TRUST IN NUMBERS This page intentionally left blank TRUST IN NUMBERS THE PURSUIT OF OBJECTIVITY IN SCIENCE AND PUBLIC LIFE Theodore M. Porter PRINCETON UNIVERSITY PRESS PRINCETON,NEW JERSEY Copyright 1995 by Princeton University Press Published by Princeton University Press, 41 William Street, Princeton, New Jersey 08540 In the United Kingdom: Princeton University Press, Chichester, West Sussex All Rights Reserved. Library of Congress Cataloging-in-Publication Data Porter, Theodore, 1953– Trust in numbers : the pursuit of objectivity in science and public life / Theodore M. Porter. p. cm. Includes bibliographical references and index. ISBN 0-691-03776-0 1. Science—Social aspects. 2. Objectivity. I. Title. Q175.5.P67 1995 306.4′5—dc20 94-21440 This book has been composed in Galliard Princeton University Press books are printed on acid-free paper and meet the guidelines for permanence and durability of the Committee on Production Guidelines for Book Longevity of the Council on Library Resources 13579108642 Contents Preface vii Acknowledgments xiii Introduction Cultures of Objectivity 3 PART I: POWER IN NUMBERS 9 Chapter One A World of Artifice 11 Chapter Two How Social Numbers Are Made Valid 33 Chapter Three Economic Measurement and the Values of Science 49 Chapter Four The Political Philosophy of Quantification 73 PART II: TECHNOLOGIES OF TRUST 87 Chapter Five Experts against Objectivity: Accountants and Actuaries 89 Chapter Six French State Engineers and the Ambiguities of Technocracy 114 Chapter Seven U.S. Army Engineers and the Rise of Cost-Benefit Analysis 148 PART III: POLITICAL AND SCIENTIFIC COMMUNITIES 191 Chapter Eight Objectivity and the Politics of Disciplines 193 Chapter Nine Is Science Made by Communities? 217 Notes 233 Bibliography 269 Index 303 This page intentionally left blank Preface SCIENCE is commonly regarded these days with a mixture of admiration and fear. Until very recently, though, English-language historians of sci- ence were more likely to resent its pretensions than to fear its power. Here resentment grew out of reverence. Karl Popper and Alexandre Koyré, who gave form to brilliant traditions in the philosophy and his- tory of science beginning especially in the 1950s, agreed that science was about ideas and theories. Koyré gave priority to thought experi- ments over the work of hands and instruments, and wondered, fa- mously, if Galileo had ever performed any experiments at all. Popper allowed that experimentation could falsify theories, but held that the real work was done when the theory was adequately articulated. Experi- menters had no more than to carry out what the theory dictated. Both praised science as a model of intellectual and philosophical achievement. Neither provided any reason for thinking that science could have much to do with technology. Still less could the hierarchical imagination of the historian or philosopher of science conceive that social science was authentically powerful. This problem of the relations of science to technology inspired noth- ing like the heated (and, it now seems, empty and incoherent) contro- versy over the relative merits of “externalist” and “internalist” explana- tions of scientific change. Rather than arguing, much of the profession took for granted that science had the loosest connections with the prac- tical world of engineering, production, and administration. In retro- spect, I can see that my graduate training provided ample opportunity to form a more judicious view. My teachers learned earlier than I did to appreciate the limitations of seeing the scientific enterprise mainly as a pursuit of theory. Still, I think I was not unusual among historians of science of my generation in thinking that the widespread linking of science and technology or of science and administrative expertise in- volved something fundamentally spurious, that these supposed connec- tions brought undeserved credit to each enterprise by making science seem more practical and its “applications” more intellectual than either really is. A critique of this nature underlay my original formulation of this proj- ect. I planned to examine the history of neoclassical economics, the most mathematical of social science disciplines—indeed, possibly the most mathematical of all disciplines. Economics values most highly this supremely abstract mathematics, yet somehow economists sustain the viii PREFACE image of a discipline capable of telling businesses and governments how to manage their affairs more effectively. I expected to show through an analysis of the relations of economics to policy that academic economics was a kind of sport, empty of implications for economic practice. That is not the book I have written. It didn’t take long to realize that neoclassical economics has had many critics who were better informed than I was likely to become. I found also that the economics discipline involves a greater variety of tools, aims, and practices than I had appreci- ated, and while I still think there is need for a more profound consider- ation of the relations