back for a general look at tie dynamics The evolution of patterns across types of tie reflects incidence and distribution of control struggles The results of this evolution can be modeled crudely through the differing frequencies of emerging compounds in types of tie on the population Such is the rationale of the DavisHollandLeinhardt triad census approach MAN from chapter 2 Such pattern can also be assessed by examining the incidences of various compoundings across all types of tie Boorman and White 1976 for example clustering of friends enemies ties may overlap enemy but not friend ties the balance hypothesis that was also laid out in the MAN section of chapter 2 In this assessment structural equivalence suggests possible partitions into identities Each partition specifies a split into sets each of which pressured by similar influences being compounded from the other sets might trigger perception as itself an identity in social action A blockmodel is a conjecture as to such a partition see chapter 2 again Each of these sets of nodes making up a blockmodel may become a corporate identity under some mechanism of process some genre of discipline But how is clustering into specific discipline identities to be understood There is no necessary implication that such compound actor is cohesive is bound together positively or is even aware of itself At some given time character as an identity may not have been demonstrated in participants eyes despite structural equivalence as to interrelations with other corporate actors across all the sorts of tie This means that embedding remains an open issue The disciplines cumulate control pressures such as are implied by the pattern of observed ties see Burt 1978b Faulkners 1983 account of evolutions of moviemaking teams in Hollywood is an example These teams combine different sorts of specialists and Faulkner reports sequences of memberships for project after project The basic fact is the astronomical number of possible dynamic configurations Rather than searching out which ones lead to this or that discipline I argue for identifying the disciplines that prove able to reproduce themselves The real point is not how a particular example of a discipline came into being but rather the mechanism for that species which then inhibits breaking out of the mold Padgett and McLean 2006a b have recently reported on a massive study of changing sociopoliticaleconomic organization of Renaissance Florence over a century and more They interpret it as the emergence of capitalism a somewhat similar study for medieval Genoa is taken up in detail in chapters 4 and then 6 They offer a wealth of circumstantial detail to support their interpretation enough so that one could look and test for discipline configurations Their point is also on a larger scale that the significant organization is one that reproduces itself 313 Other Perspectives First a brief review of the genres An interface gains identity via production and continuing delivery of a tangible flow of social productions For instance in a hunting group in tribal context members are identified by quality of the meat they bring back whereas in each em 56 bodiment of the arena the ordering can sift itself apart and together into new identities characterized in purity valuation The embodiment for the council is the formation of alliances and counteralliances in mobilization to retain existing valuation in prestige This focuses dynamics around valuation order Other sorts of dynamic should be considered Each type of tie from chapter 2 may be distinctively involved in the formation of some genre of discipline as a species of identity We may find some correspondence between on the one hand the three universal ways or modes of induction for tie types suggested there from case studies and on the other hand the three genres of discipline we sketch here in chapter 3 Asymmetric tie suggests the interface species overlap tie suggests the council species and indirect tie suggests the arena species It is also instructive to match this abstract partitionbased on asymmetry overlap and catenation of tieswith a substantive partition common in the literature eg Laumann and Pappi 1976business honor and professional respectively But it is a distortion of the social material to argue for the neat and tidy in either discriminations or mappingsexcept as imposed by the analyst for purposes of inferential computation Disciplines are not freefloating molecules and we now need to specify how they embed into context during struggles But before asking just what this term embedding signifies we should examine how uncertainty plays into these struggles 314 Decoupling and Contingencies Shape Uncertainty Discrimination between uncertainty in cultural contexts and uncertainty in social contexts is subtle and elusive as we saw in the previous chapter Processes in discipline emergence compound this because of the importance of contingencies from other aspects of the operational environment subsumable as biophysical geographical and technological Chapter 2 supplied an illustration each blockmodel sketches one possible partition into disciplines that results from pressures indicated in networks of ties But other contingencies biophysical ones have an impact on the outcome and are themselves affected by the changes in network among disciplines Stories emerge accordingly in sets to accommodate diverse outcomes Chance effects thus proliferate and assume interpretive guises and social forms with incidences that are correlated Decoupling is essential Decoupling is basic to networks as we saw at the end of chapter 2 Coupling is more obvious to trace in strings of ties but decoupling is equally important as it is the buffering of one chain of actions from another Chance combines with network and story to yield patterns of decoupling and embedding Decoupling is the phenomenological face of what a calculus among ambage and ambiguity and contingency can assert analytically The general source for decoupling is attempts by identities to establish comparability as they seek footings These efforts have a paradoxical character comparability is established for perceivers other identities only through the most strenuous efforts at superiority Attempts at dominance are exactly what set up the arena of contention within which comparability falls out as the unanticipated byproduct One is surrounded by examples professors vie for distinction and thereby become as peas in a pod to students in their classes physicians strive as individualsand also in much the same process as specialtiesfor prestige only to thereupon become perceived as interchangeable Burger King McDonalds Wendys and so on induce a new category of equivalence the fastfood restaurant exactly and only by striving to be better than each otherwhich requires and therefore induces as presupposition being comparable It follows that uncertainty as to contingencies in the operational environment must also enter into any equation in ambage uncertainty of social position and ambiguity uncertainty of meaning This is contingency from the physical world of work together with its social exigencies from which the social spaces have spun out initially Thus ambage and ambiguity both exist only as a followon to contingency For development of this theme see Stark 2001 Any equation must sustain three very distinct variables the equation must be in contingency as well as in ambiguity and ambage Ambage is especially associated with the connection between identities and network populations ambiguity goes with aspects of connections of production disciplines with networks and contingency in a natural environment sense is associated with how identities play off production discipline These are not assertions about single disciplines or identities These are tendencies that seem diagnostic in our social world of disorderly gels and goos as compounded by gaming 57 The subjects of these three measures for aspects of uncertainty are active processes not natural flows They are active processes among interactive identities Gaming is the current idiom for interacting manipulations Manipulations trade on the concretely contingent as well as on social maneuvers and interpretive ambiguity Manipulations often key on weather and shortages for example which provide convenient excusestories Such gamings require find ready and resupply ambage ambiguity and contingency all three of which are its raw material its medium and its product So a calculus of ambage ambiguity and contingency taken as a set is necessary Section 732 will develop these themes further Equation and calculus are being used as conventional shorthand for a much more complex interplay among variables These latter are assessments that may or may not prove to be measurable like temperatures It is difficult enough to assess uncertainty in an objective or engineering sense as Tukey 1977 tells us Shannon pioneered measurement of symbolic uncertainty here termed ambiguity in information theory but only in limited aspects for an appreciation and development see Hofstadter 1979 And the overwhelming scope and subtlety of social uncertainty of ambage is a theme seen in Knight 1921 and von Neumann and Morgenstern 1944 and Schelling 1960 From its introduction in the first chapter decoupling has been reserved for where contingency has at least some role Material situations are where time and thence stochastic process has the greatest importance Unfortunately the greatest technical gaps in social science are in models for stochastic networks for assessments see White 1973 Boorman and Levitt 1980 which are indispensable for capturing coupling and decoupling at the same time7 32 Embedding Embedding is mutual discipline that sifts out of chaotic crisscrosses of attempts at control Embeddings seed on now this happenstance and now that one but once established they can induce continuance of the discipline through repetition and imitation Embeddings take couplings to a higher level Embeddings occur at many scopes for macro 7 There are beginnings on search in networks Boorman 1975 Delaney 1989 Kleinrock 1964 on diffusion in networks Boorman 1974 Burt 1990 Bailey 1957 1982 on statistical mechanics of networks Erdos and Spencer 1974 and on parallel distributed processing models and political economy see Evans 1979 and for geographical perspective see Brenner 2004 Embeddings surround us in many varieties They overwhelm us and thereby become hard to pin down for explicit analysis Consider two examples of the embeddings in which our social lives are enmeshed The first has to do with geographical space Think back to the last snail mail envelope you addressed The first line may be an identity taken as being independently fixed no doubt a person with name and the last line similarly will have some fixed location from geography But the second line was say Apartment 231 while the third line was say 1603 Redwood Street and thus this address you wrote was exactly an embedding By itself Apartment 231 means nothing much and so too 1603 Redwood Street until each is embedded into the complete address For the second example turn to time How was your weekend What is your day like on Wednesday Again here your life is shaped around an embedding you are not even aware of now into a framing of time Here the social construction is of days as defined through being embedded into a sevenday week Only in an intensive care unit or a submarine or another of Goffmans total institutions can you partially escape The week is hegemonic in that it is utterly taken for granted And these examples remind us of the other side embedding has as complement decoupling in some other respects Recently for example I drove into a small city I was already acquainted with along what I thought was my standard approach road off the highway I circled for an hour completely disoriented because I had in fact entered along another highway connector I was embedded in one view of the street grid but that decoupled me from the view natural from the other connector Embeddings can fail and switch Decoupling is a converse to embedding Social life begins with triggerings of identities each of which comes from embedding a discipline of constituents but which then is decoupled in seeking control in its ties That is the discipline for an identity embeds its constituents while simultaneously offering them decoupling as some insulation from and brokering to the context Decoupling is also a complement to coupling as we saw in the previous chapter This can be confusing so return to the illustration given there concerning the Internet forum Logging out is decoupling Brief logins with minimal contents to a number of other accounts would count for me as couplings Whereas extended login to continue and develop ongoing discussions in that forum would count as an embedding In an influential article Mark Granovetter 1985 presents a convincing account of the gist of embeddings as being social extensions and involvements amid networks in which embedding is given overtones of emotional involvement and fuzziness Yet this does distract one from the possibility of a new level of actor emerging an embedding can be sharp and crisp and it can also disrupt as well as nest in with networks By contrast linguists are clear about the emergence of levels but neglect much of Granovetters social extension because of their focus on the dyad In the words of MAK Halliday Embedding is a mechanism whereby a clause or phrase comes to function as a constituent within the structure of a group which itself is a constituent of a clause Hence there is no direct relationship of an embedded clause and the clause within which it is embedded the relationship of an embedded clause to the outer clause is an indirect one with a group as intermediary The embedded clause functions in the structure of the group and the group functions in the structure of the clause Halliday 1994 p 242 The notion thus includes recursion since a clause may have clauses within it which in turn may have clauses in it For instance The man Sally saw was tall has the subclause in it Sally saw the man If you think of companies divisions