217 107 2MB
English Pages [272] Year 2014
Advances in Experimental Moral Psychology
Advances in Experimental Philosophy Series Editor: James R. Beebe, Associate Professor of Philosophy, University at Buffalo, USA Editorial Board: Joshua Knobe, Yale University, USA Edouard Machery, University of Pittsburgh, USA Thomas Nadelhoffer, College of Charleston, USA Eddy Nahmias, Neuroscience Institute at Georgia State University, USA Jennifer Nagel, University of Toronto, Canada Joshua Alexander, Siena College, USA Experimental philosophy is generating tremendous excitement, producing unexpected results that are challenging traditional philosophical methods. Advances in Experimental Philosophy responds to this trend, bringing together some of the most exciting voices in the field to understand the approach and measure its impact in contemporary philosophy. The result is a series that captures past and present developments and anticipates future research directions. To provide in-depth examinations, each volume links experimental philosophy to a key philosophical area. They provide historical overviews alongside case studies, reviews of current problems, and discussions of new directions. For upper-level undergraduates, postgraduates, and professionals actively pursuing research in experimental philosophy, these are essential resources. New titles in the series include: Advances in Experimental Epistemology, edited by James R. Beebe Advances in Experimental Philosophy of Mind, edited by Justin Sytsma
Advances in Experimental Moral Psychology Edited by Hagop Sarkissian and Jennifer Cole Wright Series: Advances in Experimental Philosophy
Bloomsbury Academic An imprint of Bloomsbury Publishing Plc LON DONDON • OX•F O N ELWHYO E W DK E L•HSY I • DN SY DN LON NR ED W • DE I • RNKE •WN YOR EYEY
Bloomsbury Academic An imprint of Bloomsbury Publishing Plc Bloomsbury Academic Bloomsbury Academic An imprint of Bloomsbury Publishing Plc 50 Bedford Square 1385 Broadway An imprint of Bloomsbury Publishing Plc London New York 50 Bedford Square 1385 Broadway 50 Bedford Square 1385 Broadway WC1B 3DP NY 10018 London New York London New York UK3DP USA WC1B NY 10018 WC1B NYUSA 10018 UK3DP UK www.bloomsbury.com USA www.bloomsbury.com www.bloomsbury.com Bloomsbury is a registered trade mark of Bloomsbury Publishing Plc Bloomsbury is a registered trade mark of Bloomsbury Publishing Plc Bloomsbury is a registered trade mark of Bloomsbury Publishing Plc First published 2014 Paperback edition first published 2015 First published 2014 published 2014 and Contributors 2014 © Hagop Sarkissian andFirst Jennifer Cole Wright © Hagop Sarkissian and Jennifer Cole Wright and Contributors 2014 © Hagop Sarkissian Cole asserted Wright and Contributors Hagop Sarkissian and Jennifer and ColeJennifer Wright have their right under2014 the Copyright, and Patents Act, 1988, to behave identifi ed as the of thisthe work. Hagop Designs Sarkissian and Jennifer Cole Wright asserted theirEditors right under Copyright, Hagop Designs Sarkissian and Jennifer Cole Wright asserted theirEditors right under Copyright, and Patents Act, 1988, to behave identifi ed as the of thisthe work. and Patents Act,of1988, to be identifi ed as Editors oforthis work. All Designs rights reserved. No part this publication may bethe reproduced transmitted in any form or by any or mechanical, including photocopying, All rights reserved. Nomeans, part of electronic this publication may be reproduced or transmitted All rights reserved. Nomeans, part of electronic this storage publication may be system, reproduced or transmitted recording, or any information retrieval without prior in any form or by oror mechanical, including photocopying, in any form or by means, electronic or mechanical, including photocopying, permission in writing from the publishers. recording, or any information storage or retrieval system, without prior recording, or any information storage or retrieval system, without prior permission in writing from the publishers. permission in writing the publishers. No responsibility for loss caused to anyfrom individual or organization acting on or refraining fromforaction as a result of the material this publication can No responsibility loss caused to any individual orinorganization acting on or No responsibility loss caused any individual orinorganization acting on or beaction accepted by to Bloomsbury or the author. refraining fromfor as a result of the material this publication can refraining frombeaction as a result of the material this publication can accepted by Bloomsbury or theinauthor. be accepted by Bloomsbury or the author.Data British Library Cataloguing-in-Publication A catalogue record for this book is available from the British British Library Cataloguing-in-Publication Data Library. British Library Cataloguing-in-Publication Data Library. A catalogue record for this book is available from the British A catalogue record ISBN: for thisHB: book is available from the British Library. 978-1-4725-0938-3 ISBN: HB: 978-1-4725-0938-3 978-1-4725-0938-3 ePDF: 978-1-4725-1304-5 ISBN: HB: PB: 978-1-4725-0785-3 978-1-4742-5712-1 ISBN: HB: 978-1-4725-0938-3 ePub: ePDF: 978-1-4725-1304-5 ePDF: 978-1-4725-1304-5 ePDF: 978-1-4725-1304-5 ePub: 978-1-4725-0785-3 ePUB: 978-1-4725-0785-3 ePub:Cataloging-in-Publication 978-1-4725-0785-3 Library of Congress Data Advances in experimental moral Cataloging-in-Publication psychology/edited by HagopData Sarkissian and Library of Congress Library of Congress Cataloging-in-Publication Jennifer Cole Wright. by HagopData Advances in experimental moral psychology/edited Sarkissian and Advances in pages experimental moral psychology/edited Hagop Sarkissian and cm. – (Advances in experimental philosophy) Jennifer Cole Wright. by Jennifer in Cole Wright. and Includes references index. pages cm. –bibliographical (Advances experimental philosophy) pages cm. –bibliographical (Advances in –experimental philosophy) ISBN 978-1-4725-0938-3 (hardback) ISBN 978-1-4725-0785-3 (epub) – Includes references and index. Includes bibliographical and index. Hagop ISBN 978-1-4725-1304-5 (epdf) 1. Ethics–Research. I. Sarkissian, (Professor), ISBN 978-1-4725-0938-3 (hardback) –references ISBN 978-1-4725-0785-3 (epub) – ISBN 978-1-4725-0938-3 (hardback) – ISBN 978-1-4725-0785-3 (epub) – of compilation. ISBN 978-1-4725-1304-5 (epdf) editor 1. Ethics–Research. I. Sarkissian, Hagop (Professor), ISBN 978-1-4725-1304-5 (epdf) editor 1. Ethics–Research. BJ66.A38 2014 I. Sarkissian, Hagop (Professor), of compilation. editor of compilation. 170–dc23 BJ66.A38 2014 BJ66.A38 2014 2013039795 170–dc23 170–dc23 2013039795 2013039795 Typeset by Deanta Global Publishing Services, Chennai, India Printed and bound in Great BritainChennai, India Typeset by Deanta Global Publishing Services, Typeset by Deanta Global Publishing Services, Printed and bound in Great BritainChennai, India Printed and bound in Great Britain
Table of Contents Notes on Contributors Experimental Moral Psychology: An Introduction Hagop Sarkissian and Jennifer Cole Wright Part 1
vii
1
Moral Persons
1
The Character in Competence Piercarlo Valdesolo
21
2
Spoken Words Reveal Selfish Motives: An Individual Difference Approach to Moral Motivation Jeremy A. Frimer and Harrison Oakes
36
Is the Glass of Kindness Half Full or Half Empty? Positive and Negative Reactions to Others’ Expressions of Virtue Gabriela Pavarini and Simone Schnall
55
4
What are the Bearers of Virtues? Mark Alfano
73
5
The Moral Behavior of Ethicists and the Power of Reason Joshua Rust and Eric Schwitzgebel
91
3
Part 2 6
7
8
9
Moral Groundings
Pollution and Purity in Moral and Political Judgment Yoel Inbar and David Pizarro
111
Selective Debunking Arguments, Folk Psychology, and Empirical Moral Psychology Daniel Kelly
130
The Psychological Foundations of Moral Conviction Linda J. Skitka
148
How Different Kinds of Disagreement Impact Folk Metaethical Judgments James R. Beebe
167
vi
Table of Contents
10 Exploring Metaethical Commitments: Moral Objectivity and Moral Progress Kevin Uttich, George Tsai, and Tania Lombrozo
188
11 Agent Versus Appraiser Moral Relativism: An Exploratory Study Katinka J. P. Quintelier, Delphine De Smet, and Daniel M. T. Fessler
209
Part 3
Measuring Morality
12 Know Thy Participant: The Trouble with Nomothetic Assumptions in Moral Psychology Peter Meindl and Jesse Graham
233
Index
253
Notes on Contributors Editors Hagop Sarkissian’s research is located at the intersection of cognitive science, ethics, and classical Chinese philosophy. He draws insights from the cognitive and behavioral sciences to explore topics in moral psychology, moral agency, and the status of morality, with an eye toward seeing how culture might shape cognition in these domains. In addition to drawing from the empirical sciences, he also uses the tools of experimental psychology in some of his research. He has authored or coauthored papers in these areas for Philosophical Studies, Philosopher’s Imprint, Mind & Language, The Annual Review of Psychology, Philosophy Compass, Review of Philosophy and Psychology, History of Philosophy Quarterly, The Journal of Chinese Philosophy, and for Moral Psychology, Volume I: The Evolution of Morality: Adaptations and Innateness (MIT, 2007). His work has been translated into Chinese and Korean. He teaches at The City University of New York, Baruch College, and resides in Brooklyn. Jennifer Cole Wright is an assistant professor of psychology and an affiliate member of philosophy at the College of Charleston. Her area of research is moral development and moral psychology more generally. Specifically, she studies metaethical pluralism, the influence of individual and social “liberal vs conservative” mindsets on moral judgments, and young children’s early moral development. She has published papers on these and other topics in Cognition, Mind & Language, Journal of British Developmental Psychology, Journal of Experimental Social Psychology, Oxford Studies in Experimental Philosophy, Journal of Moral Education, Philosophical Psychology, Journal of Cognition and Culture, Personality and Individual Differences, Social Development, Personality & Social Psychology Bulletin, and Merrill-Palmer Quarterly.
viii
Notes on Contributors
Contributors Mark Alfano is an assistant professor of philosophy at the University of Oregon. In 2011, he received his doctorate from the Philosophy Program of the City University of New York Graduate Center. He has been a postdoctoral fellow at the Notre Dame Institute for Advanced Study and the Princeton University Center for Human Values. Alfano works on moral psychology, broadly construed to include ethics, epistemology, philosophy of mind, and philosophy of psychology. He also maintains an interest in Nietzsche, focusing on Nietzsche’s psychological views. Alfano has authored papers for such venues as the Philosophical Quarterly, The Monist, Erkenntnis, Synthese, and the British Journal for the History of Philosophy. His first book, Character as Moral Fiction (Cambridge University Press, 2013), argues that the situationist challenge to virtue ethics spearheaded by John Doris and Gilbert Harman should be co-opted, not resisted. He is currently writing an introduction to moral psychology and a research monograph on Nietzsche, as well as editing three volumes on virtue ethics and virtue epistemology. James R. Beebe is an associate professor of philosophy and a member of the Center for Cognitive Science at the University at Buffalo. He has published papers on reliabilism, skepticism, and the a priori in mainstream epistemology and is actively engaged in the empirical study of folk epistemic intuitions. He also has a keen interest in the psychology of moral, political, and religious beliefs. Jeremy A. Frimer is an assistant professor of psychology at the University of Winnipeg in Winnipeg, Canada. He completed a PhD at the University of British Columbia in 2011. His research investigates how the words that individuals tend to speak reveal information about their personality and lead to congruent behaviors. Harrison Oakes is in the final year of a BAH in Psychology at the University of Winnipeg. Yoel Inbar is an assistant professor of social psychology at Tilburg University in the Netherlands. His research focuses on the interplay between two general mental processes that influence judgment, namely: rational, deliberate analysis
Notes on Contributors
ix
and intuitive, emotional reactions. Yoel has applied his research to study how intuition affects our choices, how our moral beliefs determine our actions and judgment of others, and how the emotion of disgust can predict our moral and political attitudes. David Pizarro is currently an associate professor in the Department of Psychology at Cornell University in Ithaca, NY. His primary research interest is in how and why humans make moral judgments, such as what makes us think certain actions are wrong, and that some people deserve blame. In addition, he studies how emotions influence a wide variety of judgments. These two areas of interest come together in the topic of much of his recent work, which has focused on the emotion of disgust and the role it plays in shaping moral, social, and political judgments. Daniel Kelly is an associate professor of philosophy at Purdue University. His research interests are at the intersection of the philosophy of mind and cognitive science. He has published papers on moral judgment, social norms, implicit bias and responsibility, racial cognition, and social construction, and is the author of Yuck! The Nature and Moral Significance of Disgust. Peter Meindl is a doctoral student in the social psychology department at the University of Southern California. Prior to matriculating at USC, Pete received a BA and an MA in psychology from Stanford University and Wake Forest University, respectively. He primarily conducts research pertaining to moral cognition and morally relevant behavior, but his research also often touches on issues related to self-regulation. Jesse Graham received his PhD (psychology) from the University of Virginia in 2010, a Master’s (Religious Studies) from Harvard University in 2002, and a Bachelor’s (Psychology) from the University of Chicago in 1998. He is currently assistant professor of psychology at the University of Southern California, where he hovers menacingly over the Values, Ideology, and Morality Lab. His research interests are in moral judgment, ideology, and implicit social cognition. Gabriela Pavarini is a PhD candidate in psychology at the Embodied Cognition and Emotion Laboratory at the University of Cambridge. Her
x
Notes on Contributors
research interests focus on the psychological mechanisms underlying prosocial behavior and the formation of interpersonal bonds. Before moving to Cambridge, she received her BA in Psychology from the Federal University of Sao Carlos, Brazil. She then completed her MPhil in Social and Developmental Psychology at Cambridge, specializing in emotional reactions to expressions of virtue. Currently, she is interested in the social functions of physiological and behavioral synchrony, as well as other-focused emotions such as admiration and compassion. Simone Schnall is a senior lecturer of psychology at the University of Cambridge and the director of the Cambridge Embodied Cognition and Emotion Laboratory. She studies the relationship between cognitive and affective processes. In particular, she is interested in how embodiment informs and constrains thought and feeling. Recent topics have included the effect of emotion on morality and the role of physical ability on perceptual judgments. Her work has been funded by numerous grants from public and private sources, including the Economic and Social Research Council (UK), the National Science Foundation (USA), and the National Institute of Mental Health (USA). Schnall’s research is routinely covered in the popular media such as the New York Times, The Economist, Newsweek, New Scientist and Psychology Today. Katinka J. P. Quintelier received her PhD in philosophy from Ghent University. She was a postdoctoral research fellow at the Konrad Lorenz Institute for Evolution and Cognition Research in Altenberg, Austria, at the time this chapter was written, and she is now a researcher at the University of Amsterdam. Her research interests are reputation, individual differences in morality, and empirically informed normative ethics. Daniel M. T. Fessler is an associate professor of anthropology and director of the Center for Behavior, Evolution, and Culture at the University of California, Los Angeles. Combining experiments, ethnography, and published data, he explores the determinants of behavior, experience, and health in domains such as emotions, disease avoidance, aggression, cooperation, morality, food and eating, sex and reproduction, and risk taking.
Notes on Contributors
xi
Delphine De Smet obtained her master’s in moral sciences at Ghent University, where she is now a PhD student on disgust and moral emotions. She does research on the cultural and biological origins of the incest taboo and is currently working on disgust and fascination related to blood. Joshua Rust is associate professor of philosophy at Stetson University, Florida. In addition to a number of books and papers focused on the writings of John Searle he has coauthored (with Eric Schwitzgebel) a number of papers concerning the moral behavior of ethicists. Eric Schwitzgebel is professor of philosophy at UC Riverside, working at the intersection of philosophy of mind and empirical psychology—especially on issues of self-knowledge, consciousness, moral psychology, attitudes such as belief, and the role of intuition or common sense in philosophical method. Among his works are Perplexities of Consciousness (MIT Press, 2011), Describing Inner Experience? Proponent Meets Skeptic (with Russell T. Hurlburt, MIT Press, 2007), “The unreliability of naïve introspection” (Philosophical Review 2008), “The moral behavior of ethics professors: Relationships among self-reported behavior, expressed normative attitude, and directly observed behavior” (with Joshua Rust, Philosophical Psychology in press), “Acting contrary to our professed beliefs” (Pacific Philosophical Quarterly 2010), and “A dispositional approach to attitudes: Thinking outside the belief box” (in Nottelmann, ed., New Essays on Belief 2013). He also blogs at The Splintered Mind: http://schwitzsplinters.blogspot.com. Linda J. Skitka is a professor and associate head of psychology at the University of Illinois at Chicago. Her research interests bridge social, political, and moral psychology. She has over 100 academic publications on topics such as the psychology of moral conviction, the cognitive and motivational underpinnings of ideological disagreements, and the psycho-political consequences of terrorism. Her research on these topics has received grant support from the National Science Foundation and NASA-Ames. She serves on multiple editorial boards as a consulting editor, is currently an associate editor for the Journal of Personality and Social Psychology, and is a former president of the International Society for Justice Research.
xii
Notes on Contributors
Tania Lombrozo is an associate professor of psychology at the University of California, Berkeley, an affiliate of the Department of Philosophy, and a member of the Institute for Cognitive and Brain Sciences. She received her PhD in psychology from Harvard University in 2006 after receiving a BS in symbolic systems and a BA in philosophy from Stanford University. Dr Lombrozo’s research addresses foundational questions about cognition using the empirical tools of cognitive psychology and the conceptual tools of analytic philosophy, focusing on explanation and understanding, conceptual representation, categorization, social cognition, and causal reasoning. She is the recipient of numerous awards, including the Stanton Prize from the Society for Philosophy and Psychology, a Spence Award from the Association for Psychological Science, a CAREER award from the National Science Foundation, and a McDonnell Foundation Scholar Award. She blogs about psychology, philosophy, and cognitive science for NPR’s 13.7: Cosmos & Culture. Kevin Uttich received his PhD in psychology from the University of CaliforniaBerkeley in 2012 after receiving a BA in psychology from the University of Chicago. Dr Uttich’s research examines issues at the intersection of social cognition and moral psychology with a particular focus on how people understand prescriptive norms. His work has received the poster prize from the Society for Philosophy and Psychology and has been recognized by the publication Science as an Editor’s Choice. Piercarlo Valdesolo is an assistant professor of psychology and director of the Moral Emotions and Trust lab at Claremont McKenna College in Claremont, CA. He teaches courses in social and moral psychology and his research explores the psychological bases of trust, cooperation, and character. He is coauthor of the book Out of Character and he is a member of the Editorial Board of the journals Emotion, Journal of Experimental Social Psychology and Journal of Personality and Social Psychology. He is also a regular contributor to Scientific American and Psychology Today.
Experimental Moral Psychology: An Introduction Hagop Sarkissian and Jennifer Cole Wright
This volume is titled Advances in Experimental Moral Psychology and it is part of a series addressing recent advances in the field of experimental philosophy more generally. Thus, it behooves us to say at least something about both moral psychology and its relationship to experimental philosophy. Although moral psychology could certainly be seen as a subdiscipline within experimental philosophy, it would be wrong to consider it its child—if anything, we might consider it its inspiration. After all, moral psychology, which we will define loosely as the interdisciplinary investigation of how human beings function in moral contexts, has been around for a long time. Though initially a parallel endeavor taken on by philosophers on the one hand, and psychologists on the other, these two research agendas have increasingly become one, as scholars from both disciplines have begun to not only seek insight and support from the other’s theoretical and methodological contributions, but also to work together as interdisciplinary teams with common goals. Contemporary moral psychology keeps its investigative eye equally on two questions. The first question is: How do people actually (as opposed to theoretically) function in moral contexts? This question breaks down into a whole host of additional questions, such as: How does our awareness of morality develop (what is “innate” vs. learned, and what does the learning process look like)? What does this moral awareness entail and of what is it composed (e.g., perception, emotions, automatic vs. effortful cognition)? What produces morally good behavior? What are the main differences between morally good and bad people—and to what extent do those differences involve something psychologically stable, such as “character” or “moral identity”? And so on.
2
Advances in Experimental Moral Psychology
The second question is: In what ways are the answers to the first set of question philosophically interesting? Do they inform philosophical theory—insuring that our theoretical conceptions of morality properly line up with the empirical data? Do they help us adjudicate between competing philosophical views? Do they raise problems for long-standing philosophical commitments? Of course, this second question presupposes that what we learn about actual moral functioning is meaningful to philosophical theorizing, something that is not universally accepted but, as evidenced by increasing interdisciplinary activity, is nonetheless maintained by a substantial group of researchers. Examples of the fruitfulness of this mindset can be found not only in this volume, but also in other interdisciplinary volumes on moral psychology published in recent years, such as those edited by Walter Sinnott-Armstrong (2007), Darcia Narvaez and Daniel K. Lapsley (2009), Thomas Nadelhoffer et al. (2010), and John Doris and the Moral Psychology Research Group (2012). And while philosophers have much to gain through their collaborations with psychologists and their use of empirical data to inform their projects, psychologists’ empirical endeavors likewise can only benefit from an increased awareness of and appreciation for the theoretical models arising out of philosophical work on moral normativity, moral epistemology, and metaethics. The rich history of philosophical reflection on these questions can serve as grounds to generate new hypotheses for testing and new avenues for research. Given this interdisciplinary interaction, we see moral psychology as a sort of role model for the more recent developments in what people are calling “experimental philosophy”—which is, broadly speaking, the use of empirical and experimental methods to investigate philosophical questions. The early (turn of this century) focus on using empirical methods to largely challenge or undermine philosophical theories (e.g., the “situationist” critique of virtue ethics) and philosophical methodology (e.g., the role of intuitions in theory formation) has been labeled the “negative program.” However, philosophers (and others) have increasingly begun to shift the focus of experimental philosophy to other, more constructive endeavors (see Systma and Livengood in press, for a discussion of what they call the developing “positive,” “naturalist,” and “pragmatist” programs). These endeavors embrace the exploration of people’s intuitions, judgments, and cognitive process more broadly in order to clarify what they are, when they count as philosophical
Experimental Moral Psychology: An Introduction
3
evidence, and also more generally what they reveal about human cognition, language, and behavior. It is here that we think moral psychology has much to offer as a role model.
Moral persons In the early stages of conceiving this volume, when we were considering which sorts of contributions would best represent the field’s most exciting current developments, we decided to cast a wide net—one that would include not only people already identified with the experimental philosophy family (as evidenced, for example, through participation in panels, conferences, and volumes bearing that designation), but also other philosophers and psychologists whose research makes a meaningful contribution to the field. This means that our invitations went out to people engaged in a wide variety of different projects, exploring several enduring questions concerning moral psychology. In what follows, we will attempt to highlight some of the ways these researchers’ work relate and inform one another along a number of dimensions. For example, a long-standing question that has preoccupied philosophers and psychologists alike concerns the nature of moral persons. What are morally good people like (and is there a difference between what we think they are like and what they are really like)? How do they become moral? Piercarlo Valdesolo observes in his chapter that we are sensitive to moral qualities in people, and that judging or appraising individuals along moral dimensions is a natural, inescapable part of life. We find individuals fair or cruel, kind or cold, affable or scheming. What’s more, it seems clear that we favor certain types of individuals over others. We tend to seek company with those we find fair, kind, and giving, and impart unto them moral value. Yet, this tendency may, according to Valdesolo, be biased. According to much of the research on “person perception,” we commonly evaluate people along two basic dimensions: (a) their social skills or “warmth” (e.g., the degree to which people are sincere, compassionate, friendly, trustworthy, caring, helpful, etc.) and (b) their intellectual skills or “competence” (e.g., the degree to which people are determined, intelligent, creative, focused,
4
Advances in Experimental Moral Psychology
hardworking, etc.—the dimension of “competence”). Valdesolo argues that our assessment of what morally good people are like tends to lean heavily on the dimension of warmth, while discounting or ignoring entirely the dimension of competence. Why is this? It might be that our primary interest in evaluating someone’s moral status is determining his or her intentions toward us, and the extent to which we might benefit from them (warmth), and not necessarily the extent to which they are able to carry out and benefit from their own intentions (competence). It seems to be in our self-interest to seek company with those who are warm and kind, as we might benefit from their values, dispositions, and behavior, and we might take such qualities to be definitive of virtue and moral worth. But, as Valdesolo points out, this tendency to value warmth, kindness, and generosity is likely to miss important features that constitute truly moral individuals. Achieving morally valuable ends or bringing about morally desirable outcomes requires more than personality traits of warmth and generosity. Oftentimes, and in addition to these, individuals will require a range of other competencies in order to succeed in their moral projects. Traits such as grit, discipline, and industriousness are thus also important and should be equally valued—yet they aren’t. In the end, Valdesolo suggests a revision to our evolved tendency to favor warmth over competence; we ought to value both. Valdesolo’s argument finds support in the fact that many theoretical models of virtue regard both dimensions as important aspects of the morally good person. Along these lines, Jeremy Frimer and Harrison Oakes’ contribution may provide additional insight into why we are inclined toward the asymmetry (problematic though it might seem) as well as why, under some circumstances, it may be justified. According to Frimer and Oakes, people typically claim to have both “agentic” (self-promoting, competence-based) values and “community” (other-promoting, warmth-based) values. The only problem— and it is a problem specifically from the perspective of moral development—is that people spend way more time pursuing their agentic values as opposed to their community values, and often view the two as being at odds with one another. This is familiar enough. Most people readily see how the two values can come into conflict, and it would not be unusual to hear someone say “I can’t
Experimental Moral Psychology: An Introduction
5
both do what’s best for me and help other people—I have to choose.” (Indeed, much ink has been spilled in the history of moral philosophy trying to defend the view that it actually is in our self-interest to be moral, which might be seen as an attempt to accommodate these competing tendencies.) Thus, it is not surprising that we tend to take people’s expression of community values (or “warmth”) as indicators of their moral goodness, while taking any expression of agentic values (or “competence”) as indicators of their self-interested motives. Frimer and Oakes argue, however, that these two value systems do not necessarily have to be in conflict. Indeed, for some individuals—namely, recognized moral exemplars—they are not. Frimer and Oakes’s research provides evidence that the chasm between people’s agentic and community values decreases along the developmental trajectory until, at the limit, it may disappear entirely. Morally good people have learned to synthesize their agentic and community values such that they are able to pursue both simultaneously; promoting their own well-being becomes effectively linked with the well-being of others. And as their intellectual and social skills become oriented toward and focused on the same ends, they also become more effective moral agents—a finding consonant with Valdesolo’s point that warmth and competence are both important for moral goodness. In ancient Greece, Aristotle described the virtuous person as someone who elicits others’ admiration and respect, and a notable feature of classical Confucian virtue ethics is its emphasis on the magnetic qualities of capable moral leaders—their ability to gain the assent and loyalty of others in an effortless fashion. It is natural to assume that virtuous conduct is something that is universally recognized and esteemed, especially since there never seems to be enough of it going around. However, according to Gabriela Pavarini and Simone Schnall, individuals exemplifying moral virtue are sometimes denied feelings of approbation and approval. Indeed, they can instead be subjected to ridicule and censure, or otherwise disparaged. Pavarini and Schnall highlight the paradoxical nature of people’s reactions to displays of genuine virtue by others. As they point out, morally good behavior rarely occurs in a social vacuum, and being the witness to another’s morally good deeds has implications beyond the immediate act itself. On the one hand, witnessing displays of virtue can lead to a sense of elevation, and a desire to
6
Advances in Experimental Moral Psychology
praise, reward, and cooperate or associate with those who act virtuously. Morally good people elicit elevation, respect, and gratitude. Through their actions they can raise the standards of an entire group, spurring individuals to greater levels of prosociality, and reminding them of the resources that lie untapped within them that could be marshaled toward improving their communities and making the world a better place. Moral exemplars can renew in others a belief that they can shape the world around them in positive ways. Yet, on the other hand, such elevation of moral standards can also be perceived as a threat to a one’s integrity and sense of self-worth, and to one’s standing among one’s peers; after all, one might be seen by others as falling short of expectations when someone else has just shown that morally excellent behavior is within reach. When one is threatened in this way, the reaction may be to denigrate the moral exemplar, or to explain away her good deeds as products of self-interest, situational demands, or other extraneous factors. In short, moral goodness can engender envy, hostility, and suspicion just as easily as it can inspire awe, gratitude, and respect. At its extreme, the motivation to avoid comparison with the moral exemplar and/or save face can lead to hatred and a desire to belittle, demean, and even destroy the source of this reputational threat. Pavarini and Schnall observe (as Valdesolo does) that qualities of warmth and kindness make us feel safe, protected, uplifted, grateful, while qualities of competence or resourceful make us feel challenged, inadequate, and threatened. We feel united with those who display warmth—and perceive more warmth in those with whom we are united—and feel in competition with those who display competence. Given Frimer and Oakes’s suggestion that moral exemplars come to exemplify traits related to both warmth and competence, it would make sense that people could react either positively or negatively to their example depending upon their relationship to the exemplar and which of her traits (her warmth or competence) is most salient to them at the time. However we respond to and evaluate morally good people, we commonly think of the virtues they display as residing within them, constituting part of their identities. Indeed, philosophers have long emphasized that laudable traits of character are important features of moral persons. Mark Alfano’s chapter raises a skeptical worry about whether there is such a thing as a “good” person
Experimental Moral Psychology: An Introduction
7
separate from the social and asocial environments in which people display good behavior. Rather than argue (as many have) that moral agents’ goodness is keyed to their possession of stable internal character traits that manifest in displays of virtuous behavior across diverse contexts, Alfano argues that we need to rethink the nature of character itself. Specifically, we need to stop thinking about it as a disposition internal to the moral agent and start thinking of it as a trifold, relational property, composed of the interactions between (a) a person’s internal states/capacities, (b) the social environment in which the person is imbedded, and (c) a variety of asocial features of the physical environment (e.g., noise levels, smells, lighting) that impact the person’s thoughts, feelings, and behavior in specific ways. The upside to this view is that what constitutes a morally good person is not just the internal states/capacities that she possesses, but also the sort of social and asocial environment she finds herself in and/or has taken part in creating. This means that when it comes to moral development, as much (if not more) of the burden falls on the world around developing moral agents as it does on the agents themselves, for the environment figures into the very structure of moral character they possess. Individuals themselves are not bearers of virtue. People with the right sorts of internal states/capacities represent just one leg in the trifold relation; if stuck within a morally corrupt social environment, and/or an asocial environment filled with hardship, danger, or distractions, virtue will be incomplete. This highlights an interesting link between Frimer and Oakes’s and Alfano’s contributions—namely, that as people’s values and motives become more integrated and synchronized, so do their social/asocial environments. This may be because the synchronization of their values results in the active selection and creation of social and asocial environments that promote and protect those values—environments that positively reflect the rewards and benefits of their chosen lifestyle—allowing for increased dispositional stability and expression. (Of course, exemplars must often construct and integrate such environments where they do not previously exist.) Finally, if something along these lines is correct (viz., that virtue consists in features extrinsic to individuals to some considerable extent), then it suggests that certain kinds of environments—namely, those in which you are socially rewarded for thinking hard about morality, but not necessarily behaving morally—are not going to be enough to generate virtue. And this
8
Advances in Experimental Moral Psychology
might help us to explain the rather dispiriting findings reported in Joshua Rust and Eric Schwitzgebel’s chapter that those of us who think and reflect deeply about the nature of morality (such as all of you reading this book right now) are not on average morally better behaved. We might hold out hope that our academic interest in morality and ethics will pay off in noticeable and measurable differences in our moral behavior, that having dedicated our lives to reflecting on all matters moral we should have generated a healthy level of moral goodness. It turns out that all our efforts may be for naught. Or so say the studies conducted by Rust and Schwitzgebel, which measure professional ethicists, moral philosophers, and other academics along a number of moral dimensions. Whether rated by their peers or observed at conferences, whether measured by their charitable donations, voting behavior, responsiveness to student needs, or other dimensions, Rust and Schwitzgebel have repeatedly found that professional ethicists and moral philosophers are no better than their colleagues in other academic fields, despite reflecting deeply about moral reasoning and the structure of morality, despite grappling with the greatest figures in the history of moral thought, and despite expressing more stringent moral attitudes. (For example, while ethicists were much more likely than non-ethicists to rate meat-eating as morally bad, the evidence suggested no correspondingly significant difference in how often they ate meat). This is in line with Frimer and Oakes’s finding that most people—ethicists or not— report the personal importance of prosocial, community-oriented values to be much higher than their actual daily activities reflect; they say such values are highly important to them yet spend far less time actually engaging in related activities. Instead, people tend to do that which promotes their own selfinterested, agentic values. What conclusion should we draw? Rust and Schwitzgebel consider five alternative explanations—including Jonathan Haidt’s “rational tail” view in which most of our reasoning is employed post hoc to rationalize attitudes we already possess, and the provocative view that philosophers of ethics are making up for a natural deficiency in their moral intuitions (so, they actually are better than they would have otherwise been)—and conclude that the data do not champion one explanation over the other. In the end, regardless of the explanatory model one favors, their research suggests that moral goodness requires more than just theoretical reflection and critical thinking to develop.
Experimental Moral Psychology: An Introduction
9
Moral groundings Thus far, the contributions we’ve canvassed have largely focused on the nature of moral persons and moral exemplars—how they are constituted, how they are motivated, and how others evaluate/respond to them. But, we haven’t said much about the nature of morality itself. Here, several of the contributions help to enrich our understanding of the psychological mechanisms that may constitute part of moral life, as well as the standing that we take morality to have. A widely experienced facet of morality is its importance and weightiness relative to other evaluative practices and domains. We might disagree with others across a wide range of domains, including matters of convention and aesthetics. However, disagreement about, say, standards of physical beauty seldom seem as pressing or compelling to us as disagreements about basic moral issues such as racial equality, reproductive rights, or distributive justice. When it comes to these latter topics, we tend to think that we are arguing over issues of central importance to human life—issues that must ultimately admit of correct answers even in the face of entrenched disagreement. Similarly, when we condemn certain acts as right or wrong, virtuous or depraved, it seems as though these judgments have a certainty—an objectivity—that elides judgments concerning matters of convention or taste. Why is this so? Evolutionary accounts have been featured in both the Valdesolo and Pavarini and Schnall contributions above. For Valdesolo, the promise of immediate gains and the need to properly understand others’ dispositions and intentions toward us may have fostered a favorable disposition toward those who are kind and warm as opposed to persistent and focused. For Pavarini and Schnall, our divergent reactions to moral exemplars can be understood as facilitating very different evolutionary demands—to cooperate and build cohesive communities on the one hand, and to maintain one’s reputation and status within the group on the other. Other contributions also link current moral psychological tendencies to evolved capacities. For example, Yoel Inbar and David Pizzaro note the widespread role that disgust reactions play in our moral lives, where moral transgressions evoke not only moral condemnation but also a visceral feelings of repulsion. In particular, they focus on research suggesting both that disgust can arise as a consequence
10
Advances in Experimental Moral Psychology
of making certain types of moral appraisals (e.g., when confronted with taboo behavior or unfair treatment) and that disgust can amplify moral judgments when it is elicited in the formation of a moral evaluation. Yet why should this be so? Why should disgust be recruited in moral judgment at all? Inbar and Pizarro argue that disgust is part of an evolved general motivational system whose function is to distance us from potential threats—namely, diseasebearing substances and individuals. Disgust was also co-opted by higher order systems to serve as a warning mechanism against socially and morally prohibited behaviors, as well as any potential contaminant or contagion. Sexual acts, food taboos, physical abnormalities or deformities, individuals from strange or foreign cultures-each of these triggers our core behavioral immune system, which serves to create distance between the individual and these potential sources of physical and moral threat. One upshot of this is that moral transgressions often repulse us in ways that other sorts of transgressions don’t, eliciting from us a feeling of undeniable wrongness. But if Inbar and Pizarro are correct, then it leads us inexorably to a question: Should we trust our moral judgments when they involve disgust reactions? Or should we instead recognize such responses as likely to be biased or erroneous, pushed around by a mechanism that was shaped by forces not designed to reliably track morally relevant considerations? Daniel Kelly, in his contribution, argues for the latter claim. Disgust cannot, according to Kelly, be treated as a reliable indicator of moral transgressions; it is overly sensitive to cues related to its older and more primitive function of protecting the individual against poisons and contaminants. What is more, it is a system for which false positives are much more advantageous than false negatives, so we are prone to find things disgusting even when nothing actually “disgust worthy” is present. And given the particularly phobic response that disgust generates (the experience of nausea and/or the intense desire to remove oneself from the presence of the triggering stimulus), Kelly worries that disgust has the potential to cause more harm (particularly in the form of outgroup denigration and rejection of physical/cultural abnormalities) than good, even when it is “on track.” The general form of this debunking story is familiar: psychological mechanisms that are involved in moral judgment are sensitive to irrelevant considerations and should therefore be viewed with suspicion. However,
Experimental Moral Psychology: An Introduction
11
while Kelly acknowledges that many of the psychological mechanisms that are recruited in moral life will have an evolutionary past that may render them suspect, he argues that each particular mechanism needs to be examined on its own; there can be no straightforward debunking of the entirety of moral psychology from the basic fact that many of the psychological mechanisms underwriting it were shaped by forces and pressures whose chief function was not to track moral truth. As noted, feelings of disgust can strengthen one’s moral judgments, rendering them more severe or certain in character. However, this tendency is not limited to judgments that have obvious connections with disgust. Put another way, we need not experience disgust in particular to feel as though certain moral transgressions are obviously—perhaps even objectively—wrong. Whether we are reflecting on general moral principles, more specific moral rules, or even judgments about particular cases, it is a familiar feature of moral cognition to feel as though it is imbued with objectivity—that is, with a commitment to moral questions having right and wrong answers independent of any given person’s or society’s beliefs or practices. In her contribution, Linda Skitka points out that our moral attitudes play an important role in generating this phenomenology. Moral attitudes are stronger, more enduring, and more predictive of a person’s behavior than other attitudes or preferences they might hold. Moral attitudes are distinguished by the fact that they are highly resistant—even impervious—to other desires or concerns, and have the force of imperatives for those who hold them. They have authority independent of others’ opinions or social conventions, and have particularly strong ties to emotions. People experience moral attitudes, convictions, or mandates as tracking objective features of the world that apply to all universally rather than subjective facts about themselves. Such convictions are inherently motivating and accompanied by strong affect. Indeed, Skitka’s contribution creates an interesting wrinkle in our thinking about the morally good person. It is natural to think that morally good people have strong moral convictions—that they are willing to stand on principle and fight for what they believe is morally right. Yet, according to Skitka, this represents a potential dark side to our moral psychology. Strong moral convictions come with a price. Specifically, stronger moral convictions are often accompanied by intolerance of different moral beliefs, values, and practices;
12
Advances in Experimental Moral Psychology
the stronger the conviction, the greater the intolerance. Indeed, strong moral convictions—more so than any other strong attitudes—predict people’s lack of tolerance for different cultures, their unwillingness to interact with and help people with moral beliefs different from their own, their propensity to engage in vigilante justice against perceived wrongdoings, and their imperviousness to clear social disagreement with their views. In sum, Skitka’s contribution is a cautionary tale for moral psychologists; while we tend to focus on the positive aspects of moral development, there are pitfalls as well, including intolerance to differences and insensitivity to the rich complexity of moral life. We must keep in mind that morally good people walk a fine line between integrity and conviction on the one hand and intolerance or dogmatism on the other. Most of us have such strong moral experiences, supported by moral attitudes that seem particularly compelling, unshakeable, and rooted in some set of objective moral facts about the world around us. Nevertheless, we may wonder whether there are such things as moral facts and, if so, whether they are actually objective in ways suggested by our moral attitudes—that is, whether they are independent from any person’s or any group’s beliefs, values, or ways of life. Metaethicists have long sought to answer this question. Do moral judgments refer to objective moral facts or do they merely express subjective moral attitudes? Do participants in moral disputes argue over claims that can aspire to truth, or are they merely trading opinions with no objective basis? In pursuit of these questions, metaethicists have sought to capture the essence of morality as reflected in ordinary practice—how we as moral creatures experience morality in our daily lives. Recent experimental work has helped to reveal the mechanisms that may underlie our ordinary commitments to objectivism about morality, and how they might be related to other aspects of our psychological lives. James Beebe points out in his chapter that while people tend to attribute more objectivity to moral issues than other social or personal issues, their beliefs concerning the objectivity of morality do not stand alone. Rather, they are affected by a number of rather surprising factors. For example, people seem to be sensitive to the perceived existence (or absence) of consensus concerning moral issues in judging whether there is an objective truth underlying them. Moreover, Beebe discovered that people tend to ground issues more objectively when they consider them concretely—for example, as being contested by
Experimental Moral Psychology: An Introduction
13
particular individuals (individuals with names and faces)—than when they consider them abstractly—for example, as contested by nameless or faceless individuals. Beebe suggests that the very ways in which philosophers normally discuss metaethical issues—that is, abstractly and in rarefied fashion—may impact how they and others think of them. Finally, people display more objectivism when it comes to moral wrongs than moral rights—that is, they might be more objectivist concerning the wrongness of racial discrimination as opposed to the rightness of anonymous charitable donation. Individuals thus display a range of attitudes and commitments concerning the status of morality, belying treatments of ordinary moral practice that depict it as unified in any interesting respect. Beebe’s contribution invites future researchers to examine the nature of these commitments—and the things that influence them—more closely. Complementing these studies are others reported by Kevin Uttich, George Tsai, and Tania Lombrozo. As opposed to Beebe, who looks at how various external factors may affect folks’ views concerning the status of morality, Uttich et al. seek to better understand how beliefs in moral objectivity might be related to other beliefs individuals may hold. Specifically, they explore how belief in moral objectivism may be related to beliefs concerning moral progress and beliefs concerning whether or not we live in a “just world.” One believes in moral progress to the extent that one believes that the world is trending toward moral improvement over time (even if such change is slow, uneven, or not necessarily assured or certain), and one believes in a just world to the extent that one believes that virtue is rewarded and vice punished— that good things happen to good people and bad things to bad people. What Uttich et al. find is that while all three of these beliefs were correlated with one another, each of these beliefs nonetheless has some independent force or role in people’s conceptions of morality. Hence, folk views about the standing of morality might form part of a web of beliefs about how morality inheres and functions in the world, including beliefs such as whether good deeds are rewarded and bad deeds punished, or whether there is a tendency of moral progress in history. Both these contributions seek to understand folk views concerning the nature of morality by looking at how folk understand the nature of moral disagreement. This is in line with other recent studies in this area. Yet, Katinka
14
Advances in Experimental Moral Psychology
Quintelier, Delphine De Smet, and Daniel Fessler raise another possibility. Suppose X commits an action—say, cheating on her income taxes—and two people disagree as to whether this instance of cheating was moral or immoral: one maintains that the action was morally permissible, whereas another maintains that the action was morally wrong. If one thinks that at least one of these two appraisers must be wrong, then one reflects a commitment to moral objectivism. Yet are their perspectives, or the perspectives of any other people judging X’s action (a perspective referred to as appraiser relativism), the only relevant ones to consider when deciding whether what X did was wrong or right? Most existing studies seem to imply as much, as they focus on disagreement among appraisers in order to measure the extent to which folk may be committed to objectivity about morality. Yet it seems that there is another perspective worth considering—namely, X’s own. After all, it is easy to imagine X finding her own act (cheating on her income taxes) as either morally wrong or morally permissible. Does X’s assessment concerning the moral status of her own action matter when considering its actual status? If one takes X’s assessment of her own action as relevant to judging the moral status of the action itself, then one embraces an agent relativist stance toward morality. Quintelier et al. find that people do indeed take the agent’s own assessment of her action as relevant to assessing the actual status of the action, as well as the truth of statements about that action. They also found differences in the degree to which people expressed moral relativism depending upon whether they asked about agent relativism or appraiser relativism—a distinction most research in this area fails to make. Thus, whether people are relativists depends, in part, on whether they are asked agent or appraiser relative questions.
