17,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Geisteswissenschaft
- Sprache: Englisch
Social structure is arguably the central concept of sociology, and in recent years a much wider public has taken up with fresh vigor the sociological idea that persistent inequalities are rooted in social structures. Yet there seem to be as many definitions of the term as there are sociologists, and we often struggle to articulate accessible yet precise accounts of structures that can guide empirical research and other kinds of action.
Jonathan Eastwood offers a set of pragmatic strategies for thinking about social structures, emphasizing ways in which we can approach them as complex lacings of relationships, representations, and rules. He then teases out a variety of implications of these strategies for qualitative and quantitative research, the analysis of social problems, and the implementation of social policies. Written for advanced undergraduate and graduate students as well as fellow scholars, this insightful book contributes to our understanding of this fundamental and dynamic ingredient of social life.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 424
Veröffentlichungsjahr: 2024
Ähnliche
CONTENTS
Cover
Table of Contents
Quote
Title Page
Copyright
Acknowledgments
Preface
Notes
Part I: The Basic Ideas
Notes
1 The Three Rs
Relationships
Representations
Rules
Basic actions that interdependently give rise to the three Rs
Are relationships (or representations, or rules) fundamental?
What some other social science words mean when translated into the terms of the three Rs
On the reality of relationships, representations, and rules
Notes
2 Other Meanings of Social Structure
An incomplete menagerie of definitions and characterizations
Cognitive associations and potential sources of confusion
Are structures “real” or are they “models”?
What do we need from “social structure”?
Notes
3 Interlacings and Intralacings
A brief discussion of intralacings
The nature of interlacings
Some possibly common types of interlacing
Basic actions that can interdependently give rise to interlacings
Common kinds of complex structures made of interlaced, simpler ones: roles and organizations
How can we study interlacings?
Toward a proliferation of (models of) intra- and interlacings as a strategy for understanding sociological mechanisms
Notes
Part II: Working with Social Structures
Embracing the complexity
Avoiding the complexity: shining a small flashlight in a vast, dark space
Notes
4 Explaining with Structures
Questioning structural explanations
Thinking about mechanisms
From theories about mechanisms to data about mechanisms
Qualitative research, abductive inference, and social structural explanations
Notes
5 Data Structures and Social Structures
What regression models think
Conventional statistical models of contexts
Individuals, contexts, and structures
Models of social structures
Data structures as social structures
Notes
6 Social Structures and Social Policies
The very basics of causal identification
Social structural thinking in work at the boundaries of policy analysis
Social structures and “external validity”
Social structure and policy implementation
Theories of contextual heterogeneity of policy effects
Notes
Conclusion
What has the book argued?
Some implications beyond sociology
Looking forward
Notes
Questions for Reflection and Discussion
References
Index
End User License Agreement
List of Illustrations
Part I
Figure 0.1
Chapter 1
Figure 1.1
Figure 1.2
Figure 1.3
Chapter 2
Figure 2.1
Chapter 3
Figure 3.1
Chapter 5
Figure 5.1
Chapter 6
Figure 6.1
Guide
Cover
Table of Contents
Quote
Title Page
Copyright
Acknowledgments
Preface
Begin Reading
Conclusion
Questions for Reflection and Discussion
References
Index
End User License Agreement
Pages
ii
iii
iv
vi
vii
viii
ix
x
xi
1
2
3
4
5
6
7
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
“… there can be no question that a few clear ideas are worth more than many confused ones.”
Charles Sanders Peirce
Social Structure
Relationships, Representations, and Rules
JONATHAN EASTWOOD
polity
Copyright © Jonathan Eastwood 2025
The right of Jonathan Eastwood to be identified as Author of this Work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988.
First published in 2025 by Polity Press
Polity Press65 Bridge StreetCambridge CB2 1UR, UK
Polity Press111 River StreetHoboken, NJ 07030, USA
All rights reserved. Except for the quotation of short passages for the purpose of criticism and review, no part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the publisher.
ISBN-13: 978-1-5095-6194-0
A catalogue record for this book is available from the British Library.
Library of Congress Control Number: 2024939641
The publisher has used its best endeavours to ensure that the URLs for external websites referred to in this book are correct and active at the time of going to press. However, the publisher has no responsibility for the websites and can make no guarantee that a site will remain live or that the content is or will remain appropriate.
Every effort has been made to trace all copyright holders, but if any have been overlooked the publisher will be pleased to include any necessary credits in any subsequent reprint or edition.
For further information on Polity, visit our website:politybooks.com
Acknowledgments
I have many people to thank for the ways in which they have contributed to this work, indirectly or directly. These include my students, my colleagues, my own teachers, my favorite coffee place, external reviewers and editors, and my family.