teams and so on as simply organizations then they can nest as well because each is just an organization within an organization Likewise disciplines can become identities that in turn can be disciplined Further exploration of the nexus of embedding with decoupling is needed 321 Embedding with Decoupling Failure invokes the dramatic demise of an identity In this sense failure couples control to identity whereas forgetting and referral are less dramatic erosions of identities Failure is a basic social invention which turns boundary condition into source of action Any process of triage supposes and allows for failures as components Failure is key to boundaries but that is secondary to its generating change in embedding among levels Embeddings surround us They indeed overwhelm us and thereby become hard to discern in explicit analyses Overtones of the term embed as used in ordinary discourse incline us to miss the very induction of a new level of actor that is central Granovetter 1985 presents a convincing account of social extension and involvement as the gist of embedding yet this is a twodimensional portrayal as it were a portrayal that neglects the emergence of new levels of actors from embedding By contrast linguists are clear about the emergence of levels but neglect most of Granovetters concept of social extension because most of them focus on the dyad Embedding is a mechanism whereby a clause or phrase comes to function as a constituent within the structure of a group which is itself a constituent of a clause The relationship of an embedded clause to the outer clause is an indirect one with a group as intermediary Analogously the relation of a downstream actor in production market economy to the upstream supplier of the producer it buys from is indirect mediated by a market as will be developed further in chapter 5 The puzzle of generalized exchange which traces to Durkheim but is more specifically rooted in theories of LéviStrauss has motivated studies upon which my present line of inquiry depends Notable studies Rose 1955 Hart and Pilling 1960 that delved into aboriginal kinship as economic organization culminate in Bearman 1997 Recognizing this analogy supplies further insight into the possible emergence of distinct new level of actors Production markets from chapter 5 and later in this chapter are analogous to sections in classificatory kinship systems White 1963a but need not have analogies in every kinship terminology Let us look more closely at mechanisms Embeddings can result from each actors not only knowing entailments and warranties in his relationships but also being known to so know Thus the firm may become subject to and be known to be subject to the hegemonic pressure exerted by the others engaged in the continuing reproduction of a distinct identity as market Such a market is itself embedded within networks in ways that at the same time constrain the actions of firms Such a market is an actor on a distinct level and yet firms are still partially decoupled Embedding can build a public subculture across actors at both levels within a particular sector of markets even though commitments and decisions remain tied into catenations and compoundings of particular ties The subculture recognizes and personalizes actors of different levels while at the same time providing customs of switching from the calculus for one to the calculus for another Participants in the various markets act as if embedded in a curious web of cocoons see Tilly 1996 for an analogous term among markets that also come to constitute a selfconsistent field such as discussed in section 454 Yet the quality ordering is a property of the mechanism of the market as a whole core to its identity There is a transmutation of hitherto entailed ties into market parameters The market is in effect projecting a new more abstract sort of tie This is the asymmetric envelope that is recognized as a market profile Warranty and entailment on the level below are thereby abrogated as to addressee if not interpretive content 322 Embedding in Operational Environment Any species of discipline can be analyzed abstractly but its actual functioning requires embedding in an operational environment By its identity a discipline induces designation from outside perhaps by literal naming enhancing this awesome integral of embedding The components of a discipline and their interactions are taken for granted invisible ordinarily they become visible to participants or others only in failed disciplines A name for a discipline establishes the commonness in action perceived among the entities making it up which is one side of embedding but a name also establishes commonness in relations to the setting of the discipline and that is the other side of embedding Disciplines are perceived and characterized as embodiments of processes but they presuppose and generate a larger setting with which they interact and wherein they embed I have called this setting the operational environment Measures that encompass this second side must be developed in order to go on and explore still larger social formation among identities at a higher level as we begin to do at chapters end but the measures must reflect the first side as well Each identity becomes a joint formation that reconciles the social spaces with whatever is the ecological impetus It is this dual process that motivates the term embedding Talcott Parsons in his later work see chapter 6 for the earlier Parsons developed a positive theory of social system as selfsimilar system He specified nested sets of functional structures as universals This was the AGIL scheme AdaptationGoal attainmentIntegrationLatent pattern maintenance Mark Granovetter suggests that there should be a mapping into my discipline species A possible mapping to the three species of disciplines is G to interface I to council and L to arena The correspondent to Parsons Adaptation is of another kind it is the adaptation of social organization taken as a whole to the biophysical ecology One can argue for a more general correspondence of AGIL to my principal constructs G to production I to network population and L to identity However there appear to be no analogues in Parsons to mechanisms of decoupling apart from his generalized media of exchange Any identity embeds to a new level of identity only through its discipline but it continues being subject to ecological incident as well as to social unravelings Struggles by identities take place in space as well as in netdoms Geography must play some role too A big wedding say will roll up many hunks of social networks into the same hall and there may engender new ties and clusters But it is all too common for social theorists to treat space as unproblematic For example in a collection of views on sociology of culture Desan Ferguson and Griswold 1989 both Pierre Bourdieu and Alain Viala cf p 292 accept physical space as the obvious arena of social action just with the forces changed from their physical form Field investigators in social science fortunately do better Hackmans 1993 studies of interactions among threeperson crews in airliner cockpits can serve as a prototype for interrelating social with physical space in small scale And physical settings do give an easy start for visualizing social networks for example the playground in chapter 1 Yet distant pen pals may share only their love of certain authors Social networks need be controlled by ordinary physical space only episodically Understanding disciplines requires social specification of context And analysis seeks abstraction of any given embedding according to several aspects 323 Involution Differentiation and Dependency Any discipline survives only as it settles into its context both social and technical This settling is an embedding wherein the context provides the support enabling the discipline to reproduce itself I see three dimensions of this embedding into the operational environment the same set of dimensions for any of the three genres dependency differentiation and involution Any discipline that is accomplishing a joint task manifests some degree of dependence with respect to the operational environment This dependence may be an absolute parameter of separation as often applies for an arena discipline Dependency can be seen as the extent to which the particular discourse in stories within the discipline as well as physical activities interdigitate with the external For the interface discipline the measure of dependence becomes some ratio parameter of contrast between upstream and downstream yet at the same time there will also be some degree of differentiation in the embedding of context across particular members of the discipline Differentiation is how spread out the constituents embedding into an identity become on the appropriate valuation Members channel attention toward those of higher prestige who also may attract more attention from the outside so differentiation determines visibility The third aspect of embedding is the involution of the discipline among other disciplines observed the extent to which its embedding is shaped by ricochets from network processes in and around other disciplines It reflects the stringing together of identities by chains of ties that close back on their origins and so impact the valuation ordering in the given discipline Involution can also be viewed as specialization The place of any particular discipline can be plotted as a point in the threedimensional space that is implied by these three measures Examples will be plotted in such spaces in chapter 8 Note that although the valuation order itself can be seen as an embedding it is in a different key It embeds the members with each other in a single dimension which however is different for each genre To recast the argument the three dimensions will have different importance in one genre from another Within each genre the outcomes of discipline embedding will be less sensitive on one dimension than the other two The focus of the interface for example is pumping production and this commitment is not so much affected by change in involution which falls outside their focus Whereas a council discipline by its collegial nature tends to be opaque to the outside so that decrease in the extent of differentiation has only limited impact on the council As to the arena its raison detre is separation from the impure outside so that changes in dependence have less import on the arena It may help to think of the embedding of a given discipline species in terms of a triangle with each of the three dimensions as one vertex See figure 31 There are three triangles one for each species The triangle for interface disciplines shows a perpendicular from the vertex for involution to the line between differentiation and dependence think of that line as really the edgeon view of the whole plane defined by dependence and differentiation variations for a given value of involution Since interface mechanism is least sensitive to involution one can approximately plot as in figure 82 examples in just three planes for high and medium and low levels of involution The same logic applies to the other two Readers may wish to sketch out plottings of particular examples for a discipline either their own or examples from these chapters in actual threedimensional graphs such as in figures 82 83 and 84 There is stricter justification for such placements of disciplines themselves rather than of the larger formations about which I hazard conjectural placements in figures 82 83 and 84 The number of constituent identities in a discipline may be small as for a work group But in some operational environments of a given genre the valuation order may support discipline across a large set Empirical application can accommodate idiosyncratic placements of identities along the valuation order of that discipline in predicting accomplishments generality for the genre mechanism comes through representing placements in terms of a gradient a profile These remarks should illuminate the placements of diverse examples in figures 82 83 and 84 in section 831 To recapitulate a discipline presupposes some orderliness of perceptions by its participants The control struggles that create and surround these identities and to which their disciplines contribute settle out variously in networks and also in styles institutions and regimes Observedly pressures from those larger contexts reach in to affect some orderliness of perception by a particular discipline as an entity It is this dual orderliness that is reflected in the embedding measures The three genres of disciplinethe interface the arena and the councilare the configurations that proved robust to all these perception pressures above and below across varied contexts Now I expound each genre separately One variety of each genre will be developed in some detail around a published study But the main emphasis in each of the three sections is the sheer variety of applications possible with that same core mechanism The three genres together can illuminate much of social life 33 Interfaces Physical production and biophysical reality matter and work is the human adaptation to them Much of life is work Udy 1970 and work takes place in physical ecologies Work and its contingencies provide a major coupler between the purely social and the biophysical But much of work also involves coping with other identities and their control efforts often also through interface disciplines in social context But war too matters and is endemic In military operations the basic interface is with the enemy and the classic though not sole way of dealing across the interface is the engagement One commits to the engagement but in seeking disequilibrium may turn either upstream or downstream Asymmetry underlies all the variations of this commit interface Asymmetry is built into the form On one side individual flows are being induced drawing on suppliers amid jockeying for relative niche position the other side is possibly disparate receivers appropriating the aggregate flow The flows are always into the one disaggregate side and then on to the other The social perceptions that discipline