Measuring morality Of course, since this is a volume on advances in the empirical study of moral psychology, we would be remiss to not include some discussion about the challenges researchers face when it comes to measuring “morality.” And in their chapter, Peter Meindl and Jesse Graham do exactly this, raising an important methodological concern about one standard approach to the study of moral psychology. Most of the research conducted in this area has approached it
Experimental Moral Psychology: An Introduction
15
from a third-person (or “normative”) perspective, from which what counts as “moral” (whether morally good or bad) is determined by the researcher. Such an approach has obvious virtues, including defining morality in a way that is internally consistent and in line with current theoretical models. Nevertheless, this approach fails to take the perspective of those being studied—what they themselves find to be morally relevant, good or bad—and so fails to fully map the moral terrain. As just one example, personality researchers bemoaned the consistency of people’s behavior across different types of situations (“types” of situations being defined by the researchers), until they thought to ask people to identify for themselves the meaning they attributed to the situations to which they were responding; once they did, they found a remarkable degree of consistency across situations which, however externally different from the researcher’s or third-person perspective, were united by shared meaning from the first-person perspective. Thus, what looked like inconsistency from the outside ended up looking entirely coherent “from the inside”—an insight that is very important when trying to determine whether people possess situation-general traits, like virtues. Similarly, Meindl and Graham argue that without taking the first-person perspective into account, research into moral psychology will remain limited in its scope and application. Drawing on a number of existing studies, they call for increased use of the first-person perspective, and suggest ways in which the first- and third-person perspectives might inform and complement one another. While not their central focus or concern, several other chapters in this volume also introduce or address methodological issues. For example, Frimer and Oakes’s contribution reflects a commitment to first-person measurement, highlighting the advantages that can be obtained by using both self-report inventories and projective measures to uncover an interesting gap in people’s moral psychology between what values they explicitly endorse and what values they actually spend time pursuing. Skitka’s chapter explores the difference between moral conviction and other closely related—though importantly distinct—constructs. For example, unlike other strongly held attitudes, moral convictions uniquely predict a range of morally relevant behaviors, and unlike moral judgments, moral convictions are stable and relatively impervious to standard priming manipulations. This last finding should prompt researchers to expand upon the range of moral cognition being measured. Similarly, one of
16
Advances in Experimental Moral Psychology
the challenges faced by Rust and Schwitzgebel is the question of what behaviors to measure—that is, how to operationalize moral goodness (or, in this case, its behavioral expression). They chose a wide variety of measures, everything from the extent to which people display courtesy and engage in free-riding behavior at conferences to peer evaluations and self-report measures of behavior. This range of different kinds of measurements is useful because it allows for a sort of “triangulation” on the subject of interest—in this case, the degree to which ethics scholars’ moral behavior differs (or not) from other academics. But, their study raises the very important question of how best to operationalize and measure moral cognition and behavior. Alfano’s model encourages researchers to look beyond the person herself to the entire context of moral behavior (including social/asocial environments) when operationalizing and measuring virtue, while Valdesolo’s encourages researchers to expand on the kinds of virtues (caring/other-oriented vs. competence/self-oriented) they include in an assessment of people’s moral psychology. Finally, Quintelier et al. raise yet another an important methodological consideration, arguing that those researching the metaethical commitments of ordinary folk need to be careful to specify (among other things) what type of relativism they are investigating, or whose perspective is being taken into account when assessing the objectivity of moral claims. This serves as just one important example of how collaboration between philosophers and psychologists would aid in the development of methodological approaches that are both scientifically rigorous and appropriately sensitive to important philosophical distinctions.
Conclusion The papers in this volume all speak to the vibrancy of research in moral psychology by philosophers and psychologists alike. Enduring questions concerning the nature of moral persons, the motivations to become moral, how to measure morality, and even the status and grounding of morality itself are each the focus of considerable research activity. This activity is driven both by theoretical commitments and by a sensitivity to empirical data that might shed light on the subject. We’ve highlighted the ways in which the research included here informs (and in some cases problematizes) our understanding of morality.
Experimental Moral Psychology: An Introduction
17
And while the contributions to this volume fall fairly evenly across the disciplines of philosophy and psychology, we hope it will be apparent that, at some level, these disciplinary categories seek to be of interest on their own. The questions at the heart of this research program have long histories in both disciplines, and the methods between them have begun to blur. Philosophers now use experimental methods, and experimental psychologists draw from (and contribute to) philosophical theorizing. The field is expanding, and we are delighted to mark some of its direction and vigor with this volume.
References Doris, J. (2010). The Moral Psychology Handbook. Oxford: Oxford University Press. Nadelhoffer, T., Nahmias, E., and Nichols, S. (2010). Moral Psychology: Historical and Contemporary Readings. Malden, MA: Wiley-Blackwell. Narvaez, D., and Lapsley, D. K. (2009). Personality, Identity and Character: Explorations in Moral Psychology. New York: Cambridge University Press. Sinnott-Armstrong, W. (2007a). Moral Psychology: The Evolution of Morality: Adaptations and Innateness (Vol. 1). Cambridge, MA: Bradford Book. —(2007b). Moral Psychology: The Cognitive Science of Morality: Intuition and Diversity (Vol. 2). Cambridge, MA: Bradford Book. —(2007c). Moral Psychology: The Neuroscience of Morality: Emotion, Brain Disorders, and Development (Vol. 3). Cambridge, MA: Bradford Book. Sytsma, J., and Livengood, J. (in press). The New Experimental Philosophy: An Introduction and Guide. Peterborough, Ontario, Canada: Broadview Press.
Part One
Moral Persons
1
The Character in Competence Piercarlo Valdesolo*
Perseverance, grit, efficiency, focus, determination, discipline, industriousness, fortitude, skill. According to existing models of person perception and most experimental studies of moral cognition, these kinds of traits are typically not considered to be relevant to evaluations of moral character (c.f. Pizarro and Tannenbaum 2011). Goodness and badness tend to be defined according to whether or not an individual is likely to be hostile or threatening toward the self. As such, traits and states associated with this perceived likelihood (e.g., compassion, empathy, and trustworthiness) dominate the literature in moral psychology. This flies in the face of a long tradition in virtue ethics that identifies the competence-based qualities listed above as belonging to a broader set of “intellectual virtues,” conceptually distinct but no less important to character than “moral virtues” (Aristotle, fourth-century B.C.E; Grube and Reeve 1992). Of course, studies of person perception are simply interested in describing the qualities that contribute to overall assessments of others, and not at all concerned with philosophical accounts of what kinds of traits ought to compose moral character. That said, it might strike some as odd that such intellectual virtues are not, according to existing evidence, considered by most perceivers to be morally relevant. And it might strike some as odd that investigations into the processes by which we evaluate others along these dimensions are relatively rare in moral cognition research. This chapter intends to argue that such evaluations might, in fact, be relevant to individuals’ assessments of moral character and that their categorization as amoral in person perception is misleading. Furthermore, their absence from work on moral cognition represents a gap in the literature that might be profitably filled in the future. I will begin by describing the most
22
Advances in Experimental Moral Psychology
well-supported model of person perception (Stereotype Content Model), and how the core components of this model are thought to relate to moral judgments. I will then argue that the underemphasis of these “intellectual virtues” results from a tendency to see the moral virtues as other-oriented and the intellectual virtues as self-oriented. This distinction, however, might be dependent on a prioritization of short-term interests as compared to longterm interests in orienting ourselves to others. Specifically, I will suggest ways in which the behavior promoted by the “intellectual virtues” might be equally important to other-oriented outcomes. As such, intellectual virtues might be relevant to evaluations of moral character that consider long-term outcomes, and their importance might not be reflected in individuals’ first impressions of interaction partners. In other words, the relevance of competence-based traits to moral character may track the impact that those traits have on others’ well-being. Since ultimate human flourishing requires societies composed of individuals who are both warm and competent, the moral relevance of competence will increase as perspective shifts from the short term to the long term. I will conclude with suggestions for how moral cognition could benefit from the study of the processes underlying the intellectual virtues.
Person perception and moral character Research in person perception has identified two broad dimensions of social cognition that guide our global impressions of others, as well as our emotional and behavioral responses to interaction partners: warmth and competence (Fiske et al. 2007). Some form of these two dimensions can be traced back throughout much of the literature in social cognition, though they have taken on different labels depending on the particular theory. For example, Rosenberg et al. (1968) instructed participants to sort 64 trait words into categories that were likely to be found in another person. These participants generated two orthogonal dimensions of person perception: intellectual good/bad (defined by traits such as determined, industrious, skillful, intelligent), and social good/bad (defined by traits such as warm, honest, helpful, sincere)—two dimensions that are conceptually similar to warmth and competence. Recent research in face perception has also demonstrated the ease and speed with which participants
The Character in Competence
23
will judge trustworthiness and competence from short exposures to faces (Todorov et al. 2006; Willis and Todorov 2006)—again, two dimensions that overlap significantly, if not completely, with warmth and competence. These dimensions reflect basic and adaptive categories of evaluations: the need to anticipate actors’ intentions toward oneself (warmth) and the need to anticipate an actor’s ability to act on their intentions (competence). In other words, these evaluations allow us to answer the questions “Does the other intend help or harm?” and “can the other carry out this intent?” (Cuddy et al. 2008). Though this distinction in the literature on person perception maps very closely onto the two separate categories of virtues identified by philosophers (intellectual vs. moral), theorists have drawn a sharp divide between the moral relevance of these traits. Put simply, traits that communicate warmth are morally relevant, while traits that communicate competence are not. Virtue ethicists, on the other hand, see merit not only in character traits associated with what most contemporary models of person perception identify as “warmth,” but also in traits identifiable as relevant to “competence” (intelligence, fortitude, perseverance, skill; c.f. Dent 1975; Sherman 1989). This moral distinction is evident throughout much of the literature in social cognition and is presumed to be due to the self- versus other-focused nature of the traits associated with each dimension. Traits related to warmth, such as friendliness, honesty, and kindness, tend to motivate other-oriented behavior (e.g., altruism), whereas traits associated with competence, such as efficacy, perseverance, creativity, and intelligence, tend to motivate self-oriented behavior (e.g., practicing a skill; Peeters 2001). Indeed, past theories that have posited a similar kind of two-dimensional approach to person perception have made the distinction more explicit by using the labels of “morality” and “competence” to describe the two kinds of evaluations (Phalet and Poppe 1997). Some work even conceptualizes these two domains as operating in tension with one another, arguing that traits promoting other-oriented concerns interfere with the development of traits promoting self-interest, and vice versa (Schwartz 1992). Why this asymmetry in the moral relevance between self- versus otherfocused traits? One possible interpretation is that perceivers value otheroriented traits because they are thought to be more likely to directly benefit
24
Advances in Experimental Moral Psychology
themselves. Warmth motivates others to care for us whereas behaviors motivated by others’ competence do not seem to directly impact our fortunes. The kinds of appraisals that are thought to underlie judgments of warmth and competence seem to corroborate such an interpretation. Specifically, evaluations of competition and status have been found to predict the degree to which individuals judge others to be competent and warm, respectively. I will discuss these in turn. Individuals and groups are “competitive” if they have goals that are perceived to be incompatible with the goals of the perceiver. For example, competitive others would desire to maximize their own resources at the expense of others’ ability to acquire resources. These assessments inform our judgments of others’ social intents, carving the social world up into those who intend to facilitate the achievement of our own goals, and those who seem to have no such intention. Perceptions of warmth follow directly from this evaluation. Appraisals of “competition” track closely to group membership. Because ingroup members tend not to compete with a perceiver for resources (though this may vary depending upon the ingroup in question), they are judged to be low in competitiveness and therefore trigger perceptions of warmth, while outgroup members are judged to be higher in competitiveness and, therefore, colder. Similar effects would be expected regardless of the dimensions along which group membership is perceived. Perceived similarity to another is dynamically evaluated along multiple dimensions of identity (Tversky 1977). Any such perceived similarity should trigger appraisals of low competition and high warmth. In sum, targets with whom we share identity, and consequently from whom we can expect to benefit, are considered to be high in warmth. These considerations, in turn, determine perceptions of moral character. Evaluations of the “status” of individuals and groups inform judgments of competence given the assumption that status is a reliable indicator of ability. The degree to which an individual is capable of pursuing and achieving her goals is presumed to be reflected in their place in society. As such, high-status targets are judged to be more highly competent than low-status targets. These considerations are not considered to be morally relevant. Taking the effect of status and competition on person perception together, it seems that the moral relevance of a trait is largely defined by the degree to which that trait motivates behavior that confers benefits on anyone other than
The Character in Competence
25
the actor—in other words, behavior that is more likely to bring about behavior that profits the self. The valence of the emotional responses to targets categorized along these dimensions supports this view. The stereotype content model posits specific sets of emotional responses triggered by the various combinations of these two dimensions and, in thinking about their relevance to moral judgments, it is instructive to examine the content of these emotions. The perception of both warmth and competence in targets elicits primarily admiration from others (Cuddy et al. 2008). These individuals are evaluated as having goals that are compatible with those of the perceiver, and they have the skills requisite to help perceivers achieve those goals. In other words, individuals high in both warmth and competence are our most socially valued interaction partners. The perception of warmth without competence elicits pity, competence without warmth triggers envy, and the absence of both triggers contempt and disgust (Cuddy et al. 2008). The organization of these emotional responses with regard to self- versus other-benefiting actions also squares nicely with recent research showing the paradoxical nature of perceiver’s responses to moral behavior on the part of individuals perceived to have traits that seem similar to warmth or competence (Pavarini and Schnall 2014). Admiration is elicited in response to the moral behavior of warm (i.e., low competition) others, while the same behavior by those considered to be low in warmth (i.e., high competition) elicits envy. Indeed, the fact that perceivers discount the moral relevance of competence traits relative to warmth traits could be a simple function of an ingroup bias. We have negative emotional responses toward outgroup competent others, because they might not be favorably oriented toward us. The bias against the importance of general competence in judgments of moral character, compared to general warmth, seems to be a reflection of a self-interested motivation to maximize the likelihood of resource acquisition. Evidence in line with this interpretation shows that perceivers value competence more positively in close others (“a close friend”) compared to less close others (“distant peers”). Though warmth judgments still carry more weight in predicting positivity toward others compared to competence, competence only becomes relevant to character when it is perceived to have consequences for the self (Abele and
26
Advances in Experimental Moral Psychology
Wojciszke 2007). For example, the positivity of an employee’s evaluations of his boss tracks competence only when the employee’s fate is tied to the decisions of the boss.
Competence as long-term other-orientation Literature on person perception has largely agreed that, while evaluation of others proceeds along both dimensions, warmth holds primacy over competence. The rationale for such an asymmetry in value has typically been explained as follows: “From an evolutionary perspective, the primacy of warmth makes sense because another’s intent for good or ill matters more to survival than whether the other can act on those goals” (Cuddy et al., p. 89). These kinds of statements imply that warmth and competence differ in their immediate relevance to a perceiver’s well-being. As such, the importance of a trait to judgments of others’ character is proportional to the relevance of that trait to one’s own survival. Though there seems to be no dispute that warmth is more relevant to immediate survival, could it be that competence has similar consequences as time passes? Could it be that the behaviors motivated by competence, though they do not benefit others in the short term, benefit others in the long term? This section will argue (a) that competence-related traits in targets do increase the likelihood of benefits to perceivers over the long term and (b) that evidence arguing for the relevance of warmth to character judgment might be failing to detect the importance of competence to character evaluations because of its focus on evaluations of short-term impressions. The final section will argue for the importance of incorporating how people assess competence in general—and over time—into the study of moral cognition, as well as offer some initial ways in which this pursuit might move forward. Though warmth in interaction partners might matter more for achieving the short-term interests of perceivers, there are reasons to believe that longterm benefits would depend on valuing the competent as well. Indeed, the argument for the importance of traits that cultivate individual skills becomes even more central to others’ long-term well-being if you consider the unit of analysis to be the group as opposed to the individual.
The Character in Competence
27
The idea that groups flourish when individuals are motivated to pursue their own interests is not new. Adam Smith argued in the Wealth of Nations for the pursuit of immediate self-interest as the key to flourishing societies. His theorizing on the power of free markets suggests that it is precisely the drive for self-interest through which societies advance. This idea was captured in Smith’s metaphor of the invisible hand: collective well-being is best achieved by groups of individuals who pursue their own advancement without concern for others. The engine of this process is specialization. Focusing individuals’ efforts on skills/domains in which they have a comparative advantage ultimately benefits a community by maximizing the collective capabilities of group members, allowing for a potentially wider and richer distribution of resources as well as a competitive advantage relative to other groups. Consequently, cultivating traits that foster such an end may ultimately benefit the community by enhancing the collective competence of a population. This argument assumes that collective value is, at least in part, created by self-focused motivational states associated with competence-based traits. Indeed, I largely agree with Smith’s sentiment that “by pursuing his own interest he frequently promotes that of the society more effectually than when he really intends to promote it” (Smith 1776/1937). That said, societal flourishing cannot be achieved through these kinds of motivations alone. Specialization only pays off when a collective defined by the free-flowing exchange of resources has been established. In other words, societies flourish when composed of individuals who (a) maximize their individual potential in terms of skills/abilities and (b) are willing to exchange those resources with others. What drives this willingness? Smith initially offers a strong answer: “it is not from the benevolence of the butcher, brewer or baker that we should expect our dinner, but from a regard for their self-interest.” The argument put forward in this chapter, however, suggests an alternative. It is not solely through regard to self-interest that people in groups should expect the beneficence of others—it is also through the benevolence of the butcher, brewer, and baker. Societies composed of the warm and the competent should ultimately thrive, and structural and legal constraints should reflect the collective value inherent in both traits. A critical insight provided by sociobiology has been the adaptive value of other-interested drives—those that cultivate perceptions of warmth in others.
28
Advances in Experimental Moral Psychology
If competence-based traits motivate behaviors that contribute to specialization, then warmth-based traits motivate behaviors that contribute to the desire to exchange the fruits of such specialization. A balance of these motivations maximizes long-term well-being. Societies composed of individuals with intentions to act warmly toward others as well as the capacity to act on those intentions will best achieve long-term collective well-being. Theories arguing for the importance of competence without warmth ignore the social function of other-interested states (particularly their role in mediating the emergence of reciprocal altruism, c.f. Trivers 1971), and those arguing for the importance of warmth without competence ignore the importance of the process through which individual resources contribute to collective value. Other-interested motivations solidify social groups by establishing and maintaining mutually beneficial relationships and by providing the proximal mechanisms that motivate the exchange of resources individuals have accrued. As such, the traits that underlie perceptions of warmth and competence are essential in the ultimate creation of flourishing societies. This perspective fits well with the finding that person perceivers admire those who are both warm and competent: these are the kinds of individuals who ultimately contribute to the flourishing of social groups. It also fits well with the finding that people think warmth is morally relevant, since immediate and long-term intentions will have consequences for the self. It raises the question, however, as to whether people also think that, under some circumstances, competence can also speak to moral character (c.f. Valdesolo and DeSteno 2011). Because competence-based traits in targets might, over time, increase the likelihood of benefits to perceivers, and because this seems to be a crucial criterion for evaluating another’s character, then it’s possible that competence traits might take on more importance for moral character over the long term. Importantly, the effect of competence on perceived character might operate independently from perceptions of warmth. Smith draws a distinction between the efficacy of promoting societal interest incidentally (through focusing on the self) or intentionally (through focusing on others). For the purposes of judging moral character, it’s possible that judgments are sensitive to not just the likelihood that traits will have immediate positive effects on well-being, but perhaps also whether traits will have long-term positive effects on well-being. It is neither
The Character in Competence
29
surprising nor particularly controversial to argue for the moral relevance of other-interested traits to those who study social, or moral, cognition. But it is indeed a departure from the norm to posit the moral relevance of competencebased traits. Those who might worry that this argument champions the moral value of self-focused states over other-focused states need not worry. It does not challenge the view that perceptions of warmth will always remain morally relevant to perceivers. Instead, it simply proposes a change in the relative moral relevance of warmth and competence judgments as a function of the time horizon over which these judgments are made. As perspective shifts from the short to the long term, and the effects of being surrounded by the competent become more tangible, competence traits might have a greater impact on judgments of character. Suggestive evidence of this idea comes from work evaluating the traits displayed by moral exemplars (Frimer et al. 2011; Frimer et al. 2012). Twenty-five recipients of a national award for extraordinary volunteerism were compared to 25 demographically matched comparison participants with the specific aim of comparing the degree to which these moral exemplars displayed traits associated with both agency (i.e., competence) and communion (i.e., warmth). Results suggested that exemplars were consistently higher not only in communion but also in agency, as well as in the tendency to incorporate both these core dimensions into their personality. In particular, this work provides an empirical basis for why competence should be included in the study of moral psychology. One response to this argument might be that if it were the case that competence traits are relevant to moral judgments, there would be more evidence of it in person perception and moral cognition. Much of the research into the two dimensions of person perception relies on perceivers “spontaneous” interpretations of behavior or “impressions” of others (c.f. Cuddy et al. 2008, p. 73). Studies of person perception deal with immediate evaluations of others’ character—snap decisions based off minimal amounts of information. If judgments of others’ moral character are tied to inferences about the degree to which another’s behaviors might profit the self, then it follows that we should asses others’ character along dimensions of warmth during first impressions. Someone’s competence is less immediately relevant to well-being compared to whether they desire to hurt or harm you.
30
Advances in Experimental Moral Psychology
In this context, warmth might be more central to character judgments only because it is the most obviously other-oriented dimension of perception. However, in contexts where perceivers are judging the importance of cultivating particular traits over the long term, competence might become more central to character judgments. This idea leads to quite simple and testable predictions about how the composition of our evaluations of moral character might change depending on the time horizon under consideration. In the same way that men and women have been found to value different traits when judging the attractiveness of potential short- versus long-term relationship partners, perceivers might value different traits when judging the character of shortversus long-term interaction partners. Characteristics such as efficiency, perseverance, and determination might be considered more morally relevant in contexts when the behaviors that those traits motivate are more relevant to outcomes for the perceiver. Previous work, as described earlier, has shown that competence becomes more relevant to people’s global evaluations of others as their fate becomes more tied to the target (Woczjiske and Abele 2008); however, no study to my knowledge has examined the consequences of these kinds of fate-dependent manipulations on moral judgments. In line with the idea that evaluations of character seem yoked to the degree to which individuals see others as helping facilitate their goals, fate dependence might have an impact on character judgments via inferences of competence. Indeed, there is already evidence suggesting that competence traits form an important part of one’s own moral identity. Recent studies of the composition of moral identity find that adjectives such as “hardworking” are considered central to the moral self and predictive of moral cognition and behavior (Aquino and Reed 2002). This makes the discounting of such traits in others’ character even more interesting and worthy of research. Why is it that most existing theories of person perception define “moral” traits as orthogonal to traits rooted in competence, even though we seem to readily acknowledge the moral value of competence for the self? The relative weight given to warmth and competence in defining the moral identity of the self and others could be another interesting avenue to test the degree to which judgments of character depend on the self-interested preference for traits that help achieve one’s goals.
The Character in Competence
31
In sum, if the import, and perceived moral relevance, of warmth traits is due to the immediacy of the effects of such traits on others’ well-being, then it may be the case that competence-related traits would increase in importance, and moral relevance, if the decision context were framed differently. Though someone’s industriousness might not matter for immediate evaluations of moral character, such a trait might take on moral meaning if participants felt, for example, that their fate were connected to this individual in some way over the long term.
New direction for moral cognition This perspective suggests several interesting avenues for future research in the study of moral cognition, judgment, and behavior. First, it implies that the centrality of traits to moral character during person perception might be distinct from those considered central to one’s own moral identity. The moral relevance of competence-related traits to one’s own identity, but not others, speaks to this possibility. What other differences might there be? More generally, what are the processes underlying assessments of others’ abilities to carry out their goals? To what degree do we weight an individual’s determination, grit, or perseverance in assessing moral character and under what conditions? Does it matter if they are ingroup or outgroup members? Does it matter whether we are making judgments that seem to only have immediate or also long-term implications for the self? Two current areas of interest in moral psychology to which this perspective might be fruitfully applied include (1) work testing the role of intent and outcome on moral judgments, and (2) the relationship between mind perception and moral judgment. With regard to the former, work from the lab of Fiery Cushman (Cushman 2008; Cushman et al. 2009) has posited two processes along which moral evaluations proceed: one which is sensitive to the causal relationship between an agent and an outcome, and the other which is sensitive to the mental states responsible for that action. It is tempting to see processes associated with the former as relevant to competence-based evaluations. An individual’s ability to achieve a goal (competence) might serve as input into a determination of whether an agent is, or is likely to be,
32
Advances in Experimental Moral Psychology
causally responsible for morally relevant outcomes, whereas an individual’s intentions for harming or helping others (warmth) might be more directly related to determinations of the mental states of actors. Given that this research identifies distinct patterns in moral judgments associated with evaluations of intent and causal connection, it is possible that inferences of competence and warmth might show similar patterns of relationships to particular kinds of moral judgments. For example, causal responsibility seems to be more directly relevant to judgments of punishment and blame while intent matters more for judgments of wrongness or permissibility. Might the moral relevance of perceived warmth and competence follow a similar pattern? Could perceiving general competence in individuals make them more morally blameworthy for outcomes? In exploring this possibility it would be important to distinguish how competence judgments might influence moral evaluations of behavior (responsibility, intent, blame) from moral evaluations of character (how good/ bad is this person). It may be that interesting patterns of responses emerge from considering these kinds of moral judgments separately. A distinction between act-centered models of moral judgment and person-centered models of moral judgment has recently been emphasized in the literature (e.g., Pizarro and Tannenbaum 2011). On this account, moral judgments are often formed by considering the moral character of the individual involved. And assessments of intent, control, responsibility, and blame might be unified in their relationships to underlying assessments of “who the actor is and what he or she values” (Pizarro and Tannenbaum). The second area of research in moral cognition for which the moral relevance of competence might have interesting implications is recent theorizing on the relationship between mind perception and morality (Gray et al. 2007, 2012). These theories posit that all moral judgments require perceiving two distinct kinds of interacting minds: agents and patients. Agents have the capacity to intend and carry out action, while patients are the recipients of agents’ actions. Interestingly, agents are defined by traits that seem more conceptually associated with competence (self-control, planning, thought) but also “morality.” In models of mind perception the distinction between warmth and competence in terms of moral competence seems to disappear. How can theories of person perception, which draw a sharp
The Character in Competence
33
distinction between morally relevant capacities of warmth and the amoral capacities of competence, be reconciled with emerging theories of mind perception in moral psychology? As I have argued throughout this chapter, the answer may be that the distinction drawn in person perception is misleading and, perhaps, a function of the contexts in which such studies have been conducted.
Conclusion In conclusion, this chapter serves as a call for increased attention toward the processes underlying the evaluation of others’ competence in moral judgments of them and, consequently, renewed attention to the role of such traits in theories of morality more generally. Moral cognition has focused almost exclusively on traits related to warmth (kindness, altruism, trustworthiness) and has paid relatively little attention to how we assess others’ capacities to achieve their goals. These self-focused traits—discipline, focus, industriousness—have long been considered relevant to moral character by virtue ethicists, and their absence from psychological theories of person perception is, at the very least, worthy of more direct empirical attention.
Note * Author’s Note: Piercarlo Valdesolo, Department of Psychology, Claremont McKenna College. Correspondence should be addressed to Piercarlo Valdesolo, 850 Columbia Avenue, Claremont, CA 91711. Email: [email protected].
References Abele, A. E., and Wojciszke, B. (2007). Agency and communion from the perspective of self versus others. Journal of Personality and Social Psychology, 93(5), 751. Aquino, K., and Reed, A. (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83(6), 1423–40. Aristotle (4th Century, B.C.E./1998). The Nicomachean Ethics. Oxford: Oxford University Press.
34
Advances in Experimental Moral Psychology
Cuddy, A. J., Fiske, S. T., and Glick, P. (2008). Warmth and competence as universal dimensions of social perception: The stereotype content model and the BIAS map. Advances in Experimental Social Psychology, 40, 61–149. Cushman, F. (2008). Crime and punishment: Distinguishing the roles of causal and intentional analyses in moral judgment. Cognition, 108(2), 353–80. Cushman, F., Dreber, A., Wang, Y., and Costa, J. (2009). Accidental outcomes guide punishment in a “trembling hand” game. PloS ONE, 4(8), e6699. Dent, N. J. H. (1975). Virtues and actions. The Philosophical Quarterly, 25(101), 318–35. Fiske, S. T., Cuddy, A. J., and Glick, P. (2007). Universal dimensions of social cognition: Warmth and competence. Trends in Cognitive Sciences, 11(2), 77–83. Frimer, J. A., Walker, L. J., Dunlop, W. L., Lee, B. H., and Riches, A. (2011). The integration of agency and communion in moral personality: Evidence of enlightened self-interest. Journal of Personality and Social Psychology, 101(1), 149–63. Frimer, J. A., Walker, L. J., Riches, A., Lee, B., and Dunlop, W. L. (2012). Hierarchical integration of agency and communion: A study of influential moral figures. Journal of Personality, 80(4), 1117–45. Gray, H. M., Gray, K., and Wegner, D. M. (2007). Dimensions of mind perception. Science, 315(5812), 619. Gray, K., Young, L., and Waytz, A. (2012). Mind perception is the essence of morality. Psychological Inquiry, 23(2), 101–24. Grube, G. M. A., and Reeve, C. D. C. (1992). Plato: Republic. Hackett: NJ. Pavarini, G., and Schnall, S. (2014). Is the glass of kindness half full or half empty? In J. Wright and H. Sarkissian (eds), Advances in Experimental Moral Psychology: Affect, Character, and Commitments. Continuum Press. Phalet, K., and Poppe, E. (1997). Competence and morality dimensions of national and ethnic stereotypes: A study in six eastern-European countries. European Journal of Social Psychology, 27(6), 703–23. Pizarro, D. A., and Tannenbaum, D. (2011). Bringing character back: How the motivation to evaluate character influences judgment of moral blame. In M. Mikulincer and Shaver, P. (eds), The Social psychology of morality: Exploring the causes of good and evil. APA Press: Washington DC. Rosenberg, S., Nelson, C., and Vivekananthan, P. S. (1968). A multidimensional approach to the structure of personality impressions. Journal of Personality and Social Psychology, 9(4), 2. Schwartz, S. H. (1992). Universals in the content and structure of values: Theoretical advances and empirical tests in 20 countries. Advances in Experimental Social Psychology, 25(1), 1–65.
The Character in Competence
35
Sherman, N. (1989). The Fabric of Character: Aristotle’s Theory of Virtue (Vol. 6). Oxford: Clarendon Press. Smith, A. (1937). The Wealth of Nations (1776). New York: Modern Library, p. 740. Todorov, A., Pakrashi, M., and Oosterhof, N. N. (2009). Evaluating faces on trustworthiness after minimal time exposure. Social Cognition, 27(6), 813–33. Trivers, R. L. (1971). The evolution of reciprocal altruism. Quarterly Review of Biology, 46(1), 35–57. Tversky, A. (1977). Features of similarity. Psychological Review, 84(4), 327–52. Valdesolo, P., and DeSteno, D. (2011). Synchrony and the social tuning of compassion. Emotion-APA, 11(2), 262. Willis, J., and Todorov, A. (2006). First impressions making up your mind after a 100-ms exposure to a face. Psychological Science, 17(7), 592–8. Wojciszke, B., and Abele, A. E. (2008). The primacy of communion over agency and its reversals in evaluations. European Journal of Social Psychology, 38(7), 1139–47.
2
Spoken Words Reveal Selfish Motives: An Individual Difference Approach to Moral Motivation Jeremy A. Frimer and Harrison Oakes*
While running for the US presidency in 2012, Mitt Romney made numerous promises. To various audiences, he accumulated at least 15 major pledges of what he would accomplish on his first day in office, should he be elected. These included approving an oil pipeline, repealing Obamacare, sanctioning China for unfair trading, submitting five bills to congress, increasing oil drilling, and meeting with Democrat leaders. By any realistic account, this collection of pledges was unfeasible for a single day.1 Romney’s ambitious avowals raise the question: Is such unrealistic over-promising out of the ordinary? Perhaps Romney’s promises are revealing of the situational pressures that politicians face when trying to appeal to voters. More broadly, perhaps most people— politicians and the populus alike—regularly feign a desirable exterior to garner social approval. Then again, perhaps Romney’s campaign vows are also revealing of something specific about Romney’s personality. When characterizing Romney’s policies after the primary elections, his advisor commented, “Everything changes. It’s almost like an Etch-A-Sketch. You kind of shake it up and restart all over again” (Cohen 2012). In other words, Romney’s team did not see his pledges as necessitating congruent actions. Perhaps some people, like Romney, are more duplicitous than other people. In this chapter, we present a case for both possibilities: that feigning a moral self is the norm and that some people do it more than others. We begin by reviewing an apparent paradox, that most people claim to be prosocial yet
Spoken Words Reveal Selfish Motives
37
behave selfishly. We interpret this inconsistency as evidence that humans have two distinct motives: (a) the desire to appear prosocial to others (appearance motives) and (b) the desire to behave in a way that benefits the self (behavioral motives). Self-report inventories thus paint an unrealistically rosy impression of human nature; this divergence has likely contributed to a widespread skepticism about self-reports in the social sciences. The overarching goal of this chapter is to discuss recent efforts to develop a more subtle motivation measure that accesses private behavioral motives— in a sense, a “metal detector of the soul.” This new measure relies on the projective hypothesis: a person’s spontaneously produced words are revealing of their inner psychological world. Projective methods have the potential to circumvent certain biases endemic to self-reports, augment the prediction of behavior, and detect novel and morally significant individual differences. We describe recent efforts to make projective methods less subjective and more expedient. And we conclude by exploring how these new efforts may open up new areas of moral psychology research.
Selfish or moral? A paradox A quick glance at the news headlines or textbooks on evolution, economics, or social psychology gives the impression that humans are primarily selfish (Haidt 2007). The view that human nature is primarily selfish has been popular among scholars throughout history (e.g., Hobbes, Spinoza, Adam Smith, and Ayn Rand). Empirical research has largely been supportive of this theory. As but one illustration within the social sciences, consider the dictator game wherein one person has a fixed amount of money to unilaterally and anonymously divide between the self and a partner. In reviewing studies on the dictator game, Engel (2011) found that most people take more for themselves than they give to their partner (see Figure 2.1). As a rough, first estimate of human nature, much theory and evidence suggest that selfishness is the rule. This broad impression of human nature as selfish generalizes to behaviors in many contexts. However, it does not manifest when people describe themselves. Self-descriptions tend to be more prosocial than selfish. To illustrate this, we examined three of the more prevalent self-report inventory measures that tap
38
Advances in Experimental Moral Psychology
Prosocial 13% Evenhanded 17% Selfish 70%
Figure 2.1 Most people behave selfishly. The pie chart shows the percentage of people who behave in three different ways in the dictator game (N ⫽ 20,813). The majority of players selfishly take more money than they give. Minorities of people prosocially give more than they take or even-handedly divide the money equally between the self and partner. Adapted from Engel (2011).
Traits
0.84
Values
1.18
Goals –1.5
1.47 –1 (Selfish)
–0.5
0
0.5
Effect size, Cohen’s d
1
1.5
(Prosocial)
Figure 2.2 Most people claim to be moral. Bars represent the effect sizes of selfreported inventories of goals, values, and traits, based on published norms. Across all three psychological constructs, people see prosocial items as being more selfdescriptive than selfish items. Calculated from published norms in Trapnell and Broughton (2006), Schwartz et al. (2012), and Schmuck et al. (2000), for traits, values, and goals, respectively.
prosocial and selfish tendencies—goals (Schmuck, Kasser, and Ryan 2000), values (Schwartz et al. 2012), and traits (Trapnell and Broughton 2006). For each construct, we calculated a score reflecting how prosocial versus selfish the population is, operationalized by an effect size (Cohen’s d; prosocial minus selfish).2 The results are shown in Figure 2.2. Across all three psychological constructs, participants’ self-reports portray a distinctly prosocial human nature.