Students played important roles in thinking through these ideas with me. The book itself began with an advanced undergraduate seminar on social theory, taught in Fall 2020. Students in that course asked great questions, prompting me to try more clearly to articulate the idea of social structure, central to so many visions of sociology, and helping me realize that my own understanding of the idea, or perhaps our shared understanding of it in sociology, was vague and contradictory.
Following the conclusion of that course, one student, Claire Smith, spent a month working with me to coauthor a paper, the core ideas of which form the basis for Chapters 1 and 3 here, which she presented virtually at the 2021 American Sociological Association (ASA) meeting as part of a Theory Section roundtable. Beyond that initial collaboration, Claire and I had a number of subsequent conversations about the project, and she provided detailed and helpful feedback about a rough draft of the book manuscript. This book might never have been written without that month of early work, and I am grateful to Claire for her many contributions.
In Fall 2022, following a version of the same social theory course, Meg Graham, another student, read an early draft of the book. In addition to providing detailed comments on the manuscript, she worked with me for several months, in particular thinking through the materials on abductive theorizing and qualitative research in Chapter 4, reading and discussing with me the ethnographies considered there and sharing insights about them and the book’s argument, for which I am grateful. Several other students – Teresa Aires, Tanajia Moye-Green, Allie Stankewich, Mansi Tripathi, and Jamie Winslett – also provided helpful comments on an early manuscript draft, and I thank them for their help. Some of the ideas in Chapters 4–6 were first worked out in memos I wrote for students in my “Data Science Tools for Social Policy” course, taught in the winter of 2022. I thank the students in that course for their patience and insight. A near-final version of the book was read by students in the winter 2024 iteration of the same social theory course in which this all began. They made helpful comments and suggestions about the text (I am especially grateful to Guangpu Chen, Trip Wright, and Zhengyi Ke for thoughtful feedback) and also showed me that motivated undergraduates could make good use of, and expand on, the ideas contained here.
Several colleagues, too, played important roles. In particular, Howard Pickett and Chris Handy were generous with their thoughts about my ideas, discussing them with me at length, reading some draft materials, and raising important questions. In addition, Howard taught a paper version of the ideas to several classes, and his feedback about how students responded to the readings was very helpful. He also organized a discussion with the Shepherd Program Faculty Cohort, and I benefited greatly from feedback on an extended outline of the book plan from many colleagues there. Jeff Kosky is always willing to talk about ideas and made several timely and important suggestions. So too was the late Tyler Dickovick, and though he passed away before work on this book began, I’d never have written it without years of thinking and talking with him about social science and how best to teach it. Francesco Duina provided invaluable feedback about the first paper version of the ideas. In many ways, Chapter 2, and in particular the distinctions drawn there between social structure and form, grew out of thinking about his comments. Hafsa Oubou co-taught with me a version of the manuscript. An early version of Chapter 2 was presented at one of the Social Theory Panels (“Structures, Levels, and Time”) at the ASA meetings in Los Angeles in August 2022, and I appreciate the audience feedback I received there.
Colleagues in the Department of Sociology and Anthropology at Washington and Lee have been wonderful in so many ways, in terms of both sharing ideas and just being a congenial community. Above all, I want to recognize the late Ken White, who spent countless hours with me discussing topics in social theory, among many other things, and who was a model of intellectual generosity and selflessness, clear thinking, and curious engagement with the core questions of social theory.
I thank my university for its support of this work, including several summer Lenfest grants and a semester’s sabbatical leave.
Our own teachers, of course, shape our thinking well beyond the moment we leave their classrooms. I benefited from many great ones. Above all, I am thankful to Liah Greenfeld, the most brilliant person I have ever met, and the most exceptionally dedicated and influential teacher. She showed me the fundamental questions of sociology and her own formidable answers to some of them. She taught me how to think about social life. She also taught me how to think with students. My ideas have traveled a bit since those years, as should happen, but I hope she is happy with these results, and recognizes in them an effort to answer the call. I would like also to acknowledge several of my teachers who have now passed on and who taught me many important things that find their way into the pages that follow: Michael Martin, John Stone, and especially Chuck Lindholm.
Thanks also to the folks at Roadmap CoffeeWorks, where substantial portions of this book were written, fueled by masterfully roasted and prepared brews.
Jonathan Skerrett and Karina Jákupsdóttir at Polity were wonderful to work with and I’m grateful for their feedback on this project at multiple stages in its development. I would like also to thank the anonymous peer reviewers – those that reviewed the original proposal and those that reviewed the full manuscript – for their very helpful comments. I believe these improved the book considerably.
My parents, Randy and Marilyn, and my brother Dave have been consistently supportive and kind. My children – Gabi, Sammy, and Alex – are my joy. I appreciate their patience and kindness and how much fun they are. In addition, each has discussed ideas from the book with me, in their own ways, and done so insightfully. They are amazing kids. Above all, my wife Mari deserves my thanks. She has been with me on this project from the beginning, talking it through over and over, endlessly patient, contributing both intellectually and personally in too many ways to count.