producers into order come from both sides of the interface but behavioral cues to specific niches are on one side only Embedding into a new identity which for the other two discipline species is a byproduct here in the interface is the focus A set of actors can become comparable become peers through jostling to join in an interface on comparable terms They commit by joining together to pump downstream versions of a common product which are subjected to comparison by them and by the downstream Children competing in hopscotch or reciting for a teacher mathematicians in a test for a prize manufacturers of recreational aircraft for the US market actors in a playall can be examples of commit interface The word interface connotes passing through and transformation as does the word membrane But interface discipline is without the latters implication of a sharply demarcated material body instead an interface is a mutually constraining array of contentions for control that yield as the net result a directed flow a committed flow Interfaces are the most transparent of the three species The key parameters in explicit models prove to be variances so that the average sizes of the flows being generated through this interface are divorced from the selfreproduction of the interface Quality captures connotations of the transitive ordering induced in such interface disciplines This valuation ordering cannot be sustained by the induction and routing of average flows Instead such valuations provide scaffoldings for dispersions in social formations that then prove able to reproduce themselves 331 Supervision and Identities The span of control in modern bureaucracies is an unusually explicit formulation of the mechanism side of the commit interface as distinct from the quality ordering A superior is aside from concrete abilities and achievements also a placeholder a symbolic representation of subordinates embedding into a new joint identity Their identity is as comparable peers with common or rather parallel but differentiated goals embedded within a common social production Embodiment like this of the interface as a tangible distinct actor who represents the new embedded identity indicated by the designation supervisor lays the ground for subsequent further embeddings In these the supervisor can fold in together with others so as to become comparable peers embedded in yet another new identity also a commit interface Or the chain of embeddings can be initiated from above in which case it can and usually does switch from one to another of the three species of discipline Pressures from contending efforts at control are so strong as to also support socalled inverted supervision Here a set of peers interact with no common structuring beyond having a common subordinate This common underling may be a research assistant or a secretary Or at a different level the underling of several personality components may be a sexual drive In any case the underling may increasingly come to cue actions of all and thereby achieve de facto apportionment and so supervision The commit interface is robust to both external and internal control projects The equivalency in peer positions subjects insiders to very strong discipline by the comparable others Effective oversight comes from those similarly located and thus conversant with the information and perspective the subject is bringing Supervisors symbolize authority but are substantively more important in linking to other levels of Reference group theory long ago came to the view that it was dispersions in rewards among actors not averages that drove social action in small groups In the classic formulation from the Stouffer et al 1949 study of World War II military anticipation over time was equally important with dispersion Recently Tversky and Kahneman Kahneman Slovic and Tversky 1982 among others Lindenberg 1989a have revived this notion 62 disciplines and context Even the accumulation of authority through a supervisor is difficult Interfaces are directly concerned with identities Since identities in action are refractions of what does not fit neatly into social organization of network and discipline there normally is a multivalent correspondence as regards authority between identity and any tangible actor There is more than one correspondence between interface and identity even fresh identity Interfaces yield substantive outputs as well as comparative standings and this distracts attention from the new integral identity on different level Competition is about the importance of doing slightly better than your peers who in the larger context are so very similar to oneself what is not necessarily signified is the strength of the new identity created by the competition Within a supervisory structure as example a tale is induced of potential antipathy toward the supervisor That tale exactly conceals the fundamental effect of commitment through the interlocking externality of peers impacts upon each one of them as the production of social action proceeds 332 Production Market and Quality Order Production markets can be seen as extensions of simpler situations Material production quite generally comes in these interface disciplines Here the receivers are a distinct set and the context is not relaxed and sociable The hunting or gathering groups described for tribal contexts Firth 1935 1978 Lee 1979 Udy 1959 Rose 1960 are early realizations that have analogues in sports teams Leifer 1990 and in childrens games Fine 1979 Opie and Opie 1969 The basic mechanism does not require or presuppose distinct roles among the producers with explicit cues and assignments Rather a spread of performances is induced by attention of producers to differential preferences by the other side who can turn off their attention or more tangible payments for production The same phenomenon of commitment can be seen on a much more massive scale in modern markets for manufactured products The term product whether light aircraft or frozen pizzas Leifer 1985 has no independent reality as a technical or engineering matter Its reality is induced only through the commitment of producing firms into being peers in a differentiated set that organizes terms of trade around an This example will be picked up in later chapters As more scope and depth are introduced into the theory production market in its larger contexts will provide a running example like the playground of chapter 1 induced order of quality among the producers Note the analogue to the humble example of children competing in hopscotch or before their teacher The industrial production markets from the previous century up through today are exemplars of commit interfaces Take some production market as a particular institutional realization of the commit interface Such a market is an ongoing social act These markets accomplish the feat of reproducing themselves to continue month after month just by their coherence as social acts Leifer 1985 Leifer and White 1988 Nothing is passive about this The producer firms which usually are the actors in modern markets are giant pumps expensively committed to spouting continuing flows of products more or less unchanged The set of pumps the market as superpump is built up only in interaction with and confidence in provision of an orderly and continuing social setting with buyers The production market must induce at the same time as it renders comparable distinctive flows from a tobedetermined set of producers into the hands of an array of buyers becoming accustomed and committed here This social process is what induces a definition of product from the common properties of this flow Some decoupling of relations lubricates choices and bargainings made in a production market But in any interface market the transaction flow with upstream procuring is distinct from but coordinated with that directed downstream selling so that market process invokes not only the peer producers but also two other layers of actors There emerges a collective identity associated with rather than opposed to decoupling Supply and demand are not operational concepts here Supply equals demand after the fact each time as a tautology But it is the variation among producers in qualities and the difficulties each confronts in production that shape the interface that motivates and sets the terms of trade I have laid out explicit models for this variety of commit discipline the production market White 2002 1981a b The interface here consists in the observable spread of terms of trade being achieved by various producers with their distinctive flows At the simplest level these terms are revenue for volume shipped leaving aside variant models for a line of related products that any given producer may supply Gossip can supply to each producer an estimate of most of the terms achieved by peers For the market to reproduce itself each producer must continue to see its pair revenue and volume as its optimal choice from the menu of observed terms of trade Only this menu is known to be sustainable by the buyers who themselves are comparisonshopping Terms of 63 trade lie on a commonly observable shape that cues actors into niches by their own preferences that yet are agreeable across the interface Call this shape a market profile Myriad requests and searches each perhaps minor to its originator cumulate into an overall pressure on some market of producers who are each baffled as to the overall response from across that profile to their particular product Each firm seeks a footing for itself from some location along a profile that is traced through the set of choices volume revenue already made by all the firms So the profile is an interpolation through the array of prices according to total volume of production Thus an industry builds an identity that is perhaps as binding upon its constituent firms as is authority within a clan or bureaucracy Location along a market profile can be designated as a niche a footing for that firm The breadth of a firms niche translates into some spread in quality around its index value on quality and perhaps also some spread in time horizon Firms caught up in a given market are also decoupled from having to be bound to habitual ties downstream This is because the firms rivalry in a market and as a part of a market insulates their commitments to particular deliveries of production flows downstream All this has to find expression in a common idiom stories circulating among them Each firm is unconcerned that other markets which are home to those with whom they transact are cushioning those transactors in like cocoons for transactions still further along I return to this decoupling in chapter 5 333 Embedding a Profile The interface is being accepted by the embedding side The latter is the arbiter of the competition of the relative performances of the producer firms and thus of the shape of the profile But the interface as a whole is not in competition with other interfaces like its firms are in competition with each other The ironic implication is that a production market thus governs only the relative sizes of differentiated flows not the aggregate size of flows combined The aggregate size is a byproduct of accidentthat is of the detailed incidents and situations that figure in the process of establishing the curvature of the profile So aggregate demand is also so to speak an arbitrarily induced byproduct That need not be true of all interfaces Return to the earlier illustration of disciplines in mealtaking Like the cafeteria a fastfood restaurant will be an interface with probably a rather flat profile across its workers The fastfood restaurant from one chain will as an entity also be caught up in an interface of larger scope with competing restau rants Fastfood restaurants will work to establish that they are comparable to others with which they compete that is their point So one sees that interfaces like other disciplines may recur and thus get embedded at successive levels There is varied basis in detailed phenomenology in the perceptions of actors for estimating parameters within a model across the niches of the firms within the interface embedded in network Some especially concern the pull and push of flows of intermediate goods Estimation is not easy How many paths of access downstream and upstream are afforded by the observed networks This must be assessed and predicted not just from the density of connections but also from their involutionOther parameters map comparative assessments of different firms as to perceived quality Estimates of those parameters will derive not only from the involution and density of ties in a network but also from overlaps in the incidence of different types of ties representing distinct bases of relation The set of values on a quality index identify firms at a given period This pegging by quality sets off further mulling over identities by some or all firms Yet the quality ordering is a property of the mechanism of the market as a whole core to its identity Processes of decoupling and embedding supplant birth and death of particular actors as foci Demography is not the point Decoupling is essential to the paradoxical duality of markets and firms as being both embedded in tangible networks among concrete actors while yet simultaneously being actors with scripts for relations that are transposable and interpretable 334 Other Examples and Control Profiles There is a strippeddown variety of this commit species that is just the reflexive form of a human group putting on a performance for itself One modern example is the discussion group there has been a tradition of systematic measurement of interaction profiles Bales 1970 across the group that permits quantitative analysis Breiger and Ennis 1979 A village meeting in India Mayer 1960 repartee in a barber shop lunch table conversation all illustrate relaxed social contexts in which this or another genre of discipline can be found Standard computer software packages are available for making such estimations eg UCINET and STRUCTURE The greater difficulty is devising fieldwork techniques for efficient specification or estimation of a network which will require