Spoken Words Reveal Selfish Motives
39
By conventional standards, effect sizes are in the large or very large ranges. Most people claim to be more prosocial than selfish. The general impression from the social sciences (e.g., from the dictator game)—that people are selfish—appears to contradict the general impression from personality psychology—that people claim to be moral. We will make a case that this apparent paradox is in fact revealing of a complexity (to put it nicely) or a hypocrisy (to put it bluntly) built in to human nature: a desire to appear prosocial while behaving selfishly.
Moral hypocrisy How do these disparate motives play out in human interaction? Daniel Batson’s coin-flipping experiments provide a compelling account of how, for most people, morality is primarily for show (Batson et al. 1997). Research participants were asked to choose one of two tasks to complete. One task offered participants a chance to win money; the second was boring and rewardless. Individuals were told that the next participant would have to complete whichever task they did not choose. However, this participant would remain unaware of the assignment process. The experimenter described the situation facing participants as a kind of moral dilemma, and explained that most people think the fair way to decide is by flipping a coin. However, participants were not required to flip the coin, nor were they required to adhere to the coin toss results should they choose to flip the coin. Participants were then left alone in a room with a coin and a decision to make. This set up a zero-sum situation in which one person’s benefit meant another person’s loss (essentially a variation of the dictator game). Moreover, the situation was effectively anonymous, with reputational forces stripped away. What would participants do? As we might expect from Figure 2.1, unapologetic selfishness was common in these studies. Roughly half of participants never bothered to toss the coin. Almost all (90%) of these participants immediately assigned themselves to the favorable task. The other half of the sample, however, submitted to the fair procedure of tossing a coin. Probabilistically speaking, about 50 percent of these participants would have won the coin toss and assigned themselves to the favorable task. However, 90 percent of participants who tossed a coin assigned
40
Advances in Experimental Moral Psychology
themselves to the favorable task, a full 40 percent more than probability odds would predict. Given the anonymity of their decision, participants who lost the coin toss found themselves in a bind. They were caught between the desire to act upon the result of the fair procedure and the desire to win some money. In the end, most (80%) of the people who lost the coin toss ignored the results and assigned themselves to the favorable task. Batson interpreted these data to suggest that among the corpus of human motives are distinct desires to behave selfishly and appear moral. Flipping a coin to fairly adjudicate the assignment of unequal tasks is a display of the desire to appear moral. Not only do we try to convince others of our good nature, we try to convince ourselves too by internalized and generalized selfbeliefs. In the Batson studies, participants completed a self-report measure of their moral responsibility. The measure predicted whether participants would flip the coin, but it did not predict how they actually behaved, meaning that participants’ self-proclamations were more closely linked to how they wanted others to see them than they were to private behavior (viz., assigning the task). Having discussed the majority of participants in Batson’s studies who exhibited moral hypocrisy or unabashed selfish behavior, we are left with the few laudable participants who assigned the other person to the good task, either with or without a coin toss. With no one watching, this important minority gave of themselves to benefit another person. Introducing these “givers” into iterated economic games directly benefits fellow players (in terms of payouts) and encourages generosity from them to one another (Weber and Murninghan 2008). Whereas these givers may appear to be self-sacrificial, over time, they tend to reap rewards for their generosity. What sets them apart from the selfish and the hypocrites is that their self-interest is interwoven with the interests of those around them. Givers are probably the sort of people one would prefer as a babysitter, colleague, or government representative, given their honorable behavior. Society would benefit from an efficient means of detecting this minority of the population, which would also raise the likelihood of catching hypocrisy, thus making prosocial behavior more attractive to would-be hypocrites. We next explore whether and how moral psychology might develop a personality tool that detects these honorable individuals.
Spoken Words Reveal Selfish Motives
41
Moral is as moral does? The protagonist of the 1995 film, Forrest Gump, played by Tom Hanks, was not a bright man. He had an IQ of 75. He was inarticulate and had a poor grasp of social rules, cues, and expectations. Yet his behavior was mysteriously brilliant. To name a few of his accomplishments, he was a football star, taught Elvis Presley to dance, was a Vietnam war hero, started a multimillion dollar company, and met three US presidents. Was Forrest smart? Or was he stupid, as his IQ and peers suggested? When asked directly, Forrest retorted, “Stupid is as stupid does.” In other words, behavior—not thoughts or words—is the true measure of a person. Forrest’s ontological stance coincides with the general feeling in social psychology: the best way to know a person is by observing their actions, not their words. A person is as a person does, not as he/she claims. This assertion may be grounded in the notion that people have poor insight about the causes of their own behavior (Nisbett and Wilson 1977). The self-congratulatory impression emerging from self-reports (see Figure 2.2) in conjunction with self-serving behavior in economic games (see Figure 2.1) might seem to add to the skepticism. We believe that this degree of skepticism about the validity and utility of self-reports in understanding behavior is overly dismissive. Reports from the person (self-report inventories and projective measures, combined) can offer a reasonably accurate picture of human nature and improve predictions of behavior. The key to prediction is to expand the toolset beyond inventories.
Limitations of self-report inventories I (the first author) came to psychology after completing a degree in engineering physics, wherein the tools of the trade were sophisticated machines housed in metal boxes with names like “oscilloscope.” I anticipated that psychology had its own raft of sophisticated measurement devices, which it does. However, most tools for measuring the person are astonishingly simple. The most common personality measure involves self-reporting one’s traits (e.g., extraversion) by ticking boxes on a Likert scale. This seemed imprecise, biased
42
Advances in Experimental Moral Psychology
by self-presentation demands, and missing what felt to me to be the essence of a person (e.g., a person’s history, aspirations, fears, and desires). Sixty years earlier, Allport (1937) arrived at a related sentiment: “Nor can motives ever be studied apart from their personal settings, they represent always the strivings of the total organism toward its objective” (p. 18). Self-report inventories do have their benefits; indeed, no one has more information about any particular person than that person him/herself. Nevertheless, these self-reports seem untrustworthy, especially if one end of a Likert scale represents something socially desirable, such as prosocial motivation. An alternative means of knowing a person is through the words that he/she spontaneously speaks. Premising this approach is the projective hypothesis, which essentially states that the more a person cares about something, the more the person will speak about that thing. Projective methods usually begin in an interview, either about some ambiguous picture (e.g., Rorschach inkblots, Thematic Apperception Test) or one’s own life. After recording and transcribing the interview, trained researchers code each story for the presence of absence of a particular theme (e.g., power, achievement, or intimacy). The scores from this coding process tend to be highly predictive of behavior. Why then do mainstream empirical researchers tend to avoid studying projective methods? We propose two pragmatic reasons.
Objectivity Researchers may be weary of projective methods because science demands objective, replicable measurements with researcher bias minimized. The moment an interview begins, an avalanche of conflating factors compromise the validity of the data. Among these are the interviewer’s personal views, preferences, and knowledge of the status of the individual.
Expedience Conversely, researchers may be attracted to self-report methods owing to expedience. Self-report measures require few resources, can be collected online or from groups of participants at the same time, and can be analyzed the same day. In the current era, expedience is a prerequisite to feasibility. In contrast, interviewing, transcribing, and coding require a significant resource investment.
Spoken Words Reveal Selfish Motives
43
The revealing nature of the spoken word Both expedience and objectivity are important concerns, but so is prediction. We suggest that neither of the two pragmatic reasons are sufficient grounds for neglecting the richness of projective methods, especially in the twenty-first century. Technological advances of recent years have opened new opportunities to using spoken and written words, expediently and objectively. Later in this chapter, we describe our current efforts to develop a new projective measure to better understand a person’s moral character. But first, we briefly outline the traditional projective method. Analyzing spoken words is founded on the projective hypothesis: when trying to make sense of a stimulus that has no clear meaning, people create meaning, thereby projecting the thoughts that are chronically accessible in their mind. Rorschach inkblots and the ambiguous picture scenes of the Thematic Apperception Test were early tests built upon the projective hypothesis (Frank 1939). Respondents were asked to make sense of what they saw; in doing so, they projected motives such as achievement, power, or intimacy to tell a story about the stimuli. Complex scoring systems (Exner 1993; McClelland 1975) detailed how coders should assign scores based on the respondent’s answers. Dan McAdams (2001) adapted and expanded upon these scoring systems to empirically study personal life stories. In an interview context, participants describe various life events, such as earliest memories or a turning point event wherein some major life change takes place. Later, the interviews are transcribed and each event is coded for the presence or absence of various themes, and then tallied to form a narrative metric. Among the menu of available coding themes are (a) affective tone, ranging from optimistic to pessimistic; (b) redemption, wherein a negative event gives rise to a positive one; (c) contamination, wherein a positive event gives rise to a negative one; (d) agency, which includes themes of power, achievement, and empowerment; and (e) communion, which includes themes of help, love, and friendship. A founding premise of the projective method is that data garnered from spontaneous words are non-reducible to scale/inventory data. Personal stories contain ideographic information, structure, and meaning that—qualitatively— scales cannot capture. Knowing that a person is highly extraverted or emotionally stable does not tell you which stories they will share about their
44
Advances in Experimental Moral Psychology
earlier life, or what meaning those stories hold in the present. Additionally, personal stories contain important quantifiable information that is not accessible via self-report questionnaires. To illustrate this point in the moral domain, we consider whether personality measures can distinguish bona fide moral heroes from the general population. Walker and Frimer (2007) studied the personalities of 25 recipients of the Caring Canadian Award, a national award for sustained prosocial engagement. The authors also recruited a set of demographically matched comparison individuals, drawn from the community. All participants completed a battery of measures including inventories of self-reported traits (Wiggins 1995) and a projective measure—an individual Life Story Interview (McAdams 1995). The interview includes questions about high point events, low point events, turning point events, and so on. Common among the variety of stories that people told were weddings, the birth of children, the death or illness of friends or family, and work transitions. Awardees scored higher than comparisons on many of the measures, both inventory and projective. As we will show next, projective measures were more distinguishing of exemplars from comparisons than were inventories. Imagine reviewing the personality scores of the participants without knowing whether each participant was an awardee or comparison individual. The first line, for example, would contain an array of numbers representing the scores of a particular individual, say “Joe.” How accurately could you guess whether Joe is a moral exemplar or a comparison subject, based on the data at hand? To find out, we performed a logistic regression on the original data set, predicting group status (exemplar or comparison). With no predictors, correct classification was at chance levels—specifically, 50 percent. In the first step of the regression, we entered self-report personality data (the Big 5) for all participants. Correct classification improved from 50 percent to 72 percent, a significant increase above chance, ΔNagelkerke R2 ⫽ 0.27, p ⫽ 0.04. In the second step, we added the projective measure data listed above. Figure 2.3 shows how correct classification increased to near perfection (72% → 94%) with the addition of projective data, a further significant increase, ΔNagelkerke R2 ⫽ 0.55, p ⬍ 0.001. We tested which of inventory or projective data is a more powerful predictor by entering the variables in the reverse order (projective then inventory). In the
Spoken Words Reveal Selfish Motives Inventory then Projective Data
45
Projective then Inventory Data Step 2. Add Inventory Data 6%
Step 2. Add Projective Data 22% Step 1. Add Inventory Data 22%
Step 0. Chance 50%
Step 1. Add Projective Data 38%
Step 0. Chance 50%
Figure 2.3 Projective data adds to the predictive power of moral behavior. Correct classification of moral exemplars and ordinary comparison individuals in two logistic regression analyses. The left panel shows the correct classification based on chance (Step 0), Big 5 trait inventories alone (Step 1), and inventory with five projective measures (Step 2). The right panel shows correct classification based on the reverse ordering—projective then inventory. Calculated from Walker and Frimer (2007).
first step, projective data increased prediction above chance, from 50 percent to 88 percent, a significant increase, ΔNagelkerke R2 ⫽ 0.74, p ⬍ 0.001. In the second step, inventory data did not significantly augment the differentiation of exemplars from comparisons (88% → 94%), ΔNagelkerke R2 ⫽ 0.08, p ⫽ 0.19. By knowing a person’s self-report inventory scores and projective scores (and nothing else), one could correctly guess whether or not Joe was a moral exemplar, 19 times out of 20. Projective data, if anything, is the more powerful predictor of moral behavior. Telling a story is inherently different than making global attributions about oneself. For starters, telling a story is an ambiguously defined task. Even if a particular episode is specified (e.g., a turning point), one still needs to select a particular memory, determine a starting point, and then build a coherent story from there forward. The result is that each person tells a rather unique story. Consider one of the comparison participants in Walker and Frimer’s (2007) study. In response to a question about a high point event in his life, this
46
Advances in Experimental Moral Psychology
comparison participant began his story by describing how he had prepared for a vacation to Europe: I was going on a vacation to Italy. I invested around $4000 for the whole tour and whatnot. . . . Getting the Canadian passport was easy enough because I had one before, and it only took two weeks . . .
Surprisingly, this was not leading to fond memories of vineyards, beaches, friends and family, sunshine, relaxation, or traditional foods. The story quickly turned sour as the participant revealed that he had lost his passport. The story became labored, as he told of his struggles with government institutions and officials to try to recover his passport in time for his vacation. Several minutes later, he continued: . . . I went down, I showed them my income tax forms, and . . . that I’d paid my taxes, and this, that, and the other. And if you could speak to a Canadian government guy, and he could get you on the computer and talk to you just like you and I, it makes sense. But there’s no sense. You’re talking to a number. . . .
The sob story continued for roughly 10 minutes. Eventually the interviewer interjected, asking the participant to return to the high point of his story by asking whether he had received his passport in time for his vacation. His “high point” event ended in disappointment. No. . . . I didn’t want to go to no doctor and, you know, have to lie about being sick and all that. . . . As far as I was concerned, the holiday was over. You know, I’d spent that money. That was it.
This comparison participant’s “high point” story is of a tragic form. It had a contaminating tone, wherein a positive event (vacation) gave rise to a negative outcome (a lost passport and lost money); the affective tone was pessimistic; it was nearly devoid of prosocial communion themes; and most of the characters were antagonistic “villains.” This kind of story was common in the comparison group, but not in the exemplar group. The high points of moral exemplars were the kinds of generative stories that one would expect. As an example, one moral exemplar described a touching story in which a disadvantaged child savored a Christmas gift: Christmas Eve one year . . . [my wife and I] looked at all the gifts under our tree. . . . It was a true mass of gifts to be opened. And yet we still looked
Spoken Words Reveal Selfish Motives
47
at each other, and asked, sincerely, “Is there enough for the kids . . . to be happy?” We realized how fortunate our kids were, and how fortunate we were, that regardless of how the impact was going to be, or how minimal or how large it was going to be, we were going to start a program the following year. . . . It evolved very, very slowly from going to local stores asking for a hockey stick and a baseball glove, to donated wrapping paper. . . . Six hundred and fifty gifts, the first year . . . evolving to literally 80,000 gifts one year. . . . [We would take the gifts] into these small communities. . . . Very isolated, and exceedingly poor. . . . I can remember this one little girl . . . sat on Santa’s knee. . . . She was nervous. We provided her with a gift, which I knew was a doll from the shape of it. . . . [I was] quite anxious for her to open the gift; I wanted to see her reaction. But she didn’t. . . . After all the kids had received their gifts, I took one of the people from the community aside, and I said, “I was quite anxious for this one young girl to open her gift, but she didn’t.” I said. . . . “I wonder if she felt embarrassed, or if she felt awkward, or maybe she doesn’t understand the tradition of Christmas. . . .” And they said, “No, she fully understands, but this is December 22nd. That will be the only gift that she has. She will wait until Christmas morning to open that gift. . . .” And [that’s] the true essence of what that program is all about.
The contrast between these two stories (about the lost passport vs. the disadvantaged child receiving a cherished gift) illustrates both the profound richness and also the predictive utility of spoken words. Personal stories reveal a great deal about a person.
Motives in stories Having discussed broader issues concerning how self-report traits and specific life stories add to our understanding of the person, we now retain this distinction and return to specific questions about moral motivation. How can traits and spoken words help us understand the role of selfishness and morality in human functioning? One way of measuring a person’s moral motivation from their stories is by examining how often themes of agency and communion arise (Bakan 1966). A story with much agency (power, achievement) conveys a tone of selfishness. Conversely, a story rich in communal themes (benevolence, care, universalism) communicates prosociality. Given that prosociality is more socially desirable
48
Advances in Experimental Moral Psychology
than selfishness, one would expect themes of communion to be more prevalent than themes of agency in most self-proclamations. This is the case with selfreport inventories (see Figure 2.2). However, the opposite is found with spoken words; when people speak about their lives, they communicate selfishness. For example, when college students tell stories about peak experiences and earliest memories, agentic themes are more than twice as prevalent as communal themes (McAdams et al. 1996, Study 1). And both ordinary adults and moral exemplars have more than twice as many themes of agency (than communion) in their stories (Frimer et al. 2011). The relative strength of selfish agency and moral communion depends on the kind of measure used (viz., self-report endorsements vs. projected narratives), and may be revealing of differences in the psychological processes they each measure. Most people claim to be more communal than they are agentic on self-report measures of the importance of goals, values, and the description of traits (see Figure 2.2). Yet, these same people tell stories that are primarily about money, achievement, status, and recognition (agency), and less often about taking care of friends and family members, or contributing to the greater good (communion). In striking contradistinction to self-report measures, the themes emergent from narratives are primarily selfish. In other words, the impression arising from projective methods—that people are selfish—coincides with the predominant view (and corroborating evidence) that began this chapter. Why? One possible reason for this disparity between responses from inventories and projective measures is the frame of reference. Perhaps responding to self-report inventories prompts the respondent to take a third-person perspective—the Jamesian (1890) “me-self ”—to examine what the self is like from the perspective of another. Thus, reports reveal socially desirable features that are for public viewing, which tend to be moral. In contrast, narrating requires taking a first-person perspective—the Jamesian “I-self ”—to assemble memories and construct a coherent story from “the driver’s seat.” Thus, personality assessments derived from narratives are revealing of the agent’s private desires, which tend to be selfish. Perhaps the rarity of communion in life stories is not a fair test of whether people are innately selfish. The quantity (or frequency) of motives may not be an appropriate metric. Perhaps quality is what really matters. To project
Spoken Words Reveal Selfish Motives
49
a prosocial persona, perhaps people communicate (in one way or another) that the ultimate purpose for their agentic strivings is some communal end. Societal leaders may be particularly adept at explaining how numerous, proximal agentic goals (e.g., changing laws) serve a more distal, singular communal purpose (e.g., advancing the greater good). To test whether iconic leaders thus frame agency as a means to an end of communion, or simply dwell on agency, Frimer et al. (2012) examined speeches and interviews of some of the most influential figures of the past century, as identified in Time magazine’s lists (“Time 100,” 1998, 1999). In Study 1, experts—social science professors at Canadian universities—rated the moral character of each target to identify highly moral and less-moral leaders. Among the top 15 moral leaders were Nelson Mandela, Mohandas Gandhi, Aung San Suu Kyi, The Dalai Lama, Mother Teresa, and Martin Luther King, Jr. The bottom 15 less-moral leaders included Kim Jong Il, Eliot Spitzer, Vladimir Putin, Donald Rumsfeld, Mel Gibson, George W. Bush, and Adolf Hitler. In Study 2, trained coders examined each influential individual’s speeches and interviews, ascertaining both the implied means (agency or communion) and ends (agency or communion) of each speech or interview. Unsurprisingly, Gandhi and the other moral leaders treated agency as a means to an end of communion. Perhaps these icons garnered such public approval, in part, because of their ability to connect pragmatic agency with communal purpose. However, not all leaders did this. The speeches and interviews of Hitler and the other less-moral leaders were of a primarily agentic nature, with agency as a means to more agency. Agency and communion may be distinct, mentally segregated motives early in the life span, with the integration of agency and communion being a developmental achievement. Moreover, some people, such as moral exemplars, may be more likely to realize integration (Frimer and Walker 2009). These findings that words are revealing of the moral character of leaders contradicts the common-sense notion that the public words of leaders are a means of social persuasion—the product of the calculating minds of advisors and ghostwriters. This common sense seems to be overly dismissive of the wealth of information that spoken words communicate. We suggest that most people cannot help but project their own deeply held motives when speaking.
50
Advances in Experimental Moral Psychology
What is needed is an expedient, objective measure for detecting these inner motives, and then experimentally testing if and when people can fake high levels of prosociality.
How to spot a hypocrite: Toward an expedient, objective measure We conclude this chapter by describing ongoing efforts in our lab to concentrate the “active ingredients” of the projective hypothesis into “pill form”—an expedient, objective projective measure. The new projective measure assesses the density of agentic and communal words in texts that people produce, using computer software such as Linguistic Inquiry and Word Count (LIWC; Pennebaker et al. 2007). LIWC is a transparent, well-validated computer program (available from http://liwc.net) that counts all the words within a text file that matches the words in a specified dictionary. LIWC is expedient—it processes texts in seconds—and objective—no human coders are involved. Frimer and Oakes (2013) created agency and communion dictionaries for LIWC, and validated them against human coding. Using these dictionaries, LIWC produces density scores for agency and communion from a given text. These scores are then corrected to account for different dictionary sizes, and then used to calculate moral motivation scores. Moral motivation is calculated as follows: Moral Motivation ⫽ Communion ⫺ Agency. Positive moral motivation scores imply that a text is richer in communal words than it is in agentic words. Along with the usual reliability and validity concerns, we expect that the successful moral motivation measure will predict human behavior. For this objective to succeed, at least two specific hypotheses concerning general tendency and individual differences will need to be supported.
Hypothesis 1: Projected moral motives reveal selfish general tendency We predict that the projective measure will characterize human nature as selfish. To do so, projected moral motivation should be negative in the population. This criterion assumes that the theory of human nature is selfish
Spoken Words Reveal Selfish Motives
51
and the results from economics, biology, and social psychology—that people tend to behave selfishly—are accurate. Which content—agentic or communal—emerges most frequently when people speak about topics that matter to them? We predict that agentic content will be more common than communal content. Preliminary findings are confirming hypothesis 1: when describing important goals, people produce more agentic content than communal content (Frimer and Oakes 2013). This effect is not attributable to base rates of agency and communion in the dictionaries or typical English. When people talk about what matters to them, they selectively use more agentic words than communal words, communicating/revealing a selfish lifestyle. This selfish portrait emergent from the projective measure was the opposite of the impression emerging from a comparable endorsement inventory. Participants also rated the importance of their goals using the Aspiration Index (Grouzet et al. 2005), a standard self-report inventory of goals. The effect reversed: participants now rated their communal goals as more important than their agentic goals. These results support a dualistic theory of motivation. Results from the projective measure coincide with the general conclusion that people are selfish; results from the endorsement measure suggest the opposite, and may tap socially desirable appearance motives. In Batson’s studies, hypocrisy existed between two opposing behaviors (moral coin-tossing vs. selfish task assignment), with self-deception keeping the two at bay. In the present study, people demonstrated the coming apart of their own motives by acknowledging the primacy of their own selfish goals on a projective measure, then declaring their moral goals as most important while endorsing items on a goal inventory. On self-report inventories, people tend to present a moral self (see Figure 2.2); on projective measures, they tend to reveal their selfishness. Thus, the first criterion of a successful projective measure is supported: mean-level estimates from projective methods coincide with the general interdisciplinary conclusion that people are selfish.
Hypothesis 2: Projected moral motives predict moral behavior One of the limitations of the Batson studies is their limited utility in an applied setting. For example, the coin-flipping task could not realistically be used to screen prospective employees. One of the benefits of the projective measure
52
Advances in Experimental Moral Psychology
is adaptability and unobtrusiveness—it could be used in a variety of contexts. To be useful as such, however, the measure would need to meet a second criterion—predicting moral behavior. Lending initial support, the moral motivations found in the Nobel Peace Prize lectures are positive (viz., prosocial; Frimer and Oakes 2013). In contrast, the Nobel Literature Prize lectures scored neutrally (viz., equally prosocial and self-interested). These findings replicated with a variety of interviews and speeches of moral exemplars like Gandhi and leaders with the opposite moral reputation like Rumsfeld, Putin, and Hitler (from Frimer et al. 2012). Preliminary evidence is thus far supportive of the claim that the projective measure of moral motivation predicts moral behavior, and functions somewhat like a metal detector for private intentions.
Conclusion Projective measures have the potential to augment our understanding of human motives and enhance our ability to detect moral character in the real world. Individual differences in the strategies of pursuing selfishness remain of the utmost concern to building civil society. If the projective hypothesis is as useful as we are supposing, the possibilities for predicting prosocial behavior naturalistically are virtually endless. Remaining to be seen is whether this tool could have predicted the unscrupulous behaviors of the likes of Tiger Woods, Lance Armstrong, and Bernard Madoff, and the progressive thinking of the Dalai Lama, Aung San Suu Kyi, and Bono.
Notes * Author’s Note: Jeremy A. Frimer, Harrison Oakes, Department of Psychology, University of Winnipeg, Winnipeg, MB, Canada. Corresponding Author: Jeremy A. Frimer, Department of Psychology, 515 Portage Avenue, Winnipeg, MB, Canada R3B 2E9. Email: [email protected]. 1 Regardless of whether voters interpreted his commitments literally or figuratively (as general indicators of his intentions), Romney’s pledges illustrate the wiggle room that often exists between specific proclamations and their corresponding behavior.
Spoken Words Reveal Selfish Motives
53
2 For goals, we compared means for community (prosocial) against the average of financial success, appearance, and social recognition (selfish). For values, we contrasted an aggregate of benevolence and universalism (prosocial) against an aggregate of achievement, power, and face (selfish). For traits, we contrasted nurturance for both genders (prosocial) against the orthogonal assertiveness for both genders (selfish).
References Allport, G. W. (1937). Personality: A Psychological Interpretation. Oxford England: Holt. Bakan, D. (1966). The Duality of Human Existence: An Essay on Psychology and Religion. Chicago: Rand McNally. Batson, C., Kobrynowicz, D., Dinnerstein, J. L., Kampf, H. C., and Wilson, A. D. (1997). In a very different voice: Unmasking moral hypocrisy. Journal of Personality and Social Psychology, 72, 1335–48. doi:10.1037/0022-3514.72.6.1335 Cohen, T. (21 March 2012). Romney’s big day marred by Etch A Sketch remark. CNN. Retrieved from http://CNN.com Engel, C. (2011). Dictator games: A meta study. Preprints of the Max Planck Institute for Research on Collective Goods. Exner, J. E. (1993). The Rorschach: A Comprehensive System, Volume 1: Basic Foundations (3rd ed.). New York, NY: Wiley. Frank, L. K. (1939). Projective methods for the study of personality. Transactions of The New York Academy of Sciences, 1, 1129–32. Frimer, J. A., Walker, L. J., Dunlop, W. L., Lee, B. H., and Riches, A. (2011). The integration of agency and communion in moral personality: Evidence of enlightened self-interest. Journal of Personality and Social Psychology, 101, 149–63. doi:10.1037/a0023780 Frimer, J. A., Walker, L. J., Lee, B. H., Riches, A., and Dunlop, W. L. (2012). Hierarchical integration of agency and communion: A study of influential moral figures. Journal of Personality, 80, 1117–45. doi:10.1111/j.1467-6494.2012.00764.x Frimer, J. A., and Oakes, H. (2013). Peering into the Heart of Darkness: A Projective Measure Reveals Widespread Selfishness and Prosocial Exceptions. Manuscript under review. Frimer, J. A., and Walker, L. J. (2009). Reconciling the self and morality: An empirical model of moral centrality development. Developmental Psychology, 45, 1669–81. doi:10.1037/a0017418 Grouzet, F. E., Kasser, T., Ahuvia, A., Dols, J., Kim, Y., Lau, S., and Sheldon, K. M. (2005). The structure of goal contents across 15 cultures. Journal of Personality and Social Psychology, 89, 800–16. doi:10.1037/0022-3514.89.5.800
54
Advances in Experimental Moral Psychology
Haidt, J. (2007). The new synthesis in moral psychology. Science, 316(5827), 998–1002. doi:10.1126/science.1137651 James, W. (1890). The Principles of Psychology. New York, NY: Holt. McAdams, D. P. (1995). The Life Story Interview (Rev.). Unpublished manuscript, Northwestern University, Illinois, USA. McAdams, D. P., Hoffman, B. J., Mansfield, E. D., and Day, R. (1996). Themes of agency and communion in significant autobiographical scenes. Journal of Personality, 64, 339–77. doi:10.1111/j.1467-6494.1996.tb00514.x McAdams, D. P. (2001). The psychology of life stories. Review of General Psychology, 5, 100–22. doi:10.1037/1089-2680.5.2.100 McClelland, D. C. (1975). Power: The Inner Experience. Oxford England: Irvington. Nisbett, R. E., and Wilson, T. D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84, 231–59. doi:10.1037/0033295X.84.3.231 Pennebaker, J. W., Booth, R. J., and Francis, M. E. (2007). Linguistic Inquiry and Word Count: LIWC [Computer software]. Austin, TX: LIWC.net. Schmuck, P., Kasser, T., and Ryan, R. M. (2000). Intrinsic and extrinsic goals: Their structure and relationship to well-being in German and U. S. college students. Social Indicators Research, 50, 225–41. doi:10.1023/A:1007084005278 Schwartz, S. H., Cieciuch, J., Vecchione, M., Davidov, E., Fischer, R., Beierlein, C., and Konty, M. (2012). Refining the theory of basic individual values. Journal of Personality and Social Psychology, 103, 663–88. doi:10.1037/a0029393 Time 100: Heroes and icons. (14 June 1999). Time, 153(23). Time 100: Leaders and revolutionaries. (13 April 1998). Time, 151(14). Trapnell, P. D., and Broughton, R. H. (2006). The Interpersonal Questionnaire (IPQ): Duodecant markers of Wiggins’ interpersonal circumplex. Unpublished data, The University of Winnipeg, Winnipeg, Canada. Walker, L. J., and Frimer, J. A. (2007). Moral personality of brave and caring exemplars. Journal of Personality and Social Psychology, 93, 845–60. doi:10.1037/0022-3514.93.5.845 Weber, J. M., and Murninghan, J. K. (2008). Suckers or saviors? Consistent contributors in social dilemmas. Journal of Personality and Social Psychology, 95, 1340–53. doi:10.1037/a0013326 Wiggins, J. S. (1995). Interpersonal Adjective Scales: Professional Manual. Odessa, FL: Psychological Assessment Resources.
3
Is the Glass of Kindness Half Full or Half Empty? Positive and Negative Reactions to Others’ Expressions of Virtue Gabriela Pavarini and Simone Schnall*
Mahatma Gandhi is one of the world’s most famous and influential symbols of peace. His philosophy of nonviolence has moved, transformed, and inspired individuals and communities. Yet, he was accused of racism (e.g., Singh 2004), and was never awarded a Nobel Peace Prize, despite having been nominated five times. Mother Teresa, an equally remarkable symbol of compassion and altruism, dedicated her life to helping the poor and the dying in over a hundred countries. Her funeral procession in Calcutta brought together thousands of people who lined the route in expression of admiration and respect. Yet, the entry “Mother Teresa was a fraud” returns 65,300 results on Google. Indisputably, people are strongly affected by witnessing the good deeds or heroic actions of exceptional individuals, but at the same time, such actions invoke sentiments that vary from appreciation and warmth to cynicism and bitterness. The central goal of this chapter is to address this paradox: Under what conditions does the kindness of others inspire and move individuals to tears, or invoke envy and a desire to derogate the other person’s intentions? We review what is known about both reactions and present a functional analysis, suggesting that assimilative and contrastive reactions to virtuous others serve distinct purposes: whereas feeling moved or uplifted binds individuals together in cooperative contexts and communities, contrastive responses serve to regulate one’s own social status within a group.
56
Advances in Experimental Moral Psychology
The two sides of human prosociality Human beings have a remarkable capacity to set aside self-interest, help one another, and collaborate (Becker and Eagly 2004). Prosocial behavior enables groups to achieve feats that could never be achieved by individuals alone. Some have proposed that the formation of large cooperative communities that include genetic strangers is possible only through a series of affective mechanisms (McAndrew 2002; Burkart et al. 2009). Even before the age of 2, toddlers derive more happiness from giving treats to others than from receiving treats themselves, and find it more emotionally rewarding when giving is costly—that is, when they give away their own treats rather than a treat that was found or given to them (Aknin et al. 2012). Similarly, adults derive greater happiness from spending money on others than spending on themselves (Aknin et al. 2013; Dunn et al. 2008). Finally, the most prosocial individuals are the least motivated by the pursuit of status among peers (Willer et al. 2012). Although engagement in prosocial behavior may not be necessarily motivated by the pursuit of status, prosocial others are nonetheless often preferred and, as a consequence, ascribed greater status. Infants as young as 6 months show a preference for characters who help others over mean or neutral characters (Hamlin et al. 2007). Later on, adults tend to affiliate with kind rather than attractive others when they find themselves in stressful situations (Li et al. 2008), and usually prefer morally virtuous others as potential mates (Miller 2007). Evidence also suggests that individuals who build a reputation as generous by giving more in a public goods game are more likely to be chosen as a partner in subsequent games, as well as to receive “social rewards” (i.e., “honor”) than those who do not (Dewitte and Cremer 2004; Sylwester and Roberts 2010). In other words, displays of altruistic behavior signal one’s moral quality and desirability as a potential partner and thus induce the tendency in others to praise and affiliate (Miller 2007; Roberts 1998). Because of the individual benefits of being generous, people may also behave altruistically to improve their own reputation. This strategic route to prosociality has been widely documented. When reputational concerns are at stake—for example, when all participants have access to individual contributions in an economic game—people behave more prosocially (Barclay
Is the Glass of Kindness Half Full or Half Empty?
57
and Willer 2007; Hardy and Van Vugt 2006). Similarly, after having been primed with status motives individuals are more likely to purchase products that benefit the environment (Griskevicius et al. 2010). Even minimal cues of being observed and therefore evaluated by others, such as images of eyes in one’s surroundings, lead to more prosocial choices in economic games and greater charitable giving (Bateson et al. 2006; Bereczkei et al. 2007; Haley and Fessler 2005). Thus, by displaying prosocial behavior when one is likely to be seen, one enhances the chances of receiving status benefits (e.g., Hardy and Van Vugt 2006). From an individual perspective, however, all members of a group wish to maintain their reputational status and optimize their chances of being chosen as future cooperation partners. In this context, others’ exemplary moral behavior represents an increase in the standard for prosociality and challenges those who observe it to display equally costly prosocial behavior to upregulate and protect their moral status (Fessler and Haley 2003). In other words, witnessing somebody else acting prosocially might be threatening because it “raises the bar” for others. Therefore, observers defuse the threat imposed by the “morally superior” other by engaging in prosocial acts that would strategically improve their own status. Alternatively, they may try to reduce the standard for prosocial behavior by derogating the virtuous other or excluding him or her from the group (Monin 2007). In any of these cases, witnessing others’ generosity can possibly lead to negative emotions such as fear and envy, followed by efforts to regulate one’s own status in a group.
The half empty glass of kindness: When virtuous others are suspected and derogated The extensive literature on social comparisons addresses a range of reactions to successful others. In general, if the individual feels outperformed by a similar other and perceives the other’s success as unachievable, the social comparison leads to contrastive reactions. Such reactions include self-deflection and the activation of self-defensive strategies—for example, feelings of aversion toward outperformers and derogation or punishment of them (Fiske 2011; Salovey and Rodin 1984). Even though this literature has mainly focused on people’s
58
Advances in Experimental Moral Psychology
reactions to highly skillful or competent individuals, analogous reactions have been reported in reaction to exceedingly virtuous others (e.g., Monin 2007). To start with, there seems to be a general asymmetry in how individuals react to seemingly selfless versus selfish acts. A change from being moral to immoral is more notable to others than a change from being immoral to moral (Reeder and Coovert 1986). Similarly, after reflecting about a prosocial deed participants voluntarily generate selfish reasons for the act, whereas they rarely seek altruistic reasons for a selfish act (Critcher and Dunning 2011). Further, individuals tend to underestimate the likelihood that others would respond to a request for help (Flynn and Lake 2008) and to underrate the trustworthiness of their counterparts in economic games (Fetchenhauer and Dunning 2010). This asymmetry makes functional sense, since the costs associated with trusting an immoral character are greater than those associated with falsely suspecting a good person. This general cynicism surrounding other people’s moral actions does not necessarily imply any self-threat. However, individuals do seem to derogate virtuous others as a result of an unfavorable comparison in the moral domain possibly as a response to a sharp rise in the standard for prosocial behavior that threatens an observer’s reputation among group members. If that is the case, these reactions would be more likely to occur in the following contexts: (a) when another person’s behavior is appraised from an individual, competitive perspective, and therefore contrasted with the observer’s own morals; (b) when the virtuous other and the observer compete for the same cooperation partners; (c) when reputational benefits are at stake; and (d) when others’ behavior is unattainable by the observer. Initial research on “do-gooder” derogation supports these predictions. Monin, Sawyer, and Marques (2008) observed that participants who were already engaged in a negatively perceived experimental activity, such as a counter-attitudinal speech or a racist task, disliked individuals who refused to participate in the same tasks. Non-involved observers either preferred moral rebels to obedient others (Study 1) or liked and respected them equally (Studies 2 and 3), indicating that only participants who had performed the negative activity themselves were motivated to derogate others. Moreover, whereas non-involved observers judged the rebel as more moral than the obedient other, threatened individuals evaluated the obedient person as just as
Is the Glass of Kindness Half Full or Half Empty?
59
moral as the moral rebel. These results show that whereas a non-threatening prosocial other triggers tendencies to affiliate and praise, exposure to a similar other that does a good deed for which you have missed your opportunity leads to derogation and cynicism. This pattern applies to meat-eaters’ reactions to vegetarians. Minson and Monin (2012) asked participants to indicate whether they themselves were vegetarians or not, to report whether they thought vegetarians felt morally superior, and to freely list three words that came to mind when thinking of vegetarians. Meat-eaters expected vegetarians to feel morally superior to themselves and to non-vegetarians in general, and nearly half of them listed at least one negative quality, generally referring to negative traits (e.g., arrogant, weird, self-righteous, opinionated). Indeed, the more they expected vegetarians to feel morally superior, the more negative words participants listed. In a second study, participants who were first asked to rate how they would be seen by vegetarians rated vegetarians more negatively in comparison to those who were not primed with threat of being morally judged. These studies are a compelling demonstration of how engaging in a stark comparison between oneself and a morally superior other triggers defensive reactions, which may serve to regulate one’s own sense of morality. Another defensive reaction consists of expressing a desire to expel excessively generous members from a cooperative group. Fair participants, who receive proportional rewards for their contributions, are significantly more popular than both extremely benevolent and selfish participants (Parks and Stone 2010). Ironically, unusually benevolent members are ironically rated just as unfavorably as selfish members in terms of the extent to which they should be allowed to remain in the group. When asked to explain the desire to expel unselfish individuals from the group, 58 percent of participants used comparative reasons (e.g., “people would ask why we can’t be like him”). These findings suggest that people attempt to reduce the standard for prosocial behavior when reputational demands are present and members compete for cooperation partners. Further, large-scale cross-cultural evidence for the existence of punishment of prosocial others was provided by Herrmann et al. (2008). Participants from 16 countries participated in an economic game in groups of four. They were given tokens to contribute to a group project, and contributions were distributed
60
Advances in Experimental Moral Psychology
equally among partners. After each round, players could punish other players by taking tokens away from them. Beyond the well-known punishment of selfish participants, the authors observed that participants across the world also punished those who were more prosocial than themselves. Unlike altruistic punishment, levels of antisocial punishment were highly variable across communities, and greatly influenced by economic and cultural backgrounds: more equalitarian societies, with high levels of trust, high GDP per capita, strong norms of civic cooperation, and a well-functioning democracy were the least likely to punish virtuous others. These results as a whole indicate that in conditions of high competition and demand for reputation maintenance, individuals tend to react defensively to virtuous others. These reactions assume a number of configurations that include attributing negative traits to virtuous others, punishing or excluding them from the group, and denying the ethical value of their acts. The exclusion and derogation of the extremely generous members of one’s group might be effective in regulating one’s moral reputation by decreasing the competitive and comparative standard for prosocial behavior to a less costly level. Such contrastive reactions may also help regulate the stability of one’s group morality and establish achievable norms of prosocial behavior. When one derogates another person’s moral status, it reduces the standard for prosocial behavior for all members of the group. We have so far discussed negative and defensive reactions that arise from unfavorable social comparisons in the moral domain. There are, however, circumstances under which negative reactions take place for reasons other than comparative ones. One example is when exceedingly benevolent behavior by an ingroup member is interpreted as deviant from the group norm. Previous research has shown that ingroup members are highly sensitive to behavior that differs from the norms set for members of one’s ingroup, and so derogate such behavior in an attempt to maintain group cohesiveness (Marques et al. 1988; Abrams et al. 2000). In fact, in Parks and Stone’s (2010) study, 35 percent of the participants did use normative reasons to justify their desire to expel the selfless member of the group (e.g., “He’s too different from the rest of us”). The authors suggest that this shows a desire for equality of participation even from participants who are willing to give more, as well as a resistance against changing the group norm in an undesirable direction.