Preface
Nowadays, references to social structure are common, a welcome change for sociologists. For years, we have argued that individuals’ actions cannot be viewed in isolation. People make choices in contexts that unfold in time. Their choices feed back onto those contexts, potentially reshaping them.1 Social life is not just the individuals who make it, but is a set of “things” that emerge from those individuals’ interactions. As one of our field’s founders taught us, “we must treat social facts as things”2 – even if it’s not always clear what kinds of things they are.
To invoke social structure, though, is not merely to acknowledge that interactions scale up to make social facts. Rather, the idea of social structure is used to argue that people face constraints.3 At root, this is often about responsibility.4 We deploy the concept of structure because we don’t want those who are disadvantaged to be blamed for their disadvantage. We put those two sets of ideas together: choices are conditioned by features of contexts and thus people are not fully responsible for their fates.
Relatedly, popular discussion reflects a growing awareness that our societies are characterized by pronounced inequalities.5 People from many walks of life have a new or renewed sense that such inequalities are rooted in structures and systems.6 These two terms – structures and systems – are often used interchangeably, though we have good reason to think of them as being distinct, but related (on which, more in the pages that follow). Sociology contributes to this in important ways. The field is a key source of ideas about social structures, and sociologists have done empirical work that illuminates the workings of social structures in particular contexts as well as structural sources of common kinds of inequality.7
On the other hand, sociology arguably has not yet put together a practical and accessible set of strategies for thinking about social structures. That’s not for lack of insight or effort. There are difficulties familiar to social scientists, challenges we face with respect to almost any subject upon which we focus. There is no periodic table of the sociological elements.8 We cannot just look at the social world and induce unambiguously from it the categories through which we think about it. Instead, we have to develop most of our concepts pragmatically, judging them by what they do and don’t allow us to see, the questions and potential answers they afford, and those they obscure.9 Different thinkers make different choices in these respects, leading to the potential for miscommunication. Many of the most insightful ideas presuppose different terminologies. Resulting from this, we have some distinctions, like structure versus culture, that on reflection make little sense. Polysemy is great for poetry but less useful for clear and consistent thinking about social life.10
The goal of this book is selectively to assemble some pieces of the strategies developed by sociologists and other social scientists into an accessible, and hopefully coherent, whole: a provisional set of strategies for thinking about social structures.11 It’s provisional because the work of understanding social structures is not done, and likely never will be.12 Since the structures humans build interdependently are constantly changing in form, sociology probably is a science that has “eternal youth.”13
What is presented here is also hardly the only possible framework.14 It may not be the best one, and it won’t be the last one, but at least it strives for clarity and intelligibility rather than trying to dazzle its readers.15 My hope is that students, general readers, scholars in adjacent fields, and even professional sociologists will find some of those strategies for thinking about social structures helpful. Helpful for what? First, for making sense of what we might mean when we say that something is social structural. Then, for trying to think about how complex structures are built, and might be changed, by focusing on their parts and how these parts fit together. The ways in which social structures are built need to be better understood if we wish to unmake or remake them. Finally, for clarifying how theories of structure are involved when we build new explanations, examine data, and evaluate policies: bread and butter tasks of social science.
The idea for this book emerged from working with students. I teach at a liberal arts college with very strong undergraduates. One of the things I like best about my job is that I get to think with smart, engaged young people who haven’t yet been fully professionalized into the role of the social scientist. Arguments about ideas are, for them, mostly just arguments about ideas. The stakes are not the same as those held by graduate students and professors who may already be publicly identified with, or otherwise personally invested in, particular theoretical commitments. It is easier to imagine something different with such students. They also seem less embarrassed than those others might be to (a) think simply and slowly and (b) acknowledge what we don’t understand. We should all try to be more like them.
Above all, my students seem readily to recognize that, however brilliant the theorists whom we have read together, our field has no clear and accessible consensus account of its central concept.16 The project began with my attempts to explain theories of social structure to my students. As I thought with them, I discovered that those theories didn’t fully make sense to me.17 Working with my students, I tried to get these ideas to make sense. This project is an attempt to work out further and express those attempts at sense-making, an incomplete task, but one that seems far enough along to put into print. My hope is that the results will be congenial for students and colleagues both. I beg the patience of colleagues when I explain for students’ sake things that colleagues may already know. I beg the patience of students with passages that might more directly address colleagues’ questions and concerns, and I encourage them to persist with and through those passages.
Social structures are complex and often hidden, thus poorly understood. Faced with their complexity and partial hiddenness, we may be tempted to mystify them, as others have done. But mystification is the opposite of scientific understanding.18 The most basic principle: we won’t understand social structures until we can break our accounts of them down into stories about things that are more easily grasped and that our conversational partners will recognize as real and intelligible. Because they are complex, we need to think about them in simple ways.