astute devisings of sampling frames for networks Wasserman and Faust 1994 By their construction interfaces do not control for averages and cannot be programmed to yield prespecified flows Instead interfaces build their dynamics around the spread of contributions across the comparable set The commitment characterizing an interface is best portrayed by how variation in members properties correlates to the curvature in profile Mutual attention of peers is directed toward jockeying for relative positions that yield each one a distinctive niche It follows that the interface can exert control through the shape of its profile only given skill in manipulation of social organization Only variances and their ratios constrain the curvature of the interface when it is operating autonomously White 2002 But rewards separately and in aggregate depend upon the average somewhat accidental outcomes so there is a latent motivation to try to shift interface in concert The shift can be accomplished only if the acceptable shapes of profile are retained so that they become envelopes for achieved control profiles Participants can make systematic use of these facts see Eccles and White 1985 for how chief executive officers use these interfaces in achieving control over leading subordinates The interface comes in many other varieties other institutional embodiments Star systems in entertainment and elsewhere grow out of interfaces where embedding induces perceptions of events that are greatly exaggerated from the view of actors producing them Faulkner 1983 Even where the differentiation or dependence is limited as among starlets in entertainment there is the same pressure to generate events sufficient to embed them with a skewed distribution of fame despite undetectable differences as judged within the interface These star systems can be seen as closely analogous to the industrial markets The interface is a species with still more dresses than these few special types of competition or formal supervision or the analogous pair of production market and Hollywood scene Sitting around in a bull session or other conversation group is being in a commit interface The institutional costume may be similar but the process shifted or the reverse Actors ordinarily do not perceive and react to higherorder measures like variances so this like the other two mechanisms must be realized through forms that are perceived and estimated directly in everyday terms A model and its context must be specified in some detail to examine how this can occur 34 Councils The council genre of discipline is centered on a process of balancing contending but evershifting coalitions Such is the focus for example in the logistics of mounting war perhaps its toughest aspect One escapes control through single centers by preset formulae Instead preexisting strings of dependency feed into an unending process of corralling and allocating all the while the dynamics of contention keep this discipline going up and down in scale14 Mediation predominates over straightforward mobilization this is mediation between identities of course but also between goals and regarding amounts of resources An early place to see the workings of this discipline is in the kinship lineage an extended kin group with corporate rights whose allocations are balanced and rebalanced in a mutual discipline of mediation Our starting point is not the organized corporate group or the fabric of continuity and stability in the social structure part 1 cf Fortes 1945 but the individual the standard forms and processes of persontoperson relationships the gradation of rights and duties in jural and ritual relations in which lineage segments emerge as corporate units every person belongs to a hierarchy of lineage segments lying between the maximal and minimal limits of his maximal lineage relevant to his conduct in different degrees and according to variations in the social situation even when only a segment of a lineage emerges in any corporate activity its status and functions are influenced by the total lineage field including at the limit the field of clanship Conversely a lineage always functions as a combination of segments not as a collection of individuals of common descent Every lineage segment represents a dynamic equilibrium of mutually balancing segments It must be emphasized that these distinctions are not made by the natives The Tallensi have no term for the lineage units of Tale social organization can only be defined by reference to the way in which they emerge in corporate action in relation to other like units Fortes 1949 pp 1011 14 In the production economy context discussed at the end of this chapter as well as in later chapters the mediator discipline is visible in procurement and supplier networks Corey 1978 Porter 1976 B Shapiro 1980 Walker 1985 In that context too this discipline evidences the clientelist qualities developed in political contexts in section 65 There regime as larger context is seen as shaping constituents as well as the reverse etiology of buildup to which the present chapter is devoted Clientelism is a term for dependency especially in political institution built around strings of connections such as are latent chains in mobilizations Since the mediation discipline has indefinite boundaries it is natural to liken it to clientelist control regime a more intricate system Trust is more prominent and problematic in the larger system than in a mediation discipline but the logic is similar The discussion here and later oversteps the bounds of a discipline proper and will be taken up in the further discussion of uncertainty tradeoffs in sections 522 561 and 571 and in sections 613 685 and 686 These lineage disciplines are not preset but rather are invoked as to membership and type of tie and occasion The discipline in this genre is from an ongoing and interconnected and changing set of mobilizations and conciliations The stories and identities being induced are about claims and the content of social process activates chains of these claims Thus council disciplines revolve around conflicts and contradictions spread over time Theorizing these conflicts and contradictions will help in specific technical modeling see for example chapter 9 in Luhmann 1995 341 Mediation through Prestige Council discipline is the closest to the purely social Stories associated with this discipline center on the allocation of resources both material and social The concern with mediation is seen in some gatheringandforaging party as much as in a formal council with purely political concerns The social processes are urgent ones though the overall identity being induced and embedded by this discipline often aims at timerooted claims and an appearance of immutability Though not as urgent the playground of chapter 1 also provides illustrations of council discipline Children do not just appear at random and where they appear is significant over here they move into particular productions but over there it is unstructured play activities and mediations are required Named street gangs give further color and recognition to what is a universalnamely the interactive the riposte nature of councils Correlation among involved identities is key as both the source and outcome of mediation To mediate is to induce similarity in actual social enactment even though the story told may be a unique claim This is so because objectively mediation is always a reciprocating process of inducing like claims and commitments in responses to those of other actors these claims and commitments need not be verbalized One can best equate mediations according to whether they yield equal differentiation across all the claimants Council discipline is what anthropologist Adrian Mayer discusses as actionsets and quasigroups 1960 Any effort at mediation for an innocuous purpose as well as a weighty one is latent mediation of what others might mediate Mediation feeds upon itself induces counteraction and structure to any triggering action This is as true among the Nuer of old EvansPritchard 1940 as among our big business folk Bower 1986 or in government Bearman 1989 Namier 1961 It is just as true on playgrounds with their diverse yet persistent subcultures of conciliation Opie and Opie 1969 Thus mediation cannot exist as an actual social process except in a set of foils and counterfoils Mediation means tangible wrestling about claims that ostensibly are fixed and abstract Mobilization also means inducting other actors via commitments that are unrelated to claims Rather these commitments focus on successive scopes of alliance and opposition from concrete commitments The arena discipline is an inverse of sorts to this mediation discipline the council Participants in an arena are not embodiments of the rooted interests of factions as in a council Selection within the arena has actors functioning off to one side in obscured cliques of matchers and there is mystification by systematic doctrine in the purifier format eg committee or profession of the arena In the mediatorcouncil and only there can each direct participant also invoke ties of dependence 342 Factions and Autocracy Factions and their endless maneuverings around control of substantial outcomes are the substance of the council mechanism There is never a solution a permanent alignment of factions for that would contradict the central process Mediation and realignment goes on routinely in continuous adjustment It is a context in which changes are embedded as routine Members enter into and leave particular dependencies as well as the council structure itself whatever the stories told An identitys standing within the council is tallied according to how many successes are being accumulated in close encounters what some Native American groups referred to as counting coup The Roman Senate of Republican days Syme 1939 Gruen 1974 Badian 1958 1985 is an apt illustration of how this counting of coups worked out persisting over centuries In its heyday the Senate was remarkable in its marshalling of small initiatives leery of grand gestures as it encroached its way into dominance in Italy and beyond this rested exactly on subtle remediations among evershifting confrontations of honorseeking aristocrats caught in the hold of mediation pattern that lasted for centuries In the special case of autocracy any external or public differentiationspread in standing among all but the autocratis squeezed to nil This remains true no matter how much effort is expended in jockeying for standing among a servile elite around the autocrat be it Holmess autocrat of the breakfast table or an Ottoman sultan Findley 198015 As a further example the council power mechanism is central to the segmentary lineage systems exemplified by the Nuer who exist in a homogeneous context near the upper Nile that offers little scope for involution Here it resembles action in the Roman Senate but in simpler clothing It can also be seen in a whole array of segmentary kinship systemsalong with their analogues in boards of directors Burt 1983 Levine 1972 1989 Mintz and Schwartz 1985 Mizruchi 1982 1984 Mizruchi and Schwartz 1988 Palmer 1984 In EvansPritchards words The outstanding structural characteristic of Nuer political groups the distinction and individuality of a political group is in relation to groups of the same kind is a generalization that embraces all Nuer local communities from the largest to the smallest The relation between tribes and between segments of a tribe which gives them political unity and distinction is one of opposition Fortes and EvansPritchard 1949 pp 28283 An essential mechanism among the Nuer is the feud which is endlessly particular to pairs and need not induce any recognizable status ordering There is endless breaking and recalling of old alliances and identifications all as circumstances change Council disciplines are more stochastic in operation the more complex the overall economy and technology in their context Among the Nuer exactly the same genealogical descent lines and groupings can continue indefinitely as the grammar in which power adaptations are made through fission and fusionseen as moving further back up a descent line or further down In more complex environments there is a greater density of disciplines being worked up among a comparable number of actors This greater density will have the effect especially in the council form with its dependent strings of shocks that unsettle particular alignments without changing the overall discipline Power is often the idiom of the story set for the council discipline The explicit demonstration of power is only occasional in the actual operation of this discipline in any of its variants While displays of power are only occasional they are awesome in actual encounter 15 This contrast with exemplars of counting coup seems to be illustrated by the distinction between French and German intellectual cliques as sketched by Lepenies 1988 p 268 The ecclesia invisibilis of the Georgeans followers of the charismatic Stefan George in fact had about it as much of the French chapelle the demonstrative academiccultural defensive alliance congregated around a patron as it did of the informal indirect and thus all the more influential coterie of an invisible college This occurs when there is temporarily fully joint action and perception by the constituent and competing strings of dependency whose representatives or embodiments constitute the council This is overwhelming when it occurs not just when there is a great imbalance among constituents resources but because of hegemony of the council as integral identity with monopolization of the reality perceived by all involved Trust presupposes power power is built out of trust in council disciplines16 But power over engrossment and disbursement of social and physical material can be seen as what mediation points toward Without mediation discipline there is no power but very often without anticipation of power there is no mediation 343 Lazegas Law Practice This case study will be reported in some detail and then discussed again under other rubrics in sections 37 and 442 and in sections 632 773 824 and 852 This