Is the Glass of Kindness Half Full or Half Empty?
61
A second example is when either the prosocial act benefits an outgroup but provides an ingroup disadvantage, or the other’s behavior is not considered virtuous from the perspective of the observer. For example, a teacher who lectures chastity-based sex education at school may be considered virtuous by some people, but a violation of teenager’s freedom of conscience by others. People’s moral values vary (Graham et al. 2009; Schwartz 2006) and so their emotional reactions to actions that either support or undermine different values should vary as well. In these cases, derogating the “virtuous” other is not a reaction to a threat to one’s reputation but rather to a threat to the one’s personal or political interests.
A half full glass of kindness: When virtuous others become partners, saints, and heroes On the flip side, others’ generosity can trigger positive reactions that include feelings of respect and admiration. Haidt (2000, 2003) employed the term elevation to refer to this “warm, uplifting feeling that people experience when they see unexpected acts of human goodness, kindness, and compassion” (Haidt 2000, p. 1). Elevation is generally associated with feelings of warmth in the chest and feeling a “lump in the throat.” The distinctive appraisal, physical sensations, and motivations related to elevation differentiate it from happiness and other positive moral emotions, such as gratitude or admiration for skill (Algoe and Haidt 2009). To date, the most remarkable evidence in this field has been a positive relationship between elevation and prosocial behavior. Participants exposed to stories showing expressions of forgiveness or gratitude are more likely to donate money for charity (Freeman et al. 2009), or volunteer for an unpaid study, and spend time helping an experimenter by completing a tiresome task compared to neutral or mirth-inducing conditions (Schnall et al. 2010). Importantly, the more participants report feelings relating to elevation, such as warmth in the chest and optimism about humanity, the more time they engage in helping behavior (Schnall et al. 2010). Similar effects have been observed in real-life settings outside of the laboratory. Employees who evaluate their boss as highly fair and likely to self-sacrifice report greater feelings of
62
Advances in Experimental Moral Psychology
elevation, and are in turn more likely to show organizational citizenship and affective commitment (Vianello et al. 2010). Further, self-reported frequency of feelings of elevation during a volunteering service trip predicts trip-specific volunteerism 3 months later. This effect holds above and beyond the effect of personality traits, such as empathy, extroversion, and openness to experience (Cox 2010). Although the emotional correlates of these prosocial initiatives substantially differ from those that arise from strategic prosocial behavior, this does not rule out the possibility that they represent a reaction to a threatening comparison. In other words, after being confronted with somebody “more moral” than they are, participants may have felt motivated to act prosocially in order to restore their self-worth and moral reputation. Recent evidence, however, has convinced us otherwise. We have found, for example, that individuals who feel elevation after exposure to morally upstanding others rarely report feelings of envy or engage in contrastive comparisons between their moral qualities and the ones of the protagonist. Rather, they often justify the magnificence of the other person’s act by referring to general standards (e.g., “what she did was a very sort of selfless act. I don’t think many people would have chosen to do that”) suggesting little self-threat (Pavarini et al. 2013). In another study, we explored the effects of self-affirmation before the exposure to an elevating video clip and the opportunity to engage in helping (Schnall and Roper 2012). Previous research has shown that self-affirmation can reduce defensive responses to self-threatening information (McQueen and Klein 2006; Sherman and Cohen 2006). Therefore, we predicted that being reminded of one’s qualities would make participants more receptive to being inspired by a virtuous other, increasing prosocial responding. Indeed, our results suggested that participants who affirmed their personal qualities before watching an uplifting clip engaged in more helping behavior than participants who self-affirmed before watching a neutral clip. Further, those who specifically affirmed moral self-qualities showed the highest level of helping, more than participants in the elevation condition who affirmed a more selfish value or no value at all. Affirming one’s prosocial qualities possibly reminded participants of their core values, as well as their ability to do good. The exposure to a prosocial other under these conditions had an empowering effect on prosocial responding.
Is the Glass of Kindness Half Full or Half Empty?
63
These results suggest that prosocial motives that accompany elevation are not a reaction to self-threat; rather, they seem to be grounded in moral, self-transcendent values. Indeed, elevating stories lead participants to spontaneously declare their values and beliefs (Pavarini et al. 2013) and, from a neurological perspective, uplifted participants show activation of brain areas (e.g., posterior/inferior sectors of the posteromedial cortex) implicated in highlevel consciousness and moral reasoning (Englander et al. 2012; ImmordinoYang et al. 2009; Immordino-Yang and Sylvan 2010). Furthermore, individuals for whom moral values are highly self-defining report stronger feelings of elevation after reading an uplifting story than people for whom moral values are less central. These feelings, in turn, predict greater engagement in prosocial behavior (Aquino et al. 2011). The strong self-transcendent nature of elevation helps illuminate its functions. We propose that positive emotional reactions to others’ kindness facilitate two processes that are inherent to human sociality: humans’ tendency to affiliate with prosocial others (Li et al. 2008; Hamlyn et al. 2007), and people’s tendency to provide social rewards to generous third parties (Willer 2009). Thus, feeling moved and uplifted in the face of virtuous others play two distinct, yet interconnected roles. First, it supports the formation of strongly bonded dyads and cohesive cooperative communities. Second, it enables ascriptions of status in cooperative communities. If elevation is about identifying truly prosocial partners, one would expect it to be modulated by the extent to which others’ actions signal genuine moral qualities. The witnessed prosocial deed would then lead to a positive shift in one’s perceptions of the virtuous other. There is initial support for these predictions. Participants who are asked to remember a highly costly witnessed prosocial deed report higher feelings of elevation and stronger prosocial motives than those who recall a good deed requiring minimal effort (Thomson and Siegel 2013). In turn, witnessing such actions changes observers’ attitudes toward the agent. After recalling a situation when somebody acted virtuously, 41 percent of the participants report having identified positive qualities of the other person and 48 percent express a desire for affiliation with him or her (Algoe and Haidt 2009). Spontaneous verbal reports during the experience of elevation typify these motivations, such as “he is just extremely selfless!,” “it’s almost like you wanna go give them a pat on the back or like hug them.
64
Advances in Experimental Moral Psychology
Be like, you’re such an awesome person sort-of-hug,” and “although I have never seen this guy as more than just a friend, I felt a hint of romantic feeling for him at this moment” (Haidt 2000; Immordino-Yang, personal communication). Beyond a desire for affiliation, moral elevation is accompanied by feelings of respect and a tendency to praise the virtuous other. Forty-eight percent of participants who recall a virtuous act report having gained respect for the virtuous other (Algoe and Haidt 2009), and spontaneous verbal reports also suggest a tendency to enhance the person’s status (e.g., “I felt like telling everyone about his good deed”; Haidt 2000). Outside the emotion literature, people’s tendency to ascribe rewards and status to virtuous others has been widely documented (e.g., Hardy and Van Vugt 2006; Milinski et al. 2000; Willer 2009). Willer (2009), for example, observed that both players and observers of an economic game rate generous participants as more prestigious, honorable, and respected than low-contributing participants. Further, when given the opportunity to freely allocate $3 between themselves and a game partner, benevolent others are given a greater share than those who had contributed less in the previous game. Moral elevation may be the emotional underpinning of a tendency to ascribe social and material rewards to highly prosocial members of a group. It is important, however, to differentiate the prestige attributed to prosocial members from related constructs such as power, authority, and dominance. A prestigious other is defined as somebody who is respected and listened to, normally due to socially desirable skills, whereas dominance implies use of intimidation and coercion (Henrich and Gil-White 2001). Thus, positive emotional reactions to acts of virtue possibly support attributions of prestige, but do not necessarily lead to attributions of dominance. Strong and recurring feelings of elevation, respect, and honor toward virtuous others may transform them into role models or heroes. There is evidence that virtuous others are indeed viewed as role models for children and youths (Bucher 1998) and their stories of bravery, courage, and compassion used as a tool to foster moral development (e.g., Conle 2007; Puka 1990). After exposure to prosocial members of their own community, students generally report having identified positive qualities of the “local hero” and mention feelings of inspiration (e.g., “he can deal with everything so well”; “I think everyone should be told this. It touched me so much”; Conle and Boone 2008, pp. 32–3).
Is the Glass of Kindness Half Full or Half Empty?
65
In a similar manner, adults frequently perceive certain cultural heroes as images of their ideal selves (Sullivan and Venter 2005) and occasionally experience strong feelings of awe toward them (Keltner and Haidt 2003). In this context, moral elevation seems to serve as an affective cue that positively modifies one’s views of the observed other. As Haidt and Algoe (2004) argue, individuals tend to rank social targets along a vertical dimension of morality with goodness on top and badness on the bottom. This vertical hierarchy helps individuals understand and organize their moral world. Saints (who are normally regarded as virtuous) are perceived as more sacred than humans (Brandt and Reyna 2010). In this context, feelings of moral elevation may facilitate locating a virtuous target at the top of this vertical dimension (Haidt and Algoe 2004; also see Brandt and Reyna 2011). The meaning of being ascribed a top position in this “chain of being” is still unclear, but it possibly implies greater chances of being chosen as a cooperative partner, being treated well, and given an influential position in a group. From a broad perspective, experiences of moral elevation may signal safety in the social domain. Most individuals wish to live in a community that values and respects cooperative norms (e.g., kindness, reciprocity, honesty; Brown 1991). As suggested by Haidt (2006), feelings of elevation may signal that the current environment involves generosity. Individuals might therefore appraise it as a safe and reliable place for prosocial exchanges and bonding. In fact, moral elevation has been shown to have a relaxing effect; individuals feeling uplifted report muscle relaxation (Algoe and Haidt 2009) and nursing mothers are more likely to release milk after watching an inspiring clip than an amusing video (Silvers and Haidt 2008). This association with milk letdown suggests a possible involvement of oxytocin, a hormone implicated in relaxing states (Uvnas-Moberg et al. 2005), prosocial orientation (Israel et al. 2009; Zak et al. 2007), and social trust (Zak et al. 2005). In short, feelings of moral elevation may support a shift from a competitive mindset to thinking and acting in a safe cooperative environment as a dyad or community. Elevation is a state of self-transcendence and social focus. Yet, as we have attempted to demonstrate, different flavors of elevation lead to different motivational tendencies. When blended with love and compassion, feeling uplifted may emerge as an affective element of caregiving, mating, and general affiliation systems. When blended with respect and awe, elevation may support
66
Advances in Experimental Moral Psychology
attributions of prestige and status to generous members of a group. These different reactions may depend on by whom and in which circumstances the good deed is performed—for example, whether the agent represents a potential cooperative partner or a more distant other. Such differential effects deserve further investigation. Beneath all particularities, however, underlies the remarkable observation that simply watching other people’s good deeds can be emotionally rewarding, and that this affective reaction crucially sustains human prosociality.
Concluding remarks Individuals are strongly affected by witnessing the kindness of strangers; however, these reactions can be either positive or negative. We here attempted to shed light onto some of the conditions under which one denigrates or elevates a virtuous other. In general, positive reactions take place in conditions of little self-threat, when individuals adopt a more collective mindset and focus on their values and identity. In competitive situations, though, where the virtuous other and the observer compete for the same resources and future interaction partners, defensive reactions become more likely. Research on morality has clarified how a person’s moral actions are affected by either self-regulatory or identity concerns. On the one hand, there is evidence that people self-regulate their morality over time. After doing a bad deed, individuals tend to do good deeds in order to regain their moral worth, and when they are reminded of positive self-qualities, they feel “licensed” to act immorally (Merritt et al. 2010; Sachdeva et al. 2009). However, when individuals are reminded of positive self-qualities in an abstract manner (e.g., a good deed in the past), identity concerns are activated and they tend to act in a consistent instead of compensatory fashion—that is, engaging in further prosocial behavior (Conway and Peetz 2012). The present chapter expands this literature by exploring how analogous processes take place in a dyadic or group context. Exposure to a virtuous other may lead people to engage in contrastive comparisons and regulatory behavior (e.g., derogating the person’s intentions), or it may lead to the activation of moral values that are central to one’s identity, inspiring individuals to put these values
Is the Glass of Kindness Half Full or Half Empty?
67
into action. Interestingly, although these are individual processes, such reactions may influence how the virtuous other reacts and may ultimately regulate a group morality. For example, rewarding others’ generosity may encourage them to engage in further prosocial acts, and help to establish stronger prosocial bonds, whereas derogating their intentions may prevent an increase in the general standard for prosocial behavior for all members of the group. The capacity to evaluate and react emotionally to other’s moral behavior is essential for navigating the social world. Such evaluations help observers to identify who may help them and who might be a foe. Yet, prosocial others are not always judged positively. Our review suggests that reactions to expressions of uncommon goodness vary strikingly. The “glass of kindness” can be perceived as half empty or half full, depending on whether the act is appraised from a competitive or cooperative mindset, and on whether the virtuous other is seen as a suitable social partner, a potential leader, or a possible rival.
Note * Author’s Note: Gabriela Pavarini and Simone Schnall, Department of Psychology, University of Cambridge. The preparation of this chapter was supported by ESRC Grant RES-000-22-4453 to S. S. Correspondence: Simone Schnall, University of Cambridge, Department of Psychology, Downing Street, Cambridge, CB2 3EB, Email: [email protected].
References Abrams, D., Marques, J. M., Bown, N., and Henson, M. (2000). Pro-norm and antinorm deviance within and between groups. Journal of Personality and Social Psychology, 78, 906–12. doi:10.1037//0022-3514.78.5.906 Aknin, L. B., Hamlin, J. K., and Dunn, E. W. (2012). Giving leads to happiness in young children. PLoS ONE, 7, e39211. doi:10.1371/journal.pone.0039211 Aknin, L. B., Barrington-Leigh, C. P., Dunn, E. W., Helliwell, J. F., Burns, J., Biswas-Diener, R., Kemeza, I., Nyende, P., Ashton-James, C. E., and Norton, M. I. (2013). Prosocial spending and well-being: Cross-cultural evidence for a psychological universal. Journal of Personality and Social Psychology, 104, 635-52. doi: 10.1037/a0031578
68
Advances in Experimental Moral Psychology
Algoe, S., and Haidt, J. (2009). Witnessing excellence in action: The other-praising emotions of elevation, admiration, and gratitude. Journal of Positive Psychology, 4, 105–27. doi:10.1080/17439760802650519 Aquino, K., McFerran, B., and Laven, M. (2011). Moral identity and the experience of moral elevation in response to acts of uncommon goodness. Journal of Personality and Social Psychology, 100, 703–18. doi:10.1037/a0022540 Barclay, P., and Willer, R. (2007). Partner choice creates competitive altruism in humans. Proceedings of the Royal Society of London, Series B, 274, 749–53. doi:10.1098/rspb.2006.0209 Bateson, M., Nettle, D., and Roberts, G. (2006). Cues of being watched enhance cooperation in a real-world setting. Biology Letters, 2, 412–14. doi:10.1098/ rsbl.2006.0509 Becker, S. W., and Eagly, A. H. (2004). The heroism of women and men. American Psychologist, 59, 163–78. doi:10.1037/0003-066X.59.3.163 Bereczkei, T., Birkas, B., and Kerekes, Zs. (2007). Public charity offer as a proximate factor of evolved reputation-building strategy: An experimental analysis of a real life situation. Evolution and Human Behavior, 28, 277–84. doi:10.1016/j. evolhumbehav.2007.04.002 Brandt, M. J., and Reyna, C. (June 2010). Beyond infra-humanization: The perception of human groups, the self, and supernatural entities as more or less than human. Poster presented at the annual meeting of the Society for Personality and Social Psychology, Las Vegas, NV. —(2011). The chain of being: A hierarchy of morality. Perspectives on Psychological Science, 6, 428–46. doi:10.1177/1745691611414587 Brown, D. E. (1991). Human universals. New York: McGraw-Hill. Bucher, A. A. (1998). The influence of models in forming moral identity. International Journal of Educational Research, 27, 619–27. doi:10.1016/S0883-0355(97)00058-X Burkart, J. M., Hrdy, S. B., and van Schaik, C. P. (2009). Cooperative breeding and human cognitive evolution. Evolutionary Anthropology, 18, 175–86. doi:10.1002/ evan.20222 Conle, C. (2007). Moral qualities of experiential narratives. Journal of Curriculum Studies, 39, 11–34. doi:10.1111/j.1467-873X.2007.00396.x Conle, C., and Boone, A. (2008). Local heroes, narrative worlds and the imagination: The making of a moral curriculum through experiential narratives. Curriculum Inquiry, 38, 7–37. doi:10.1111/j.1467-873X.2007.00396.x Conway, P., and Peetz, J. (2012). When does feeling moral actually make you a better person? Conceptual abstraction moderates whether past moral deeds motivate consistency or compensatory behavior. Personality and Social Psychology Bulletin, 6, 907–19. doi: 10.1177/0146167212442394
Is the Glass of Kindness Half Full or Half Empty?
69
Cox, K. S. (2010). Elevation predicts domain-specific volunteerism 3 months later. Journal of Positive Psychology, 5, 333–41. doi:10.1080/17439760.2010.507468 Critcher, C. R., and Dunning, D. (2011). No good deed goes unquestioned: Cynical reconstruals maintain belief in the power of self-interest. Journal of Experimental Social Psychology, 47, 1207–13. doi:10.1016/j.jesp.2011.05.001 Dewitte, S., and Cremer, D. (2004). Give and you shall receive. Give more and you shall be honored. Experimental evidence for altruism as costly signaling. Working Paper, Department of Applied Economics 0458, University of Leuven, Belgium. Dunn, E. W., Aknin, L. B., and Norton, M. I. (2008). Spending money on others promotes happiness. Science, 319, 1687–8. doi:10.1126/science.1150952 Englander, Z. A., Haidt, J., and Morris, J. P. (2012). Neural basis of moral elevation demonstrated through inter-subject synchronization of cortical activity during free-viewing. PLoS ONE, 7, e39384. doi:10.1371/journal.pone.0039384 Fessler, D. M. T., and Haley, K. J. (2003). The strategy of affect: Emotions in human cooperation. In P. Hammerstein (ed.), The Genetic and Cultural Evolution of Cooperation. Cambridge, MA: MIT Press, pp. 7–36. Fetchenhauer, D., and Dunning, D. (2010). Why so cynical? Asymmetric feedback underlies misguided skepticism in the trustworthiness of others. Psychological Science, 21, 189–93. doi:10.1177/0956797609358586 Fiske, S. (2011). Envy up, scorn down: How status divides us. New York: Russel Sage Foundation. Flynn, F. J., and Bohns Lake, V. K. (2008). If you need help, just ask: Underestimating compliance with direct requests for help. Journal of Personality and Social Psychology, 95, 128–43. doi:10.1037/0022-3514.95.1.128 Freeman, D., Aquino, K., and McFerran, B. (2009). Overcoming beneficiary race as an impediment to charitable donation: Social Dominance Orientation, the experience of moral elevation, and donation behavior. Personality and Social Psychology Bulletin, 35, 72–84. doi:10.1177/0146167208325415 Graham, J., Haidt, J., and Nosek, B. A. (2009). Liberals and conservatives rely on different sets of moral foundations. Journal of Personality and Social Psychology, 96, 1029–46. doi:10.1037/a0015141 Griskevicius, V., Tybur, J. M., and Van den Bergh, B. (2010). Going green to be seen: Status, reputation, and conspicuous conservation. Journal of Personality and Social Psychology, 98, 392–404. doi:10.1037/a0017346 Haidt, J. (2000). The positive emotion of elevation. Prevention & Treatment, 3, 1–4. doi:10.1037//1522-3736.3.1.33c —(2003). Elevation and the positive psychology of morality. In C. L. M. Keyes and J. Haidt (eds), Flourishing: Positive Psychology and the Life Well-lived. Washington, DC: American Psychological Association, pp. 275–89.
70
Advances in Experimental Moral Psychology
—(2006). The Happiness Hypothesis: Finding Modern Truth in Ancient Wisdom. New York: Basic Books. Haidt, J., and Algoe, S. (2004). Moral amplification and the emotions that attach us to saints and demons. In J. Greenberg, S. L. Koole, and T. A. Pyszczynski (eds), Handbook of Experimental Existential Psychology. New York: Guilford, pp. 322–35. Halevy, N., Chou, E. Y., Cohen, T. R., and Livingston, R. W. (2012). Status conferral in intergroup social dilemmas: Behavioral antecedents and consequences of prestige and dominance. Journal of Personality and Social Psychology, 102, 351–66. doi:10.1037/a0025515 Haley, K. J., and Fessler, D. M. T. (2005). Nobody’s watching? Subtle cues affect generosity in an anonymous economic game. Evolution and Human Behavior, 26, 245–56. doi:10.1016/j.evolhumbehav.2005.01.002 Hamlin, J. K., Wynn, K., and Bloom, P. (2007). Social evaluation by preverbal infants. Nature, 450, 557–9. doi:10.1038/nature06288 Hardy, C., and Van Vugt, M. (2006). Nice guys finish first: The competitive altruism hypothesis. Personality and Social Psychology Bulletin, 32, 1402–13. doi:10.1177/0146167206291006 Henrich, J., and Gil-White, F. J. (2001). The evolution of prestige: Freely conferred status as a mechanism for enhancing the benefits of cultural transmission. Evolution and Human Behaviour, 22, 165–96. doi:10.1016/S1090-5138(00)00071-4 Herrmann, B., Thöni, C., and Gächter, S. (2008). Antisocial punishment across societies. Science, 319, 1362–7. doi:10.1126/science.1153808 Immordino-Yang, M. H., and Sylvan, L. (2010). Admiration for virtue: Neuroscientific perspectives on a motivating emotion. Contemporary Educational Psychology, 35, 110–115. doi:10.1016/j.cedpsych.2010.03.003 Immordino-Yang, M. H., McColl, A., Damasio, H., and Damasio, A. (2009). Neural correlates of admiration and compassion. Proceedings of the National Academy of Sciences USA, 106, 8021–6. doi:10.1073/pnas.0810363106 Israel, S., Lerer, E., Shalev, I., Uzefovsky, F., Riebold, M., Laiba, E., Bachner-Melman, R., Maril, A., Bornstein, G., Knafo, A., and Ebstein, R. P. (2009). The oxytocin receptor (OXTR) contributes to prosocial fund allocations in the dictator game and the social value orientations task. PLoS ONE, 4, e5535. doi:10.1371/journal. pone.0005535 Keltner, D., and Haidt, J. (2003). Approaching awe, a moral, spiritual, and aesthetic emotion. Cognition and Emotion, 17, 297–314. doi:10.1080/02699930302297 Li, N. P., Halterman, R. A., Cason, M. J., Knight, G. P., and Maner J. K. (2008). The stress-affiliation paradigm revisited: Do people prefer the kindness of strangers or their attractiveness? Journal of Personality and individual Differences, 44, 382–91. doi: 10.1016/j.paid.2007.08.017
Is the Glass of Kindness Half Full or Half Empty?
71
Marques, J. M., Yzerbyt, V. Y., and Leyens, J. P. (1988). The ‘black sheep effect’: Extremity of judgments towards ingroup members as a function of group identification. European Journal of Social Psychology, 18, 1–16. doi:10.1002/ejsp.2420180102 McAndrew, F. T. (2002). New evolutionary perspectives on altruism: Multilevelselection and costly-signaling theories. Current Directions in Psychological Science, 11, 79–82. doi:10.1111/1467-8721.00173 McQueen, A., and Klein, W. M. P. (2006). Experimental manipulations of self-affirmation: A systematic review. Self and Identity, 5, 289–354. doi:10.1080/15298860600805325 Merritt, A. C., Effron, D. A., and Monin, B. (2010). Moral self-licensing: When being good frees us to be bad. Social and Personality Psychology Compass, 4/5, 344–57. doi:10.1111/j.1751-9004.2010.00263.x Milinski, M., Semmann, D., and Krambeck, H. (2000). Donors to charity gain in both indirect reciprocity and political reputation. Proceedings of Royal Society, 269, 881–83. doi:10.1098/rspb.2002.1964 Miller, G. F. (2007). Sexual selection for moral virtues. Quarterly Review of Biology, 82, 97–125. doi:10.1086/517857 Minson, J. A., and Monin, B. (2012). Do-Gooder derogation: Disparaging morallymotivated minorities to defuse anticipated reproach. Social Psychological and Personality Science, 3, 200–7. doi:10.1177/1948550611415695 Monin, B. (2007). Holier than me? Threatening social comparison in the moral domain. International Review of Social Psychology, 20, 53–68. Monin, B., Sawyer, P. J., and Marquez, M. J. (2008). The rejection of moral rebels: Resenting those who do the right thing. Journal of Personality and Social Psychology, 95, 76–93. doi:10.1037/0022-3514.95.1.76 Parks, C. D., and Stone, A. B. (2010). The desire to expel unselfish members from the group. Journal of Personality and Social Psychology, 99, 303–10. doi:10.1037/a0018403 Pavarini, G., Schnall, S., and Immordino-Yang, M. H. (2013). Verbal and Nonverbal Indicators of Psychological Distance in Moral Elevation and Admiration for Skill. Manuscript in preparation. Puka, B. (1990). Be Your Own Hero: Careers in Commitment. Troy, NY: Rensselaer Polytechnic Institute. Reeder, G. D., and Coovert, M. D. (1986). Revising an impression of morality. Social Cognition, 4, 1–17. doi:10.1521/soco.1986.4.1.1 Roberts, G. (1998). Competitive altruism: From reciprocity to the handicap principle. Proceedings of the Royal Society of London, Series B: Biological Sciences, 265, 427–31. doi:10.1098/rspb.1998.0312 Sachdeva, S., Iliev, R., and Medin, D. L. (2009). Sinning saints and saintly sinners: The paradox of moral self-regulation. Psychological Science, 20, 523–8. doi:10.1111/ j.1467-9280.2009.02326.x
72
Advances in Experimental Moral Psychology
Salovey, P., and Rodin, J. (1984). Some antecedents and consequences of socialcomparison jealousy. Journal of Personality and Social Psychology, 47, 780–92. doi:10.1037//0022-3514.47.4.780 Schnall, S., and Roper, J. (2012). Elevation puts moral values into action. Social Psychological and Personality Science, 3, 373–8. doi:10.1177/1948550611423595 Schnall, S., Roper, J., and Fessler, D. (2010). Elevation leads to altruistic behavior. Psychological Science, 21, 315–20. doi:10.1177/0956797609359882 Schwartz, S. H. (2006). A theory of cultural value orientations: Explication and applications. Comparative Sociology, 5, 136–82. doi:10.1163/156913306778667357 Sherman, D. K., and Cohen, G. L. (2006). The psychology of self-defense: Selfaffirmation theory. In M. P. Zanna (ed.), Advances in Experimental Social Psychology (Vol. 38). San Diego, CA: Academic Press, pp. 183–242. Silvers, J., and Haidt, J. D. (2008). Moral elevation can induce nursing. Emotion, 8, 291–5. doi:10.1037/1528-3542.8.2.291 Singh, G. B. (2004). Gandhi: Behind the Mask of Divinity. New York: Prometheus. Sullivan, M. P., and Venter, A. (2005). The hero within: Inclusion of heroes into the self. Self and Identity, 4, 101–11. doi:10.1080/13576500444000191 Sylwester, K., and Roberts, G. (2010). Cooperators benefit through reputationbased partner choice in economic games. Biology Letters, 6, 659–62. doi:10.1098/ rsbl.2010.0209 Thomson, A. L., and Siegel, J. T. (2013). A moral act, elevation, and prosocial behavior: Moderators of morality. Journal of Positive Psychology, 8, 50–64. doi:10.1080/17439760.2012.754926 Uvanas-Moberg, A., Arn, I., and Magnusson, D. (2005). The psychobiology of emotion: The role of the oxytocinergic system. International Journal of Behavioral Medicine, 12, 59–65. doi:10.1207/s15327558ijbm1202_3 Vianello, M., Galliania, E. M., and Haidt, J. (2010). Elevation at work: The effects of leaders’ moral excellence. Journal of Positive Psychology, 5, 390–411. doi:10.1080/ 17439760.2010.516764 Willer, R. (2009). Groups reward individual sacrifice: The status solution to the collective action problem. American Sociological Review, 74, 23–43. doi:10.1177/000312240907400102 Willer, R., Feinberg, M., Flynn, F., and Simpson, B. (2012). Is generosity sincere or strategic? Altruism versus status-seeking in prosocial behavior. Revise and resubmit from Journal of Personality and Social Psychology. Zak, P. J., Kurzban, R., and Matzner, W. T. (2005). Oxytocin is associated with human trustworthiness. Hormones and Behavior, 48, 522–27. doi:10.1016/j.yhbeh.2005.07.009 Zak, P. J., Stanton, A. A., and Ahmadi, S. (2007). Oxytocin increases generosity in humans. PLoS ONE, 2, e1128. doi:10.1371/journal.pone.0001128
4
What are the Bearers of Virtues? Mark Alfano*
Despite the recent hubbub over the possibility that the concepts of character and virtue are empirically inadequate,1 researchers have only superficially considered the fact that these concepts purport to refer to dispositional properties.2 For the first time in this controversy, we need to take the dispositional nature of virtue seriously. Once we do, one question immediately arises: What are the bearers of virtues? In this chapter, I argue for an embodied, embedded, and extended answer to this question. It is generally hopeless to try to say what someone would do in a given normative state of affairs without first specifying bodily and social features of her situation. There’s typically no fact of the matter, for instance, about whether someone would help when there is sufficient reason for her to help. However, there typically is a fact of the matter about whether someone in a particular bodily state and social environment would help when there is sufficient reason to help. If that’s right, it puts some pressure on agent-based theories of virtue, which tend to claim or presume that the bearers of virtue are individual agents (Russell 2009; Slote 2001). Such theories hold that a virtue is a monadic property of an individual agent. Furthermore, this pressure on agent-based and agentfocused theories suggests a way of reconceiving of virtue as a triadic relation among an agent, a social milieu, and an asocial environment (Alfano 2013). On this relational model, the bearers of virtue are not individual agents but ordered triples that include objects and properties outside the agent. Here is the plan for this chapter. Section 1 summarizes the relevant literature on dispositions. Section 2 sketches some of the relevant psychological findings. Section 3 argues that the best response to the empirical evidence
74
Advances in Experimental Moral Psychology
is to revise the concept of a virtue. A virtue is not a monadic property of an agent, but a triadic relation among an agent, a social milieu, and an asocial environment.
Virtues as dispositional properties The subjunctive conditional analysis The most straightforward way to approach dispositions is through the simple subjunctive conditional analysis: (S-SCA) object o is disposed to activity A in condition C if and only if o would A if C were the case (Choi and Fara 2012)
The A-term refers to the characteristic manifestation of the disposition; the C-term refers to its stimulus conditions. To say that OxyContin is an analgesic for humans is to attribute a dispositional property to it: OxyContin is disposed to relieve pain when ingested by a human. This statement would then be analyzed as: OxyContin would relieve pain if it were ingested by a human. According to the standard semantics for subjunctive conditionals, this analysis means that all close possible worlds at which a pained person ingests OxyContin are worlds at which that person’s pain is subsequently relieved. Many of the dispositions we refer to on a regular basis, such as being fragile, soluble, or poisonous, do not wear their manifestation and stimulus conditions on their sleeves. Following Lewis (1997), the standard “two-step” strategy for dealing with them is first to spell out these conditions, and then to translate them into the subjunctive conditional schema. So, for example, in the case of a fragile vase, the first step would be to spell out the A-term (chipping, cracking, shattering) and the C-term (being struck, dropped on a hard surface, abraded). The second step would be to slot these into the schema: “the vase is fragile” is translated as “the vase is disposed to chip, crack, or shatter when struck, dropped, or abraded,” which is true if and only if it would chip, crack, or shatter if it were struck, dropped, or abraded. Furthermore, it’s now recognized that most dispositions have characteristic masks, mimics, and finks. A fragile vase might not break when struck because
What are the Bearers of Virtues?
75
its fragility is masked by protective packaging. A sugar pill can mimic a real analgesic via the placebo effect. An electron’s velocity, which it is naturally disposed to retain, inevitably changes when it is measured—a case of finking.3 Such possibilities are not evidence against the presence or absence of the disposition in question; instead, they are exceptions to the subjunctive conditional. The vase really is fragile, despite its resistance to chipping, cracking, and shattering. The sugar pill is not really an analgesic, despite the pain relief. The electron really is disposed to follow its inertial path, despite Heisenberg’s uncertainty principle. Let’s stipulate that finks, masks, and mimics be collectively referred to as disrupters. Since it is possible to possess a disposition that is susceptible to finks and masks, and to lack a disposition that is mimicked, the simple subjunctive conditional analysis fails. In my view, the most attractive response to the constellation of disrupters is Choi’s (2008) anti-disrupter SCA: (AD-SCA) object o is disposed to activity A in condition C if and only if o would A if C were the case and there were no disrupters present.
If the object fails to A in C when there are finks or masks present, the righthand side of the definition is false. For instance, if the fragile vase were encased in protective bubble wrap, its fragility would be masked: it is not disposed to chip, crack, or shatter when struck, dropped, or abraded. But a disrupter is present, which means that the biconditional is still true. Likewise, if the object As in C when a mask is present, the right-hand side of the definition is again false, which would not yield the undesirable conclusion that the object has the disposition in question.
Woulda, coulda, shoulda The subjunctive conditional analysis, even in the sophisticated form that allows for disrupters, expresses a very strong notion of what it takes to be a disposition. Consider, for example, a loaded die that has a 0.5 probability of showing ace and a 0.1 probability of showing each of the other results. Surely, one might argue, the die is disposed to show ace, even though there are plenty of close possible worlds at which it shows two, three, four, five, or six.
76
Advances in Experimental Moral Psychology
In light of such cases, it’s helpful to provide weak and comparative analyses of dispositions. For instance, we can analyze a weak disposition as follows: (W-AD-SCA) o is weakly disposed to A in C if and only if o could A if C were the case and there were no disrupters present.
In the standard semantics, this means that o As at some undisrupted nearby C-worlds. The same object can be both weakly disposed to A in C and weakly disposed to not-A in C. This makes weak dispositions less informative than we might like. W-AD-SCA can be supplemented with the comparative analysis: (C-AD-SCA) o is more disposed to A than to A* in C if and only if o is significantly more likely to A than to A* if C were the case and no disrupters were present.
In the standard semantics, this means that there are more nearby undisrupted C-worlds where o As than nearby undisrupted C-worlds where o A*s. Which, if any, of these notions is appropriate to an analysis of virtues? Elsewhere (Alfano 2013), I have argued for an intuitive distinction between high-fidelity and low-fidelity virtues. High-fidelity virtues, such as honesty, chastity, and loyalty, require near-perfect manifestation in undisrupted conditions. For these, AD-SCA seems most appropriate. Someone only counts as chaste if he never cheats on his partner when cheating is a temptation. Lowfidelity virtues, such as generosity, tact, and tenacity, are not so demanding. For them, some combination of the W-AD-SCA and C-AD-SCA seems appropriate. Someone might count as generous if she were more disposed to give than not to give when there was sufficient reason to do so; someone might count as tenacious if she were more disposed to persist than not to persist in the face of adversity.4 If this is on the right track, the analysis of virtuous dispositions adds one additional step before Lewis’s two. First, determine whether the virtue in question is high fidelity or low fidelity. For instance, it seems reasonable to say that helpfulness is a low-fidelity virtue whereas loyalty is a high-fidelity virtue. Second, identify the stimulus conditions and characteristic manifestations. The most overt manifestation of helpfulness is of course helping behavior, but more subtle manifestations presumably include noticing opportunities to
What are the Bearers of Virtues?
77
help, having occurrent desires to help, deliberating in characteristic ways, and forming intentions to help. The most overt manifestation of loyalty is refusal to betray, but again there are more subtle manifestations. The stimulus condition for helpfulness is a normative state of affairs: that there is adequate reason to help. For loyalty, too, the stimulus condition appears to be a normative state of affairs: that there is temptation but not sufficient reason to betray whomever or whatever one is loyal to. Finally, the stimulus conditions and characteristic manifestations are slotted into the relevant schema. To be helpful, then, is to be weakly disposed to help (among other things) when there is adequate reason to do so, whereas to be loyal is to be strongly disposed not to betray (among other things) when there is a temptation to do so.
The psychology of dispositions In this section, I argue that both high-fidelity and low-fidelity virtues, as traditionally conceived, suffer from an indeterminacy problem. There’s often no fact of the matter about whether someone would exhibit the characteristic manifestations of a high-fidelity virtue in undisrupted stimulus conditions. There’s often no fact of the matter about whether someone is more disposed than not to exhibit the characteristic manifestations of a low-fidelity virtue in undisrupted stimulus conditions. This is because both bodily and social factors partially determine whether such a disposition is manifested. Someone might be strongly disposed to tell the truth if others think of her as an honest person, but strongly disposed to deceive if others think of her as a dishonest person. For such a person, it would be incorrect to say that she would tell the truth in undisrupted stimulus conditions, but also incorrect to say that she would lie in undisrupted stimulus conditions. However, once the asocial and social contexts are specified, it generally will be possible to assert true subjunctive conditionals about her. Similarly, someone might be significantly more disposed to persist than desist in the face of adversity if she is in a good mood, but significantly more disposed to desist than persist if she is in a bad mood. For such a person, it would be incorrect to say that she is weakly disposed to persist in undisrupted stimulus conditions, but also incorrect to say that she is weakly disposed
78
Advances in Experimental Moral Psychology
to desist in undisrupted stimulus conditions. As before, once the asocial and social contexts are specified, it generally will be possible to assert true subjunctive conditionals.
Asocial situational influences Asocial situational influences are asocial features of the immediate environment that partially determine which subjunctive conditionals are true about how someone will act in a given virtue’s stimulus conditions. As I explore in more detail elsewhere (Alfano 2013), two of the main varieties are ambient sensibilia and mood modulators. For instance, the volume of ambient sound influences both helping behavior and physical aggressiveness (Matthews and Cannon 1975; Donnerstein and Wilson 1976). Subtle shifts in lighting partially determine whether people cheat: they cheat more in an almost-imperceptibly darker room and act more selfishly when wearing shaded glasses rather than clear ones (Zhong et al. 2010). These are just a few of literally hundreds of relevant experiments. Together, they suggest that there are stable, though weak, connections between seemingly morally irrelevant sensibilia and the manifestation of virtue. The connections don’t all run in the same direction, and they interact. Loud environments don’t uniformly dispose toward morally bad behavior, nor do pleasant smells invariably dispose toward morally good behavior. Whether a given sensory input will tend to produce good, bad, or neutral results depends on what kind of behavior is normatively appropriate and which other sensibilia are present, among many other things. I should also note that these connections are not crudely behavioristic, an accusation often unfairly leveled at situationist critics of virtue theory (Kamtekar 2004; Kristjansson 2008). According to behaviorism, “Behavior can be described and explained without making ultimate reference to mental events or to internal psychological processes” (Graham 2010). But ambient sensibilia influence behavior in large part by modifying the agent’s cognitive and motivational set. Loud noises, for instance, result in attentional focusing (Cohen 1978), while pleasant smells induce openness to new experiences (Baron and Thomley 1994). These internal states mediate the connection between the asocial environment and the agent’s behavior.
What are the Bearers of Virtues?