Notes
1.
This has been emphasized by many sociologists, one of the clearest being Robert Merton (1968), whose arguments were well summarized by Stinchcombe (1975).
2.
This is, as sociologists know, the fundamental maxim from Émile Durkheim’s
The Rules of Sociological Method
(1982 [1895]).
3.
See, for example, the discussion of “What Is Structural Inequality?” on the webpage of the University of Pennsylvania’s Center for High Impact Philanthropy:
https://www.impact.upenn.edu/what-is-structural-inequality/
.
4.
This point is well made in, among other places, Iris Marion Young’s
Responsibility for Justice
(2011).
5.
Examples are many, and include the widespread coverage of Thomas Piketty’s
Capital in the Twenty-First Century
(2014), the focused attention on protests against persistent racial inequality after the killing of George Floyd in 2020, and the #MeToo movement.
6.
For one of hundreds of possible examples, see Karen Zraick’s 2021
New York Times
article, “Racism is Declared a Public Health Crisis in New York City.” For an example from recent sociology, see Homan et al., 2021.
7.
At the same time, as Shelby (2016) argues, ultimately such questions are about justice, and require intellectual tools beyond those of social science. Social science can contribute to those efforts by providing cogent concepts, empirical descriptions, and explanations.
8.
I take this to be an implication of Weber (1949).
9.
Collier and Adcock (1999).
10.
Merton (1975) saw the plurality of conceptions of structure as a virtue. Levine (1985) extolled the more general virtues of “ambiguity.” Clearly, ambiguity has its uses, and in the human sciences we may never fully escape it, but too much ambiguity about core concepts in social science makes it easy to talk past each other – or ourselves! I thank Xiangyu Ma for bringing Levine’s beautiful book to my attention.
11.
It’s not a systematic or historical study of those pieces, but instead a rather idiosyncratic selection based on what seems most useful to me. For an accessible and short history, see López and Scott (2000). For an alternative set of basic conceptual tools, see Crothers (1996: 82–125).
12.
For some reasons why, see Taylor (1971).
13.
This is how Weber describes it (1949: 104).
14.
For an excellent recent alternative, see Rawlings et al. (2023).
15.
A model in this regard is Rojas (2017), which I recommend to students interested in an accessible overview of contemporary sociological theory.
16.
Jonathan Turner (2013) has pointed this out as well. It is indeed, as Charles Crothers (1996: 21) describes it, an “absent concept,” often used but seldom clearly defined.
17.
Making sense of the social world – even while we do not simply accept or reconstruct the meanings people assign to things and consider our task complete – is what I take social science to be for; see Taylor (1971: 16).
18.
On this point, see Hedström (2005), though we may not fully agree on which theories count as mystifying.
Part I:The Basic Ideas
There are two words in the phrase social structure. Let’s take the second one first.
A structure is a thing that is built. The word has had this meaning since the Romans.1 In Latin, the word is structura, a building. Imagine an ancient Roman building, made of brick. A structure, it is constructed. It enables and constrains the people in its midst.2 Some enablements that such a structure affords: people can find shelter in it; they can congregate there; it can serve as a focal point as they try to coordinate action (“let’s meet at the forum, by the basilica!”). Some constraints: they cannot walk through its walls; some cannot enter; some may be imprisoned there; it is not easily moved.
As humans do, we have extended this term out through shades of application and metaphor, and we speak also of some structures that were not built in the strict sense of the word.3 Think, for example, of molecular structure, or physiological structures, or geophysical structures. Perhaps the unspoken cognitive linkage here is that some have thought of these things as built too, though not by human hands. Or perhaps our sense of structures as built disappeared from our minds’ view: we came to think of these as structures because we saw in them things that reminded us of the structures we made. The word resonated when we used it to describe them. When we ask about the structure of a molecule, or a skeleton, or a geological formation, we are asking about something similar to what we see when we look at a building. It is made up out of parts, assembled through some process, however complex. We want to know how it is made, what it is made of, how its parts fit together.
Different kinds of structures are made in different kinds of ways, of different kinds of things, of different parts assembled in different kinds of combinations. What distinguishes social structures is what they are made of.
A building is made through a social process. People construct it. But it is made of straw, wood, or stone. The principles that govern its persistence have much to do with what it is made of. Yes, social processes can impinge on this too (we can huff and puff and some buildings will fall), but if we think about a building, in an immediate sense we see that its persistence is shaped by its material features. Stone endures, wood slowly decays, straw properly assembled is much more persistent than old stories suggest.