is parallel to the treatment of other sections marked with an asterisk An American big city law firm centers on providing advice Advice to its clients is shadowed by the pervasive seeking and offering of advice from one partner to another None of these are isolate events advice is developed and repeated has history and one line crosses with another so that resolution of some sort is needed It exemplifies a council discipline in operation but close scrutiny is required Close scrutiny is just what Lazega 2001 supplies in his monograph17 He uses several idioms which are compatible with council discipline He begins with the problem of fission articulating part of how this council discipline works Wellknitted teamsthat is subsets of members who recurrently belong to the same task forcesbecome a threat to the organization because they can defect and take away with them valued members and clients Collegial organizations solve this problem by allowing some niche members to reach firmwide status through the accumulation of one type of resource and the establishment of a balance of power between these 16 Compare Baldwin 1978 and Lasswell and Kaplan 1950 for accounts that are however not referred to explicit social formations 17 Besides the detailed analysis of one partnership Lazega surveyed a whole range of firms in this city with help from various law professors and the like So the book could also be invoked in the next chapters on style and rhetoric And indeed the firm is large and diverse so that complete analysis must embrace features not tied to council discipline may include the biophysical The complementarities assessed can range from simple pair matching on over to a full team of specialties For a largescale example turn to Pierre Bourdieus analysis of the process of examining candidates for the Grandes Écoles of France in The State Nobility Bourdieu 1996 parts 1 and 2 For smaller scale return to the playing field of chapter 1 in choosing up sides the straggle of children are sorting out from network context and proffered identities the sorts of contributions and degrees of expertise that different kids can bring The actual team may be a commitinterface a reification and symbolization of the new identity being created But the choosing is based on a preceding selectmechanism of the arena discipline 351 Acquaintance Dance For council discipline there was one extended empirical exemplification Lazegas law firm For arena the concrete example of acquaintance dance will be followed by two others 352 and 353 An acquaintance dance or mixer provides an example of the arena discipline22 Whole entering classes at one or more colleges may be brought together exactly to generate still further ties and networks out of random seeding of dyads An acquaintance dance invokes some rotation of partners for the dance floor perhaps with opposite circles and then switching of dyads at the stopping of music It is instructive to contrast the dynamics here with those for neighborhood bars discussed in the previous chapter Clientele of the neighborhood bar evolve from working networks toward corporate forms In contrast in the acquaintance dance corporates are worked about to generate ties which modify networks In this there is a duality The acquaintance dance is a limiting case that invokes the explicit coordination required for bureaucracy and the like It is but a first step in a proliferation of introduction settings of adult life nested within one another and across one another These go on among elites and in esoteric settings as well as everyday with gender as only one example of attribute selected on Individuals within an acquaintance dance may seek more restrictive corporates within the initial mass More specialized mixers are likely to follow More spontaneous versions are common Boys and girls use any occasion of corporate commonality as a basis for exploring and forming couples Men and women do so as well usually in more complex contexts of organization 22 This ceremonial form may be disappearing its replacement not yet clear 97 352 Gibson on TurnTaking The most potent social discipline you experience is one that you cant articulate because you so much take it for granted This is even though or rather because it is part of your habitus in Bourdieus formulation David Gibson 2003 2005 has established that turntaking in oral communication is such discipline and I will argue that it fits the arena genre in particular As with Lazegas study earlier I offer considerable detail The valuation implicit I argue is decorum good manners keeping face and letting others do the same This valuation is a form of purity and therefore the genre of discipline is the arena Gibson observed fifty thousand turntakings in committee meetings many committees each in many sessions He concerted these into successive pairs of turns and then converted those into an inventory of participation shifts The key is what the participation shift suppresses what it does not code first the participation shift omits the names of speaker and of target in a given turn being concerned only with the switching between that turn and the succeeding one It looks only to whether a party in the second turn and if so which of the two there coincides with previous speaker or target in the first turn Second this participation shift PS also ignores the topic and content of either turn except for any throat clearings discourse markers like Oh Well by the second speaker in seizing the spot And third PS suppresses information on the duration of either turn although that is stored in the computer along with of course Gibsons list of names for members but without any transcript of topic Gibson thus created a grammar of turntaking akin to the grammar of pronouns in regular language These are momentary pronouns just S marking speaker and T target in the first turn Interest initially focuses on whether the second turn just switches S to target and T to speaker on the second turn and indeed in a good 40 percent of the PS such answering shift is observed There is another piece in this grammar Gibson also coded whether a turn had all conversants the whole group G as target These are committees of bank managers who dont brawl or all talk at once Gibson does not code any instance of G being the speaker on a turn About 30 percent of all the individual turns are undirected that is with the group G as target so that 70 percent are directed In the PS grammar for pairs of turns whereas only onetenth of the PS had G as the target on both turns onefifth of the PS had G on the first turn only23 23 Indeed the percentage of PS with G as target in the first turn only is about the same 40 percent for PS with an individual as first target who then reciprocates as speaker back to the first speaker as target 92 members with different forms of firmwide status consists in systematizing the division of leadership workthat is status inconsistency particularly through a process of destabilization of task forces and circulation of associates members but not partners p 39 And he later amplifies Members competing for status also reached out of their niches By doing so they both gained some autonomy and created at the overall level a cohesionbased integration process of stitches bridging sides separated by internal boundaries p 199 And he reports the following results First members influential in policy discussions in partnership meetings tended to be cited as important professional advisers and important protectors of the common good and were considered by many to be friends Secondly membersincluding young and entrepreneurial partners considered important to others as strong coworkers tended also to be sought out often as professional advisers but did not put in many billable hours These were often rainmakers or finders who were in a position to distribute large amounts of work to others Finally there were solo operators who happened to bill and collect individually enormous amounts during the year but tended to be prima donnas p 262 Lazega backs up this portrayal using several bodies of systematic data plus extensive confidential interviewing He extracts meaning from this large corpus through use of several standard techniques of network analysis18 and network diagrams using structurally equivalent sets of actors as the nodes eg figures 31 and 41 pp 103 and 163 Lazega also collaborates with Pattison in devising powerful statistical framing Lazega and Pattison 1999 to test how incidences of several types of tie influence one another also allowing for effects from attributes of individual members Cooperation is systematically amenable to structural analyses at the dyadic triadic and overall levels p 131 Finally he developed an imaginative new sort of sociometric question about who a respondent thought had what kind of control and then what sorts of levers he would use to influence action there Krackhardt was his collaborator Lazega and Krackhardt 2000 on new scaling techniques to portray these results as in figure 73 p 214 18 Some were described in chapter 2 But Lazega does not employ blockmodeling for the reasons he lays out in note 12 p 308 I assess this divergence in section 442 on professionalism 93 Lazega presents summary assessments that fit this firm into council discipline Status is not only based on seniority and money it has a particularly strong dimension of prestige of symbolic recognition of a members contribution and of ongoing critical judgment about members quality p 272 And Lazega also invokes embedding and decoupling as well as identities The existence of social niches means that one must learn to personalize embed and depersonalize disembed work relationships and business transactions p 272 This is not only a convincing portrayal of a council discipline but also a model as was the Sampson monastery study of chapter 2 of research both rigorous and insightful The quality of the data is central it was by persistence over years that Lazega won unusual access to confidential data including access to policy and individual member financial performance19 But there is more to say and this case will be carried forward through subsequent chapters along with the examples of childrens playground and of diffusion A tangible detailed study like this makes it obvious that reality comes neither cut up into separate cookies nor laid out in cleanly separate levels And as to discipline species we argue in 564 for Lazegas firm as a member of a production interface as a big law mill competing to churn out cases of various sorts Then when it attends more to the judgeship core of law the firm orients around purity and is also an arena discipline 344 Ambiguity in Council Disciplines Discipline in councils is highly ambiguous There is little in the way of systematic doctrine Connectivity seems essential to power the key is to be tied to the right faction through the right connection whether in the Senate or in a family business However when normal reshufflings of factions are traced over time it becomes clear that the essential aspect of structure is structural equivalence No factions and alliances ever stay the same In the end what counts is the likelihood that your connections will be mobilized because of parallel hostilities and re 94 flected interests with dominant sets of other factions The combination of fissionfusion with stringingdependency tends to generate aversive pillars as a byproduct of structural equivalence The awesome yet ambiguous power of council disciplines is concurrent with an openness and flexibility of agenda Each council miniscule and local as it might be in some broader survey is sovereign in its own concerns These concerns are limited in and by concrete fact but they are not limited by any systematic doctrine or rationale20 In particular there is little interpenetration of one council by another in dependencies and in various wrangles and quarrels As with any discipline context shapes the process Imagine a context with maximum differentiation maximum contrast of upstream over downstream appreciation mountains out of molehills the context insisting on imputing great differences in status between actors and strings even though their attributes are not much different as separately measured An example is clientelism as described by Riggs 19661969 common at all levels in Thailand Riggs brings out that the strings of dependency are normally short with dependency being very much personalized rather than leading to retinues Thai clienteles are not specialized Another example of context shaping process in councils is seen in the commissaires of Renaissance French absolutism These councils make and carry out wartime military requisitions Hintze 1919 see Eccles and White 1988 and chapter 6 in this book Dependency is higher indexing the much higher attention that commissaires devote to producing what the fighting forces need as compared with insouciance of Thai patrons to clients Take another example that is medium on dependency and also tends to show high involution rather than sensitivity to context Hollywoods production system is built around the councilmediation mechanism This was analyzed and documented by Faulkner 1971 1983 using a modern system of social structure measures derived from blockmodels and compatible here At first sight it might appear to be a marketplace an arenapurifier but it is the same specialists who come together again and again They get rearranged into new packagings of skills for each new production Major figures find minor ones from the same or other expertises and bring them in tow into these packages There is heavy inflation of tiny initial differences in standings In council disciplines trust is required and power is constructed The two presuppose each other Mobilization spreads from some be 20 In Nadels terms from chapter 4 role summation is common 95 ginning link which at the same time will trigger other apparently distant links whose actors perceive relevance to their own social locales This is what underlies the mediations that reconcile divergences As a result some distribution tangible or symbolic of inputs and outputs is coming into place Only a