79
Another primary asocial influence is the set of affect modulators, very broadly construed to include mood elevators, mood depressors, emotion inducers, and arousal modifiers. There are documented effects for embarrassment (Apsler 1975), guilt (Regan 1971), positive affect (Isen 1987), disgust (Schnall et al. 2008), and sexual arousal (Ariely 2008), among others. As with sensibilia, affect modulators are connected in weak but significant ways to the manifestation (or not) of virtue. Fair moods don’t necessarily make us fair, nor do foul moods make us foul. The valence of the effect depends on what is normatively appropriate in the particular circumstances. It’s important to point out, furthermore, that while asocial influences tend to have fairly predictable effects on behavioral dispositions, they by no means explain action all by themselves. Indeed, any particular factor will typically account for at most 16 percent of the variance in behavior (Funder and Ozer 1983).
Social influences In addition to the asocial influences canvassed above, there are a variety of social influences on the manifestation of virtue. Two of the more important are expectation confirmation and outgroup bias. In cases of expectation confirmation, what happens is that the agent reads others’ expectations off explicit or implicit social cues, and then acts in accordance with the expectations so read. People often enough mistake or misinterpret others’ expectations, so what they end up doing isn’t necessarily what others expect, but what they think others expect. In cases of outgroup bias, the agent displays unwarranted favoritism toward the ingroup or prejudice toward the outgroup. Since everyone belongs to myriad social groups, whether someone is perceived as in or out depends on which group identities are salient at the time; (Tajfel 1970, 1981); hence, the perceived social distance of a given person will vary over time with seemingly irrelevant changes in the salience of various group identities. In this section, I have room to discuss only social expectations. Much of the groundbreaking social psychology of the second half of the twentieth century investigated the power of expectation confirmation. The most dramatic demonstration was of course the Milgram paradigm (1974),
80
Advances in Experimental Moral Psychology
in which roughly two-thirds of participants were induced to put what they thought was 450 volts through another participant (actually an actor who was in on the experiment) three times in a row. While there were many important features of this study, the key upshot was that the participants were willing to do what they should easily have recognized was deeply immoral based on the say-so of a purported authority figure. They performed exactly as expected. Blass (1999) shows in a meta-analysis that Milgram’s results were no fluke: they have been replicated all around with the world with populations of diverse age, gender, and education level. Another example of the power of social expectations is the large literature on bystander apathy (Darley and Latané 1968; Latané and Nida 1981). It turns out that the more bystanders are present in an emergency situation, the lower the chances that even one of them will intervene. What seems to happen in such cases is that people scan others’ immediate reactions to help themselves determine what to do. When they see no one else reacting, they decide not to intervene either; thus, everyone interprets everyone else’s moment of deliberation as a decision not to intervene. Reading off others’ expectations and acting accordingly doesn’t always lead to bad outcomes. Recent work on social proof shows that the normative valence of acting in accordance with expectations depends on what is expected. For instance, guests at a hotel are 40 percent more likely to conserve water by not asking for their towels to be washed if they read a message that says, “75% of the guests who stayed in this room participated in our resource savings program by using their towels more than once” than one that says, “You can show respect for nature and help save the environment by reusing towels during your stay” (Goldstein et al. 2008) Psychologists and behavioral economists have also investigated the effect of subtle, thoroughly embodied, social distance cues on moral behavior. In a string of fascinating studies, it’s been shown that people are much more willing to share financial resources (Burnham 2003; Burnham and Hare 2007; Rigdon et al. 2009), less inclined to steal (Bateson et al. 2006), and less disposed to litter (Ernest-Jones et al. 2011) when they are “watched” by a representation of a face. The face can be anything from a picture of the beneficiary of their behavior to a cartoon robot’s head to three black dots arranged to look like eyes and a nose.
What are the Bearers of Virtues?
81
Revising the metaphysics of virtue In the previous section, I argued that both social and asocial factors shape how people are disposed to think, feel, and act. What we notice, what we think, what we care about, and what we do depend in part on bodily and social features of our situations. This is not to deny that people also bring their own distinctive personalities to the table, but it suggests that both high-fidelity and low-fidelity virtues, as traditionally conceived, are rare. To see why, let’s walk through the three-step analysis of a traditional virtue: honesty. For current purposes, I’ll assume that it’s uncontroversial that honesty is high fidelity. Next, we specify the stimulus conditions and characteristic manifestations. I don’t have space to do justice to the required nuances here, but it wouldn’t be too far off to say that the stimulus conditions C are temptations to lie, cheat, or steal despite sufficient reason not to do so, and that the characteristic manifestations A are behavioral (not lying, cheating, or stealing), cognitive (noticing the temptation without feeling too much of its pull), and affective (disapprobation of inappropriate behavior, desire to extricate oneself from the tempting situation if possible, perhaps even prospective shame at the thought that one might end up acting badly). Finally, we slot these specifications into the schema for high-fidelity virtue: (AD-SCA-honesty) The agent is disposed to activity A in condition C if and only if she would A if C were the case and there were no disrupters present.
More longwindedly, at all nearby undisrupted worlds where she is tempted but has sufficient reason to resist temptation, she thinks, feels, and acts appropriately. It’s only reasonable to assume that at some of these undisrupted temptation-worlds, however, the lighting will not be bright; at others, she might not feel watched. What she would do depends in part on these factors, and so the subjunctive conditional is false, which in turn means that she is not honest. It could be objected that these factors are disrupters, and so should be ruled out by fiat. This objection is unmotivated, however. We can see how protective packaging masks an object’s fragility. Does it make sense to say that being in slightly dim conditions would mask someone’s honesty? What good is honesty if it gives out so easily? Does it make sense to say that not being watched would
82
Advances in Experimental Moral Psychology
fink someone’s honesty? What good is honesty if honest people need constant monitoring? Ruling out all of the asocial and social influences described in the previous section as disrupters isn’t just ad hoc; it threatens to disqualify honesty from being a virtue. Furthermore, social and asocial influences are ubiquitous. Indeed, it’s difficult even to think of them as influences because they are so common. Should we say that very bright lighting is the default condition, and that lower levels of light are all situational influences? To do so would be to count half of each day as disrupted. Or should we say that twilight is the default, and that both very bright and very dark conditions are situational influences? It’s hard to know what would even count in favor of one of these proposals. Even more to the point, what one might want to rule out as a disrupter in one case is likely to contribute to what seems like a manifestation of virtue in other cases. Should we say that being in a bad mood is a situational influence? People in a bad mood give much less than other people to charities that are good but not great; they also give much more than other people to charities that are very good indeed (Weyant 1978). You can’t have it both ways. If bad moods mask generosity in the former case, they mimic it in the latter. Failure to give in the former type of case would then not be evidence against generosity, but even giving quite a bit in the latter type of case would not be evidence for it. If we try to rule out all of these factors, leaving just the agent in her naked virtue or vice, we may find that she disappears too. Strip away the body and the community, and you leave not the kernel of authentic character, but something that’s not even recognizably human. Instead of filtering out as much as possible, I want to propose including as much as possible by expanding the unit of analysis, the bearer of virtue. Instead of thinking of virtue as a property of an individual agent, we should construe it as a triadic relation among a person, a social milieu, and an asocial environment. There are two ways of fitting the milieu and the environment into the subjunctive conditional analysis. They could be incorporated into the stimulus conditions: (AD-SCA*) Person p is disposed to activity A in condition C-social-milieuS-and-asocial-environment-E if and only if p would A if it were the case that C-in-S-and-E.
What are the Bearers of Virtues?
83
Or they could be fused with the agent: (AD-SCA†) Person p-in-S-and-E is disposed to activity A in condition C if and only if p-in-S-and-E would A if C were the case and there were no disrupters present.
According to AD-SCA*, the person is still the sole bearer of the disposition; it’s just a more limited disposition, with much stronger stimulus conditions. This can be seen as a rendering of Doris’s (2002) theory of local traits in the language of disposition theory. An important problem with such dispositions is that, even if they are empirically supportable, they are normatively uninspiring. According to AD-SCA†, in contrast, the bearer of the disposition is now a complex, extended object: the person, the milieu, and the environment. What I want to suggest is that, given the sorts of creatures we are—embodied, socially embedded, with cognition and motivation extended beyond the boundaries of our own skin (Clark 2008; Clark and Chalmers 1998)—AD-SCA† is more attractive. Virtue would inhere, on this view, in the interstices between the person and her world. The object that possesses the virtue in question would be a functionally and physically extended complex comprising the agent, her social setting, and her asocial environment. The conditions under which the social and asocial environment can be legitimately included in such an extended whole are complex, but we can take a cue here from Pritchard (2010, p. 15), who argues that phenomena “that extend outside the skin of [the] agent can count as part of one’s cognitive agency just so long as they are appropriately integrated” into one’s functioning. Pritchard is here discussing cognitive rather than ethical dispositions, but the idea is the same: provided that the social and asocial phenomena outside the moral agent’s skin are appropriately integrated into her functioning, they may count as part of her moral agency and partially constitute her moral virtues. A paradigm example is the ongoing and interactive feedback we have with our friends. At least when we are at our best, we try to live up to our friends’ expectations; we are attuned to their reactive attitudes; we consider prospectively whether they would approve or disapprove of some course of action; we consult with them both explicitly and imaginatively; we revise our beliefs and values in light of their feedback.5 When we are functionally integrated with friends in this way, on the model
84
Advances in Experimental Moral Psychology
I am proposing here, they are partial bearers of whatever virtues (and vices) we might have. Or rather, to the extent that a virtue or vice is possessed in this context, it is possessed by the pair of friends together, and not by either of them on her own. Friendship is an ideal example of the kind of functional integration I have in mind here, though it may well be possible to integrate other social and asocial properties and objects into a moral agent’s functioning. This doesn’t mean that we couldn’t also continue to think of individuals as (potential) bearers of virtues, but the answer to the question, “Is there virtue here?,” might differ depending on which bearer the questioner had in mind. For example, it might not be the case that the individual agent has the virtue in question, but that the complex object constituted by the agent, her social milieu, and her asocial environment does have the virtue in question. One consequence of this view is that virtue is multiply realizable, with different levels of contribution made by each of the relata. To be honest, for example, would be to have certain basic personality dispositions, but also some combination of the following: to be considered honest by one’s friends and peers (and to know it), to consider oneself honest, to be watched or at least watchable, and to be in whatever bodily states promote the characteristic manifestations of honesty. Someone could become honest, on this view, in standard ways, such as habituation and reflection on reasons. But someone could also become honest in non-standard ways, such as noticing others’ signaling of expectations or an increase in local luminescence. This makes it both easier and harder to be virtuous: deficiencies in personality can be made up for through social and bodily support, but strength of personality can also be undermined by lack of social and bodily support. To illustrate this, consider the differences among Figures 4.1, 4.2, and 4.3. One of the salutary upshots of this way of thinking about virtue is that it helps to make sense of the diversity named by any given trait term. Different people are more or less generous, and on several dimensions. By making explicit reference to the social milieu and the asocial environment, this framework suggests ways in which partial virtue could be differently instantiated. Two people might both count, at a very coarse-grained level of description, as mostly honest, but one could do so because of personal and social strengths and despite asocial weaknesses, while the other does so because of social and asocial strengths and despite some personal weaknesses. One way to capture
What are the Bearers of Virtues?
85
Personal 1 Personal 2
Environmental 3
Personal 3
Environmental 2
Social 1
Environmental 1
Social 3
Social 2
Figure 4.1 Represents a case of perfect virtue: all three relata (personal, social, and environmental) make maximal contributions. But virtue-concepts are threshold concepts. Someone can be generous even if she sometimes doesn’t live up to the ideal of perfect generosity.
Personal 1 Personal 2
Environmental 3
Personal 3
Environmental 2
Social 1
Environmental 1
Social 3
Social 2
Figure 4.2 Represents one way of doing that, with a modest contribution from the environment and more substantial contributions from social and personal resources.
this idea is to specify, for each virtue, the minimum area of the relevant radar graph that would need to be filled for the agent-in-milieu-and-environment to be a candidate for possessing that virtue. Furthermore, the framework allows for the plausible idea that there is a kind of asymmetry among the relata that bear virtues. Someone’s personality can only be so weak before we are no longer inclined to call him (or even the
86
Advances in Experimental Moral Psychology Personal 1 Personal 2
Environmental 3
Personal 3
Environmental 2
Social 1
Environmental 1
Social 3
Social 2
Figure 4.3 Represents another way of being generous enough without being perfectly generous—this time with a primary contribution from social factors and more modest contributions from both personal and environmental factors.
complex of which he is a part) virtuous, even if that weakness is counteracted by great social and asocial strengths. This condition could be captured by further specifying a minimum area for the personal contribution. We can also make sense of the intuition that someone is extremely virtuous if he displays characteristic manifestations despite weaknesses or pitfalls in the social and asocial environment. This could be done by specifying a different (larger) minimum area for the personal contribution that would ensure that the overall area was sufficiently great. Before concluding, I want to point to two normative upshots of this view. The first is that each of us is to some extent causally and even constitutively responsible for the character of others, and in several ways. By signaling our expectations, we tend to induce expectation-confirming responses. By interacting with others, we alter their moods. When we help to construct the material and bodily environment, we also construct others’ character. Especially in relation to our friends, with whom we are likely to be functionally integrated in the appropriate way, what we do, say, and signal may partially constitute our friends’ character traits. For someone who is teetering between virtue and vice, such influences can make all the difference.6 The comforting myth of individual responsibility notwithstanding, each of us truly is our brothers’ and sisters’ keeper.
What are the Bearers of Virtues?
87
This is a heavy responsibility to bear, but it pales in comparison to the responsibility borne by those with the power to set the default expectations that govern a society and to shape the material conditions of people’s lives. On the view I am proposing, politicians, corporate leaders, reporters, and architects, among many others, quite literally have the power to make people virtuous—and prevent them from being or becoming virtuous. If this is right, we need to hold such people more accountable, and to stop pretending that it’s possible to separate political and economic power from ethics.
Notes * Author Note: Mark Alfano, Princeton University Center for Human Values & Department of Philosophy, University of Oregon. The author thanks Philip Pettit, Hagop Sarkissian, Jennifer Cole Wright, Kate Manne, Jonathan Webber, and Lorraine Besser-Jones for helpful comments, criticisms, and suggestions. Correspondence should be addressed to Mark Alfano, 321 Wallace Hall, Princeton University, Princeton, NJ 08544. Email: [email protected]. 1 The canonical firebrands are Doris (2002) and Harman (1999). Flanagan arrived at the party both too early (1991) and too late (2009) to shape the course of the debate. 2 Upton (2009) is the only book-length effort, but her work makes little use of the literature on dispositions, relying instead on her own naive intuitions. Sreenivasan (2008) also discusses the dispositional nature of virtues without reference to the literature on dispositions. 3 The concepts of masking, mimicking, and finking were introduced by Johnston (1992), Smith (1977), and Martin (1994), respectively. 4 This distinction is based only on my own hunches, but conversations with philosophers and psychologists have left me confident in it. An empirical study of its plausibility would be welcome. 5 See Millgram (1987, p. 368), who argues that, “over the course of a friendship, one becomes (causally) responsible for the friend’s being who he is,” and Cocking and Kennett (1998, p. 504), who argue that a defining feature of friendship is that “as a close friend of another, one is characteristically and distinctively receptive to being directed and interpreted and so in these ways drawn by the other.” 6 Sarkissian (2010) makes a similar point.
88
Advances in Experimental Moral Psychology
References Adams, R. M. (2006). A Theory of Virtue: Excellence in Being for the Good. New York: Oxford University Press. Alfano, M. (2013). Character as Moral Fiction. Cambridge: Cambridge University Press. Annas, J. (2011). Intelligent Virtue. New York: Oxford University Press. Apsler, R. (1975). Effects of embarrassment on behavior toward others. Journal of Personality and Social Psychology, 32, 145–53. Ariely, D. (2008). Predictably Irrational. New York: Harper Collins. Baron, R. A., and Thomley, J. (1994). A whiff of reality: Positive affect as a potential mediator of the effects of pleasant fragrances on task performance and helping. Environment and Behavior, 26, 766–84. Bateson, M., Nettle, D., and Roberts, G. (2006). Cues of being watched enhance cooperation in a real-world setting. Biology Letters, 12, 412–14. Blass, T. (1999). The Milgram paradigm after 35 years: Some things we now know about obedience to authority. Journal of Applied Social Psychology, 29(5), 955–78. Boles, W., and Haywood, S. (1978). The effects of urban noise and sidewalk density upon pedestrian cooperation and tempo. Journal of Social Psychology, 104, 29–35. Burnham, T. (2003). Engineering altruism: A theoretical and experimental investigation of anonymity and gift giving. Journal of Economic Behavior and Organization, 50, 133–44. Burnham, T., and Hare, B. (2007). Engineering human cooperation. Human Nature, 18(2), 88–108. Carlsmith, J., and Gross, A. (1968). Some effects of guilt on compliance. Journal of Personality and Social Psychology, 53, 1178–91. Choi, S. (2008). Dispositional properties and counterfactual conditionals. Mind, 117, 795–841. Choi, S., and Fara, M. (Spring 2012). Dispositions. The Stanford Encyclopedia of Philosophy, Edward N. Zalta (ed.), http://plato.stanford.edu/archives/spr2012/ entries/dispositions/ Clark, A. (2008). Supersizing the Mind: Embodiment, Action, and Cognitive Extension. New York: Oxford University Press. Clark, A., and Chalmers, D. (1998). The extended mind. Analysis, 58, 7–19. Cocking, D., and Kennett, J. (1998). Friendship and the self. Ethics, 108(3), 502–27. Darley, J., and Latané, B. (1968). Bystander intervention in emergencies: Diffusion of responsibility. Journal of Personality and Social Psychology, 8, 377–83. Donnerstein, E., and Wilson, D. (1976). Effects of noise and perceived control on ongoing and subsequent aggressive behavior. Journal of Personality and Social Psychology, 34, 774–81.
What are the Bearers of Virtues?
89
Doris, J. (2002). Lack of Character: Personality and Moral Behavior. Cambridge: Cambridge University Press. Ernest-Jones, M., Nettle, D., and Bateson, M. (2011). Effects of eye images on everyday cooperative behavior: A field experiment. Evolution and Human Behavior, 32(3), 172–8. Flanagan, O. (1991). Varieties of Moral Personality: Ethics and Psychological Realism. Cambridge, MA: Harvard University Press. —(2009). Moral science? Still metaphysical after all these years. In Narvaez and Lapsley (eds), Moral Personality, Identity and Character: An Interdisciplinary Future. New York: Cambridge University Press. Funder, D., and Ozer, D. (1983). Behavior as a function of the situation. Journal of Personality and Social Psychology, 44, 107–12. Graham, G. (Fall 2010). Behaviorism. In E. Zalta (ed.), The Stanford Encyclopedia of Philosophy, http://plato.stanford.edu/archives/fall2010/entries/behaviorism/ Harman, G. (1999). Moral philosophy meets social psychology: Virtue ethics and the fundamental attribution error. Proceedings of the Aristotelian Society, New Series, 119, 316–31. Isen, A. (1987). Positive affect, cognitive processes, and social behavior. In L. Berkowitz (ed.), Advances in Experimental Social Psychology, Vol. 20, San Diego: Academic Press, pp. 203–54. Isen, A., Clark, M., and Schwartz, M. (1976). Duration of the effect of good mood on helping: ‘Footprints on the sands of time.’ Journal of Personality and Social Psychology, 34, 385–93. Isen, A., Shalker, T., Clark, M., and Karp, L. (1978). Affect, accessibility of material in memory, and behavior: A cognitive loop. Journal of Personality and Social Psychology, 36, 1–12. Johnston, M. (1992). How to speak of the colors. Philosophical Studies, 68, 221–63. Kamtekar, R. (2004). Situationism and virtue ethics on the content of our character. Ethics, 114(3), 458–91. Kristjansson, K. (2008). An Aristotelian critique of situationism. Philosophy, 83(1), 55–76. Latané, B., and Nida, S. (1981). Ten years of research on group size and helping. Psychological Bulletin, 89, 308–24. Lewis, D. (1997). Finkish dispositions. The Philosophical Quarterly, 47, 143–58. MacIntyre, A. (1984). After Virtue: A Study in Moral Theory. Notre Dame: University of Notre Dame Press. Martin, C. (1994). Dispositions and conditionals. The Philosophical Quarterly, 44, 1–8. Matthews, K. E., and Cannon, L. K. (1975). Environmental noise level as a determinant of helping behavior. Journal of Personality and Social Psychology, 32, 571–7.
90
Advances in Experimental Moral Psychology
McKitrick, J. (2003). A case for extrinsic dispositions. Australasian Journal of Philosophy, 81, 155–74. Milgram, S. (1974).Obedience to Authority. New York: Harper Collins. Millgram, E. (1987). Aristotle on making other selves. Canadian Journal of Philosophy, 17(2), 361–76. Mischel, W. (1968). Personality and Assessment. New York: Wiley. Pritchard, D. (2010). Cognitive ability and the extended cognition thesis. Synthese, 175, 133–51. Regan, J. (1971). Guilt, perceived injustice, and altruistic behavior. Journal of Personality and Social Psychology, 18, 124–32. Rigdon, M., Ishii, K., Watabe, M., and Kitayama, S. (2009). Minimal social cues in the dictator game. Journal of Economic Psychology, 30(3), 358–67. Russell, D. (2009). Practical Intelligence and the Virtues. Oxford: Oxford University Press. Sarkissian, H. (2010). Minor tweaks, major payoffs: The problems and promise of situationism in moral philosophy. Philosophers’ Imprint, 10(9), 1–15. Schnall, S., Haidt, J., Clore, G., and Jordan, A. (2008). Disgust as embodied moral judgment. Personality and Social Psychology Bulletin, 34, 1096–109. Schwartz, S., and Gottlieb, A. (1991). Bystander anonymity and reactions to emergencies. Journal of Personality and Social Psychology, 39, 418–30. Slote, M. (2001). Morals from Motives. New York: Oxford University Press. Smith, A. (1977). Dispositional properties. Mind, 86, 439–45. Snow, N. (2010). Virtue as Social Intelligence: An Empirically Grounded Theory. New York: Routledge. Sreenivasan, G. (2008). Character and consistency: Still more errors. Mind, 117(467), 603–12. Tajfel, H. (1970). Experiments in intergroup discrimination. Scientific American, 5(223), 27–97. —(1981). Human Groups and Social Categories. Cambridge: Cambridge University Press. Upton, C. (2009). Situational Traits of Character: Dispositional Foundations and Implications for Moral Psychology and Friendship. Lanham: Lexington Books. Weyant, J. (1978). Effects of mood states, costs, and benefits on helping. Journal of Personality and Social Psychology, 36, 1169–76. Zhong, C.-B., Bohns, V., and Gino, F. (2010). Good lamps are the best police: Darkness increases dishonesty and self-interested behavior. Psychological Science, 21(3), 311–14.
5
The Moral Behavior of Ethicists and the Power of Reason Joshua Rust and Eric Schwitzgebel*
Professional ethicists behave no morally better than do other professors. At least that is what we have found in a series of empirical studies that we will summarize below. Our results create a prima facie challenge for a certain picture of the relationship between intellectual reasoning and moral behavior—a picture on which explicit, intellectual cognition has substantial power to change the moral opinions of the reasoner and thereby to change the reasoner’s moral behavior. Call this picture the Power of Reason view. One alternative view has been prominently defended by Jonathan Haidt. We might call it the Weakness of Reason view, or more colorfully the Rational Tail view, after the headline metaphor of Haidt’s seminal 2001 article, “The emotional dog and its rational tail.” According to the Rational Tail view (which comes in different degrees of strength), emotion or intuition drives moral opinion and moral behavior, and explicit forms of intellectual cognition function mainly post hoc, to justify and socially communicate conclusions that flow from emotion or intuition. Haidt argues that our empirical results favor his view (2012, p. 89). After all, if intellectual styles of moral reasoning don’t detectably improve the behavior even of professional ethicists who build their careers on expertise in such reasoning, how much hope could there be for the rest of us to improve by such means? While we agree with Haidt that our results support the Rational Tail view over some rationalistic rivals, we believe that other models of moral psychology are also consistent with our findings, and some of these models give explicit intellectual reasoning a central, powerful role in shaping the
92
Advances in Experimental Moral Psychology
reasoner’s behavior and attitudes. Part 1 summarizes our empirical findings. Part 2 explores five different theoretical models, including the Rational Tail, that are more or less consistent with those findings.
Part 1: Our empirical studies Missing library books Our first study (Schwitzgebel 2009) examined the rates at which ethics books were missing from 32 leading academic libraries, compared to other philosophy books, according to those libraries’ online catalogs. The primary analysis was confined to relatively obscure books likely to be borrowed mostly by specialists in the field—275 books reviewed in Philosophical Review between 1990 and 2001, excluding titles cited five or more times in the Stanford Encyclopedia of Philosophy. Among these books, we found ethics books somewhat more likely to be missing than non-ethics books: 8.5 percent of the ethics books that were off the shelf were listed as missing or as more than one year overdue, compared to 5.7 percent of the non-ethics philosophy books that were off the shelf. This result holds despite a similar total number of copies of ethics and nonethics books held, similar total overall checkout rates of ethics and non-ethics books, and a similar average publication date of the books. We also found that classic pre-twentieth-century ethics texts were more likely to be missing than comparable non-ethics texts.
Peer ratings Our second study examined peer opinion about the moral behavior of professional ethicists (Schwitzgebel and Rust 2009). We set up a table in a central location at the 2007 Pacific Division meeting of the American Philosophical Association (APA) and offered passersby gourmet chocolate in exchange for taking a “5-minute philosophical-scientific questionnaire,” which they completed on the spot. One version of the questionnaire asked respondents their opinion about the moral behavior of ethicists in general, compared to other philosophers and compared to non-academics of similar social background (with parallel questions about the moral behavior of specialists in metaphysics
The Moral Behavior of Ethicists and the Power of Reason
93
and epistemology). Opinion was divided: Overall, 36 percent of respondents rated ethicists morally better behaved on average than other philosophers, 44 percent rated them about the same, and 19 percent rated them worse. When ethicists’ behavior was compared to that of non-academics, opinion was split 50 percent–32 percent–18 percent between better, same, and worse. Another version of the questionnaire asked respondents to rate the moral behavior of the individual ethicist in their department whose last name comes next in alphabetical order, looping back from Z to A if necessary, with a comparison question about the moral behavior of a similarly alphabetically chosen specialist in metaphysics and epistemology. Opinion was again split: 44 percent of all respondents rated the arbitrarily selected ethics specialist better than they rated the arbitrarily selected M&E specialist, 26 percent rated the ethicist the same, and 30 percent rated the ethicist worse. In both versions of the questionnaire, the skew favoring the ethicists was driven primarily by respondents reporting a specialization or competence in ethics, who tended to avoid rating ethicists worse than others. Non-ethicist philosophers tended to split about evenly between rating the ethicists better, same, or worse.
Voting rates We assume that regular participation in public elections is a moral duty, or at least that it is morally better than non-participation (though see Brennan 2011). In an opinion survey to be described below, we found that over 80 percent of sampled US professors share that view. Accordingly, we examined publicly available voter participation records from five US states, looking for name matches between voter rolls and online lists of professors in nearby universities, excluding common and multiply-appearing names (Schwitzgebel and Rust 2010). In this way, we estimated the voting participation rates of four groups of professors: philosophical ethicists, philosophers not specializing in ethics, political scientists, and professors in departments other than philosophy and political science. We found that all four groups of professors voted at approximately the same rates, except for the political science professors, who voted about 10–15 percent more often than did the other groups. This result survived examination for confounds due to gender, age, political party, and affiliation with a research-oriented versus teaching-oriented university.
94
Advances in Experimental Moral Psychology
Courtesy at philosophy conferences While some rules of etiquette can be morally indifferent or even pernicious, we follow Confucius (5th c. BCE/2003), Karen Stohr (2012), and others in seeing polite, respectful daily behavior as an important component of morality. With this in mind, we examined courteous and discourteous behavior at meetings of the American Philosophical Association, comparing ethics sessions with non-ethics sessions (Schwitzgebel et al. 2012). We used three measures of courtesy—talking audibly during the formal presentation, allowing the door to slam when entering or exiting mid-session, and leaving behind litter at one’s seat—across 2,800 audience hours of sessions at four different APA meetings. None of the three measures revealed any statistically detectable differences in courtesy. Audible talking (excluding brief, polite remarks like “thank you” for a handout) was rare: 0.010 instances per audience hour in the ethics sessions versus 0.009 instances per audience hour in the non-ethics sessions (z ⫽ 0.3, p ⫽ 0.77). The median rate of door slamming per session (compared to midsession entries and exits in which the audience member attempted to shut the door quietly) was 18.2 percent for the ethics sessions and 15.4 percent for the non-ethics sessions (Mann-Whitney test, p ⫽ 0.95). Finally, ethicists were not detectably less likely than non-ethicists to leave behind cups (16.8% vs. 17.8% per audience member, z ⫽⫺0.7, p ⫽ 0.48) or trash (11.6% vs. 11.8%, z ⫽⫺0.2, p ⫽ 0.87). The latter result survives examination for confounds due to session size, time of day, and whether paper handouts were provided. However, we did find that the audience members in environmental ethics sessions left behind less trash than did the audience in all other sessions combined (3.0% vs. 11.9%, Fisher’s exact test, p ⫽ 0.02).
APA free riding We assume a prima facie duty for program participants in philosophy conferences to pay the modest registration fees that the organizers of those conferences typically charge. However, until recently the American Philosophical Association had no mechanism to enforce conference registration, which resulted in a substantial free-riding problem. With this in mind, we examined the Pacific Division APA programs from 2006 to 2008, classifying sessions into ethics, non-ethics, or excluded. We then examined
The Moral Behavior of Ethicists and the Power of Reason
95
the registration compliance of program participants in ethics sessions versus program participants in non-ethics sessions by comparing de-identified, encrypted lists of participants in those sessions (participants with common names excluded) to similarly encrypted lists of people who had paid their registration fees (Schwitzgebel 2013).1 During the period under study, ethicists appear to have paid their conference registration fees at about the same rate as did non-ethicist philosophers (74% vs. 76%, z ⫽ ⫺ 0.7, p ⫽ 0.50). This result survives examination for confounds due to gender, institutional prestige, program role, year, and status as a faculty member versus graduate student.
Responsiveness to student emails Yet another study examined the rates at which ethicists responded to brief email messages designed to look as though written by undergraduates (Rust and Schwitzgebel 2013). We sent three email messages—one asking about office hours, one asking for the name of the undergraduate advisor, and one inquiring about an upcoming course—to ethicists, non-ethicist philosophers, and a comparison group of professors in other departments, drawing from online faculty lists at universities across several US states. All messages addressed the faculty member by name, and some included additional specific information such as the name of the department or the name of an upcoming course the professor was scheduled to teach. The messages were checked against several spam filters, and we had direct confirmation through various means that over 90 percent of the target email addresses were actively checked. Overall, ethicists responded to 62 percent of our messages, compared to a 59 percent response rate for non-ethicist philosophers, and 58 percent for non-philosophers—a difference that doesn’t approach statistical significance despite (we’re somewhat embarrassed to confess) 3,109 total trials (c2 ⫽ 3.4, p ⫽ 0.18).
Self-reported attitudes and behavior Our most recent study examined ethicists’, non-ethicist philosophers’, and non-philosophers’ self-reported attitudes and behavior on a number of issues including membership in disciplinary societies, voting, staying in touch with
96
Advances in Experimental Moral Psychology
one’s mother, vegetarianism, organ and blood donation, responsiveness to student emails, charity, and honesty in responding to survey questionnaires (Schwitzgebel and Rust in press). The survey was sent to about a thousand professors in five different US states, with an overall response rate of 58 percent or about 200 respondents in each of the three groups. Identifying information was encrypted for participants’ privacy. On some issues—voting, email responsiveness, charitable donation, societal membership, and survey response honesty—we also had direct, similarly encrypted, observational measures of behavior that we could compare with self-report.2 Aggregating across the various measures, we found no difference among the groups in overall self-reported moral behavior, in the accuracy of the self-reports for those measures where we had direct observational evidence, or in the correlation between expressed normative attitude and either self-reported or directly observed behavior. The one systematic difference we did find was this: Across several measures—vegetarianism, charitable donation, and organ and blood donation—ethicists appeared to embrace more stringent moral views than did non-philosophers, while non-ethicist philosophers held views of intermediate stringency. However, this increased stringency of attitude was not unequivocally reflected in ethicists’ behavior. This last point is best seen by examining the two measures on which we had the best antecedent hope that ethicists would show moral differences from non-ethicists: vegetarianism and charitable donation. Both issues are widely discussed among ethicists, who tend to have comparatively sophisticated philosophical opinions about these matters, and professors appear to exhibit large differences in personal rates of charitable donation and meat consumption. Furthermore, ethicists’ stances on these issues are directly connected to specific, concrete behaviors that they can either explicitly implement or not (e.g., to donate 10% annually to famine relief; to refrain from eating the meat of such-and-such animals). This contrasts with exhortations like “be a kinder person” that are difficult to straightforwardly implement or to know if one has implemented.
Self-reported attitude and behavior: Eating meat We solicited normative attitudes about eating meat by asking respondents to rate “regularly eating the meat of mammals such as beef or pork” on a
The Moral Behavior of Ethicists and the Power of Reason
97
nine-point scale from “very morally bad” to “very morally good” with the midpoint marked “morally neutral.” On this normative question, there were large differences among the groups: 60 percent of ethicist respondents rated meat-eating somewhere on the bad side of the scale, compared to 45 percent of non-ethicist philosophers and only 19 percent of professors from other departments (c2 ⫽ 64.2, p ⬍ 0.001). Later in the survey, we posed two behavioral questions. First, we asked “During about how many meals or snacks per week do you eat the meat of mammals such as beef or pork?” Next, we asked “Think back on your last evening meal, not including snacks. Did you eat the meat of a mammal during that meal?” On the meals-per-week question, we found a modest difference among the groups: Ethicists reported a mean of 4.1 meals per week, compared to 4.6 for non-ethicist philosophers and 5.3 for non-philosophers (ANOVA, F ⫽ 5.2, p ⫽ 0.006). We also found 27 percent of ethicists to report no meat consumption (zero meat meals per week), compared to 20 percent of non-ethicist philosophers and 13 percent of non-philosophers (c2 ⫽ 9.3, p ⫽ 0.01). However, statistical evidence suggested that respondents were fudging their meals-per-week answers: Self-reported meals per week was not mathematically consistent with what one would expect given the numbers reporting having eaten meat at the previous evening meal. (For example, 21% of respondents who reported eating meat at only one meal per week reported having eaten meat at their previous evening meal.) And when asked about their previous evening meal, the groups’ self-reports differed only marginally, with ethicists in the intermediate group: 37 percent of ethicists reported having eaten the meat of a mammal at their previous evening meal, compared to 33 percent of non-ethicist philosophers and 45 percent of non-philosophers (c2 ⫽ 5.7, p ⫽ 0.06).
Self-reported attitude and behavior: Charity We solicited normative opinion about charity in two ways. First, we asked respondents to rate “donating 10 percent of one’s income to charity” on the same nine-point scale we used for the question about eating meat. Ethicists expressed the most approval, with 89 percent rating it as good and a mean rating of 7.5 of the scale, versus 85 percent and 7.4 for non-ethicist philosophers and 73 percent and 7.1 for non-philosophers (c2 ⫽ 17.0, p ⬍ 0.001; ANOVA,
98
Advances in Experimental Moral Psychology
F ⫽ 4.3, p ⫽ 0.01). Second, we asked what percentage of income the typical professor should donate to charity (instructing participants to enter “0” if they think it’s not the case that the typical professor should donate to charity). Among ethicists, 9 percent entered “0,” versus 24 percent of non-ethicist philosophers and 25 percent of non-philosophers (c2 ⫽ 18.2, p ⬍ 0.001). Among those not entering “0,” the geometric mean was 5.9 percent for the ethicists versus 4.8 percent for both of the other groups (ANOVA, F ⫽ 3.6, p ⫽ 0.03). Later in the survey, we asked participants what percentage of their income they personally had donated to charity in the previous calendar year. Non-ethicist philosophers reported having donated the least, but there was no statistically detectable difference between the self-reported donation rates of the ethicists and the non-philosophers. (Reporting zero: 4% of ethicists vs. 10% of non-ethicist philosophers and 6% of non-philosophers, c2 ⫽ 5.9, p ⫽ 0.052; geometric mean of the non-0’s 3.7% vs. 2.6% vs. 3.6%, ANOVA, F ⫽ 5.5, p ⫽ 0.004.) However, we also had one direct measure of charitable behavior: Half of the survey recipients were given a charity incentive to return the survey—$10 to be donated to their selection from among Oxfam America, World Wildlife Fund, CARE, Make-a-Wish Foundation, Doctors Without Borders, or American Red Cross. By this measure, the non-ethicist philosophers showed up as the most charitable, and in fact were the only group who responded at even statistically marginally higher rates when given the charity incentive (67% vs. 59%; compared to 59% on both versions for ethicists and 55% vs. 52% for non-philosophers; c2 ⫽ 2.8, p ⫽ 0.097; c2 ⫽ 0.2, p ⫽ 0.64; c2 ⫽ 0.0, p ⫽ 1.0). While we doubt that this is a dependably valid measure of charitable behavior overall, we are also somewhat suspicious of the self-report measures. We judge the overall behavioral results to be equivocal, and certainly not to decisively favor the ethicists over both of the two other groups.
Conclusion Across a wide variety of measures, it appears that ethicists, despite expressing more stringent normative attitudes on some issues, behave not much differently than do other professors. However, we did find some evidence that philosophers litter less in environmental ethics sessions than in other APA sessions, and we found some equivocal evidence that might suggest slightly
The Moral Behavior of Ethicists and the Power of Reason
99
higher rates of charitable giving and slightly lower rates of meat-eating among ethicists than among some other subsets of professors. On one measure—the return of library books—it appears that ethicists might behave morally worse.
Part 2: Possible explanations The rational tail view One possibility is that Haidt’s Rational Tail view, as described in the introduction, is correct. Emotion or intuition is the dog; explicit reasoning is the tail; and this is so even among professional ethicists, whom one might have thought would be strongly influenced by explicit moral reasoning if anyone is. Our judgments and behavior—even the judgments and behavior of professional ethicists—are very little governed by our reasoning. We do what we’re going to do, we approve of what we’re going to approve of, and we concoct supporting reasons to a large extent only after the fact as needed. Haidt compares reasoning and intuition to a rider on an elephant, with the rider, reasoning, generally compelled to travel in the direction favored by the elephant. Haidt also compares the role of reasoning to that of a lawyer rather than a judge: The lawyer does her best to advocate for the positions given to her by her clients—in this case the intuitions or emotions—producing whatever ideas and arguments are convenient for the predetermined conclusion. Reason is not a neutral judge over moral arguments but rather, for the most part, a paid-off advocate plumping for one side. Haidt cites our work as evidence for this view (e.g., Haidt 2012, p. 89), and we’re inclined to agree that most of it fits nicely with his view and so in that way lends support. If moral reasoning were almost entirely ineffectual, that could explain our virtually flat results; and where our results are not entirely flat, specific secondary mechanisms could be invoked (e.g., the social awkwardness of leaving trash behind at an environmental ethics session; a tendency for people with antecedently stringent moral views to be more likely to enter professional ethics in the first place). It would be rash, however, to adopt an absolutely extreme version of the Rational Tail view (and Haidt himself does not). At least sometimes, it seems, the tail can wag the dog and the elephant can take direction from the rider. Rawls’s (1971) picture of philosophical method as involving “reflective
100
Advances in Experimental Moral Psychology
equilibrium” between intuitive assessments of particular cases and rationally appealing general principles is one model of how this might occur. The idea is that just as one sometimes adjusts one’s general principles to match one’s pretheoretical intuitions about particular cases, one also sometimes rejects one’s pretheoretical intuitions about particular cases in light of one’s general principles. It seems both anecdotally and phenomenologically compelling that explicit moral reasoning sometimes prompts rejection of one’s initial intuitive moral judgments, and that when this happens, changes in real-world moral behavior sometimes follow. How could there not be at least some truth in the Power of Reason view? So why does there seem to be so little systematic evidence of that power—even when looking at what one might think would be the best-case population for seeing its effects? Without directly arguing against Haidt’s version of the Rational Tail view or for the Power of Reason view, we present four models of the relationship between explicit moral reasoning and real-world moral behavior that permit explicit reasoning to play a substantial role in shaping the reasoner’s moral behavior, compatibly with our empirical findings above. While we agree with Haidt that our results support the Rational Tail view, our findings are also consistent with some models of moral psychology which place more emphasis on the Power of Reason. We focus on our own evidence, but we recognize that a plausible interpretation of it must be contextualized with other sorts of evidence from recent moral psychology that seems to support the Rational Tail view—including Haidt’s own dumbfounding evidence (summarized in his 2012); evidence that we have poor knowledge of the principles driving our moral judgments about puzzle cases (e.g., Cushman et al. 2006; Mikhail 2011; Ditto and Liu 2012); evidence about the diverse factors influencing moral judgment (e.g., Hauser 2006; Greene 2008; Schnall et al. 2008); and evidence from the cognitive dissonance and neuropathology literatures on post hoc rationalization of behavior (e.g., Festinger 1957; Hirstein 2005; Cooper 2007).
