Massively Parallel Globalization

David C. Earnest

Description

In this era of globalization, people organize into fluid, adaptive networks to solve complex problems and provide resources that nation-states cannot. Examples include the Grameen Bank, mHealth, and the Ushahidi open source software project. Why do these networks succeed where nation-states fail? Only recently have social scientists developed tools to understand exactly how these complex networks self-organize, emerge, adapt, and solve collective problems. Three of these tools—agent-based modeling, social network analysis, and evolutionary computing—are converging in a field known as computational social science. In this provocative book, David C. Earnest discusses how computational social science helps us understand "massively parallel globalization." Using "explorations" of global systems ranging from fisheries to banking, Earnest illustrates the promise of computer models for explaining the surprises, cascades, and complexity that characterize global politics today. These examples of massively parallel globalization contrast sharply with the hierarchical and inflexible governmental bureaucracies that are poorly suited to solve many of today's transnational and global challenges.
List of Illustrations
Acknowledgments

1. The Gyre: Rethinking Systems in World Politics

2. Agents and Networks: Complex Social Systems and Theories of World Politics

3. The Advantage of Size: Why Large groups Solve Coordination Problems Better than Small Ones

4. Dividing the Pie: How Complex networks Learn to Solve Distributive Conflicts

5. Cows Grow Trees, Nets Grow Fish: How Social Networks Manage the Commons

6. Too Big to Compromise: Did Eleven Banks Block reform during the Great Recession?

7. Nets of Insecurity: Trade Networks, Cascading Failures, and Economic Vulnerability

8. Conclusions: Self- Organization in World Politics

Notes
References
Index

Sujets

• Sociology
• Globalization
• Political economy

Informations

• Publié par: State University of New York Press
• Date de parution: 05 mai 2015
• Nombre de lectures: 0
• EAN13: 9781438456621
• Langue: English

Extrait

MASSIVELY PARALLEL GLOBALIZATION
SUNY series, James N. Rosenau series in Global Politics

David C. Earnest, editor
MASSIVELY PARALLEL GLOBALIZATION
Explorations in Self-Organization and World Politics
DAVID C. EARNEST
S TATE U NIVERSITY OF N EW Y ORK P RESS
Published by
S TATE U NIVERSITY OF N EW Y ORK P RESS , A LBANY
© 2015 State University of New York
All rights reserved
Printed in the United States of America
No part of this book may be used or reproduced in any manner whatsoever without written permission. No part of this book may be stored in a retrieval system or transmitted in any form or by any means including electronic, electrostatic, magnetic tape, mechanical, photocopying, recording, or otherwise without the prior permission in writing of the publisher.
For information, contact
State University of New York Press, Albany, NY
www.sunypress.edu
Production, Laurie D. Searl
Marketing, Anne M. Valentine
Library of Congress Cataloging-in-Publication Data
Earnest, David C.
Massively parallel globalization : explorations in self-organization and world politics / David C. Earnest.
pages cm. — (SUNY series, James N. Rosenau series in global politics)
Includes bibliographical references and index.
ISBN 978-1-4384-5661-4 (hardcover : alk. paper)
ISBN 978-1-4384-5660-7 (pbk. : alk. paper)
ISBN 978-1-4384-5662-1 (e-book)
1. Political sociology. 2. Social networks—Political aspects. 3. Globalization—Political aspects. 4. World politics—21st century. I. Title. JA76.E28 2015 306.2—dc23 2014027595
10 9 8 7 6 5 4 3 2 1
To Jim, Simon, Marty, and Bill
We have to look for power sources here, and distribution networks we were never taught, routes of power our teachers never imagined, or were encouraged to avoid … we have to find meters whose scales are unknown in the world, draw our own schematics, getting feedback, making connections, reducing the error, trying to learn the real function …
—Thomas Pynchon, Gravity’s Rainbow
CONTENTS

LIST OF ILLUSTRATIONS
ACKNOWLEDGMENTS
CHAPTER ONE
The Gyre: Rethinking Systems in World Politics
CHAPTER TWO
Agents and Networks: Complex Social Systems and Theories of World Politics
CHAPTER THREE
The Advantage of Size: Why Large Groups Solve Coordination Problems Better than Small Ones
CHAPTER FOUR
Dividing the Pie: How Complex Networks Learn to Solve Distributive Conflicts
CHAPTER FIVE
Cows Grow Trees, Nets Grow Fish: How Social Networks Manage the Commons
CHAPTER SIX
Too Big to Compromise: Did Eleven Banks Block Reform during the Great Recession?
CHAPTER SEVEN
Nets of Insecurity: Trade Networks, Cascading Failures, and Economic Vulnerability
CHAPTER EIGHT
Conclusions: Self-Organization in World Politics
NOTES
REFERENCES
INDEX
ILLUSTRATIONS