limited degree of equality through sharing is built into mobilization through the mediation processes Trust is presupposed and observably justified by some form of historicized tally This tally cumulates into a type of tie that is weighted by its history in a succession of chains of mobilization There is an institutional inflation as compared to other standards such as arena ones This is inflation in that the claims of different ties if called all together at once would bankrupt the situation more clearly even than a rundown of confidence would bankrupt a bank 35 Arenas The arena discipline selects and matches typically in an episodic fashion In the playingfield case formal teams may be only two in number but the number of clusters going into and out of selections and matchings is various and shifting Selecting here is concerned with variously perceived real tasksthrowing passes versus line play and the like and corresponding degrees of social fits This transpires before the stylization of formal teams is achieved For many sorts of tasks longerlived and more complex teams configurations enter into matching and selection for viable arena discipline These selections are also often visualized as and may in fact take place in literal arenas physical contexts given social identities from the matchings The same extends to other realms such as acquaintance dance or the production of Broadway musicals Prince 197421 Diagnosis of your current ailments legal or medical are other modern exemplars of the infinite variety of dress in which these selection disciplines come disciplines built out of exclusion for purity ordering Comparabilities are the foundation of all three genres A cohort in age as it moves along through some agegraded mobilization exemplifies how participants make themselves available through preestablished comparabilities In arena disciplines these comparabilities are probed to turn up and then match complementarities The talk tends to be about esteem a purely social aspect whereas the actual concerns in the productions being put together are with complementarities that 21 See later for the former and for the latter the previous chapter and chapter 6 in White 1993 Yet only about 5 percent of the PS had G as the target only on the second turn there is a pronounced asymmetry That is the clue to how the grammar works Gibson took the bare string of turns and coded them as simply directed d or undirected u G as target He constructs and tests a null hypothesis that the occurrences of d and u are independently distributed In the whole corpus of fifty thousand turns which differs by only one from the number of PS the null hypothesis is rejected and indeed the probability of du is substantially less than the probability of ud The real point however is that the big divergence the big effect is segregation That is ds cluster together as dd Participation Shifts a bit more than expected but especially the u turns cluster together into uu Participation Shifts very much more than expected This outcome means that there must be a lot of interaction effects back up at the level of PS which validates Gibsons intuition to look for grammar there His task was to explore these interactions and then interpret them see Gibson 2003 The notable results are how participants manage to shoehorn and piggyback into speaking via invoking just those active in immediately preceding turns A bare 3 percent of PS have two entirely new persons in the second turn The concern in this discipline manifestly is selection The valuation implicit in this discipline I proposed earlier is decorum good manners keeping face and letting others do the same which I interpret as a form of purity So the discipline uncovered is of arena genre The pattern of findings has face validity as a discipline of turntaking a discipline so engrained as to be unnoticed A discipline should also predict to other aspects of social behavior Gibson 2005a in a second piece using data on network ties shows that indeed there are strong correlations between role in PS grammar and role in networks particularly in networks of authority ties Social living is complicated and there are many aspects of the lives of these managers not captured by the discipline that governed their group interactions That does not make the discipline any less real nor does it excuse the researcher from bringing to bear other constructsthese constructs could be from the other chapters or from the other two disciplines in this chapterall of which may be visible in Gibsons settings see also Gibson 2005b 2005c 353 Arena Markets and Production Markets A selectionarena brings together actors who may be disparate and inhomogeneous into a setting that is socially constructed to emphasize formal interchangeability so that actors are viewed as comparable That is why your earliest conscious experience of this discipline may have been choosing up sides for a game amid a bunch of kids on a playground And the pureexchange market distinguished by economic theory Newman 1965 is a mediated version of choosing up Exchange markets on all scales are obvious arena disciplines The actors are there to make matchings which can be of the most variegated sorts and which can appear as flows andor stocks or services or intangibles Consider a lawn sale or for that matter a village market Money while it eases barter also obscures the underlying social form From an operational social viewpoint what is being created are dyads that can produce what they both want only by acting together By contrast within the interface discipline production is relatively unproblematic for individual actors The focus of production markets as an interface discipline is to induce orderly reception and interpretation of the flows the producers generate via a clear precedence order among the producers according to quality The arena discipline is the obverse in which the social production exclusion and recombining can be accomplished only and in diverse ways from joint activity so matching together into pairs or more various teams is induced Mating is the homeliest example Supply matching demand is a syllogism for the arena market it is at best an ex post rationale a tautology for the production market24 This is true for any scale true whether the discipline is of firms or of persons In the commitinterface discipline producers eye only each other as reflected in terms of trade achieved with the buyers adding confusion the interface operates as a oneway mirror In the selectionarena in contrast either you are in the arena or it is opaque to you You are equally in whether a seller or a buyer or both and sellers are no more active than buyers in establishing proposed terms of trade Select and exclude are obverse and complementary operations here There is also a basic difference in time construction The production market presupposes and requires unremitting attention to the flows and to the interface by the producers whereas an arena market can come in discontinuous and selfcontained sessions All those present in an arena are equivalent rather than marked by side and fixed in niche by quality Yet attendance at an arena may be fickle Note that linguistics can offer some guidance on criteria for discipline membership The issues are subtle as in the distinctions between marked and unmarked Battistella 1996 does a good job of laying out elaborations of that distinction since its early classic statement by 24 Supply matches demand is called Says law see Morishima 1973 for a trenchant dissection Roman Jakobson 1990 For a social application see Salzinger 2004 A central aspect is ambiguity not just in categorical assignment but in whether any assignment is called for intuition is helped by reading in the Brown Book by Wittgenstein 1965 Price seems like the natural idiom for matchings and selections This is so much the case that we expend great efforts to impute and realize a monetary format even when remote from any economic context But a more general way to view the ambiance of the arena discipline is as purity after selecting or excluding particulars from a whole bunch the bunch becomes more purefor instance selecting nuggets of gold from a pan of ore or picking debris out of a rice harvest 354 Fame and Chance Selectionarena disciplines are robust with respect to eruptions of control projects whether from within or from without The robustness comes from the multicentered structure combined with the fluid and stochastic nature of flows into and among clusters Size distributions of freely forming social groupings within an arena have proved robust James 1953 Coleman 1964 White 1962 and on larger scales too this is evidence of robustness across attempts at manipulation Arena disciplines are flexible for accommodating various and unexpected actors The problematic for this species is identity formation for clusters Selections into clusters of complementarities can be apt can constitute from an observers viewpoint effective social constructs in future production yet they may come with inadequate induction of identities It is for this reason that fame in various shadings and spreads so often is articulated in matchings and selections Fame can be purely social and arbitrary Fame can supply a sort of universal currency of identity a flexible spread of identities among the straggle that are matching into clusters Fame is a way to make purity ordering tangible Chance the stochastic is attracted to and helps induce the selectionarena discipline which most often we conceive as an exchange market or a garage sale This arena seems chancy and disorderly even socially problematic yet it is an effective and predictable regulator of real network flows 355 Arenas as Purifiers Purity valuation frames the operation of arena disciplines Selection is a sort of purifying Exclude is the complement of select As the metaphor arena suggests and purifier language reinforces examples of this discipline cover a vast range from an office cocktail party to diplomacy among nations Iterative realignments of alliances are arena dynamics which underscore the inevitability of compromise but also of ambivalences For example ducking frustration in arena settings may lead to successively imputing impurity to former allies Modern communication technology by proliferating audiences and possibilities of communication and affiliation more generally is also proliferating arena disciplines Take Google as a prototype in the Internet environment that affords virtually unlimited possibilities for affiliating and also for distancing Using the Internet link structure and network measures to rank value of Web pages and relevance to topics Google ranks purifies pages according to how many links it receives as well as by weighting the links according to the importance of the Web site that is doing the linking25 Google emerges as an identity shaped in arena discipline Implicit here is a distinctive degree of fluidity indeed it is weak ties in the sense of Granovetter that blend best with arena not strong ties with their more fixed architecture Membership in purifiers can vary from the most temporary and casual to the most permanent and rigid There is a tendency to concentric shells of membership The inner shells are the most pure have the most weight in evolving changing standards of matching For example when creating page rankings ties from important pages have more weight Purity is created by achieving matchings that become defined as pure Purifying occurs off to one side out of the mainstream with the most pure out of the mainstream The general purpose of purification is known and indeed disseminated On the other hand the arena of the selections and matchings is opaque from outside And just how and why particular sets of selections are made need not be visible to each other That is the rules and standards evolve and change across the disparate population that the purifier exists to make comparable A village caste council in India Mayer 1960 is a selectionarena that explicitly emphasizes purity ordering Its business is endless small matchings and adjustments of ceremonial requirements This example like exchange markets and indeed disciplines of any species occurs as one among many other of its type This reminds us that disciplines of all genres are analytical devices that presuppose as they also imply aspects of the larger context Disciplines do not come solo This fact of pattern goes hand in hand with the character of storyset used The caste example is one in which particular purifiers indeed 25 See Google Web site explanation httpwwwgooglecomtechnologyindexhtml sources given to this or that situation Slack is shorthand for uncertainty in social context One further observes that not even the overall total consumption of attention and other resources is subject to much constraint We have all noticed that committees tend to go on for unpredictable lengths of time with unpredictable partition of attention by item They are subject even against formal rules to being convened exceptionally High slack is a corollary of the purifiers being out of the mainstream of everyday business available for unpredictable claims for social fixes Openair markets are another institutional embodiment of purifiers here where the hurt to be fixed is a material one The pure theory of exchange applies Newman 1965 But this pure theory concerns only the formation of prices that are internal to the purifier operation Left out is the huge slack built into the arbitrary times occasions and membership in the pure exchange arenathe swap meets and lawn sales of current times or the village fairs of Skinners Asia 19641965 or early Europe or Africa Smith 1975 It may be because of high slack that one finds purifiers so often in whole setssay lattices of committees for a larger environment On a longer timescale careers are being negotiated more or less explicitly High slack permits the subtle and endless probing and estimation that lead to invitations and acceptances and possibly to careers But this points beyond particular disciplines to the rhetorics of chapter 6 36 Households Family and Gender Bringing It All Together A family in