between economic mathematics and the practices that support forecasting and policy advice, I am not the one to under- take it. In any case, my earlier suspicion that mathematics and policy were almost independent worked badly as a way of formulating a histor- ical project. Its validity was even more damaging than its shortcomings. If, indeed, neoclassical mathematics is irrelevant to the economic world, my history of the relations between economics and policy would turn into the history of nothing at all. So I have taken here a different tack. The interpenetration of science and technology, I now concede, is unmistakable, especially in the cur- rent century. That of social knowledge and social policy is only slightly less so. How are we to account for the prestige and power of quantita- tive methods in the modern world? The usual answer, given by apolo- gists and critics alike, is that quantification became a desideratum of so- cial and economic investigation as a result of its successes in the study of nature. I am not content with this answer. It is not quite empty, but it begs some crucial questions. Why should the kind of success achieved in the study of stars, molecules, or cells have come to seem an attractive model for research on human societies? And, indeed, how should we understand the near ubiquity of quantification in the sciences of nature? I intend this book to display the advantages of pointing the arrow of explanation in the opposite direction. When we begin to comprehend the overwhelming appeal of quantification in business, government, and social research, we will also have learned something new about its role in physical chemistry and ecology. My approach here is to regard numbers, graphs, and formulas first of all as strategies of communication. They are intimately bound up with forms of community, and hence also with the social identity of the re- searchers. To argue this way does not imply that they have no validity in relation to the objects they describe, or that science could do just as well without them. The first assertion is plainly wrong, while the latter is ab- surd or meaningless. Yet only a very small proportion of the numbers and quantitative expressions loose in the world today make any pretense of embodying laws of nature, or even of providing complete and accu- PREFACE ix rate descriptions of the external world. They are printed to convey re- sults in a familiar, standardized form, or to explain how a piece of work was done in a way that can be understood far away. They conveniently summarize a multitude of complex events and transactions. Vernacular languages are also available for communication. What is special about the language of quantity? My summary answer to this crucial question is that quantification is a technology of distance. The language of mathematics is highly struc- tured and rule-bound. It exacts a severe discipline from its users, a disci- pline that is very nearly uniform over most of the globe. That discipline did not come automatically, and to some degree it is the aspiration to a severe discipline, especially in education, that has given shape to modern mathematics.1 Also, the rigor and uniformity of quantitative technique often nearly disappear in relatively private or informal settings. In public and scientific uses, though, mathematics (even more, perhaps, than law) has long been almost synonymous with rigor and universality. Since the rules for collecting and manipulating numbers are widely shared, they can easily be transported across oceans and continents and used to co- ordinate activities or settle disputes. Perhaps most crucially, reliance on numbers and quantitative manipulation minimizes the need for intimate knowledge and personal trust. Quantification is well suited for commu- nication that goes beyond the boundaries of locality and community. A highly disciplined discourse helps to produce knowledge independent of the particular people who make it. This last phrase points to my working definition of objectivity. It is, from the philosophical standpoint, a weak definition. It implies nothing about truth to nature. It has more to do with the exclusion of judgment, the struggle against subjectivity. This impersonality has long been taken to be one of the hallmarks of science. My work broadly supports that identification and tends to the view that this, more than anything else, accounts for the authority of scientific pronouncements in contempo- rary political life. Once again, though, I am reluctant to make science the unmoved mover in this drive for objectivity. In science, as in political and administrative affairs, objectivity names a set of strategies for dealing with distance and distrust. If the laboratory, like the old-regime village, is the site of personal knowledge, the discipline, like the centralized state, depends on a more public form of knowing and communicating. Quantification is preeminent among the means by which science has been constructed as a global network rather than merely a collection of local research communities.