Figure 0.1
A building, though made through a social process, is not made of a social process. Social structures are. This is the “social” in social structure. Social here means interdependent actions. Social structures are things that are made of interdependent actions of multiple people.4
This concept of interdependent actions is central and deserves our attention right from the start. Let us try to understand it with a simple example. Think of a portion of a friendship network, looking something like what we see in Figure 0.1. You can see three groups here, joined together by a couple of ties (J–E and E–B) that make this a single component. The social structure, in the simplest sense, is this pattern of ties. This structure is, quite literally, made of interdependent actions. It is nothing else besides the ongoing decisions on the part of the individuals involved to affiliate (or not affiliate) with each other, yielding the pattern represented in the graph. That’s the actions part. The “interdependent” part is that people’s actions depend on each other’s actions. The potential for J and G to form a tie depends, in part, on their shared tie to E, to pick one of many possible examples.5 Another example: we might expect a tie between A and C to form, since they share common ties. For network sociologists, this is elementary.
The point, again, is that the relational structure in the figure is made of those actions and the actions are interdependent. Even if we know nothing about A and C as individuals, we can see that A and C are likely to form a tie, because this likelihood is entirely due to their embeddedness in patterns of ties that depend on the actions of B and D as well as their own. B and D, presumably, are not choosing their connections based on the implications of these connections for the possibility of A forming a tie with C.
You can see this idea in a different way if you consider K’s reachability from C (or from anyone else in the graph). K can transmit information to C even without knowing C or intending to make that transmission. This depends on affiliation choices made by others (above all by B, E, and J). B and E, presumably, are not trying to make C reachable from K. They are pursuing their own objectives, whatever those may be (e.g., spending time with people whose company they enjoy; trying to build connections from which they can later benefit).
The structure emerges as people’s individual-level actions. To understand the structure, we need theories about those actions. This means trying to understand what people want, what they can see and know as they pursue what they want, and the pattern of connections they have that might influence their decisions.
Some social theorists object to emphasizing choice, as this perspective may seem to do, and they make some good points.6 At a most general level, does “choice” imply an absence of constraints? Does it assume that all or most of human behavior is deliberate and purposeful when much may actually be due to habit, imitation, and implicit rule-following? Does it imply that individuals are rational and possess full information about their surroundings, selecting optimal strategies given their choice sets? Does it ignore what cognitive science has shown us about “dual processes” in various cognitive domains, including those of which we are not always fully conscious (since the idea of “choice” in ordinary language seems to imply conscious and deliberative, “chosen” decision-making)?7
For some proponents of theories of action, the answer to one or more of these questions is “yes.” However, to fellow social scientists (including social science students) I recommend agnosticism with respect to philosophical questions about choice. Too often, theories about structure get hung up on the “structure-agency problem,” sometimes emphasizing the “duality” of structure and individual behavior.8 The problem has two sides, only one of which I take on. The first side is how free to choose, within structural constraints, people might actually be. But because the notion of freedom is a philosophical one, about which there have been intractable debates, and because none of the arguments of this book depends directly on any particular conception of free choice, I leave this side of the structure-agency problem mostly unaddressed. Indeed, the greatest difficulties in that side of the question concern problems with the notion of choice itself and are not specific to the relationship between action and social structure.9 Those questions won’t be resolved any time soon, and practical strategies for thinking clearly about social structures shouldn’t have to wait for the resolution of (possibly irresolvable) philosophical problems. If those philosophical problems trouble you, and doubts about action distract you, just plug in “behavior” whenever the text says “action” and the substitution should serve reasonably well.
The other side of the structure-agency problem is about how individual actions can, in the aggregate, scale up into structures that then feed back onto such actions. The perspective of this book is that models of interdependent actions as developed by social scientists already contain a viable answer to this question, and that it just needs to be drawn out and linked to clear ideas about social structure, one of this book’s main goals. In other words, other thinkers have already demystified this side of the question, and in the first half of this book I try to link their answers to a conceptual scaffolding about social structures themselves in a way that I hope is clarifying for colleagues and students alike.10
Surprisingly, these models can sometimes be more useful when they are less psychologically realistic, though that statement needs to be qualified. The psychology of behavior is complex and mostly hidden. A given act may have many motivations even within a single person, or it may spring from something beyond motivation, like a reflex.11 Comparing across people, the question grows still more difficult. A behavior may look the same from the outside but be very different on the inside, a strategic choice for one person, an imitation for another, a habit for a third. In the broader social world around us, our ability to sort one from the other is limited and uncertain. Thankfully, for the analysis of social structures, that’s usually not what we need. Indeed, unnecessary psychological realism is likely to interfere with our efforts to think about social structures in the simplest possible terms.