the everyday sense around its meal table exemplifies council discipline Yet also gender is sorting out according to arena discipline emerging from ongoing networks And above all so too is production of children along with work amalgamated as interface discipline Of course all three figure in later chapters too gender and family build into and from the institutional system of kinship see chapter 5 and often also with the control regimes of chapter 6 see Hamilton and Biggart 1990 Household discipline undergirds much of traditional work institutions Udy 1959 which exemplifies historicity in disciplines So this section is in every sense evoking how disciplines indeed all the constructs we work with invoke and affect context of all kinds This section is a preview of the general look at contextualizing in chapter 8 Gender calls up gendering of all sorts of items in stories Gender is a major symbolic resource the grammar of most languages attests this and for some like French and Spanish scream it out At the same time gender also gets stretched out as if taffy into one cline and anotherfor example along the degree of femininity or masculinity with age for both sexes 361 Meld of All Three Disciplines Patriarchy and domination call forth each other across some meld of gender and household with family discipline And all three often figure in the institutional system of production economy Udy 1970 and of polity control Padgett and Ansell 1993 Demography when it is guided by theory and history Hammel and Laslett 1974 beyond mere bean counting needs to give an account that is integrated across all three disciplines Network population itself melds all three Macro redundancies such as across ageperiodcohort build out of all three Ryder 1965 Ours is not like an ant society Wilson 1970 ours is a species with continuous rather than discrete generational spread All our social formats as proposed in chapters 17 are pushed and stretched by the persistent and yet erratic injections of babies and thus new identities As we shall see in chapter 8 the network population context feeds back into shaping the meld of disciplines This is true as to network topology but also as to structure of spread over time Family discipline for example has to stretch over extreme fluctuation in relative age and kinship transitivity And of course the geography and ecological setting of the network population correlate with variants of each discipline 37 Inventory of Disciplines Yet the melding of disciplines is not unique to family setting Lazegas monograph on a law firm supplies tangible exemplification Yes it was seen earlier as council discipline But Lazega goes out of his way to bring out the playing out of the other two valuation orders The firm depends on bringing in new business and elaborating existing counsel which can support interface discipline around such commitments Nonetheless this seems secondary to council discipline since the rainmakers are by no means hegemonic And certainly this set of lawyers is concerned with excellence in legal thinking Some partners win honor from that so one can argue some presence of arena discipline but a Supreme Court panel is not what they are emulating Beyond the focus thus far on a single genre of discipline I claim that there is nothing esoteric about them They are everywhere in all refer to and imply particular others so that they imply a larger rhetorical structure The caste example is extreme in the explicitness of concern with purity to the extent that there is an ideology tied to a religious institution to undergird it Another sort of example is the office of chief executive officer CEO which evolved in business over a fiftyyear period in the last century Chandler 1962 Fligstein and Fernandez 1988 Vancil 1979 Like all real evolutions it was blindfold At first it seemed but an innocent alternative to president it seemed at most an example of title inflation where both CEO and president could bask in apex glow Perhaps it also served to ease retirement transitions President itself was an uneasy concept President was a temporary managerial servitor in an environment of owners unlimited in term and too many tried to assimilate to the time horizon of owner26 Analytically the CEO indeed is a selectionarena discipline The CEO was not the handson manager and decider of operations but exactly the fixer the healerand thereby the controller given appropriate networks of committees and offices below27 which is to say that the CEO is a purifier Over a single decade Vancil and Green 1984 the CEO came to be more and more commonly designated as a small committee of interchangeables who are to function in what we have described as the purifier way as the selectionarena discipline selecting for issues choices of strategic importance At another extreme consider committees with which you are familiar in your own practical life They too are purifiers in a broad range of situations The committee enables placements to be made in much less constrained ways that is committees may serve to pull from out of the main flows streams of problems and opportunities and bring them to one side where matchings can be made cf March and Olsen 1976 Because the functioning although not the provenance of a committee can be private existing memberships and networks can be temporarily suspended in devising rationales for matchings and selections The formal equality within committees which neutralizes age and status and tribal distinctions and the like is important to the flexibility Common also as purifiers are arenas defined as uncommon centered on an inner core of purification specialists A gathering of elders may 26 It eventually became clear that this mere change in terms went hand in hand with a shift to a very different a larger physiology of control to the multidivisional form from the scalar functional one see discussion in chapter 7 27 In chapter 7 this shift to CEO is also shown to be exactly a shift to use of arena disciplines in a new lattice of control scopes all realms at all levels The trick is to penetrate the bland mirror of common sense to track them in motion I will practice what I preach by looking for disciplines right at home in the most familiar aspects of my lifeor rather at work not home But first I take up general difficulties in taking inventory even if just for the playground of chapter 1 Already with networks there are puzzles about how to inventory The previous chapter described 234 an elaborate parsing of larger pattern in a network in terms of constituent triads Perhaps that can be brought to bear on disciplines each of which obviously must invoke triads selectively in distinctive ways Each discipline folds into the contexts of any other disciplines within the social process being observed At the same time netdoms among its constituents are influenced and reshaped The sets of stories that characterize ties borrow from valuation orders in environing disciplines just as disciplines get hitched to one another through network ties preexisting or new between constituents And any such hitchings between disciplines are only one part of the larger configurations taken up in subsequent chapters Networkpopulations and their disciplines interpenetrate through migration and conquest and many other processes from which control struggles emerge on a new scale These struggles settle into reproducible social formations that are so far not describable from constructs and measures developed in these first chapters Three new approaches that probe different aspects of robust articulations of network populations are required At the very end of the chapter I examine tournaments and liminal occasions which furnish two extremes that bound the larger social formations in the succeeding three chapters extreme constraint and extreme looseness One general difficulty in analysis is fuzziness in the disciplines to be inventoried A closer look at catnet will help clarify this issue 371 Catnet as Residual of Disciplines A discipline as actor comes in and out of activity being buffeted by thrusts of control efforts by self and others When can a discipline maintain identity as integral actor with energy shaped from its embedding In disciplines as in netdoms meaning gets shaped through switchings Discipline once emerged is tracked as narrative within discourse So discipline pulls on and is pulled by ties in and across networks The resulting potpourri sometimes manages to keep reconstituting itself in distinctive profiles of meaning such as we turn to in the next chapter function as a purifier in dealing with ill health matching complaints to treatments A gathering similarly may mediate conflicts seen as a purification with wisdom A gathering may similarly be consulted and consult concerning sin and matching sin with apt contrition In otherwise very different societies and rhetorical systems purifier arenas of any of these kinds can be given explicit formal standing In this era such uncommon purifier arenas are professional settings see Abbott 1981 1988 and chapter 6 in this book Doctors heal judges judge priests offer sacraments each in formalized settings fitting arena discipline A joint opaque arena of matchings underlies the formal setting but the potential interventions from many and varying actors are not obvious In each there is a logic of purity akin to that in the caste situation keyed to an inner because more permanent body of allowed practitioners There is little tangible connection between the body of doctrine kept up to sustain purity definitions on the one hand and the tangible matchings of victim to remedy or of actor to another actor or to material parcel Thus the setting can be understood as a discipline rather than some mere offshoot of overall arcane cultural prescriptions 356 Ambiguity versus Slack in Arena Disciplines Professions and committees are examples of the arena discipline Arenas orient around purifying In all purifiers there is a decoupling of external status from internal standing Within the purifier the practitionerstemporary or permanentare operating in a collegial mode There is no tendency toward inducing status discrimination within the purifier Practitioners as such tend to have special standing when viewed from outside and this is associated with a tendency to impute differential status among practitionerswhich need have no relation to any internal standings The opaqueness of a purifier permits it to seek matchings that can reduce social flexibility and unrootedness At the same time it provides rationale for ambiguity being kept fixed or even fixed up as seen from outside The very same committee members who may have just artfully patched up a social tie may back outside the committee not see that or similarly not see how they have rigged a generally accepted cultural rule The unusual degree of social confusion within the purifier is necessary to obfuscate the ambiguity that according to the formal cultural rules is being introduced by purifier action A purifier is a structure whose system of operation offers a high degree of slack That is there is not a tight constraint on timing there is not a tight constraint on the amount of attention and other re In some contexts there may be no sustainable variety of any species as modeled here according to thencurrent observation Disciplines themselves can be identities with social ties and at the same time participants within such a discipline can be identities as well Thus some network of ties can appear connecting entities embedded on different levels It is only while a discipline is in action and hegemonic that its context and contingency minutiae in the form of ties and networks of ties may fade from the focus of attention When clearcut disciplines do not emerge one can expect not only a profusion of identities of limited durations but also much more prominence for looser forms of social organization There are other representations of social space and contingency that surface where identities cannot be grounded in terms of niches in disciplines Attributes of actors or of events become more prominent in the absence of place in an articulated mechanism of discipline Stereotypesthat is to say categorical attribution of characterappear among actors and events Catnets appear in the absence of stable wellformed disciplines But Lazega goes out of his way to bring out the playing out of the other two valuation orders In earlier work Schwartz 1966 1967 the term catnet was coined to capture the involutions among network interconnections and personalization of attributes Chapter 2 already introduced catnets Persons recognize indirect connections that are implied by the set of pair relations assumed common knowledge which can be represented as a network But these indirect connections are recognized only in part and over a limited number of removes These indirect connections are reacted to in concrete terms rather than as welldefined new types of relations The principal result of the evolution of a catnet is the definition in the eyes of participants of structural equivalence as a guide to the perceived system A simple example is the development of cliques in a network of friendship The rather diffident theory of social interaction elaborated by Niklas Luhmann 1995 chapter 10 can be operationalized in terms of catnets One can see that the catnet construct is a path around the portrayal implicit in this chapter that disciplines have disjunct memberships It is a path toward the style construct of the next chapter as we will see 372 In My Own Experience Many years ago as a young faculty member my views remained commonsense ones with lots of stories of villains and heroes and shenanigans and especially goals along with elaborate plans to achieve these goals Underneath I slowly have come to see the reality is selfperpet coupling and embedding in an extreme form that is the opposite of that for tournaments Any and all network populations may but need not impact encounters during Mardi Gras Liminality is signaled by ceremonial boundaries that is there is a selfconscious embedding within an explicit culture usually by rejection At the same time the period of