Narrow principles Professional ethicists might have two different forms of expertise. One might concern the most general principles and unusually clean hypothetical cases—the kinds of principles and cases at stake when ethicists argue about
The Moral Behavior of Ethicists and the Power of Reason
101
deontological versus consequentialist ethics using examples of runaway trolleys and surgeons who can choose secretly to carve up healthy people to harvest their organs. Expertise of that sort might have little influence on one’s dayto-day behavior. A second form of expertise might be much more concretely practical but concern only narrow principles—principles like whether it’s okay to eat meat and under what conditions, whether one should donate to famine relief and how much, or whether one has a duty to vote in public elections. An ethicist can devote serious, professional-quality attention to only a limited number of such practical principles; and once she does so, her behavior might be altered favorably as a result. But such reflection would only alter the ethicist’s behavior in those few domains that are the subject of professional focus. If philosophical moral reasoning tends to improve moral behavior only in specifically selected narrow domains, we might predict that ethicists would show better behavior in just those narrow domains. For example, those who select environmental ethics for a career focus might consequently pollute and litter less than they otherwise would, in accord with our results. (Though it is also possible, of course, that people who tend to litter less are more likely to be attracted to environmental ethics in the first place, or that the context of an environmental ethics session is such that even non-specialists would be moved to litter a bit less.) Ethicists specializing in issues of gender or racial equality might succeed in mitigating their own sexist and racist behavior. Perhaps, too, we will see ethicists donating more to famine relief and being more likely to embrace vegetarianism—issues that have received wide attention in recent Anglophone ethics and on which we found some equivocal evidence of ethicists’ better behavior. Common topics of professional focus tend to be interestingly difficult and nuanced. So maybe intellectual forms of ethical reflection do make a large difference in one’s personal behavior, but only in hard cases, where our prereflective intuitions fail to be reliable guides: The reason why ethicists are no more likely than non-ethicists to call their mothers or answer student emails might be because the moral status of these actions is not, for them, an intuitively nonobvious, attractive subject of philosophical analysis and they take no public stand on it. Depending on other facts about moral psychology, the Narrow Principles hypothesis might predict—as we seem to find in the vegetarianism and charity
102
Advances in Experimental Moral Psychology
data—that attitude differences will tend to be larger than behavioral differences. It will do so because, on this model, the principle must be accepted before the behavior changes, and since behavioral change requires further exertion beyond simply adopting a principle on intellectual grounds. Note that, in contrast, a view on which people embrace attitudes wholly to rationalize their existing behaviors or behavioral inclinations would probably not predict that ethicists would show highly stringent attitudes where their behavior is unexceptional. The Narrow Principles model, then, holds that professional focus on narrow principles can make a substantial behavioral difference. In their limited professional domains, ethicists might then behave morally better than they otherwise would. Whether they also therefore behave morally better overall might then turn on whether the attention dedicated to one moral issue results in moral backsliding on other issues, for example due to moral licensing (the phenomenon in which acting well in one way seems to license people to act worse in others; Merritt, Effron, and Monin 2010) or ego depletion (the phenomenon according to which dedicating self-control in one matter leaves fewer resources to cope with temptation in other matters; Mead et al. 2010).
Reasoning might lead one to behave more permissibly but no better Much everyday practical moral reasoning seems to be dedicated not to figuring out what is morally the best course—often we know perfectly well what would be morally ideal, or think we do—but rather to figuring out whether something that is less than morally ideal is still permissible. Consider, for example, sitting on the couch relaxing while one’s spouse does the dishes (a very typical occasion of moral reflection for some of us!). One knows perfectly well that it would be morally better to get up and help. The topic of reflection is not that, but instead whether, despite not being morally ideal, it is still permissible not to help: Did one have a longer, harder day? Has one been doing one’s fair share overall? Maybe explicit moral reasoning can help one see one’s way through these issues. And maybe, furthermore, explicit moral reasoning generates two different results approximately equally often: the result that what one might have thought was morally permissible is not
The Moral Behavior of Ethicists and the Power of Reason
103
in fact permissible (thus motivating one to avoid it, e.g., to get off the couch) and the result that what one might have thought was morally impermissible is in fact permissible (thus licensing one not to do the morally ideal thing, e.g., to stay on the couch). If reasoning does generate these two results about equally often, people who tend to engage in lots of moral reflection of this sort might be well calibrated to permissibility and impermissibility, and thus behave more permissibly overall than do other people, despite not acting morally better overall. The Power of Reason view might work reasonably well for permissibility even if not for goodness and badness. Imagine someone who tends to fall well short of the moral ideal but who hardly ever does anything that would really qualify as morally wrong, contrasted with a sometimes-sinner sometimes-saint. This model, if correct, could be straightforwardly reconciled with our data as long as the issues we have studied—except insofar as they reveal ethicists behaving differently—allow for cross-cutting patterns of permissibility, for example, if it is often but not always permissible not to vote. It would also be empirically convenient for this view if it were more often permissible to steal library books than non-ethicists are generally inclined to think and ethical reflection tends to lead people to discover that fact.
Compensation for deficient intuitions Our empirical research can support the conclusion that philosophical moral reflection is not morally improving only given several background assumptions, such as (i) that ethicists do in fact engage in more philosophical moral reflection than do otherwise socially similar non-ethicists and (ii) that ethicists do not start out morally worse and then use their philosophical reflection to bring themselves up to average. We might plausibly deny the latter assumption. Here’s one way such a story might go. Maybe some people, from the time of early childhood or at least adolescence, tend to have powerful moral intuitions and emotions across a wide range of cases while other people have less powerful or less broad-ranging moral intuitions and emotions. Maybe some of the people in the latter group tend to be drawn to intellectual and academic thought; and maybe those people then use that intellectual and academic thought to compensate for their deficient moral intuitions and
104
Advances in Experimental Moral Psychology
emotions. And maybe those people, then, are disproportionately drawn into philosophical ethics. More or less, they are trying to figure out intellectually what the rest of us are gifted with effortlessly. These people have basically made a career out of asking “What is this crazy ethics thing, anyway, that everyone seems so passionate about?” and “Everyone else seems to have strong opinions about donating to charity or not, and when to do so and how much, but they don’t seem able to defend those opinions very well and I don’t find myself with that same confidence; so let’s try to figure it out.” Clinical psychopathy isn’t what we’re imagining here, nor do we mean to assume any particularly high uniformity in ethicists’ psychological profile. All this view requires is that whatever positive force moral reflection delivers to the group as a whole is approximately balanced out by a somewhat weaker set of pretheoretical moral intuitions in the group as a whole. If this were the case, one might find ethicists, even though no morally better behaved overall, more morally well behaved than they would have been without the crutch of intellectual reflection, and perhaps also morally better behaved than non-ethicists are in cases where the ordinary intuitions of the majority of people are in error. Conversely, one might find ethicists morally worse behaved in cases where the ordinary intuitions of the majority of people are a firmer guide than abstract principle. We hesitate to conjecture about what issues might fit this profile but if, for example, ordinary intuition is a poorer guide than abstract principle about issues such as vegetarianism, charity, and environmentalism and a better guide about the etiquette of dayto-day social interactions with one’s peers, then one would expect ethicists to behave better than average on the issues of the former sort and worse on issues of the latter sort.
Rationally driven moral improvement plus toxic rationalization in equal measure A final possibility is this: Perhaps the Power of Reason view is entirely right some substantial proportion of the time, but also a substantial proportion of the time explicit rational reflection is actually toxic, leading one to behave worse; and these two tendencies approximately cancel out in the long run. Such tendencies needn’t only concern permissibility and impermissibility, and the
The Moral Behavior of Ethicists and the Power of Reason
105
consequence of these countervailing forces needn’t involve any improvement in overall moral calibration. Perhaps we sometimes care about morality for its own sake, think things through reasonably well, and then act on the moral truths we thereby discover. And maybe the tools and habits of professional ethics are of great service in this enterprise. For example: One might stop to think about whether one really does have an obligation to go to the polls for the mayoral runoff election, despite a strong preference to stay at home and a feeling that one’s vote will make no practical difference to the outcome. And one might decide, through a process of explicit intellectual reasoning (let’s suppose by correctly applying Kant’s formula of universal law), that one does in fact have the duty to vote on this particular occasion. One rightly concludes that no sufficiently good excuse applies. As a result, one does something one would not have done absent that explicit reasoning: With admirable civic virtue, one overcomes one’s contrary inclinations and goes to the polls. But then suppose that also, in equal measure, things go just as badly wrong: When one stops to reflect, what one does is rationalize immoral impulses that one would otherwise not have acted on, generating a superficially plausible patina of argument that licenses viciousness which would have been otherwise avoided. Robespierre convinces himself that forming the Committee of Public Safety really is for the best, and consequently does evil that he would have avoided had he not constructed that theoretical veil. Much less momentously, one might concoct a superficial consequentialist or deontological story on which stealing that library book really is just fine, and so do it. The tools of moral philosophy might empower one all the more in this noxious reasoning. If this bivalent view of moral reflection is correct, we might expect moral reflection to produce movement away from the moral truth and toward one’s inclinations where common opinion is in the right and our inclinations are vicious but not usually acted on, and movement toward the moral truth where common opinion and our inclinations and unreflective behavior are all in the wrong. When widely held norms frustrate our desires, the temptation toward toxic rationalization can arise acutely and professional ethicists might be especially skilled in such rationalization. But this misuse of reason might be counterbalanced by a genuine noetic desire, which—perhaps especially with the right training—sometimes steers us right when otherwise we would have steered wrong. In the midst of widespread moral misunderstanding that
106
Advances in Experimental Moral Psychology
accords with people’s pretheoretic intuitions and inclinations, there might be few tools that allow us to escape error besides the tools of explicit moral reasoning. Again, one might make conditional predictions, depending on what one takes to be the moral truth. For example, if common opinion and one’s inclinations favor the permissibility of single-car commuting and yet singlecar commuting is in fact impermissible, one might predict more ethicist bus riders. If stealing library books is widely frowned upon and not usually done, though tempting, we might expect ethicists to steal more books.
Conclusion We decline to choose among these five models. There might be truth in all of them; and still other views are available too. Maybe ethicists find themselves increasingly disillusioned about the value of morality at the same time they improve their knowledge of what morality in fact requires. Or maybe ethicists learn to shield their personal behavior from the influence of their professional reflections, either to improve the objectivity of their reasoning or as a kind of self-defense against the apparent unfairness of being held to higher standards because of their choice of profession. In short, we believe the empirical evidence is insufficient to justify even tentative conclusions. We recommend the issues for further empirical study and for further armchair reflection.
Notes * Author’s Note: Joshua Rust, Stetson University, and Eric Schwitzgebel, University of California at Riverside. For helpful discussion of earlier drafts, thanks to Gunnar Bjornnson, Jon Haidt, Linus Huang, Hagop Sarkissian, and Jen Wright. Correspondence should be sent to: Eric Schwitzgebel, Department of Philosophy, University of California at Riverside, Riverside, CA 92521-0201, Email: eschwitz@ ucr.edu or Joshua Rust, Stetson University, Department of Philosophy 421 North Woodland Boulevard, DeLand, Florida 32723, Phone: 386.822.7581, Email: jrust@ stetson.edu. 1 The APA sent the list of names of APA paid registrants to a third party (U.C.R.’s Statistical Consulting Collaboratory) who were not informed of the nature of the
The Moral Behavior of Ethicists and the Power of Reason
107
research or the significance of the list of names. To them, it was just a meaningless list of names. Separately, we (2nd author and an Research Assistant (RA)) generated a list of names of people listed as participants on the APA program. Finally, a 2nd RA generated a mathematical formula unknown to us (but using certain guidelines) that would convert names into long number strings. This 2nd RA then converted the list of program participants into those number strings according to that formula and told the formula to the Collaboratory, who then separately converted their name lists into number strings using that same formula. Finally, the 2nd author received both encrypted lists and wrote a program to check for encrypted name matches between the lists. Names were matched just by last name and first initial to reduce the rate of false negatives due to different nicknames (e.g., Thomas vs. Tom), and common or repeated last names were excluded to prevent false positives, as were names with spaces, mid-capitals, diacritical marks, or in which the person used only an initial as the first name. Although the APA Pacific Division generously supplied the encrypted data, this research was neither solicited by nor conducted on behalf of the APA or the Pacific Division. 2 Survey recipients were among the people whose voting and email responsiveness we had examined in the studies reported above. The other observational measures were collected in the course of the survey study.
References Brennan, J. (2011). The Ethics of Voting. Princeton, NJ: Princeton University Press. Confucius. (5th c. BCE/2003). Analects. (E. Slingerland, trans.). Indianapolis, IN: Hackett Publishing Company. Cooper, J. (2007). Cognitive Dissonance. London: Sage. Ditto, P. H., and Liu, B. (2011). Deontological dissonance and the consequentialist crutch. In M. Mikulincer and P. R. Shaver (eds), The Social Psychology of Morality. Washington, DC: American Psychological Association. Festinger, L. (1957). A Theory of Cognitive Dissonance. Evanston, IL: Row, Peterson. Green, J. D. (2008). The secret joke of Kant’s soul. In W. Sinnott-Armstrong (ed.), Moral Psychology, vol. 3. Cambridge, MA: MIT Press. Haidt, J. (2001). The emotional dog and its rational tail: A social intuitionist approach to moral judgment. Psychological Review, 108, 814–34. doi:10.1037/0033295X.108.4.814 —(2012). The Righteous Mind. New York: Pantheon. Hauser, M. D. (2006). Moral Minds. New York: Ecco/HarperCollins.
108
Advances in Experimental Moral Psychology
Hirstein, W. (2005). Brain Fiction. Cambridge, MA: MIT Press. Mead, N. L., Alquist, J. L., and Baumeister, R. F. (2010). Ego depletion and the limited resource model of self-control. In R. Hassin, K. Ochsner, and Y. Trope (eds), Self-control in Society, Mind, and Brain. Oxford: Oxford University Press. Merritt, A. C., Effron, D. A., and Monin, B. (2010). Moral self-licensing: When being good frees us to be bad. Social and Personality Psychology Compass, 4, 344–57. doi:10.1111/j.1751-9004.2010.00263.x Mikhail, J. (2011). Elements of Moral Cognition. Cambridge: Cambridge University Press. Rawls, J. (1971). A Theory of Justice. Harvard: Harvard University Press. Rust, J., and Schwitzgebel, E. (2013). Ethicists’ and non-ethicists’ responsiveness to student emails: Relationships among self-reported behavior, expressed normative attitude, and directly observed behavior. Metaphilosophy, 44, 350–71. doi:10.1111/ meta.1203 Schnall, S., Haidt, J., Clore, G. L., and Jordan, A. H. (2008). Disgust as embodied moral judgment. Personality and Social Psychology Bulletin, 34, 1096–109. doi:10.1177/0146167208317771 Schwitzgebel, E. (2009). Do ethicists steal more books? Philosophical Psychology, 22, 711–25. doi:10.1080/09515080903409952 —(2013). Are ethicists any more likely to pay their registration fees at professional meetings? Economics & Philosophy, 29, 371–80. Schwitzgebel, E., and Rust, J. (2009). The moral behaviour of ethicists: Peer opinion. Mind, 118, 1043–59. doi:10.1093/mind/fzp108 —(2010). Do ethicists and political philosophers vote more often than other professors? Review of Philosophy and Psychology, 1, 189–99. doi:10.1007/s13164009-0011-6 —(in press). The moral behavior of ethics professors: Relationships among expressed normative attitude, self-described behavior, and directly observed behavior. Philosophical Psychology. Schwitzgebel, E., Rust, J., Huang, L. T., Moore, A. T., and Coates, J. (2012). Ethicists’ courtesy at philosophy conferences. Philosophical Psychology, 35, 331–40. doi:10.1080/09515089.2011.580524 Stohr, Karen (2012). On Manners. New York: Routledge.
Part Two
Moral Groundings
6
Pollution and Purity in Moral and Political Judgment Yoel Inbar and David Pizarro*
Disgust, an emotion that most likely evolved to keep us away from noxious substances and disease, seems especially active in our moral lives. People report feeling disgust in response to many immoral acts (e.g., Rozin et al. 1999), make more severe moral judgments when feeling disgusted (e.g., Wheatley and Haidt 2005), and are more likely to view certain acts as immoral if they have a tendency to be easily disgusted (Horberg et al. 2009). Yet, despite the wealth of evidence linking disgust and morality, the reason for the link remains unclear. This may be because the bulk of empirical work on the topic has been aimed at simply demonstrating that disgust and moral judgment are connected—a claim that, given the influence of rationalist models of moral judgment such as Kohlberg’s (1969), is novel and surprising. Fewer researchers have attempted to explain why disgust and moral judgment should be so connected (for recent exceptions, see Kelly 2011 and Tybur et al. 2012). Here, we present an attempt to do so. Our primary claim is that disgust functions as part of a general motivational system that evolved to keep individuals safe from disease. As such, disgust motivates negative evaluations of acts that are associated with a threat of contamination (e.g., norm violations pertaining to food and sex); negative attitudes toward unfamiliar groups who might pose the threat of contamination through physical contact (e.g., outgroups characterized by these norm violations, or who are unfamiliar); and greater endorsement of certain social and political attitudes that minimize contamination risk (such as increased sexual conservatism, reduced contact between different social
112
Advances in Experimental Moral Psychology
groups, and hostility toward foreigners). This account provides a theoretical rationale for the observed relationship between disgust and moral judgment, and it is able to unify findings from two literatures that, until now, have been largely separate: research examining the role of disgust in moral judgment, and research examining the effects of pathogen threat on political and social attitudes. One of the conclusions to emerge from this review is that that the link between disgust and morality may be different from what has been assumed by many researchers. Rather than a response to moral violations per se, disgust may instead be linked more generally to judgments about acts, individuals, and groups that pose a pathogen threat.
Disgust and moral judgment: Three claims In order to defend this conclusion, it is necessary to first review the evidence linking disgust to moral judgment, and to distinguish between the various ways disgust has been hypothesized to play a role in moral judgment. We have argued previously (Pizarro et al. 2011) that researchers have made three distinct claims regarding the relationship between disgust and moral judgment: (1) that the emotion of disgust is a consequence of perceiving moral violations; (2) that disgust serves to amplify judgments of immorality; and (3) that disgust acts as a moralizer, pushing previously non-moral issues into the moral domain. These claims are not mutually exclusive—all three could be true. However, there are varying degrees of empirical evidence to support each. According to the “disgust as consequence” view, disgust is the emotional output of a certain kind of moral appraisal. For instance, researchers have found that disgust is elicited by violations of moral “purity” (Rozin et al. 1999), “taboo” moral violations (Gutierrez and Ginner-Sorolla 2007), or being treated unfairly (Chapman et al. 2009). On this view, disgust might drive reactions to immorality—for example, by motivating people to reject or distance themselves from those seen as immoral—but does not play a causal role in determining whether an action is seen as immoral. In contrast, the “disgust as amplifier” view characterizes disgust as a causal influence on moral judgment, arguing that the presence of disgust during a moral evaluation makes wrong things seem even more wrong. This is a stronger
Pollution and Purity in Moral and Political Judgment
113
claim regarding the role of disgust, and has been made by researchers who have experimentally manipulated disgust independently of the act being evaluated, for example by inducing disgust with a post-hypnotic suggestion (Wheatley and Haidt 2005), with a foul odor, or with disgusting film clips (Schnall et al. 2008). Finally, the strongest causal claim regarding the influence of disgust on moral judgment is that of “disgust as moralizer.” On this view, morally neutral acts can enter the moral sphere by dint of their being perceived as disgusting. For instance, an act (such as smoking) can move from “unhealthy” to “immoral” if reliably accompanied by the emotion of disgust. This claim has the least empirical support of the three, although it is consistent with the finding that “morally dumbfounded” participants defend their self-admittedly irrational moral judgments with an appeal to the disgusting nature of an act (Haidt and Hersch 2001). Our argument here relies primarily on evidence for the disgust-asconsequence and disgust-as-amplifier views, for which the evidence is strongest (see Pizarro et al. 2011). In particular, the view we will defend here is a combination of these two approaches that takes into account additional research on the specificity of these effects—that disgust is more likely to arise and amplify judgments within a particular domain (viz., when the threat of pathogens is involved).
Why disgust? Why should disgust be involved in moral judgment at all, whether as a consequence, amplifier, or moralizer? Theoretical justifications have come largely in the form of broad statements that disgust is an emotional impulse to reject certain objects, individuals, or ideas that, for a variety of reasons, happen to overlap with objects, individuals, and ideas that are morally objectionable. For example, Schnall et al. (2008, p. 1097) write that disgust is “an emotion of social rejection” that is misattributed to many targets of judgment. Cannon et al. (2011, p. 326) write that “disgust is a reaction to offensive objects as well as offensive actions,” and Koleva et al. (2012) describe disgust as a response to “social contaminants.” Wheatley and Haidt (2005, p. 780) write that disgust is “a kind of information” that influences moral judgments.
114
Advances in Experimental Moral Psychology
Many of these theoretical explanations are simply restatements of the link between disgust and morality, and do not offer much by way of explanation for it. Rozin et al. (2008, p. 764) offer a more detailed argument, stating that disgust at immoral behavior results from “an opportunistic accretion of new domains of elicitors to a rejection system already in place”—in other words, that moral disgust piggybacks on an older, more basic food rejection response. Along the same lines, Kelly (2011) argues that disgust first evolved to motivate food rejection and pathogen avoidance, and was later “co-opted” to motivate moral judgment and intergroup attitudes. Finally, Tybur et al. (2012) propose an entirely different account, arguing that disgust in response to immorality is an evolved solution to a social coordination problem— namely, the need to coordinate condemnation of specific actions with others. On this account, expressions of disgust function as condemnation signals to others in the vicinity. All these accounts point to the possibility that moral judgments may be built on more simple psychological systems of avoidance and rejection. But why should the emotion of disgust in particular be involved in reactions to immorality? Kelly (2011) argues that disgust has two features that make it particularly suited to this role: (1) it entails a strong rejection response; and (2) its antecedents (i.e., elicitors) are, at least in part, learned (and therefore flexible). However, humans (and other animals) also show non-disgust-based aversive responses to unpleasant stimuli such as extreme heat or cold, loud or high-pitched sounds, dangerous predators, and so on. In fact, such responses are phylogenetically older than disgust—which is found in its full form only in humans—and are quite flexible, in that people (and other animals) can readily acquire aversions to novel stimuli (Staats and Staats 1958; Tully and Quinn 1985). In contrast, the elicitors of “core” disgust are in fact fairly circumscribed when compared to these other emotional responses, and tend to be limited to food, certain animals, and human body products (Rozin et al. 2008). If moral judgments needed to be built on top of an existing affective response, a more basic rejection system would thus be an equally if not more plausible candidate. Similarly, any number of emotions (such as anger) could be used to signal moral condemnation. Why would natural selection have favored disgust—an emotion that likely had its origins in a gustatory response to potential oral contamination—to serve this purpose?
Pollution and Purity in Moral and Political Judgment
115
It turns out that there is a good reason that disgust, rather than a more general-purpose rejection response, would have become associated with some moral violations—namely, that disgust evolved to motivate individuals not only to avoid ingesting (or touching) poisons and contaminants, but also to distance themselves from people who posed a risk of pathogen transmission. Schaller and colleagues (Faulkner et al. 2004; Park et al. 2003; Schaller and Duncan 2007) have argued that, over the course of human evolution, people developed a “behavioral immune system” that functioned as a first line of defense against exposure to pathogens or parasites. According to this theory, individuals who show cues of infection or disease should trigger the behavioral immune system, leading to disgust and, consequently, rejection or avoidance of that individual. Because this system would have evolved independently of any explicit knowledge about pathogens, its disease detection mechanism would need to be heuristic in nature— most likely, something like “any significant anomaly in an individual’s physical appearance.” This means that the behavioral immune system can be expected to respond to any individuals who deviate from normative physical appearance, regardless of whether they actually pose a contagion risk (Schaller and Park 2011). Likewise, individuals seen as engaging in unusual (i.e., non-normative) practices regarding food, cleanliness, and sex—activities that carry an especially high risk of pathogen transmission— should also be likely to evoke disgust and rejection. Finally, strangers (i.e., members of other groups or tribes) would have been especially likely to harbor novel (and therefore particularly dangerous) infectious agents. Encountering such individuals should thus also activate the behavioral immune system, motivating hostility, rejection, and the accompanying emotion of disgust. Indeed, individuals in hunter-gatherer cultures are often intensely hostile to strangers. The anthropologist Margaret Mead wrote that “most primitive tribes feel that if you run across one of these subhumans from a rival group in the forest, the most appropriate thing to do is bludgeon him to death” (as cited in Bloom 1997, p. 74). Likewise, the geographer and anthropologist Jared Diamond wrote that for New Guinean tribesmen, “to venture out of one’s territory to meet [other] humans, even if they lived only a few miles away, was equivalent to suicide” (Diamond 2006, p. 229).
116
Advances in Experimental Moral Psychology
Importantly, this argument does not assume that all or even most of the individuals or groups evoking disgust and rejection actually pose a risk of infection. But because risks of failing to detect a contagious individual (serious illness and possibly premature death) greatly outweighed the cost of wrongly identifying a harmless individual as contagious (the foregone benefits of a positive interaction), one would expect the behavioral immune system to tend toward hypervigilance (Schaller and Duncan 2007; Shaller and Park 2011). Cues that might be associated with the risk of contamination would have become heuristics, whose mere presence would trigger disgust and rejection, but which could easily be overgeneralized.
The behavioral immune system and social attitudes Disease risk and attitudes toward the obese and disabled One prediction that follows from the behavioral immune system account is that heightened perceptions of disease risk—either chronic (i.e., dispositional) or situational—should be associated with more negative attitudes toward individuals (heuristically) associated with pathogen threat. This appears to be the case—people who are especially worried about contagious disease (as measured by a subscale of the Perceived Vulnerability to Disease scale; Faulkner et al. 2004; Park et al. 2003) are also more likely to show negative attitudes toward obese people (Park et al. 2007), and people who read bogus news articles about contagious diseases showed more negative associations with physically disabled people (as measured by the Implicit Association Test; Greenwald et al. 1998) than did those who read news articles about other health topics (Park et al. 2003). Of course, neither the obese nor the disabled are likely to actually pose a disease risk, but a perceptual system that responds to significant anomalies in appearance would likely be triggered by these individuals.
Disgust and attitudes toward homosexuals A number of researchers have found that disgust tends to be related to harsher attitudes toward gay people. Dasgupta et al. (2009) and Inbar et al. (2012) found
Pollution and Purity in Moral and Political Judgment
117
that induced disgust led to more negative implicit and explicit evaluations of gay men, respectively. Inbar et al. (2009) found that dispositional sensitivity to disgust was associated with more negative implicit evaluations of gay people, and Terrizzi et al. (2010) found a relationship between disgust sensitivity and explicit evaluations of gay people.
Disease risk and attitudes toward foreigners Concern about contagious diseases is also associated with negativity toward foreign outgroups, especially unfamiliar ones. For instance, in one study participants who were shown a slideshow highlighting disease and pathogen threats were more inclined (compared to a control group who were shown a slideshow about non-disease threats) to prefer familiar (e.g., Polish) over unfamiliar (e.g., Mongolian) immigrant groups (Faulkner et al. 2004). This claim also finds support from the finding that women in their first trimester of pregnancy (during which immune function is suppressed) are more ethnocentric and xenophobic than women in their second and third trimesters (Navarette et al. 2007).
Other sociopolitical attitudes There is also evidence that differences in the strength of the behavioral immune system are related to sociopolitical attitudes more broadly. Individuals who feel more vulnerable to disease consistently provide more conservative responses on a variety of measures tapping social conservatism (Terrizzi et al. 2013), such as right-wing authoritarianism (Altemeyer 1988), social dominance orientation (Pratto et al. 1994), and vertical collectivism (Singelis et al. 1995). Likewise, at the level of group differences, geographic variation in parasite and pathogen prevalence has been found to be associated with variation in the strength of conservative social attitudes in particular cultures. Across 71 world regions, greater historic disease prevalence is associated with more restricted (i.e., conservative) sexual attitudes and lower openness to experience (Schaller and Murray 2008); and across countries and US states, current disease prevalence is associated with greater religiosity and stronger family ties (Fincher and Thornhill 2012). Like the intergroup attitudes described above, these attitudes,
118
Advances in Experimental Moral Psychology
personality differences, and social preferences all entail greater separation between groups, less experimentation with novel cultural and sexual practices, and less contact with strangers. Although such attitudes clearly have costs (e.g., reduced opportunities for trade and slower adoption of potentially useful cultural innovations), they also have benefits, especially in environments where pathogen threat is high. Less contact with outgroups, lower mobility, and conservation of existing social practices (especially food- and sex-related) minimizes exposure to novel, potentially dangerous pathogens.
The behavioral immune system and moral judgment Disgust is the emotion most closely linked to the behavioral immune system, in that it motivates individuals to distance themselves from people or groups seen (implicitly or explicitly) as contaminated or contagious (Oaten et al. 2009). Is it possible that disgust is implicated in moral judgment for similar reasons—that is, because it arises as a reaction to perceived physical contagion threats? The most common disgust-eliciting contagion threats involve sex, food, and outgroups (Oaten et al. 2009). If disgust is involved in moral judgment primarily for violations having to do with contagion threats, moral disgust should largely be limited to these specific domains. This prediction comes close to the view endorsed by Haidt and Graham (2007) in their description of the moral domain of purity/sanctity. They write that moral disgust is “attached at a minimum to those whose appearance (deformity, obesity, or diseased state), or occupation (the lowest castes in castebased societies are usually involved in disposing of excrement or corpses) makes people feel queasy” (p. 106). Certainly, on the basis of the behavioral immune system literature one would expect avoidance of these groups. However, Haidt and Graham expand their argument, proposing that the moral domain of purity/sanctity includes a metaphorical conception of impurity as well, such that disgust (and judgments of immorality) is also evoked by “those who seem ruled by carnal passions” such as “lust, gluttony, greed, and anger” (p. 106). But how much empirical evidence is there for this more extended, metaphorical role for disgust in moral judgment? In the next section, we examine the research bearing on this question.
Pollution and Purity in Moral and Political Judgment
119
Which moral violations elicit disgust? A number of studies have examined people’s reactions to moral violations, often by asking them to read about immoral or morally neutral actions and asking them to report their emotional and cognitive evaluations. Results have reliably shown a link between certain types of violations and disgust reactions.
Sex Many of the studies showing disgust at moral violations have asked participants to evaluate sexual practices, including homosexuality, incest, and unusual forms of masturbation. Haidt and Hersch (2001), for example, asked liberal and conservative undergraduates to evaluate examples of gay and lesbian sex, unusual masturbation (e.g., a woman who masturbates while holding her favorite teddy bear), and consensual sibling incest. Haidt et al. (1993) did not directly measure disgust responses, but two of the three behaviors that they expected a priori to elicit disgust involved sex (having sex with a dead chicken and then consuming it, and consensual sibling incest). Perhaps the most commonly studied moral violation of this class has been incest—an act known to elicit disgust reliably. For instance, Rozin et al. (1994) asked participants about their responses to incest in general, Royzman et al. (2008) asked participants to evaluate parentchild incest, Gutierrez and Giner-Sorolla (2007) asked about sibling incest, and Horberg et al. (2009) used the same sibling incest vignette originally used by Haidt et al., along with the chicken sex vignette from the same source.
Repugnant foods Consumption of repugnant foods has been another commonly studied type of moral violation that appears to reliably elicit disgust. For instance, both Haidt et al. (1993) and Russell and Giner-Sorolla (2011) used a scenario in which a family ate their deceased dog. Similarly, Gutierrez and Giner-Sorolla (2007), and Russell and Giner-Sorolla (2011) presented participants with a scenario in which a scientist grew and consumed a steak made of human muscle cells.
Other moral violations Researchers have also uncovered a few moral violations that do not involve sex or food, but that nonetheless appear to elicit disgust (for a recent review, see
120
Advances in Experimental Moral Psychology
Chapman and Anderson 2013). In one notable example, Chapman et al. (2009) examined reactions to people who made unfair offers in an “ultimatum game.” This economic game involves two parties: a “proposer” and a “responder.” The proposer suggests a division of a sum (in the current study, $10) between the two, and the responder can either accept this suggestion or reject it (in which case neither party receives anything). In this study, the proposer was (unbeknownst to the participants) a computer program that sometimes made very unfair offers (i.e., $9 for the proposer and $1 for the responder). Both participants’ self-reports and their facial expressions showed that they felt disgusted by very unfair offers—and the more disgusted they were, the more likely they were to reject the offer. Similarly, when people read about unfairness (e.g., someone cheating at cards), they showed increased activation in a facial muscle (the levator) involved in the expression of disgust (Cannon et al. 2011). Other studies sometimes cited as showing that disgust can occur as a response to general moral violations are harder to interpret. Some neuroimaging studies have demonstrated overlapping regions of neural activation (as measured by fMRI) for physically disgusting acts and acts of moral “indignation” (Moll et al. 2005). However, the stimuli used in the study to evoke moral indignation often contained basic, physical elicitors of disgust (e.g., “You took your mother out to dinner. At the restaurant, she saw a dead cockroach floating on the soap pan.”). The overlapping brain regions found when participants read the “indignation” statements and the “pure disgust” statements (e.g., “One night you were walking on a street. You saw a cat eating its own excrement”) could therefore be due to the fact that both statement types contain powerful elicitors of basic disgust. One study has found that people report feeling disgust in response to pictures that depict violations such as ethnic cleansing or child abuse (but do not show physical disgust elicitors; Simpson, Carter et al. 2006). However, self-reported disgust in this study was highly correlated with self-reported anger, leaving open the possibility that participants were using the term in a metaphorical rather than literal sense (see Nabi 2002). Similarly, young children agree that moral violations such as “being very mean to someone” can be described as “disgusting,” and that a disgust face “can go with” these violations (Danovitch and Bloom 2009). However, in these studies other
Pollution and Purity in Moral and Political Judgment
121
negative emotion words and faces were not possible responses, leaving open the possibility that children simply endorsed the one negatively valenced emotion available to them.
Summary Most disgusting moral violations involve unusual sex or foodstuffs (or, in the case of the “chicken sex” vignette, both). This is what one would expect if disgust-evoking moral violations activated the behavioral immune system and negative evaluations of these acts were driven by avoidance in the same way that behavioral immune system-relevant intergroup and political attitudes are. The pattern of data is also compatible with the first part of the view advanced by Haidt and Graham (2007)—that disgust functions as the guardian of physical purity. However, empirical support for the second half of their view—that violations of spiritual purity also evoke disgust—is lacking. Furthermore, some findings are explained poorly by both accounts— namely, that unfair or selfish behavior also evokes disgust, at least under some circumstances. Such behavior is neither straightforwardly related to pathogen threats, nor to physical or spiritual purity. Of course, these findings are from only two studies, and further research is necessary to determine the robustness and generality of the relationship between witnessing unfairness or selfishness and disgust. One (admittedly speculative) possibility is that cheaters and nonreciprocators are seen as an outgroup that evokes a distancing motivation in the same way that groups seen as unfamiliar or unclean do.
Induced disgust and harsher moral judgment A number of studies have experimentally induced disgust (e.g., using bad smells, dirty surroundings, or disgusting film clips), and examined the effects of this extraneously induced disgust on people’s moral judgments. In the terminology used by Pizarro, Inbar, and Helion (2011), these studies have been used to test the “disgust as amplifier” and/or “disgust as moralizer” hypotheses. Generally, these studies have found that incidental disgust makes moral judgments harsher for a wide range of infractions, including incest, eating one’s dog, bribery, stealing library books, falsifying one’s resume, and
122
Advances in Experimental Moral Psychology
masturbating with a kitten (Schnall et al. 2008; Wheatley and Haidt 2005). Schnall et al. examined whether the type of moral infraction (“purity-violating,” e.g., dog-eating or sex between first cousins, vs. “non-purity-violating,” e.g., falsifying one’s resume) moderated the effects of induced disgust on moral judgment and found that it did not. However, Horberg et al. (2009) found that inducing disgust (as opposed to sadness) had a stronger amplification effect on judgments of “purity violations” (such as sexual promiscuity) than “harm/care violations” (such as kicking a dog). Thus, there is conflicting evidence on whether inducing disgust selectively affects certain kinds of moral judgments. However, studies that demonstrate the effects of experimental inductions of disgust on moral evaluation do not serve as evidence that disgust is naturally elicited by moral violations. An analogy to research on the effects of emotion on judgment is useful here. Research has shown that extraneously manipulating emotions such as fear, sadness, anger, or even disgust can affect a wide range of judgments and decisions (Loewenstein and Lerner 2003). But that does not mean that these emotions naturally arise when making these judgments. No one would conclude that because disgust makes one more willing to sell an item that one has been given (Lerner et al. 2004), disgust therefore also arises naturally when one is deciding whether to sell or keep an item. Similarly, showing that disgust affects judgments of certain moral violations is not informative about whether disgust is a naturally occurring response to witnessing such violations.
The effects of cleanliness on moral and political judgment If moral and political judgments are motivated at least partly by the threat of contamination, drawing attention to this threat by asking participants to wash their hands (or perhaps even by simply exposing them to washingrelated stimuli) should have similar effects on judgment as other pathogen primes. There is some evidence for this: Helzer and Pizarro (2011) found that participants who were standing next to a hand-sanitizer dispenser described themselves as more politically conservative, and that those who had just used an
Pollution and Purity in Moral and Political Judgment
123
antiseptic hand wipe were more negative in their moral judgments of unusual sexual behaviors (e.g., consensual incest between half-siblings), but not in their judgments of putatively immoral acts that did not involve sexuality. Similarly, Zhong et al. (2010) demonstrated that hand-washing made participants more conservative (i.e., more negative) on a number of social issues related mainly to sexual morality (e.g., casual sex, pornography, and adultery). However, researchers who have adopted a more metaphorical notion of purity have made exactly the opposite prediction regarding the effects of cleanliness on moral judgment, arguing that if feeling clean is psychologically the opposite of feeling disgusted, making cleanliness salient should reduce feelings of disgust and therefore make moral judgments less harsh. There is also some evidence for this view: Priming participants with purity-related words (e.g., “pure,” “immaculate,” and “pristine”) made them marginally less harsh when judging moral violations (Schnall et al. 2008, Study 1), and asking participants to wash their hands after watching a disgusting film clip attenuated the effects of the clip on moral judgments (Schnall et al., Study 2). How to reconcile these conflicting results? First, it is likely that in Schnall et al.’s (2008) Study 2, in which all participants watched a film clip showing a man crawling into a filthy toilet, physical contamination threats were salient for all participants. When contamination is salient, hand-washing may have a palliative effect, whereas when contamination is not already salient, handwashing may instead prime pathogen concerns. However, this still leaves the results of Schnall et al.’s Study 1 unexplained. It is possible that purity-related words do not prime physical pathogen threats. Such simple cognitive primes may simply not be enough to engage a motivational system built to avoid pathogens, but may be effective in reminding individuals of other cleanlinessrelated concepts. It is also possible that this single, marginally significant result from a low-powered (total N ⫽ 40) study is anomalous. This is a question that can only be settled by future research. Putting this (possibly anomalous) result aside, the account we propose here offers a parsimonious explanation why disgust and its opposite—cleanliness— would show parallel effects on people’s moral judgments and sociopolitical attitudes. Because both disgust and hand-washing make the threat of physical contamination salient, their effects on certain kinds of moral and sociopolitical judgments should be similar. In contrast, a more metaphorical view of the role
124
Advances in Experimental Moral Psychology
of disgust in moral judgment would—as outlined above—predict that physical cleansing should make moral judgments less harsh (and, possibly, make attitudes toward sexual morality-related social issues more tolerant). This, of course, is not what the bulk of the evidence shows, although more research is needed to reconcile the conflicting findings in this area.