All that our notion of interdependent action requires in terms of initial assumptions is the following, even it is only “methodological” and provisional. First, I assume people prefer some outcomes to others. They may want to make money, have power, help other people, please God, or enjoy life, among many other possible goals. Second, I assume that, given what they believe about the world and about the preferences of others, they generally do what they think is most likely to bring about their preferred outcomes.12
These are, again, provisional assumptions, not a theoretical insistence that people always do that. People sometimes behave irrationally, doing things that even they know are likely to bring about outcomes they don’t prefer. And often people aren’t thinking about what they’re doing at all. But the perspective of this book, like that of the traditions of social thought on which it rests, is that the assumptions above will get us pretty far, and that when they break down, we can then focus attention on what is causing them to do so. Habitual or imitative action might produce patterns that can’t be explained on these assumptions.13 The goal of such work, though, is not psychological realism but rather the explanation of social structural patterns.14 The overarching principle here is to keep our assumptions about individuals as simple as possible for as long as possible. We will think of “action” in this way.
The contemporary social sciences have developed techniques better to understand interdependent actions.15 In other words, mathematical or computational models can be used to see the sometimes surprising patterns that emerge through interdependence. One of the most important kinds of models comes from game theory. Another is agent-based modeling, a kind of computer simulation.16 In this book I do not try to build up formal theories of social structure of the kind that should be possible using such tools. We should not jump right to formal models when thinking about something as complex and conceptually ambiguous as social structure. That is, we need first to understand what we are trying to formalize before we can formalize it.17
For now, let’s think about simple examples in order to more fully grasp the idea of interdependent actions. Game theory is arguably the simplest approach. It tries to render abstractly the strategic situations that people face so that we can better understand them. It can be divided into games of cooperation, which mainly involve solving problems of coordination (getting people on the same page) and games of conflict (in which interests do not align).18
Some kinds of games, like the classic prisoner’s dilemma (a game of conflict), involve just two players. In a PD, with payoff matrices like the one found below, the equilibrium is defect-defect.19 The game represents a particular type of situation: one in which two people are unlikely to cooperate jointly because each fears the other will be selfish and defect.
P2 – Cooperate
P2 – Defec
P1 – Cooperate
4, 4
0, 5
P1 – Defect
5, 0
2, 2
For those new to payoff matrices, here’s how to read it: the rows are strategies for “Player 1.” The columns are strategies for “Player 2.” The cells are combinations of payoffs for the players when strategies coincide. The first, before the comma, is for Player 1, and the second is for Player 2. In other words, if P1 and P2 both cooperate, they each get a payoff of 4. But if P1 cooperates and P2 defects, P1 gets 0 and P2 gets 5.
There is nothing new here, a very simple illustration of a modeling approach that is well known to social scientists, but the example helps show what I mean by interdependent action. The payoffs to each player depend on what the other player does. If Player 1 understands the form of the game and behaves rationally, they will defect. Player 2 will too. So far, then, we see a simple example of interdependent action, but we don’t yet see a social structure. There is nothing persistent here that is “made of” interdependent action.
Structures endure and they span individuals. There is no obvious threshold of endurance beyond which something becomes structural, no precise length of time. But a one-off interaction clearly is not a social structure. To be structural, a pattern must persist at least long enough meaningfully to enable and constrain subsequent actions (i.e., to have become “a thing”). Likewise, there is no obvious threshold for how individual-spanning something has to be in order to be structural. Some have emphasized social life as possible in pairs. Others suggest triads.20 But what we ordinarily mean when referring to social structures generally involves more people than that, even if social structures do not need to span an entire society and very often do not.
This same payoff matrix can, if we consider it a game played simultaneously by many players, represent social structure. Consider the persistence of an authoritarian regime.21 One way to think about this persistence is that, at every moment, the population acquiesces to the regime by not revolting.22 That shared acquiescence is social structural: it is persistent and made of interdependent action.23
Revolt
Acquiesce
Revolt
4, 4
0, 5
Acquiesce
5, 0
2, 2
The payoff matrix is unchanged, but we are now imagining it as an iterated, n-player prisoner’s dilemma in which cooperation is joining a revolt and defection is acquiescence. If all players share the set of preferences represented here and behave rationally, none will revolt. That’s not because everyone wants shared acquiescence. On the contrary, according to this simplistic model, everyone would prefer a successful revolt. But nobody wants to revolt while the rest acquiesce, and everyone would prefer to let the others do the revolting. In this way, the authoritarian regime persists.
The survival of actual authoritarian regimes is not so simple, and social science theories almost never make the naive assumptions of this illustration. First, this model notably overstates the uniformity of preferences in any given population. There are some people who always want to revolt, some who never do, and many in between. More sophisticated models of collective action stipulate that people’s preferences for revolt themselves depend on thresholds for the participation rates of others, among other factors.24 I might be ready to revolt if at least 25 percent of the rest will do so. You might want to wait until participation reaches 50 percent. Some will wait to be among the very last to join the cause.
The example helps us see that basic social structural configurations like those of a persistent authoritarian regime are all about interdependent actions. They are made of the ongoing, interdependent actions (including not acting!) of the individuals involved in them. That is likewise true about the dynamics of structures: structures change through processes of interdependent actions. They persist when people coordinate on actions that carry them forward. They change when people coordinate on new joint actions. They fall apart if and when actions become uncoordinated.