liminality serves to interrupt causal chains of agency and gaming Usually this suspension holds only for some short very specific period Liminality is a decoupler for agency even while it embeds in overall context One can see this duality in Turners Mexican fiesta in student strikes in pilgrimages Christian 1980 in masses in reunions around a stage at imperial apotheoses Cannadine and Price 1987 Liminality copes with phenomenological confusion These two extremes of tournament and liminality are perfect converses In the tournament the social standing of each actor is totally clear This is a state of zero ambage and there is as a result complete ambiguity as to the cultural basis of the social standings as argued further later on In liminal formations on the other hand there is zero ambiguity because there is an agreement on an extremely simple new culture of rules Usually this is just an erasure of previous rules At the same time there is complete indefiniteness in social patterns of relation and thus extreme high ambage While liminal occasions mush together disparate and remote swirls of action tournaments break apart larger contexts Ambage is suppressed at the cost of erasing verifiable cultural content as criterion of dominance Byproducts such as hierarchies may emerge in the larger context around the tournaments Various tournaments each with relatively arbitrary cultural bases can be fitted together in resultants of projects for control Chickens do their flocks pecking order sincerely but unlike humans they have no capacity to concatenate tournaments into ladders of mobility and hierarchies of control Utopias are the imagined longterm versions of liminality compromised to various degrees for plausibility The dictatorial utopias the Orwellian 1984s and the Walden IIs are as much liminal as the happily communal ones It is not clear that liminal formations are observed in any other species Mathematical modeling by stochastic process and encounter matrix should be feasible along lines analogous to Landaus cited in chapter 8 but there is no set of empirical findings to give bite to the analysis in which symmetric ties would replace the asymmetric dominance ties What is clear is that liminality is common if not universal in known human societies28 It appears to be an episodic formation29 Why the contrast with dominance hierarchies Liminality appears endogenous in that no external pressure or trigger seems needed Yet liminality is episodic presumably from endogenous pressures of decay Dominance tournaments are often extremely stable in structure and even in exact mapping of member to rank yet except in stupid species where there are dominance hierarchies as a coupling between biophysical and social tournaments are ceremonial contrivances not engendered in the routine operation of a social formation Liminality and tournaments both emphasize discontinuity Time in social life has a picaresque quality overall pickerup in ambiguity and ambage as at a cocktail party but moment by moment it is remorselessly Markovian Feller 1968 White 1973 Memory whether as vengeance or forethought interrupts the Markovian quality only occasionally with most of the generation of the picaresque coming for each actor from interruptions through gamings by others Liminality produces the illusion of an enlarged present in which Markovian chains are broken Tournaments impose an enlarged present by fiat Tournaments make mock of preceding continuities of interdependent action Complete linear orders of precedence complete sets of pair dominances without ambage are antithetical to experiences of time and sequence and interdependence 28 See Turners crosscultural survey 1969 29 However sleep it can be argued Aubert 1965 White and Aubert 1959 is a liminal social state FOUR STYLES A SENSIBILITY is how style presents itself in experience For the observer style may become apparent as an interpretative tone deployed by native expertise As such it exhibits a distinctive texture in social process sensed by those immediately involved Analytically style can be of any scope and level it is a scaleinvariant concept And whatever the scope two basic aspects of style come intertwined a the interpretive tone along with b the feedback dynamics Style is not transposable though it may get reconstituted Style is immediately available through attending to the sensibility that goes with texture in life Some family mealtimes are homely in style Graduation ceremonies are formal in style I will always remember the style around Harvard Yard in midApril 1976 where Elysian weather accompanied both extreme social stress and mellow discourse to form a unique texture But style is not to be seen everywherenot for example in administrative routine of a business day or deployment of an army battalion Style generates its own context and so is involute constituting a boundary in contrast with the network population of chapter 2 see 213 Style also differs from the disciplines of chapter 3 whose survival and properties depended on how they are embedded into contexts Networks need not persist just as they happen to be thrown up by stochastic eruption any more than a particular discipline with its projected identity will persist and reproduce itself independent of social context which is in turn made up of networks These social contexts are necessarily stochastic Style transcends and commingles network populations and disciplines via peculiar patterns of switchings Despite being a selfreproducing context and a selfcontained identity style can change through stochastic social processes across diverse constituents among networks and disciplines Social temporality emerges basically from a profile of switchings across netdoms Whereas identities can be like musical notes that struggle for a melody and discipline embeds these identities in the larger context of a genre style is the rhythm of social life Metaphorically identities contextualized in disciplines make up the melody to which style adds temporality Couples dancing close for instance are in a style and so are teenagers jumping around on the dance floor Styles encompass a wide range in scope scale and level Style can characterize strategic actors as well as whole social contexts Hence a style is scaleinvariantor scalefree in the sense laid out by Abbott 2001 Like those Russian dolls that are identical in shape pattern and colors but differ slightly in size style offers a nested structure Thus for example a person in the ordinary sense that we avoided in chapter 1 is a style Pithy conversation may establish a style such as in a Paris salon or a research discussion group Successful mobilization toward a political end embodies a style perhaps differently around the cause of human rights than around the cause of ethnic autonomy Expertise attends to style as interpretive tone as will be argued later however rational choice theory seeks mobilization of rationality as the style for expertise My initial examination of sensibility will end with two largescale studies of womankind Then we turn to three studies of style emergent in networks of commerce Next comes an examination of person as style followed by the argument that rationality itself is a style These discussions motivate then taking up general ways to appraise and measure social spaces and profiles One main focus is how best to observe styles personal or other and to locate them in some ecology The divisions in this chapterand the bookare not conventional The numerous and diverse examples spread across conventional framings of social life as found in textbooks In particular this central chapter is cited in the next three chapters as well as having been foreshadowed in the previous three chapters Indeed analysis in each chapter of the book must given the nature of social process presuppose and draw on findings from the others as I show visually in the final chapter see figure 81 Not surprisingly many studies to be reported will be large and heterogeneous not neat and focused After thus spelling out approaches to recognizing ecology I will explicate more general selves and also communities as styles Then we return to observing emergence and change with three further studies of style where the cultural is intertwined with the social To cap that I will offer a general proposition The whole range of studies we will cover flesh out the initial definition of style and illustrate how style and control interpenetrate which is the main topic of the final section 41 Sensibility Style is a profile of the commingling of network relations and discursive processes across switchings that result from and also shift situations Temporality in the social world comes from style from the rhythm of profiles of switchings This temporality is not time as we ordinarily think of it in the count of hours and minutes displayed by a clock Rather it is closer to a tempo in music A conversation can exhibit a style which may be bland as in making talk or just chatting or may instead track and trigger vivid episodes through domains across some situation As in music a conversation can have a specific rhythm a specific sensibility that is the signature of its style similarly the issue is whether a field of recurring discourse sustains and reproduces a distinctive rhythm of social interaction Relevant here at a completely different scope is Randall Collinss magisterial survey 2002 of philosophies social constitutions spread over millennia and continents His study documents successions of what I have termed council disciplines as well as cross sections within those disciplines but not an overall style Instead he shows that a sensibility establishes itself again and again which one now sees from a long perspective as the hallmark of a civilization I shall analyze styles with much the same tools whatever the scale size or level of living whether in a large organization or in the playground example from chapter 1 At the micro end Bourdieus habitusas a matrix of perceptions appreciations and actions Bourdieu 1977 p 83is analogous to style as it also predetermines interpretive tone Conversely styles can shape bend and mark bodies as well Styles can be thought of as specification of how individuals live their lives in this view individual lives emerge through an ongoing process of combining understandings of situations with sets of practices arrayed across lives embedded in social networks Bourdieu proposes habitus as the signature of a person 1996ab whereas later in this chapter I argue for person as style At the macro end toward the close of this chapter come hieratic styles observed across fields of army church and colonialism Such styles encompass vast numbers of networks across disciplines and unite identities around unique sensibilities1 Bourdieus fieldin some of its aspectsrefers to manifestations of such broad style Styles both couple and decouple actions among networkpopulations that overlap in physical space Styles are set off by and thus ap1 However the concept of style is not to be confounded with the concepts of habitus and field Although all three concepts invoke orderliness of perception and action they differ in the way they qualify orderliness First styleat its core a stochastic concept is in itself a source of innovation and change Style itself does not change easily yet offers options for getting action Second stylebecause it is a scalefree conceptcan be applied to a whole range of contexts and social formations that the other two concepts do not allow taking into account pear in the midst of complex overlappings and switchings of networks and also of disciplines In that sense they are analogous to identities But styles are envelopes as I specify in a later section created from innumerable attempts at control by identitiesenvelopes that once created limit and funnel control The environment in which they appear is one of stochastic incident and process among actors In the next section I present exemplifications of sensibility I begin by examining the potential for style in the small and everyday events such as conversation after which I turn to organizational settings to show how interpretive tone shapes and is shaped around expertise The section winds up with two evocations of massive yet focused change in sensibility in the historical record of the United States 411 Style as Texture of Social Dynamics The contexts in which identities lie are shaped by their attempts at control whichif the identities survive and concatenatecan be represented by ties in networks along with disciplines What matters is the texture of the process as reflected in the context created from contingencies surrounding their disciplines and ties Social organization becomes heterogeneous not tidy as it survives stochastic flows of contingencies All this leads to fluid social contexts only some of which prove able to reproduce themselves as styles Identities are differently positioned in one and another network so that multiple networks must impact each other Two actors who are joined by one type of tie may have no connections in other types or they may have connections there to third parties that appear complementary to their joint tie Thus in chapter 2 I investigated impacts from multiple networks deploying the concept of structural equivalence which gives little purchase with a single network I argue that most of the realtime dynamics play out for the overall network of multiplex or general ties as in Small World searching So the continuing texture of social living appears in styles Style is a generalization of network that traces along strings of ties and may reshape the connectivity aspects of network Like stories style characterizes the rich phenomenological texture the fabric of lived experience of identities among populations in networks and disciplines Style may supply a particular idiom for the orderings by values in disciplines but it also presupposes and specifies some complex layering of networks within and across levels Networks made up alongside disciplines are a new realization of network at a new level of analysis Here levels characterize the different kinds of control patterns in social process