Disgusting but permissible actions One potential objection to the account we defend here is that there are many behaviors that are judged by most as disgusting yet morally permissible, such as picking one’s nose in private (see also Royzman et al. 2009). However, our argument does not require that all disgusting acts be seen as immoral (or, for that matter, that all immoral acts be seen as disgusting). Rather, we argue that reactions to certain moral violations (primarily those involving sex or food), certain sociomoral attitudes (primarily toward individuals seen as physically abnormal, norm-violating, or foreign), and certain political attitudes (primarily those related to sexual conservatism, reduced contact between different social groups, and hostility toward outsiders) rely on a shared motivational system; that this system evolved due to the adaptive benefits of responding to disease or contamination threats with rejection and avoidance; and that its primary motivating emotion is disgust. This account allows, but does not require, that disgust might extend to other kinds of moral violations as well (as we have described above, evidence for such extension is scarce). One way that such an extension could happen is that disgust may become “attached” to some behaviors for which there already exist non-moral proscriptive norms (e.g., smoking or eating meat; Nichols 2004). In these cases, the pairing of disgust with (or the tendency to be disgusted by) the behavior might cause it to be “pushed” into the moral domain—especially if the behavior can be construed as harmful (see Rozin 1999). Such a moralization process might be observed with longitudinal data comparing moral attitudes toward disgusting and non-disgusting behaviors that either have an existing (but non-moral) proscriptive norm and those which do not. If our account is correct, one would expect moralization over time to occur only in the disgusting behaviors for which there are already conventional norms in place.
Pollution and Purity in Moral and Political Judgment
125
Conclusion Reviewing the evidence linking moral violations and disgust shows that with a few exceptions, the moral violations that elicit disgust involve food, sex, or both. This is consistent with the view that seeing such acts as immoral and feeling disgust in response to them result from activation of the behavioral immune system, an evolved motivational system that responds to physical contamination threats. We believe that this account parsimoniously explains disgust’s connection with moral judgments, sociomoral attitudes, and political beliefs. It also suggests that the link between disgust and morality may be different from what has been assumed by many researchers. Although there is an empirical connection between disgust and seeing a variety of acts as immoral, this may be due to the specific content of the acts in question rather than to a more general relationship between disgust and judgments of immorality. A great deal of research points to a reliable connection between disgust and acts, individuals, or groups that are threatening because of the potential for physical contamination, whereas there is as yet little evidence that disgust is a reaction to immoral behaviors per se.
Note * Author’s Note: Yoel Inbar, Tilburg University, and David Pizarro, Cornell University. Corresponding Author: Yoel Inbar, Department of Social Psychology, Tilburg University, Email: [email protected].
References Altemeyer, R. A. (1998). The other “authoritarian personality.” In M. P. Zanna (ed.), Advances in Experimental Social Psychology (Vol. 30). New York: Academic Press, pp. 47–91. Bloom, H. (1997). The Lucifer Principle: A Scientific Expedition into the Forces of History. New York: Atlantic Monthly Press. Cannon, P. R., Schnall, S., and White, M. (2011). Transgressions and expressions: Affective facial muscle activity predicts moral judgments. Social Psychological and Personality Science, 2, 325–31.
126
Advances in Experimental Moral Psychology
Chapman, H. A., and Anderson, A. K. (2013). Things rank and gross in nature: A review and synthesis of moral disgust. Psychological Bulletin, 139, 300–27. Chapman, H. A., Kim, D. A., Susskind, J. M., and Anderson, A. K. (2009). In bad taste: Evidence for the oral origins of moral disgust. Science, 323, 1222–6. Danovitch, J., and Bloom, P. (2009). Children’s extension of disgust to physical and moral events. Emotion, 9, 107–12. Dasgupta, N., DeSteno, D. A., Williams, L., and Hunsinger, M. (2009). Fanning the flames of prejudice: The influence of specific incidental emotions on implicit prejudice. Emotion, 9, 585–91. Diamond, J. M. (2006). The Third Chimpanzee: The Evolution and Future of the Human Animal. New York: Harper Perennial. Faulkner, J., Schaller, M., Park, J. H., and Duncan, L. A. (2004). Evolved diseaseavoidance mechanisms and contemporary xenophobic attitudes. Group Processes and Intergroup Behavior, 7, 333–53. Fincher, C. L., and Thornhill, R. (2012). Parasite-stress promotes in-group assortative sociality: The cases of strong family ties and heightened religiosity. Behavioral and Brain Sciences, 35, 61–79. Graham, J., Haidt, J., and Nosek, B. (2009). Liberals and conservatives use different sets of moral foundations. Journal of Personality and Social Psychology, 96, 1029–46. Greenwald, A. G., McGhee, D. E., and Schwartz, J. L. K. (1998). Measuring individual differences in implicit cognition: The implicit association test. Journal of Personality and Social Psychology, 74, 1464–80. Gutierrez, R., and Giner-Sorolla, R. S. (2007). Anger, disgust, and presumption of harm as reactions to taboo-breaking behaviors. Emotion, 7, 853–68. Haidt, J. (2001). The emotional dog and its rational tail: A social intuitionist approach to moral judgment. Psychological Review, 108, 814–34. Haidt, J., and Graham, J. (2007). When morality opposes justice: Conservatives have moral intuitions that liberals may not recognize. Social Justice Research, 20, 98–116. Haidt, J., and Hersh, M. (2001). Sexual morality: The cultures and emotions of conservatives and liberals. Journal of Applied Social Psychology, 31, 191–221. Helzer, E. G., and Pizarro, D. A. (2011). Dirty liberals! Reminders of physical cleanliness influence moral and political attitudes. Psychological Science, 22, 517–22. Horberg, E. J., Oveis, C., Keltner, D., and Cohen, A. B. (2009). Disgust and the moralization of purity. Journal of Personality and Social Psychology, 97, 963–76. Inbar, Y., Pizarro, D., Knobe, J., and Bloom, P. (2009). Disgust sensitivity predicts intuitive disapproval of gays. Emotion, 9, 435–9. Inbar, Y., Pizarro, D. A., and Bloom, P. (2009). Conservatives are more easily disgusted. Cognition & Emotion, 23, 714–25.
Pollution and Purity in Moral and Political Judgment
127
Kelly, D. (2011). Yuck! The Nature and Moral Significance of Disgust. Cambridge, MA: The MIT Press. Kohlberg, L. (1969). Stage and sequence: The cognitive-developmental approach to socialization. In D. A. Goslin (ed.), Handbook of Socialization Theory and Research. Chicago, IL: Rand McNally, pp. 347–480. Koleva, S. P., Graham, J., Iyer, R., Ditto, P. H., and Haidt, J. (2012). Tracing the threads: How five moral concerns (especially Purity) help explain culture war attitudes. Journal of Research in Personality, 46, 184–94. Lerner, J. S., Small, D. A., and Loewenstein, G. (2004). Heart strings and purse strings carryover effects of emotions on economic decisions. Psychological Science, 15, 337–41. Loewenstein, G., and Lerner, J. S. (2003). The role of affect in decision making. In R. J. Davidson, K. R. Scherer, and H. H. Goldsmith (eds), Handbook of Affective Science. New York: Oxford University Press, pp. 619–42. Moll, J., de Oliveira-Souza, R., Moll, F. T., Ignácio, F. A., Bramati, I. E., CaparelliDáquer, E. M., and Eslinger, P. J. (2005). The moral affiliations of disgust: A functional MRI study. Cognitive and Behavioral Neurology, 18, 68–78. Nabi, R. L. (2002). The theoretical versus the lay meaning of disgust: Implications for emotion research. Cognition and Emotion, 16, 695–703. Navarrete, C. D., Fessler, D. M. T., and Eng, S. J. (2007). Elevated ethnocentrism in the first trimester of pregnancy. Evolution and Human Behavior, 28, 60–5. Nichols, S. (2004). Sentimental Rules: On the Natural Foundations of Moral Judgment. New York: Oxford University Press. Oaten, M., Stevenson, R. J., and Case, T. I. (2009). Disgust as a disease-avoidance mechanism. Psychological Bulletin, 135, 303–21. Park, J. H., Faulkner, J., and Schaller, M. (2003). Evolved disease-avoidance processes and contemporary anti-social behavior: Prejudicial attitudes and avoidance of people with physical disabilities. Journal of Nonverbal Behavior, 27, 65–87. Park, J. H., Schaller, M., and Crandall, C. S. (2007). Pathogen-avoidance mechanisms and the stigmatization of obese people. Evolution and Human Behavior, 28, 410–14. Pizarro, D. A., Inbar, Y., and Helion, C. (2011). On disgust and moral judgment. Emotion Review, 3, 267–8. Pratto, F., Sidanius, J., Stallworth, L. M., and Malle, B. F. (1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67, 741–63. Royzman, E. B., Leeman, R. F., and Baron, J. (2009). Unsentimental ethics: Towards a content-specific account of the moral–conventional distinction. Cognition, 112, 159–74.
128
Advances in Experimental Moral Psychology
Royzman, E. B., Leeman, R. F., and Sabini, J. (2008). ‘‘You make me sick”: Moral dyspepsia as a reaction to third-party sibling incest. Motivation and Emotion, 32, 100–8. Rozin, P. (1999). The process of moralization. Psychological Science, 10, 218–21. Rozin, P., Haidt, J., and McCauley, C. R. (2008). Disgust. In M. Lewis, J. M. HavilandJones, and L. F. Barrett (eds), Handbook of Emotions (3rd ed.). New York: Guilford, pp. 757–76. Rozin, P., Lowery, L., Imada, S., and Haidt, J. (1999). The moral-emotion triad hypothesis: A mapping between three moral emotions (contempt, anger, disgust) and three moral ethics (community, autonomy, divinity). Journal of Personality and Social Psychology, 76, 574–86. Russell, P. S., and Giner-Sorolla, R. (2011). Moral anger, but not moral disgust, responds to intentionality. Emotion, 11, 233–40. Schaller, M., and Duncan, L. A. (2007). The behavioral immune system: Its evolution and social psychological implications. In J. P. Forgas, M. G. Haselton, and W. von Hippel (eds), Evolution and the Social Mind: Evolutionary Psychology and Social Cognition. New York: Psychology Press, pp. 293–307. Schaller, M., and Murray, D. R. (2008). Pathogens, personality, and culture: Disease prevalence predicts worldwide variability in sociosexuality, extraversion, and openness to experience. Journal of Personality and Social Psychology, 95, 212–21. Schaller, M., and Park, J. H. (2011). The behavioral immune system (and why it matters). Current Directions in Psychological Science, 20, 99–103. Schnall, S., Benton, J., and Harvey, S. (2008). With a clean conscience: Cleanliness reduces the severity of moral judgments. Psychological Science, 19, 1219–22. Schnall, S., Haidt, J., Clore, G. L., and Jordan, A. H. (2008). Disgust as embodied moral judgment. Personality and Social Psychology Bulletin, 34, 1096–109. Simpson, J., Carter, S., Anthony, S. H., and Overton, P. G. (2006). Is disgust a homogeneous emotion? Motivation and Emotion, 30, 31–41. Singelis, T. M., Triandis, H. C., Bhawuk, D. P. S., and Gelfand, M. J. (1995). Horizontal and vertical dimensions of individualism and collectivism: A theoretical and measurement refinement. Cross-Cultural Research, 29, 240–75. Staats, A. W., and Staats, C. K. (1958). Attitudes established by classical conditioning. The Journal of Abnormal and Social Psychology, 57, 37–40. Terrizzi, J. A., Shook, N. J., and McDaniel, M. A. (2013). The behavioral immune system and social conservatism: A meta-analysis. Evolution and Human Behavior, 34, 99–108. Terrizzi, J. A., Shook, N. J., and Ventis, W. L. (2010). Disgust: A predictor of social conservatism and prejudicial attitudes toward homosexuals. Personality and Individual Differences, 49, 587–592.
Pollution and Purity in Moral and Political Judgment
129
Tully, T., and Quinn, W. G. (1985). Classical conditioning and retention in normal and mutant Drosophila melanogaster. Journal of Comparative Physiology A, 157(2), 263–77. Tybur, J. M., Lieberman, D., Kurzban, R., and DeScioli, P. (2012). Disgust: Evolved function and structure. Psychological Review, 120, 65–84. Wheatley, T., and Haidt, J. (2005). Hypnotic disgust makes moral judgments more severe. Psychological Science, 16, 780–4. Zhong, C. B., Strejcek, B., and Sivanathan, N. (2010). A clean self can render harsh moral judgment. Journal of Experimental Social Psychology, 46, 859–62.
7
Selective Debunking Arguments, Folk Psychology, and Empirical Moral Psychology Daniel Kelly*
Some framing questions Reflecting on the significance of his early research on the neuropsychology of moral judgment, Joshua Greene (2007) raises an important and increasingly pressing kind of question: “Where does one draw the line between correcting the nearsightedness of human moral nature and obliterating it completely?” and goes on to more directly wonder “How far can the empirical debunking of human moral nature go?” (p. 76). The gist of such questions can be fleshed out in several ways; I attempt to distinguish different approaches in the latter half of this chapter, and situate my own in the resulting landscape. The approach I favor foregrounds the relationship between empirical cognitive science and morality,1 in order to more crisply express certain kinds of question. For example: Are their constraints on human morality that make it inflexible or resistant to transformation in certain ways? If so, what are those constraints, what imposes them, and why do they make morality rigid in whatever way they do? Are those constraints only knowable a priori, perhaps via conceptual analysis or reflection on the essence of morality, or can cognitive science help to discover them, perhaps by revealing innate features of our moral psychology? On the other hand it could be the case that human morality is relatively unconstrained, and thus fairly malleable. Is it possible—do we have it within ourselves—to transcend the types of moral judgments that are so naturally made by minds like ours? Can cognitive science show us how to most effectively do so? One virtue of this way of framing the issues is that it invites us to consider an analogy between moral theorizing and scientific theorizing and the relationship
Selective Debunking Arguments
131
each bears to its commonsensical starting place, with an eye toward where that analogy might break down. For instance, Noam Chomsky suggests that when we are doing science, theorizing can and should transcend the folk intuitions it begins with, and that departure or movement away from the commonsense concepts in which early investigation is typically couched is relatively unrestricted. For instance, while discussing scientific inquiries into the mind, and the relationship between the categories of folk psychology and those that will be taken up by cognitive science as it proceeds, he remarks: These are serious inquiries, not to be undertaken casually; our intuitions about them provide some evidence, but nothing more than that. Furthermore, whatever may be learned about folk science will have no relevance to the pursuit of naturalistic inquiry into the topics that folk science addresses in its own way. (Chomsky 1995, p. 14)
Indeed, he even suggests that in the practice of science, leaving the vernacular behind is indicative of advance, improvement—theoretic progress: “As the disciplines progress, they depart still further from the common sense and ordinary language origins of inquiry” (1995, pp. 25–6; for more recent comments in a similar vein, see Chomsky 2009). K. Anthony Appiah appears to agree with Chomsky on this picture, at least as it applies to, say, the increasingly tenuous relationship between folk physics and contemporary physical theories. However, he takes the view that scientific theorizing is importantly different from moral theorizing on this score, that is, with respect to how tightly each is tethered to the intuitive categories of the folk. In his 2007 Presidential address to the American Philosophical Association, Appiah suggests there are, in fact, limits on the extent to which morality can be detached from common-sense psychology, or significantly transformed by advances in cognitive science. Furthermore, he suggests that the presence of such limits in the moral case, and their absence in the scientific, both stem from a difference in the roles that moral and scientific theories play in human lives, and the different kinds of connections each type of theory needs to bear to our intuitive understanding to effectively play its part: It’s common to analogize folk psychology with folk physics. But, of course, professional physicists can happily leave folk physics far behind as they tinker with their Calabi-Yau Manifolds and Gromov-Witten invariants.
132
Advances in Experimental Moral Psychology
By contrast, moral psychology, however reflective, can’t be dissociated from our moral sentiments, because it’s basic to how we make sense of one another and ourselves. In a deliberately awkward formulation of Bernard Williams’s, moral thought and experience “must primarily involve grasping the world in such a way that one can, as a particular human being, live in it.” (Appiah 2007, p. 15)
Put this way, a core question that emerges is whether morality and moral theory is special or distinctive in its relation to empirical psychology and other natural sciences—roughly, whether something about human moral nature makes it more or less debunkable than other aspects of human nature, or whether something about moral judgment makes it more or less resistant to transformation than other types of judgment. These are fascinating and timely topics; they are also difficult ones. Rather than set out an overarching view or take a stand on the debunking of morality tout court, in what follows I’ll explore a divide and conquer strategy. First, I will briefly sketch a debunking argument that, instead of targeting all of morality or human moral nature, has a more narrow focus—namely, the intuitive moral authority of disgust. The argument concludes that as vivid and compelling as they can be while one is in their grip, feelings of disgust should be granted no power to justify moral judgments. Importantly, the argument is grounded in empirical advances concerning the character of the emotion itself. Next, I will step back and consider the argument’s general form. I then point to arguments that others have made that seem to share this form and selective focus, and comment on what such arguments do and do not presuppose. Finally, I locate the selective strategy with respect to approaches to debunking morality and end by reflecting on what the entire line of thought implies about Greene’s question and Appiah’s claim.
Disgust and moral justification Consider some of the following contentious, “yuck-relevant” issues: abortion, nipple piercing, same-sex marriage, circumcision (either female or male), human cloning, stem cell research, euthanasia, pornography. Also imagine that your response to one of those activities or social practices is: “yuck!”
Selective Debunking Arguments
133
You find it simply, but unequivocally, revolting and repulsive, or you just find yourself slightly disgusted by it. What follows from that yuck reaction, from the point of view of morality? Do feelings of disgust, in and of themselves, provide good enough reason to think the practice is morally wrong or problematic? Recently, such issues have come to the fore in normative and applied ethics, centering on the question of what role the emotion of disgust should play in morality, broadly construed: whether or not disgust should influence our considered moral judgments; if so, how feelings of disgust should be accounted for in various ethical evaluations, deliberations, and decisions; what sort of weight, import, or credit should be assigned to such feelings; and how our legal system and other institutions should best deal with the emotion (see Kelly and Morar manuscript for full references). Elsewhere (Kelly 2011) I have fleshed out a debunking argument designed to undermine confidence in the normative force that feelings of disgust can seem to have in moral cognition. The resulting position, which I call disgust skepticism, holds that: feelings of disgust have no moral authority; that explicit appeals to disgust, while often rhetorically effective, are morally empty; that the emotion should not be granted any justificatory value; and that we should aspire to eliminate its influence on morality, moral deliberation, and institutional operation to the extent that we can. Rather than recapitulate the argument in full, I will here mention some of its most relevant properties. First, while the argument has a normative thrust concerning the role that feelings of disgust should place in moral justification, it is firmly rooted in a descriptive and explanatory account of the nature of the emotion itself. It is worth noting that my argument shares this structural feature with arguments that others have made concerning the moral significance of disgust. All interested parties, both skeptics (Nussbaum 2004a, 2004b) and advocates (Kass 1997, 2002; Kahan 1998, 1999), base their normative conclusions on descriptive claims concerning the character of the emotion. On this score, a key advantage I claim over those competing arguments is that my account of disgust is superior to its competitors: it is more detailed, more evolutionarily plausible, and better able to explain the wealth of empirical data recently discovered by moral psychologists.
134
Advances in Experimental Moral Psychology
The two core claims of what I call the E&C view are the Entanglement thesis and the Co-opt thesis. The first holds that at the heart of the psychological disgust system are two distinguishable but functionally integrated mechanisms, one that initially evolved to protect the gastrointestinal system from poisons and other harmful food, and another that initially evolved to protect the entire organism from infectious diseases and other forms of parasites. Appeal to the operation of these two mechanisms and their associated adaptive problems can explain much of the fine-grained structure of the disgust response, its intrinsic sensitivity to perceivable cues associated with poisons and parasites, its propensity to err in the direction of false positives (rather than false negatives), and its malleability and responsiveness to social influence, which can result in variation in what triggers disgust from one group of people to the next. The second core claim, the Co-opt thesis, holds that this malleability and responsiveness to social influence was exploited by evolution, as disgust was recruited to perform auxiliary functions having nothing to do with poisons or parasites, infusing certain social norms and group boundaries with a disgustbased emotional valence. In doing so, disgust did not lose its primary functions or those properties clearly selected to allow it to perform those functions well. Rather, it retained those functions and properties, and simply brought them to bear on the auxiliary functions associated with norms and group membership (Kelly 2011, 2013). In addition to these features, the argument in favor of disgust skepticism appeals to other facts about the emotion and key elements of the picture provided by the E&C view. One is that disgust has an intrinsic negative valence, which can manifest subjectively as a kind of nonverbal authority. Intense episodes of disgust obviously have a powerful and vivid phenomenology, but even less flagrant instances can bias judgments that they influence toward negativity and harshness. However, the mere activation of disgust, in and of itself, is not even a vaguely reliable indicator of moral wrongness. The emotion remains overly sensitive to cues related to its primary functions of protecting against poisons and parasite, which results in many false positives even in those domains. There is no reason to think the situation improves when disgust operates in the sociomoral domain. Indeed, there is reason to think that disgust renders those in its grip less sensitive to the agency and intentions of others, and can make it easier to dehumanize them.
Selective Debunking Arguments
135
Moreover, triggers of disgust exhibit considerable variation from person to person and from culture to culture. This variation is found in types of cuisine that are considered edible or disgusting, but more importantly in the types of norms with which disgust becomes involved, as well as the group boundaries and markers to which it is sensitive (also see Henrich et al. 2010). Hence, when there is disagreement about the moral status of a norm, the practice it regulates, or the type of people who engage in that practice, the fact that disputants on one side of the debate denounce and feel disgust at the norm or practice, while the disputants on the other side of the debate feel no disgust and see nothing wrong with the norm or practice, may be an interesting psychological fact. But it is a psychological fact that holds no significance for the question of who is correct, or whose assessment of the moral status of the norm or practice is better justified. It is worth nothing that there will be an evolutionary story to tell about many, if not most, of the psychological mechanisms that loom large in human moral psychology. I do not hold that every evolutionary explanation is intrinsically debunking, or that the mere existence of an evolutionary account of some psychological mechanism should by itself throw suspicion on it, or lead us to doubt that it has any role to play in moral justification. However, I do hold that debunking strategies can be more selective, and that the details of the specific evolutionary story provided by the E&C view should undermine confidence in the moral significance of feelings of disgust. For, the E&C view renders most properties of the disgust system understandable, and it also allows us to see that some of the properties that are virtues when disgust is performing its primary functions become vices when disgust performs the social and morally oriented auxiliary functions. A good example is provided by the automatically activated concerns about contamination: they straightforwardly help avoid contagious diseases, but they are mismatched to the social domain, where they drive irrational worries about moral taint and spiritual pollution. The distinction between primary and auxiliary functions provided by the evolutionarily informed E&C view shows that aspects of disgust that are features in one domain are bugs in another. Hence my skepticism about the value of disgust to specifically moral justification.2 I take it that the inference “it’s disgusting, therefore it’s immoral” has prima facie intuitive force for many people, but whether or not the inference is a
136
Advances in Experimental Moral Psychology
component of folk morality and human moral nature is ultimately an empirical one. Initial evidence suggests that the inference is common among the folk, at least in some cultures, and for some segments of the population (Haidt et al. 1993; Haidt et al. 1997; Nichols 2002, 2004; c.f. Cova and Ravat 2008). Also note that my argument for disgust skepticism is designed to show that whatever the empirical facts about who tends to make that inference, or how compelling they find it, it is a bad one. It should not be accepted by anyone, and those who make it are making a mistake (c.f. Haidt 2012). Of course, to say that this kind of argument is unsound is not to say that its conclusion will always be false, or that moral judgments accompanied by disgust are never justified. Rather, some disgust involving judgments may be justified while others are not. My claim is that no moral judgments are justified by disgust; the involvement of this emotion in a judgment is just irrelevant to whether and how the judgment is justified.3
The shape of the argument: Selective debunking Now that my argument against the normative value of the yuck factor has been sketched, recall the framing questions posed at the beginning of the chapter about the relationship between morality, on the one hand, and a cognitive scientific understanding of the mind that may depart from intuition and folk psychology as it increases in sophistication, on the other. The issue is not always approached this way. Many conversations have explored related but different questions, and they have typically done so at a higher level of generality: morality and all moral judgments (or claims) are grouped together, and arguments are made that they are either all vulnerable to some sweeping form of debunking, or none of them are (Mackie 1977; Blackburn 1988; Joyce 2007; c.f. Ayer 1936; also see Street 2006; Greene 2013).4 While I have doubts about the viability of this kind of global debunking, I have just advanced what can be thought of as a selective debunking argument against the relevance of one circumscribed set of considerations, namely feelings of disgust, to moral justification.5 Here I will spell out the line of reasoning, first expressing it in condensed form before going on to elaborate and comment on different aspects of the premises and conclusion.
Selective Debunking Arguments
137
The first premise of my main argument can be expressed in the form of a conditional: 1. If some particular psychological mechanism can be shown to be problematic in a relevant way, and the intuitions or judgments influenced by that psychological mechanism can be identified, then we should disregard, discount, or discredit those intuitions and be suspicious of the judgments that they influence, to the extent that we can. Then the form of that argument can be understood as a modus ponens: 1. If some particular psychological mechanism can be shown to be problematic in a relevant way, and the intuitions or judgments influenced by that psychological mechanism can be identified, then we should disregard, discount, or discredit those intuitions and be suspicious of the judgments that they influence, to the extent that we can. 2. Disgust is problematic in a relevant way (the E&C view of disgust), and the intuitions and judgments influence by disgust can be identified (yuck-relevant issues). 3. Therefore, we should disregard, discount, or discredit those intuitions and be suspicious of the judgments that they influence, to the extent that we can (disgust skepticism). Though it is schematic, I find the general line of thought expressed in the first premise compelling, and also find this way of framing it illuminating for a number of reasons. First, formulating the conditional premise this way shows that its subject matter will be a specific psychological mechanism, but it allows the identity and details of the psychological mechanism to vary from one instantiation of the argument schema to the next. Moreover, expressing the first premise like this makes clear that it says nothing specifically about morality, let alone any particular moral theory, be it utilitarian, deontological, or otherwise (c.f. Singer 2005; Greene 2007). Nor, for that matter, does it even say anything specific about emotions or sentiments, as opposed to less affective, more coldly cognitive types of psychological mechanisms (c.f. D’Arms and Jacobson 2000, 2003). Second, the argument assumes a picture of the structure of the human mind that is now familiar in various forms from empirical work in psychology.
138
Advances in Experimental Moral Psychology
Whatever their differences in emphasis and preferred terminology, many approaches share a vision that sees the mind as comprised of many distinct, dissociable, semiautonomous psychological mechanisms, whose different operational principles and evolutionary histories can be (and are being) discovered by cognitive scientists.6 This point is also relevant to separating out and assessing the prospects of global debunking strategies as compared to selective debunking strategies. Since different psychological mechanisms may turn out to be more or less problematic, it also suggests that individual selective debunking arguments will be more or less convincing depending on the details of the particular psychological mechanism they invoke. Put another way, this picture of the mind implies that not all intuitions are created equal. In and of itself, this claim should not be anything controversial; most theorists at least tacitly accept the idea that not all intuitions are of equal value, and that in the course of theory construction some will have to be rejected or abandoned. So there is nothing revolutionary, or even very innovative, in the ruling out of some subset of intuitions. What may be novel about this line of argument is its method of identifying those intuitions that should be cast aside, and also perhaps the rationale it provides for doing so. That rationale looks to the sciences of the mind for guidance, rather than confining itself to a priori principles or general considerations of consistency and coherence.7 Different intuitions can be produced by different psychological mechanisms, and it is in virtue of this that a more sophisticated, empirically informed understanding of the mind and its component parts can reveal some intuitions to be of less value than others.8 A final reason I prefer this formulation is that it makes explicit that “problematic” is probably both the most crucial and most slippery of the notions in play. I do think that the types of considerations I have raised about disgust show it is indeed “problematic” in a relevant way. At this point, though, I do not know how to unpack that turn of phrase and remove the scare quotes. Indeed, I suspect no general recipe will be forthcoming, and that instead, each psychological mechanism and attempted selective debunk will need to be assessed on a case-by-case basis, and according to its own unique details. That said, I do think that there are some clear, perhaps paradigmatic examples that can be pointed to, in which mechanisms have been revealed as “problematic.” For example, the psychological mechanisms that underlie vision are problematic in certain circumstances (or in selective domains) because they
Selective Debunking Arguments
139
are notoriously susceptible to certain kinds of perceptual illusions. Even though the two lines in the Muller-Lyer illusion seem (to many of us) like they are the same length, we should disregard that impression, despite whatever intuitive grip it might have on us. Another example is provided by Gil Harman in his discussion of “folk physics” intuitions: Ordinary untrained physical intuitions are often in error. For example, ordinary people expect that something dropped from a moving vehicle or airplane will fall straight down to the point on earth directly underneath the place from which it was released. In fact, the dropped object will fall in a parabolic arc in the direction of the movement of the vehicle or airplane from which it was dropped. This means, among other things, that bombardiers need to be trained to go against their own physical intuitions. (Harman 1999, p. 315)
Indeed, Harman uses this as an example to soften up his reader for the main claim he makes in the paper, which is, roughly, that folk psychology consistently makes a fundamental attribution error about the determinants of people’s behavior, and that virtue ethical theories that seem to enshrine that error in the character trait-based moral psychology they advance are flawed on empirical grounds.9 Cast in my terminology, Harman is offering a debunking argument that selectively targets a specific component of folk psychology (rather than the kind of global attack on the whole conceptual framework associated with, e.g., Churchland 1981). Harman even offers an account of the psychological mechanisms that drive the fundamental attribution error, and uses it to advance his argument against those select intuitions that lead us10 to overestimate the extent to which people’s behavior is driven by internal character traits, and overlook the strong (and empirically documented) influence of external cues and situational factors. Similarly, current empirical work has shown how the psychological mechanisms underlying racial cognition can lead people to naturally, intuitively ascribe some deep and evaluatively laden racial “essence” to individuals based on their observable phenotypic characteristics like skin color or hair type. Such discoveries about the operational principles and evolutionary history of those psychological mechanisms look to be important to contemporary discussions about the nature of race itself, but also the pragmatics of racial classification. A society might decide, in light of its considered goals about
140
Advances in Experimental Moral Psychology
how to deal with racial categories and biases, and also in light of the mounting facts (genetic, biological, social, historical, etc.) about race and the source of racial differences, that its members should aspire to overcome the influence of the psychological mechanisms underlying racial cognition, and disregard the intuitions that issue from them. Indeed, empirical work on the character of those psychological mechanisms will likely point the way to the most effective methods of controlling their influence (see Kelly et al. 2010a, 2010b for discussion).11 The upshot of these examples is that arguments that have a form similar to the one I have made about disgust are not uncommon. However, there does not appear to be a single monolith or univocal notion of “problematic” that they all have in common, suggesting that there is a variety of ways in which psychological mechanisms and the intuitions that issue from them can be found to be problematic. As nice as it would be to have a single, all-purpose, or universally applicable criterion to apply to every psychological mechanism, no such clean, algorithmic test is yet in the offing, and may never be. This does not render the general argumentative strategy specious, though. Rather, it pushes us to look at and assess each instance of the argument type on a case-by-case basis, and tend to the details of the individual psychological mechanisms to which it appeals.12
Conclusion One might find reason for optimism in the themes of malleability and variation that run throughout some of the above examples, including my main example of disgust. Perhaps psychological mechanisms that are problematic in some people are unproblematic in others, suggesting that such mechanisms are plastic enough to be “fixable.” This is an interesting possibility, to be sure. However, it leaves untouched the question of what being fixed amounts to, and which mechanisms are “properly tuned” and which are not. One way to understand my point about justification is to say that in cases of disagreement about this kind of issue, members on one side of the debate cannot appeal to their own calibrated psychological mechanisms or the intuitions that issue from them to justify their position without begging the very question being raised. Even once
Selective Debunking Arguments
141
(or if) the issue of what proper tuning amounts to is settled, the argument still goes through for those improperly tuned mechanisms, and I maintain that we should continue to be on guard against their influence on judgment and action. Finally, it is also likely that different psychological mechanisms will be malleable to different extents, and in different ways. This provides more support for the divide-and-conquer strategy I advocate. Together, I think these considerations raise problems for familiar globally oriented approaches that seek to draw more encompassing conclusions in one fell swoop. I began this chapter with a passage from K. Anthony Appiah suggesting that human moral nature, morality, and moral psychology will be resistant to transformative influences originating in advances in the sciences of the mind, and with some questions raised by Joshua Greene about how far the empirical debunking of human moral nature can go. I will end not by addressing these head on, but by pointing out that in asking questions and making claims about morality as a single phenomenon and moral psychology as a uniform whole, they rely on an assumption that I think we have good reason to doubt. Rather, the argument of this chapter shows that advances in cognitive science can indeed have a transformative effect on how we think about selective aspects of morality, and how we should make sense of ourselves and some of our own moral impulses. Perhaps more importantly, the empirical work is also revealing how a more piecemeal approach is required if we are to draw any defensible normative conclusions from it. The need for a more selective focus opens up new ways to think about whether and which components of morality might be debunked by, transformed by, or even just informed and guided by our growing empirical understanding of our own moral psychology.
Notes * Author’s Note: Daniel Kelly, Department of Philosophy, Purdue University. Correspondence should be addressed to Daniel Kelly, 7126 Beering Hall, 100 N. University, West Lafayette, IN 47906. Email: [email protected]. I would like to thank Jen Cole Wright and Hagop Sarkissian for useful feedback on this chapter. 1 A similar concern animates much of Daniel Dennett’s early work on the relationship between cognitive science and propositional attitude psychology as well (see especially 1978, 1987).
142
Advances in Experimental Moral Psychology
2 In arguing that the role of disgust in the moral domain should be minimized, I realize that I am recommending that we should refrain from using what could be a useful heuristic and powerful motivation tool. However, given the risks attached to this particular emotion, namely its hair trigger sensitivity to cues that are prima facie irrelevant to morality and its susceptibility to false positives, together with its propensity to dehumanize its object, I think the costs outweigh the benefits. 3 One might imagine an individual with a “perfectly tuned” sense of disgust, whose psychological makeup is such that she feels revulsion at all and only those norms, actions, and practices that are genuinely morally wrong. My position is not undermined by this possibility. Even though, ex hypothesi, all of her judgments about those norms, action, and practices are justified, it remains open for me to claim that it is not the attendant feelings of disgust she feels that justify her judgments. Rather, the ultimate arbiter of justification is something else, above and beyond the mere presence of feelings of disgust, namely whatever standard is being appealed to in claiming that her sense of disgust is “perfectly tuned.” 4 I am particularly skeptical of the prospects of empirically motivated debunking of the entirety of morality or all moral judgments because (among other reasons) it remains unclear how to delimit the scope of such arguments. Separating the domain of morality and moral cognition off from the rest of non-moral or extra-moral cognition—identifying what moral judgments have in common that makes them moral judgments—has proven surprisingly difficult. Certainly, no consensus has emerged among practitioners in the growing field of empirical moral psychology. See Nado et al. 2009, Machery and Mallon 2010, Parkinson et al. 2011, Sinnott-Armstrong and Wheatley 2012. 5 The terminology “selective debunking” is taken from a series of thought-provoking posts on the topic by Tamler Sommers at The Splintered Mind blog (http:// schwitzsplinters.blogspot.com/2009/05/on-debunking-part-deux-selective.html). 6 I mean to cast my net widely with the first premise, but recognize that the details and preferred jargon used to discuss the distinguishable psychological mechanisms vary in different literatures. For instance, see Fodor (1983, 2000), Pinker (1997), and Carruthers (2006) for discussion in terms of different psychological modules; Evans (2003), Stanovich (2005), and Frankish (2010) for discussion in terms of dual process theory, and Ekman (1992) and Griffith (1997) for discussion of affect programs and basic emotions.
Selective Debunking Arguments
143
7 See Rawls (1971) on the method of reflective equilibrium and also David Lewis’s methodological contention that “to the victor go the spoils” (Lewis 1973). For some interesting recent discussion on the later, see (Eddon 2011; Ichikawa 2011). 8 This suggestion is very much in the spirit of some comments in Tim Maudlin’s book The Metaphysics in Physics: “if we care about intuitions at all, we ought to care about the underlying mechanism that generates them” (Maudlin 2010, pp. 146–7). In the main text, I am working with a picture similar to that implied by Maudlin’s comment, namely that one of the things psychological mechanisms that comprise the disgust system do is generate an intuition, namely the intuition that whatever triggered the system (or whatever the person thinks triggered the system, in cases of misattribution) is disgusting. 9 Also see Doris (2002) for a book-length defense of what has become known as the situationist critique of virtue ethics, and Alfano (2013) for a discussion of the current state of the debate. 10 That can lead those of us in Western cultures to commit the error, anyway. Members of Eastern Asian cultures are less prone to the mistake, suggesting it is not a universal component of folk psychology (Nisbett 2003). For another discussion about cultural variability and the fundamental attribution error, this time within the context of Confucian versus Aristotelian versions of virtue ethics, see Sarkissian (2010). 11 A final illuminating comparison, and one that might feel more apt to someone sympathetic to metaethical constructivism, is suggested by considering how intuition, on the one hand, and theoretical psychological knowledge, on the other, can best inform and guide not moral judgment but artistic creation. Reflecting on his project in Sweet Anticipation: Music and the Psychology of Expectation, cognitive musicologist David Huron offers some reasonable and intriguing comments: My musical aim in this book is to provide musicians with a better understanding of some of the tools they use, not to tell musicians what goals they should pursue. If we want to expand artistic horizons and foster creativity there is no better approach than improving our understanding of how minds work. Many artists have assumed that such knowledge is unnecessary: it is intuition rather than knowledge that provides the boundaries for artistic creation. I agree that intuition is essential for artistic production: in the absence of knowledge, our only recourse is to follow our intuitions. But intuition is not the foundation for artistic freedom or creative innovation. Quite the contrary. The more
144
Advances in Experimental Moral Psychology we rely on our intuitions, the more our behaviors may be dictated by unacknowledged social norms or biological predispositions. Intuition is, and has been, indispensible in the arts. But intuition needs to be supplemented by knowledge (or luck) if artists are to break through “counterintuitive” barriers into new realms of artistic expression. (Huron 2006, pp. ix–x, italics in original)
12 Another metaethical view that bears intriguing similarities to the one suggested by the selective debunking approach endorsed here is the “patchy realism” described by Doris and Plakias (2007).