So far, this simple example implicitly imagines that everyone is connected to everyone else, that they can observe each other.25 But in the real world the information we have about rates of participation depends on our network positions, and in large-scale societies we don’t all know each other. We make inferences about the rate of others’ participation based on what we can see. What we can see depends on who we know. The patterns in our relational networks can make a big difference.26
Look back at the simple friendship network depicted in Figure 0.1. Imagine that everyone wants to participate in the revolt if at least 1/3 of the others do. Imagine that, for whatever reason, M joins the revolt. On our assumptions, K and L will then join as well, because at least 1/3 of those they know have done so. J will then join too. But the cascade will stop there. E will not join, because only 1/5 of E’s ties will have done so. If, however, E had ties to K, L, and M, the situation would be different.
Finally, the regime might shape the probabilities of participation in the revolt through how it establishes and enforces rules that shape the payoffs of participation.27 Think of rules that impose jail time on protestors, on the one hand, but also those that promote excessive force that makes people hate the regime, becoming more willing to take risks to overturn it. The psychology of responses to such rules is not social structural, but the rules, we shall see, are.
This illustrates very simply one of the central ideas of this book: that multiple basic kinds of social structure can be laced together in complex ways. This example shows us only a very simple one. In the real world, the inter- and intralacing of different kinds of structures is generally much more complex.
How can we deal with such complexity? Again, we can only understand complex things by breaking them down into simple parts.
Notes
1.
“Structure,” Oxford English Dictionary Online,
https:www.oed.com
. On the history of the concept, see also López and Scott (2000: 7–8) and Williams (1976: 253–9).
2.
This language of enablements and constraints as features of structure was emphasized in the influential formulations of Anthony Giddens and has been picked up by many others. See, for example, Giddens (1984) and Young (2011).
3.
“Structure” may have entered sociology via the old biological distinction of structure and function (Martin 2009: ix), but the structural-functionalist conception along these lines leads to a dead end.
4.
Many social thinkers have stressed the importance of interdependent actions in social life. Perhaps the clearest such writer is Thomas Schelling (2006 [1978]), who didn’t use the concept of “social structure” as such, but who obviously had in mind patterns that are properly structural, such as residential segregation, and who showed how we can build simple models to see how interdependent actions can scale up in surprising ways. Analytical sociologists like Hedström (2005) and Hedström and Bearman (2009) pick up these ideas as well. Many formal models of micro-level behavior across the social sciences are rooted in similar ideas.
5.
As sociologists reading this know well, most of our models of network structure depend on these sorts of ideas. When we try to understand the structure of a network, we often start with a random graph and then add theoretical ideas to our model: for example, that people will disproportionately form ties with those with whom they share characteristics, or that triangles will tend to close. See, for example, Bearman et al. (2004), as well as the introductory essays in Hedström and Bearman (2009).
6.
Since, for many, what makes something an action is that it is chosen or motivated. See, for example, Weber’s discussion in “Basic Sociological Terms” (1978: 4–26).
7.
For a start on such arguments, see Vaisey (2009). More generally, see Gross (2009). As Gross et al. (2023: 15–17) discuss, some sociologists turn to pragmatism to avoid models that emphasize choice. As the reader will see, I choose a somewhat different pragmatism, and stick with “action” for as long as possible.
8.
Giddens (1984); Sewell (1992). See also helpful elaboration and development of Giddens’s perspective in Stones (2005). Such works are insightful, classics of social theory. My point is just that the conversation seems to get stuck on these points so we might instead try to bypass them, at least temporarily, if possible.
9.
For an overview, see O’Connor and Franklin (2022).
10.
See, for example, Coleman (1990); Schelling (2006 [1978]), Hedström (2005). Of course, some “structuralists” are against all such efforts to think of social structures as products of individual action. For a pugnacious and thought-provoking version of this alternative view, see Mayhew (1980, 1981).
11.
For classical sociology à la Weber, as articulated in “Basic Sociological Terms” from
Economy and Society
(1978), action is, again, meaningful behavior, with “meanings” given by motives. But Weber himself acknowledged that lots of actual behavior in the world isn’t directly and consciously motivated or fully meaningful in this sense.
12.
These are also the basic assumptions needed to use game theory to study social life, as discussed by Gintis (2009). They are also easily reconcilable with Hedström’s (2005) “Beliefs, Desires, Opportunities” model.
13.
For example, consider the classic research by John Meyer and collaborators that emphasizes institutional copying among many kinds of organizations. There is a large literature here, but to start, see Meyer and Rowan (1977) and Meyer et al. (1997).
14.
“Structural decoupling” in the case of Meyer’s work.
15.
A foundational text here is Schelling (2006 [1978]).