FIGURES
3.1. Active nonlinear test’s maximization of periods of discord, over 40 generations of the test
3.2. An example of a failure to coordinate the social choice
3.3. An example of a coordinated social choice at t = 24
4.1. Normal-form representation of the four two-person games
4.2. Effect of memory on average player score, twelve-player rules of the road game
5.1. Mean appropriator scores for each information rule, by technology and memory
5.2. Average survival time for each information rule, by technology and memory
5.3. Mean appropriator score standardized per time step and percentage of simulations in which the ecosystem collapsed, by information rule
6.1. Battle of the sexes game and extensions for the Basel simulation
7.1. Log-log plot of the distribution of out degrees
7.2. Box plot of strategy fitness by GA generation
TABLES
3.1. A Condorcet decision problem among three nominal alternatives
3.2. Parameters of theoretic interest as operationalized in the model
3.3. Pseudo code for the Condorcet model
3.4. Summary statistics for the simulations
3.5. Final parameter sets as annealed by the Active Nonlinear Test
4.1. Pseudo code for the distribution of gains model
4.2. Distribution of strategies in 1,500 tournaments of two-player iterated prisoner’s dilemma
4.3. OLS regression estimates of effect on average player score
4.4. Probit estimated effects on the likelihood of coordination
4.5. Probit estimated effects on the likelihood of pure cheating strategy
5.1. Model parameters and expectations
5.2. Pseudo code for the fisheries model
5.3. Experimental results
5.4. Estimated partial effect of parameters, interactions terms, and controls on player payoffs
5.5. Estimated partial effects of model parameters, interaction terms, and controls on measures of social network structure
5.6. Estimated partial effects of model parameters, interaction terms, and controls on likelihood of collapse
6.1. Actors’ orders of preferences
6.2. Measures of the actual and simulated interbank networks
6.3. Pseudo code for the Basel III model
6.4. Experimental results with comparisons to baseline simulations
6.5. Multinomial logit estimates of effect of simulation parameters on Basel game outcomes
7.1. Summary statistics of the U.S. air route network
7.2. Pseudo code for the air transportation network model
7.3. Results of the GA experiments
7.4. Most frequently selected strategy sets
ACKNOWLEDGMENTS

This book seeks to explain self-organization in world politics. Regrettably, book projects do not self-organize, yet one could argue that authors receive too much of the credit for what is, in the end, the enterprise of a community of scholars. Research itself is a complex adaptive social system, in which innumerable actors contribute in small yet important ways that defy easy measurement or attribution. Nevertheless, I wish to acknowledge the many colleagues and friends—a massively parallel community, if you will—to whom I have incurred numerous intellectual and practical debts.
The earliest thoughts about this project originated at the 2009 annual convention of the American Political Science Association. My good friend Matthew J. Hoffmann of the University of Toronto invited me to contribute a paper to a panel tasked with “Understanding a Complex World: Complexity Theory and Political Science.” As chair of the panel, Matt helped me formulate my thinking about the relationship between complex systems theory and network theory. Thomas F. Homer-Dixon, Ian S. Lustick, and Kenneth W. Kollman, the panel’s other participants, provided useful comments and suggestions—Ken Kollman in particular encouraged me to develop my argument about the relationship between game theory and agent-based modeling. Subsequent revisions of what became the first two chapters of this book benefited from the criticism and attention to detail of colleagues at the George Washington University. In particular, I thank Professor Susan Sell, Professor James Lebovic, and doctoral candidate Miles Townes.
At my home institution of Old Dominion University, several of my colleagues in the Department of Political Science Geography and the Graduate Program in International Studies read and commented on different portions of the manuscript. Simon Serfaty in particular carefully read and reread my introductory chapter, and offered words of patient support as the manuscript worked its way through a lengthy gestation. Kurt Taylor Gaubatz read early versions of chapters 3 and 5 . At various times Francis Adams, Jennifer Fish, Regina Karp, Aaron Karp, and Steve Yetiv also provided helpful suggestions and encouragement. Professors Maura Hametz, Austin Jersild, Michael Carhart, and Jane Merritt, all of ODU’s Department of History, graciously welcomed me to join a faculty writing group. The group’s monthly meetings taught me a great deal about communicating effectively across disciplines. As prolific scholars, my colleagues in the Department of History also helped me rid my introductory chapter (more or less) of excessive jargon. Beyond critical evaluation, various individuals at ODU provided me with the material support to complete this book. I am indebted to the College of Arts Letters for a Faculty Summer Research Fellowship in 2010 and research leave during the fall semester of 2011, both of which provided me with the time to perform the simulation experiments reported in chapters 4 through 7 . Dr. John Sokolowski, the director of the Virginia Modeling, Analysis Simulation Center (VMASC), graciously made available to me the center’s computational resources, an invaluable access that permitted me to complete in hours those experiments that otherwise would have taken weeks. Research Associate Professor Joshua Behr of VMASC was among my earliest advocates and supporters, and facilitated my access to the center’s resources. David Ralph, VMASC’s director of information technology, developed a distributed computation architecture that allowed me to run experiments seamlessly across many workstations. I also thank Jessica Jones and Aaron Sander, graduate research assistants who helped me with the experiments reported in chapters 5 and 6 respectively. The innate curiosity of Christopher Ray, an MA student who volunteered to read my first chapters and who provided trenchant commentary, motivated me to write and argue with the passion of my younger days. Erika Frydenlund, a doctoral candidate at ODU, possesses a natural understanding of complex systems that through the years she has generously shared to correct my misunderstandings.
Associate Professor Emilian Kavalski, then of the University of Wes