16.
Macy and Flache (2009); Railsback and Grimm (2019).
17.
In other words, if we don’t know what we are trying to represent with our models, they’re unlikely to be very helpful.
18.
For a good introduction, see Gibbons (1992). For an accessible introduction to game-theoretical thinking more generally, but without technical details, see Elster (2007) as well as Chwe (2013).
19.
By “equilibrium,” game theorists mean a set of joint actions where players’ “best responses” converge. See Gibbons (1992) and Gintis (2009: 35–6). In other words, it’s what people will do if they’re rational, understand the game, and if the payoff matrix accurately reflects everyone’s preferences.
20.
See discussion in Simmel (1950) as well as Berger and Luckmann (1966) and Kadushin (2012: 35–6).
21.
Many scholars have used this kind of thinking to understand authoritarian persistence. See, for example, Granovetter (1978), Lichbach (1998), and Kuran (1991), as well as discussion in Elster (2007).
22.
In simple terms, the population is following a rule, which exists as a structure precisely because the population coincides in following it.
23.
For this to make sense, we need to consider
not
doing anything to also be an action.
24.
This is widely recognized. One of the original sources of the idea is Granovetter (1978).
25.
In other words, they are members of a clique in the graph-theoretical sense, a “maximally-connected subgraph.” On cliques and other measures of groups in networks, see Wasserman and Faust (1994).
26.
There is a large literature on network structure and behaviorial diffusion, including behaviors like rebellion and acquiescence. See, for example, Marwell et al. (1988); Gould (1991, 1993); as well as Centola (2018).
27.
See, for example, Ginkel and Smith (1999).
1The Three Rs
In this book I argue that the most basic forms of social structure can be thought of in terms of three general categories, the three Rs: relationships, representations, and rules.1 These categories are theoretical tools, meant to aid in our efforts to take a complex and difficult-to-understand set of phenomena and render them cognitively tractable. And yet I think they name real things well enough for practical purposes.2 Relationships, representations, and rules designate phenomena that are not mere theoretical abstractions.3
Perhaps there are other basic forms of structure beyond those considered in this book, but I can think of no others that cannot be recast as one of the Rs or understood as built up out of Rs. Another advantage of thinking about social structures in terms of relationships, representations, and rules: people know reasonably well what these things are. Therefore, I offer here the beginnings of an account of social structures that is rooted in ideas that are accessible to common sense and that seem reasonably well understood.
Of course, relationships, representations, and rules seldom stand alone. They are most often bound together, usually in complex ways, and it can be difficult to separate them in practice. More complicated social structures can be thought of as (often quite intricate) inter- and intralacings of relationships, representations, and rules. In Chapter 3, I will explain more fully what I mean by that. Before we can turn focused attention on those interand intralacings, though, we need to think for a while about each of the basic categories. What is encompassed in each of them? How are they social structural in the sense described in the introduction to this part of the book?
Relationships
By relationships, I mean recurrent patterns of interaction, the building blocks of social networks.4 We are immersed in many of them. Sociologists and other network analysts don’t think of people as embedded in a network but rather in many overlapping and crosscutting ones.5 You might be suspended in webs of friendship, of kinship, of economic exchange, of shared organizational affiliation (attending the same school, worshiping at the same mosque, participating in the same soccer league, living in the same neighborhood) and of many other types. These networks are often not fully independent of each other. People may do business with their kin, join a soccer league with their friends, or spend social time with their fellow worshipers. Sociologists call this “multirelationality” or “multiplexity.”6 As we shall see, this overlap of webs of different kinds of relationships is one, very important kind of intralacing.
We have already encountered the idea that such networks are made of interdependent actions. A soccer league exists because of the interdependent actions of those who do and don’t join it. A friendship network shifts its form as people make and break relationships. Markets are made of the unfolding exchange relations that people enter and exit through transactions. When we think of relationships as social structural, we often try to take a bird’s eye view, to see the pattern of ties as a whole. Our emphasis is often on the patterns that emerge from people’s actions, not mainly on relationships as experienced at the dyadic level.
Social scientists have found that network structures have important implications for the capabilities of those who are embedded within them.7 Here we will review only a few of the most important findings, and those only briefly.
Many have traditionally fallen under the heading of “social capital.”8 Popular use of this term often blurs its meaning, but when sociologists refer to social capital, they mostly have in mind the resources that individuals and communities get due to the patterns of ties around them. The term has limitations. It grew out of the ideas of financial, physical, and human capital, to emphasize that relationships too matter in economic life. Unfortunately, though, the “capital” metaphor might encourage us to think in overly instrumental terms about relationships.9 At a personal level, once we think of relationships through a frame that emphasizes how we can use them, this likely changes them. From a social science point of view, we probably get some things wrong when we adopt a frame that implicitly emphasizes that relationships are resources for people, means rather than ends.