References Alfano, M. (2013). Character as Moral Fiction. Cambridge: Cambridge University Press. Ayer, A. (1936). Language, Truth, and Logic. New York: Dover Publications. Blackburn, S. (1998). Ruling Passion. London: Oxford University Press. Carruthers, P. (2006). The Architecture of the Mind. New York: Oxford University Press. Churchland, P. M. (1981). Eliminative materialism and the propositional attitudes. Journal of Philosophy, 78, 67–90. Chomsky, N. (1995). Language and Nature. Mind, 104(431), 1–61. —(2009). The Mysteries of Nature: How Deeply Hidden? The Journal of Philosophy, 106(4), 167–200. Cova, F., and Ravat, J. (2008). Sens commun et objectivisme moral: objectivisme ‘global’ ou objectivisme ‘local’? Une introduction par l’exemple à la philosophie expérimentale. (English Modified Version). Klesis - Revue Philosophique: Actualité de la Philosophie Analytique, 9, 180–202. D’Arms, J., and Jacobson, D. (2000). Sentiment and value. Ethics, 110, 722–48. —(2003). The significance of recalcitrant emotions. In A. Hatzimoysis (ed.), Philosophy and the Emotions. Cambridge: Cambridge University Press, pp. 127–46. Dennett, D. (1978). Brainstorms. Montgomery, VT: Bradford Books. —(1987). The Intentional Stance. Montgomery, VT: Bradford Books. Doris, J. (2002). Lack of Character: Personality and Moral Behavior. New York: Cambridge. Doris, J., and Plakias, A. (2007). How to argue about disagreement: Evaluative diversity and moral realism. In W. Sinnott-Armstrong (ed.), Moral Psychology, vol. 2: The Biology and Psychology of Morality. Oxford: Oxford University Press, pp. 303–32.
Selective Debunking Arguments
145
Eddon, M. (2011). Intrinsicality and hyperintensionality. Philosophy and Phenomenological Research, 82, 314–36. Ekman, P. (1992). An argument for basic emotions. Cognition and Emotion, 6, 169–200. Evans, J. (2003). In two minds: Dual-process accounts of reasoning. Trends in Cognitive Science, 7(10), 454–9. Frankish, K. (2010). Dual-process and dual-system theories of reasoning. Philosophy Compass, 5(10), 914–26. Greene, J. (2007). The secret joke of Kant’s soul. In W. Sinnott-Armstrong (ed.), Moral Psychology, Vol. 3: The Neuroscience of Morality: Emotion, Disease, and Development. Cambridge, MA: The MIT Press, pp. 35–80. —(2013). Moral Tribes: Emotion, Reason, and the Gap Between Us and Them. New York: Penguin Press. Griffiths, P. (1997). What the Emotions Really Are. Chicago: University of Chicago Press. Haidt, J. (2012). The Righteous Mind: Why Good People Are Divided by Politics and Religion. New York: Pantheon Press. Haidt, J., Koller, S., and Dias, M. (1993). Affect, culture, and morality, or is it wrong to eat your dog? Journal of Personality and Social Psychology, 65(4), 613–28. Haidt, J., Rozin, P., McCauley, C., and Imada, S. (1997). Body, psyche, and culture: The relationship between disgust and morality. Psychology and Developing Societies, 9, 107–31. Harman, G. (1999). Moral philosophy meets social psychology: Virtue ethics and the fundamental attribution error. Proceedings of the Aristotelian Society, 99, 315–31. Henrich, J., Heine, S., and Norenzayan, A. (2010). The weirdest people in the world. Behavioral and Brain Sciences, 33(June), 61–135. Huron, D. (2006). Sweet Anticipation: Music and the Psychology of Expectation. Cambridge, MA: The MIT Press. Ichikawa, J. (2011). Experimentalist pressure against traditional methodology. Philosophical Psychology, 25(5), 743–65. Joyce, R. (2007). The Evolution of Morality. Cambridge, MA: MIT Press. Lewis, D. (1973). Causation. Journal of Philosophy, 70, 556–67. Kahan, D. (1998). The anatomy of disgust in criminal law. Michigan Law Review, 96(May), 1621–57. —(1999). The progressive appropriation of disgust. In Susan Bandes (ed.), The Passions of the Law. New York: New York University Press, pp. 63–80. Kass, L. (2 June 1997). The wisdom of repugnance. The New Republic, 216(22), available online at http://www.catholiceducation.org/articles/medical_ethics/ me0006.html
146
Advances in Experimental Moral Psychology
—(2002). Life, Liberty, and the Defense of Dignity: The Challenge to Bioethics. New York: Encounter Books. Kelly, D. (2011). Yuck! The Nature and Moral Significance of Disgust. Cambridge, MA: The MIT Press. —(2013). Moral disgust and tribal instincts: A byproduct hypothesis. In R. Joyce, K. Sterelny, and B. Calcott (eds), Cooperation and Its Evolution. Cambridge, MA: The MIT Press. Kelly, D., Faucher, L., and Machery, E. (2010). Getting rid of racism: Assessing three proposals in light of psychological evidence. Journal of Social Philosophy, 41(3), 293–322. Kelly, D., Machery, E., and Mallon, R. (2010). Race and racial cognition. In J. Doris et al. (eds), The Moral Psychology Handbook. New York: Oxford University Press, pp. 433–72. Kelly, D., and Morar, N. (in press). Against the Yuck Factor: On the Ideal Role of Disgust Society. Utilitas. Mackie, J. L. (1977). Ethics: Inventing Right and Wrong. New York: Penguin Books. Maudlin, T. (2010). The Metaphysics Within Physics. New York: Oxford University Press. Nado, J., Kelly, D., and Stich, S. (2009). Moral Judgment. In John Symons and Paco Calvo (eds), The Routledge Companion to the Philosophy of Psychology. New York: Routledge, pp. 621–33. Machery, E., and Mallon, R. (2010). Evolution of morality. In J. Doris et al. (eds), The Moral Psychology Handbook. New York: Oxford University Press, pp. 3–46. Nichols, S. (2002). Norms with feeling: Towards a psychological account of moral judgment. Cognition, 84, 221–36. —(2004). Sentimental Rules: On the Natural Foundations of Moral Judgment. New York: Oxford University Press. Nisbett, R. (2003). The Geography of Thought: How Asians and Westerners Think Differently . . . And Why. New York: The Free Press. Nussbaum, M. (2004a). Hiding from Humanity: Disgust, Shame, and the Law. Princeton, NJ: Princeton University Press. —(6 August 2004b). Danger to human dignity: The revival of disgust and shame in the law. The Chronicle of Higher Education, 50(48), B6. Parkinson, C., Sinnott-Armstrong, W., Koralus, P., Mendelovici, A., McGeer, V., and Wheatley, T. (2011). Is morality unified? Evidence that distinct neural systems underlie moral judgments of harm, dishonesty, and disgust. Journal of Cognitive Neuroscience, 23(10), 3162–80. Pinker, S. (1997). How the Mind Works. New York: W.W. Norton & Co. Rawls, J. (1971). A Theory of Justice (2nd ed. 1999). Cambridge, MA: Harvard University Press.
Selective Debunking Arguments
147
Sarkissian, H. (2010). Minor tweaks, major payoffs: The problems and promise of situationism in moral philosophy. Philosopher’s Imprint, 10(9), 1–15. Singer, P. (2005). Ethics and intuitions. The Journal of Ethics, 9, 331–52. Sinnott-Armstrong, W., and Wheatley, T. (2012). The disunity of morality and why it matters to philosophy. The Monist, 95(3), 355–77. Stanovich, K. (2005). The Robot’s Rebellion: Finding Meaning in the Age of Darwin. Chicago, IL: University of Chicago Press. Street, S. (2006). A darwinian dilemma for realist theories of value. Philosophical Studies, 127(1), 109–66.
8
The Psychological Foundations of Moral Conviction Linda J. Skitka*
In a letter to the editor of the Mercury News, one reader explained his views on the death penalty as follows: “I’ll vote to abolish the death penalty . . . and not just because it is fiscally imprudent with unsustainable costs versus a life sentence without possibility of parole. More importantly, it’s morally wrong. Making us and the state murderers—through exercising the death penalty—is a pure illogicality akin to saying ‘two wrongs make a right’” (Mercury News 2012). In short, this letter writer believes murder is simply wrong, regardless of whether it is an individual or state action, and for no other reason than because it is simply and purely wrong. Attitudes rooted in moral conviction (or “moral mandates”), such as the letter writer’s position on the death penalty, represent a unique class of strong attitudes. Strong attitudes are more extreme, important, central, certain, and/ or accessible, and are also more stable, enduring, and predictive of behavior than attitudes weaker on these dimensions (see Krosnick and Petty 1995 for a review). Attitudes held with the strength of moral conviction, even if they share many of the same characteristics of strong attitudes, are distinguished by a sense of imperative and unwillingness to compromise even in the face of competing desires or concerns. Someone might experience their attitude about chocolate, for example, in extreme, important, certain, and central terms, but still decide not to order chocolate cake at a restaurant because of the calories. Vanity, or other motives such as health or cost, can trump even people’s very strong preferences. Attitudes rooted in moral conviction, however, are much less likely to be compromised or vulnerable to trade off (cf. Tetlock et al. 2000).
The Psychological Foundations of Moral Conviction
149
To better understand how attitudes that are equally strong can nonetheless differ in their psychological antecedents and consequences, we need to understand the psychological and behavioral implications of the content of attitudes as well as their structure (e.g., extremity, importance). Social domain theory (e.g., Nucci 2001; Nucci and Turiel 1978; Turiel 1998; 2002), developed to explain moral development and reasoning, provides some useful hints about key ways that attitudes may differ in substance, even when they are otherwise equally strong. Using domain categories to describe how attitudes differ represents a useful starting point for understanding the foundations of moral mandates (Skitka et al. 2005; Skitka et al. 20081; Wright et al. 2008). As can be seen in Figure 8.1, one domain of attitudes is personal preference. Personal preferences represent attitudes that people see as subject to individual discretion, and as exempt from social regulation or comment. For example, one person might support legalized abortion because she prefers to have access to a backstop method of birth control, and not because of any normative or moral attachment to the issue. She is likely to think others’ preferences about abortion are neither right nor wrong; they may just be different from her own. Her position on this issue might still be evaluatively extreme, personally important, certain, central, etc., but it is not one she experiences as a core moral conviction. Her neighbor, on the other hand, might oppose legalized abortion because this practice is inconsistent with church doctrine or because the majority of people he is close to oppose it. If church authorities or his peer group were to reverse their stance on abortion, however, the neighbor probably would as well. Attitudes that reflect these kinds of normative beliefs typically describe what “people like me or us” believe, are relatively narrow in application, and are usually group or culture bound rather than universally applied. Yet a third person might see the issue of abortion in moral terms. This person perceives abortion (or restricting access to abortion) as simply and self-evidently wrong, even monstrously wrong, if not evil. Even if relevant authorities and peers were to reverse positions on the issue, this person would nonetheless maintain his or her moral belief about abortion. In addition to having the theorized characteristic of authority and peer independence, moral convictions are also likely to be perceived as objectively true, universal, and to have particularly strong ties to emotion.
150
Advances in Experimental Moral Psychology
Preferences
Conventions
Moral Mandates
• Matters of taste
• Normative • Often codified • Group defined • Culturally narrow
• Absolute/universal • Objective • Authority independent • Motivating • Self-justifying • Strong emotions
• Subjective • Tolerant
Figure 8.1 A domain theory of attitudes.
The goals of this chapter are to review recent developments in understanding the psychology of moral conviction and related research. These developments include research on operationalization and measurement as well as testing a wide range of hypotheses about how moral convictions differ in form and implication from otherwise strong but nonmoral attitudes.
Measurement and operationalization Research on moral conviction has generally opted to use a bottom-up rather than top-down empirical approach to study this construct. Instead of defining the characteristics of what counts as a moral mandate a priori (e.g., that it be seen as universal in application or resistant to trade-offs), researchers use facevalid items2 to assess strength of moral conviction, and test whether variation in strength of moral conviction yields predicted effects (e.g., differences in perceived universal applicability). Avoiding confounds with other indices of attitude strength is important to ensure that an individual’s response is motivated by morality, rather than by some other concern such as attitude importance, extremity, and so on. For this reason, moral conviction researchers see the distinction between moral and nonmoral attitudes as something that is subjectively perceived, rather than as an objective property of attitudes, decisions, choices, or dilemmas. Although people do not always seek to maximize principled consistency when making moral judgments (Ulhmann et al. 2009), they nonetheless
The Psychological Foundations of Moral Conviction
151
appear to have a strong intuitive sense of when their moral beliefs apply to a given situation (Skitka et al. 2005). People can identify when situations engage their moral sentiments, even when they cannot always elegantly describe the processes or principles that lead to this sense (Haidt 2001). The assumption that people have some insight into the characteristics of their own attitudes is one shared by previous theory and research on the closely related concept of attitude strength. Researchers assume that people can access from memory and successfully report the degree to which a given attitude is (for example) extreme, personally important, certain, or central (see Krosnick and Petty 1995 for a review). Hornsey and colleagues (Hornsey et al. 2003; 2007) provide one example of this approach. They operationalized moral conviction with three items, all prefaced with the stem, “To what extent do you feel your position . . .” and the completions of “is based on strong personal principles,” “is a moral stance,” and “is morally correct,” that across four studies had an average Cronbach’s α ⫽ 0.75. Others have operationalized moral conviction in similar fashion, most typically using either a single face-valid item: “How much are your feelings about connected to your core moral beliefs and convictions” (e.g., Brandt and Wetherell 2012; Skitka et al. 2005), or this item accompanied by a second item, “To what extent are your feelings about deeply connected to your fundamental beliefs about ‘right’ and ‘wrong’?” (e.g., Skitka et al. 2009; Skitka and Wisneski 2011; Swink 2011). Morgan (2011) used a combination of the Hornsey et al.’s (2003, 2007) and Skitka et al.’s (2009) items to create a 5-item scale, and found α’s that ranged from 0.93 to 0.99 across three samples. The reliability scores observed by Morgan suggest that either all, or a subset, of these items work well, and will capture highly overlapping content. Some have wondered, however, if moral conviction is better represented as a binary judgment: Something that is or is not the case, rather than something that varies in degree or strength. Measuring the categorization of an attitude as moral and the relative strength of conviction both contribute uniquely to the explanatory power of the variable (Wright et al. 2008; Wright 2012). For this reason, as well as the parallelism of conceptualizing moral conviction similarly to measures of attitude strength, we advocate that moral convictions be measured continuously rather than nominally. Other ways of operationalizing moral conviction are problematic because they confound moral conviction with the things that moral convictions should
152
Advances in Experimental Moral Psychology
theoretically predict (e.g., Van Zomeron et al. 2011; Zaal et al. 2011), use items that have no explicit references to morality (e.g., “X threatens values that are important to me,”3 Siegrist et al. 2012), conflate moral convictions with other dimensions of attitude strength (e.g., centrality, Garguilo 2010; Skitka and Mullen 2006), and/or measure other constructs as proxies for moral conviction, such as importance or centrality (e.g., Besley 2012; Earle and Siegrist 2008). These strategies introduce a host of possible confounds and do more to confuse than to clarify the unique contribution of moral conviction independent of other characteristics of attitudes. Attitude importance and centrality, for example, have very different associations with other relevant variables than those observed with unconfounded measures of moral conviction (e.g., including effects that are the reverse sign, e.g., Skitka et al. 2005). To avoid these problems, researchers should therefore use items that (a) explicitly assess moral content, and (b) do not introduce confounds that capture either the things moral conviction should theoretically predict (e.g., perceived universalism) or other dimensions of attitude strength (importance, certainty, or centrality). Moral philosophers argue that moral convictions are experienced as sui generis, that is as unique, special, and in a class of their own (e.g., Boyd 1988; McDowell 1979; Moore 1903; Sturgeon 1985). This status of singularity is thought to be due to a number of distinguishing mental states or processes associated with the recognition of something as moral, including (a) universalism, (b) the status of moral beliefs as factual beliefs with compelling motives and justification for action, and (c) emotion (Skitka et al. 2005). These theoretically defining characteristics of attitudes (which taken together represent the domain theory of attitudes) are testable propositions in themselves, and have a number of testable implications (e.g., the authority independence and nonconformity hypotheses). I briefly review empirical research testing these core propositions and selected hypotheses that can be derived from them next.
Universalism and objectivism The domain theory of attitudes predicts that people experience moral mandates as objective truths about the world, much as they do scientific judgments or
The Psychological Foundations of Moral Conviction
153
facts. In other words, good and bad are experienced as objective characteristics of phenomena and not just as verbal labels that people attach to feelings (Shweder 2002). Because beliefs rooted in moral conviction are perceived as operationally true, they should also be perceived as universally applicable. The author of the letter to the Mercury News, for example, is likely to believe that the death penalty should not only be prohibited in his home state of California, but in other states and countries as well. Broad versions of the universalism and objectivism hypotheses have been tested and supported. For example, people see certain moral rules (e.g., Nichols and Folds-Bennett 2003; Turiel 1978) and values (e.g., Gibbs et al. 2007) as universally or objectively true, and that certain moral transgressions should be universally prohibited (e.g., Brown 1991). There is some evidence that people also see ethical rules and moral issues as more objectively true than, for example, various violations of normative conventions (Goodwin and Darley 2008), but other research yields more mixed results (Wright et al. 2012). Until recently, little or no research has tested the universalism hypothesis. To shed further light on the objectivism and universalism hypotheses, Morgan, Skitka, and Lytle (under review) tested whether thinking about a morally mandated attitude leads to a situational increase in people’s endorsement of a universalistic moral philosophy (e.g., the degree to which people rate moral principles as individualistic or relativistic, versus as universal truisms). Participants’ endorsements of a universalistic moral philosophy, their positions on the issue of legalized abortion, and moral conviction about abortion were measured at least 24 hours before the experimental session. Once in the lab, participants were primed to think about abortion by writing an essay about their position that they thought would be shared with an “another participant.” They were then given an essay presumably written by the “other participant,” that was either pro-choice or pro-life (essays were modeled after real participants’ essays on this topic). After reading the essay, participants completed the same universalistic philosophy measure they had completed at pretest. Strength of moral conviction about abortion was associated with increased post-experimental endorsement of a universalistic philosophy, regardless of whether participants read an essay that affirmed or threatened their own position on the topic. In short, people see moral rules in general
154
Advances in Experimental Moral Psychology
as more universally applicable when they have just thought about an attitude held with moral conviction. A second study tested the universalism and objectivity hypotheses more directly by having participants rate the perceived objectivity (e.g., “Imagine that someone disagreed with your position on [abortion, requiring the HPV vaccine, same sex marriage]: To what extent would you conclude the other person is surely mistaken?”) and universality (“To what extent would your position on [abortion/the HPV vaccine, same sex marriage] be equally correct in another culture?”) of these attitudes, in addition to providing ratings of the degree to which each reflected a moral conviction. Strength of moral conviction was associated with higher perceived objectivity and universalism of attitudes, even when controlling for attitude extremity. Finally, in a third study, participants were asked to generate sentences that articulated their own beliefs or positions with respect to “a piece of scientific knowledge,” “something that is morally right or wrong,” and “that you like or dislike.” Participants then completed the same objectivity and universalism measures used in Study 2. Scientific and moral beliefs were rated as equally objectively true and universal, and as more objectively true and universal than likes/dislikes. In sum, these three studies demonstrated that moral convictions are perceived as indistinguishable from scientific facts in perceived universality and objectivism.
Motivation and behavior Attitudes rooted in moral conviction are predicted to also be inherently motivating, and therefore should have stronger ties to behavior than those not rooted in moral conviction. A moral conviction that voluntarily terminating a pregnancy (or alternatively, interfering with a woman’s right to choose whether to sustain a pregnancy) is fundamentally wrong, for example, has an inherent motivational quality—it carries with it an “ought” or “ought not” that can motivate subsequent behavior. Moral convictions are therefore theoretically sufficient in and of themselves as motives that can direct what people think, feel, or do (Skitka et al. 2005). Implicit in this reasoning is the hypothesis that people should also feel more compelled to act on attitudes held with strong rather than weak moral
The Psychological Foundations of Moral Conviction
155
conviction. In support of this hypothesis, stronger moral convictions about salient social issues and/or presidential candidates predict intentions to vote and actual voting behavior, results that have now replicated across three presidential election cycles in the United States (Morgan et al. 2010; Skitka and Bauman 2008). The motivational impact of moral conviction was a robust effect when controlling for alternative explanations, such as strength of partisanship and attitude strength. In an ingenious study, Wright et al. (2008, Study 2) put people’s self-interests into direct conflict with their moral convictions. Participants were pretested for their moral convictions on various issues. During the lab portion of the study they were shown “another participant’s” essay about an issue (manipulated to be inconsistent with the real participants’ attitudes). People almost always divide the prizes equally in this kind of economic game (e.g., Fehr and Fishbach 2004). People with stronger moral convictions about the essay issue, however, kept most of the raffle tickets for themselves (on average, 8.5 out of 10 tickets) when dividing the tickets between themselves and the “participant” who had a divergent attitude from their own. Those who did not see the issue as a moral one, conversely, divided the tickets equally between themselves and the other “participant” (Wright et al. 2008). In summary, people are usually motivated by fairness in these kinds of economic games, but their moral convictions and disdain for someone who did not share their moral views trumped any need to be fair.
Emotion The domain theory of attitudes also makes the prediction that moral convictions should have especially strong ties to emotion. For example, Person A might have preference for low taxes. If her taxes rise, she is likely to be disappointed rather than outraged. Imagine instead, Person B, who has a strong moral conviction that taxes be kept low. He is likely to respond to the same rise in tax rates with rage, disgust, and contempt. In short, the strength and content of emotional reactions associated with attitudes rooted in moral conviction are likely to be quite different than the emotional reactions associated with otherwise strong but nonmoral attitudes. Emotional responses to given issues might also play a key role in how people detect that an attitude is a moral conviction, or in strengthening moral convictions.
156
Advances in Experimental Moral Psychology
Emotion as consequence Consistent with the prediction that moral mandates will have different, and perhaps stronger ties to emotion than nonmoral mandates, people whose opposition to the Iraq War was high rather than low in moral conviction also experienced more negative emotion (i.e., anger and anxiety) about the War in the weeks just before and after it began. In contrast, supporters high in moral conviction experienced more positive emotions (i.e., pleased and glad) about going to war compared to those low in moral conviction, results that emerged even when controlling for a variety of attitude strength measures. Similar positive and negative emotional reactions were also observed in supporters’ and opponents’ reactions to the thought of legalizing physician-assisted suicide (Skitka and Wisneski 2011).
Emotion as antecedent Other research has tested whether people use emotions as information in deciding whether a given attitude is a moral conviction. Consistent with this idea, people make harsher moral judgments of other’s behavior when exposed to incidental disgust such as foul odors or when in a dirty lab room, than they do when exposed to more pleasant odors or a clean lab room (Schnall et al. 2008). People generalize disgust cues and apply them to their moral judgments. It is important to point out, however, that moral judgments are not the same thing as moral convictions. Attitudes (unlike judgments) tend to be stable, internalized, and treated much like possessions (e.g., Prentice 1987). In contrast, moral judgments are single-shot reactions to a given behavior, actor, or hypothetical, and share few psychological features with attitudes. Learning that incidental disgust leads to harsher moral judgments, therefore, may not mean that incidental disgust (or other incidental emotions) would also lead people to have stronger moral convictions. Consistent with distinctions between judgments and attitudes, research in my lab has found no effect of incidental emotion on moral convictions (Skitka, unpublished data). We have manipulated whether data is collected in a clean versus dirty lab; in the context of pleasant (e.g., “Hawaiian breeze,”) versus disgusting smells (e.g., fart spray or a substance that smelled like a dead rat); when participants have their hands and forearms placed in an unpleasant concoction of glue and gummy worms, versus feathers and beads;
The Psychological Foundations of Moral Conviction
157
having participants write retrospective accounts about a time when they felt particularly angry, sad, happy, or disgusted; or using a misattribution of arousal paradigm. Although manipulation checks revealed that each of these manipulations had the intended effect, none led to changes in moral conviction. One possible explanation for these null results is that integral (i.e., attitude-specific) emotions tied to the attitude object itself may be trumping the potential informational influence of incidental emotions. Once a moral conviction comes to mind, so too might all the emotional associations with it, which could overwhelm and replace incidental affect in people’s current working memory. Attitude-specific emotions might therefore play a more important role than incidental emotions in how people identify whether a given attitude is one held with moral conviction. To test this idea, participants were exposed to one of four categories of pictures as part of a bogus “recognition task.” The images varied in relevance to the issue of abortion: pictures of aborted fetuses (attitudinally relevant disgust/harm); animal rights abuses (attitudinally irrelevant disgust/harm); pictures of non-bloody, disgusting images, such as toilets overflowing with feces (attitudinally irrelevant disgust, no harm); or neutral photos (e.g., office furniture; no disgust/harm). Pictures were presented at either subliminally (14 msecs) or supraliminally (250 msecs). Participants’ moral conviction about abortion increased relative to control only after supraliminal exposure to the abortion pictures. Moreover, this effect was unique to moral conviction and was not observed with attitude importance. A second study replicated this effect, and tested whether it was mediated by disgust, anger, or perceived harm. The effect was fully mediated by disgust (Wisneski and Skitka 2013). Taken together, these results suggest that emotions play a key role in how people form or strengthen moral convictions, but these processes—although fast—nonetheless require some conscious processing. In summary, it is clear that moral convictions have ties to integral emotion. The relationship between emotions and moral convictions, however, appears to be complex. Future research needs to manipulate other kinds of integral emotions, including positive emotions, to discover whether other emotional cues can also cause changes in moral conviction. Emotions not only serve as an antecedent to moral convictions, but also appear to be consequences
158
Advances in Experimental Moral Psychology
of them as well. Although more research is needed to further tease apart the complex connections between moral convictions and emotions, one thing is clear: Emotions are clearly a key part of the story.
The authority independence hypothesis A core premise of the domain theory of attitudes is that people do not rely on conventions or authorities to define moral imperative; rather, people perceive what is morally right and wrong irrespective of authority or conventional dictates. Moral beliefs are not by definition antiestablishment or antiauthority, but are simply not dependent on conventions, rules, or authorities. When people take a moral perspective, they focus on their ideals and the way they believe things ought to or should be done rather than on a duty to comply with authorities or normative conventions. The authority independence hypothesis therefore predicts that when people’s moral convictions are at stake, they are more likely to believe that duties and rights follow from the greater moral purposes that underlie rules, procedures, and authority dictate than from the rules, procedures, or authorities themselves (see also Kohlberg 1976; Rest et al. 1999). One study tested the authority independence hypothesis by examining which was more important in predicting people’s reactions to a controversial US Supreme Court decision: people’s standing perceptions of the Court’s legitimacy, or people’s moral convictions about the issue being decided (Skitka et al. 2009). A nationally representative sample of adults rated the legitimacy of the Court, as well as their level of moral conviction about the issue of physicianassisted suicide several weeks before the Court heard arguments about whether states could legalize the practice, or whether it should be federally regulated. The same sample of people was contacted again after the Court upheld the right of states to legalize physician-assisted suicide. Knowing whether people’s support or opposition to physician-assisted suicide was high versus low in moral conviction predicted whether they saw the Supreme Court’s decision as fair or unfair, as well as their willingness to accept the decision as binding. Pre-ruling perceptions of the legitimacy of the Court, in contrast, had no effect on post-ruling perceptions of fairness or decision acceptance.
The Psychological Foundations of Moral Conviction
159
Other research has found behavioral support for the prediction that people reject authorities and the rule of law when outcomes violate their moral convictions. Mullen and Nadler (2008) exposed people to legal decisions that supported, opposed, or were unrelated to their moral convictions. The experimenters distributed a pen with a post-exposure questionnaire, and asked participants to return them at the end of the session. Consistent with the prediction that decisions, rules, and laws that violate people’s moral convictions erode support for the authorities and authority systems who decide these things, participants were more likely to steal the pen after exposure to a legal decision that was inconsistent rather than consistent with their personal moral convictions. People’s moral mandates should affect not only their perceptions of decisions and willingness to comply with authorities, but should also affect their perceptions of authorities’ legitimacy. People often do not know the “right” answer to various decisions authorities are asked to make (e.g., what is best for the group, whether a defendant is really guilty or innocent), and therefore, they frequently rely on cues like procedural fairness and an authority’s legitimacy to guide their reactions (Lind 2001). When people have moral certainty about what outcome authorities and institutions should deliver, however, they do not need to rely on standing perceptions of legitimacy as proxy information to judge whether the system works. In these cases, they can simply evaluate whether authorities get it “right.” “Right” decisions indicate that authorities are appropriate and work as they should. “Wrong” answers signal that the system is somehow broken and is not working as it should. In short, one could hypothesize that people use their sense of morality as a benchmark to assess authorities’ legitimacy. Consistent with this idea, the results of the Supreme Court study referenced earlier also found that perceptions of the Court’s legitimacy changed from pre- to post ruling as a function of whether the Court ruled consistently or inconsistently with perceivers’ morally vested outcome preferences (Skitka et al. 2009).
The nonconformity hypothesis Moral convictions might inoculate people from peer as well as authority influence. People typically conform to the majority when faced with the
160
Advances in Experimental Moral Psychology
choice to accept or reject the majority position. This occurs because those who oppose the majority risk ridicule and disenfranchisement, whereas those who conform expect acceptance (Asch 1956). In addition, people may conform when they are unsure about the appropriate way to think or behave; they adopt the majority opinion because they believe the majority is likely to be correct (Chaiken and Stangor 1987; Deutsch and Gerard 1955). Therefore, people conform both to gain acceptance from others and to be “right.” Feeling strong moral convictions about a given issue should weaken the typical motives for conformity—making people more resistant to majority influence. To test this idea, Hornsey and colleagues presented student participants with feedback that their position on same-sex marriage was either the majority or minority view on campus. Surprisingly, stronger moral convictions about this issue were associated with greater willingness to engage in activism when students believed they were in the opinion minority, rather than majority—an example of counter-conformity (Hornsey et al. 2003, 2007). Another study had participants engage in what they believed was a computermediated interaction with four additional (though, in fact, virtual) “peers.” The study was scripted so that each participant was exposed to a majority of peers who supported torture (pretesting indicated that none of our study participants did). Participants were shown the other participants’ “opinions” one at a time before they were asked to provide their own position on the issue to the group. Results supported the hypothesis: Stronger moral convictions were associated with lower conformity rates, even when controlling for a number of indices of attitude strength (Aramovich et al. 2010).4 By contrast, people do show strong conformity effects in an Asch paradigm when making moral judgments about moral dilemmas, such as the trolley problem (Kundu and Cummins 2012), providing further evidence that moral judgments and moral attitudes are not the same things.
Conclusion Theorists in recent years have proposed a number of ways that attitudes rooted in moral conviction differ from otherwise strong but nonmoral attitudes. The research reviewed here supports the hypothesis that moral
The Psychological Foundations of Moral Conviction
161
mandates represent a special class of strong attitudes that do not reduce to other dimensions of attitude strength. Moreover, moral mandates differ from strong but nonmoral attitudes in ways that are predicted by a domain theory of attitudes. They are perceived as akin to facts about the world, positions that should be universally adopted, have particularly strong ties to emotion, are motivational, and predict a host of behaviors and reactions including authority independence, political legitimacy, anti-conformity, and civic engagement. With some exceptions, most research on the concept of moral conviction has focused on determining whether and how moral mandates differ from nonmoral attitudes. The challenge for future research will be to begin to gain a greater understanding of how moral mandates are developed in the first place and, once established, whether people are capable of demoralizing an attitude. Given moral mandates have the potential for motivating great good (e.g., civic engagement, willingness to fight for justice), as well as motivating acts many would label as evil (e.g., terrorism, vigilantism; see Morgan and Skitka 2009), learning more about the attitude moralization process represents an important area of inquiry going forward.
Notes * Author’s Note: Linda J. Skitka, Department of Psychology, University of Illinois at Chicago. Thanks to Brittany Hanson, G. Scott Morgan, and Daniel Wisneski for their helpful comments on earlier drafts of this chapter. Funding from the National Science Foundation #1139869 facilitated preparation of this chapter. Correspondence should be sent to: Linda J. Skitka, Ph.D., Professor of Psychology, University of Illinois at Chicago, Department of Psychology (m/c 285), 1007 W. Harrison St., Chicago, IL 60607-7137, Email: [email protected]. 1 Skitka et al. (2008) initially labeled this theoretical perspective as an “integrated theory of moral conviction” or ITMC. 2 Face validity refers to the degree to which one can infer from test items the target variable is being measured. 3 Not all values are perceived in moral terms. For example, fewer than 20 percent of participants perceived the Schwartz values associated with power, achievement, hedonism, and stimulation as moral, and fewer than 30 percent rated more than one of the self-direction items as moral (Schwartz 2007). 4 Having another dissenter in the group did not change the results of moral conviction.
162
Advances in Experimental Moral Psychology
References Aguilera, R., Hanson, B., and Skitka, L. J. (2013). Approaching good or avoiding bad? Understanding morally motivated collective action. Paper presented at the annual meeting of the Society for Personality and Social Psychology, New Orleans, LA. Aramovich, N. P., Lytle, B. L., and Skitka, L. J. (2012). Opposing torture: Moral conviction and resistance to majority influence. Social Influence, 7, 21–34. Asch, S. E. (1956). Studies of independence and conformity: A minority of one against a unanimous majority. Psychological Monographs, 70(9, No 416), 1–70. Bartels, D. M. (2008). Principled moral sentiment and the flexibility of moral judgment and decision making. Cognition, 180, 381–417. Besley, J. C. (2012). Does fairness matter in the context of anger about nuclear energy decision making? Risk Analysis, 32, 25–38. Boyd, R. (1988). How to be a moral realist. In G. Sayre-McCord (ed.), Essays in Moral Realism. Ithaca, NY: Cornell University Press, pp. 181–228. Brandt, M. J., and Wetherell, G. A. (2012). What attitudes are moral attitudes? The case of attitude heritability. Social Psychological and Personality Science, 3, 172–9. Brown, D. (1991). Human Universals. New York: McGraw-Hill. Chaiken, S., and Stangor, C. (1987). Attitudes and attitude change. Annual Review of Psychology, 38, 575–630. Cushman, F. A., Young, L., and Hauser, M. D. (2006). The role of reasoning and intuition in moral judgments: Testing three principles of harm. Psychological Science, 17, 1082–9. Darwin, D. O. (1982). Public attitudes toward life and death. Public Opinion Quarterly, 46, 521–33. Deutsch, M., and Gerard, H. B. (1955). A study of normative and informational social influences upon individual judgment. Journal of Abnormal and Social Psychology, 51, 629–36. Earle, T. C., and Siegrist, M. (2008). On the relation between trust and fairness in environmental risk management. Risk Analysis, 28, 1395–413. Fehr, E., and Fischbacher, U. (2004). Third-party punishment and social norms. Evolution and Human Behavior, 25, 63–87. Garguilo, S. P. (2010). Moral Conviction as a Moderator of Framing Effects (Master’s thesis). Rutgers University, Rutgers, NJ. Gibbs, J. C., Basinger, K. S., Grime, R. L., and Snarey, J. R. (2007). Moral judgment development across cultures: Revisiting Kohlberg’s universality claims. Developmental Review, 27, 443–550. Goodwin, G. P., and Darley, J. M. (2008). The psychology of meta-ethics: Exploring objectivism. Cognition, 106, 1139–366.
The Psychological Foundations of Moral Conviction
163
Greene, J. D., Sommerville, R. B., Nystrom, L. E., Darley, J. M., and Cohen, J. D. (2001). An fMRI investigation of emotional engagement in moral judgment. Science, 293, 2105–8. Haidt, J. (2001). The emotional dog and its rational tail: A social intuitionist approach to moral judgment. Psychological Review, 108, 814–34. Hornsey, M. J., Majkut, L., Terry, D. J., and McKimmie, B. M. (2003). On being loud and proud: Non-conformity and counter-conformity to group norms. British Journal of Social Psychology, 42, 319–35. Hornsey, M. J., Smith, J. R., and Begg, D. I. (2007). Effects of norms among those with moral conviction: Counter-conformity emerges on intentions but not behaviors. Social Influence, 2, 244–68. Hume, D. (1968). A Treatise on Human Nature. Oxford, England: Clarendon Press. Original work published 1888. Kohlberg, L. (1976). Moral stages and moralization: The cognitive developmental approach. In T. Lickona (ed.), Moral Development and Behavior: Theory, Research and Social Issues. New York: Holt, Rinehart, & Winston, pp. 31–53. Krosnick, J. A., and Petty, R. E. (1995). Attitude strength: An overview. In R. E. Petty and J. A. Krosnick (eds), Attitude Strength: Antecedents and Consequences. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 1–24. Kundu, P., and Cummins, D. D. (2012). Morality and conformity: The Asch paradigm applied to moral decisions. Social Influence, (ahead-of-print), 1–12. Lind, E. A. (2001). Fairness heuristic theory: Justice judgments as pivotal cognitions in organizational relations. In J. Greenberg and R. Cropanzano (eds), Advances in Organizational Behavior. San Francisco: New Lexington Press, pp. 27–55. Lodewijkx, H. F. M., Kersten, G. L. E., and Van Zomeren, M. (2008). Dual pathways to engage in ‘‘Silent Marches’’ against violence: Moral outrage, moral cleansing, and modes of identification. Journal of Community and Applied Social Psychology, 18, 153–67. Mackie, J. L. (1977). Ethics: Inventing Right and Wrong. New York: Penguin. McDowell, J. (1979). Virtue and reason. The Monist, 62, 331–50. Mercury News (2012). Talk back/Saturday forum letters. Retrieved 12/17/12 from http://www.mercurynews.com/top-stories/ci_21813856/oct-20-talk-backsaturday-forum-letters Moore, G. E. (1903). Principia Ethica. New York: Cambridge University Press. Morgan, G. S. (2011). Toward a Model of Morally Motivated Behavior: Investigating Mediators of the Moral Conviction-action Link (Doctoral dissertation). University of Illinois at Chicago. Morgan, G. S., Skitka, L. J., and Lytle, B. L. (under review). Universally and objectively true: The psychological foundations of moral conviction.
164
Advances in Experimental Moral Psychology
Morgan, G. S., Skitka, L. J., and Lytle, B. (in preparation). Morgan, G. S., Skitka, L. J., and Wisneski, D. (2010). Moral and religious convictions and intentions to vote in the 2008 Presidential election. Analyses of Social Issues and Public Policy, 10, 307–20. Nichols, S., and Folds-Bennett, T. (2003). Are children moral objectivists? Children’s judgments about moral and response-dependent properties. Cognition, 90, B23–32. Nucci, L. P. (2001). Education in the Moral Domain. Cambridge, UK: Cambridge University Press. Nucci, L. P., and Turiel, E. (1978). Social interactions and the development of social concepts in pre-school children. Child Development, 49, 400–7. Prentice, D. A. (1987). Psychological correspondence of possessions, attitudes, and values. Journal of Personality and Social Psychology, 53, 993–1003. Prinz, J. J. (2007). The Emotional Construction of Morals. New York: Oxford University Press. Rest, J. R., Narvaez, D., Bebeau, M. J., and Thoma, S. J. (1999). Postconventional Moral Thinking. A Neo-Kohlbergian Approach. Mahwah, NJ: Lawrence Erlbaum. Schnall, S., Haidt, J., Clore, G. L., and Jordan, A. H. (2008). Disgust as embodied moral judgment. Personality and Social Psychology Bulletin, 34, 1096–109. Schwartz, S. H. (2007). Universalism and the inclusiveness of our moral universe. Journal of Cross Cultural Psychology, 38, 711–28. Shweder, R. A. (2002). The nature of morality: The category of bad acts. Medical Ethics, 9, 6–7. Siegrist, M., Connor, M., and Keller, C. (2012). Trust, confidence, procedural fairness, outcome fairness, moral conviction, and the acceptance of GM field experiments. Risk Analysis, 32, 1394–403. Skitka, L. J. (2010). The psychology of moral conviction. Social and Personality Psychology Compass, 4, 267–81. —(2012). Understanding morally motivated behavioral intentions: A matter of consequence or conscience? Paper presented at the Cognitions vs. Emotions in Ethical Behavior Conference, University of Toronto. Skitka, L. J., Bauman, C. W., and Lytle, B. L. (2009). The limits of legitimacy: Moral and religious convictions as constraints on deference to authority. Journal of Personality and Social Psychology, 97, 567–78. Skitka, L. J., Bauman, C. W., and Mullen, E. (2008). Morality and justice: An expanded theoretical perspective and review. In K. A. Hedgvedt and J. Clay-Warner (eds), Advances in Group Processes, Vol. 25. Bingley, UK: Emerald Group Publishing Limited, pp. 1–27.
The Psychological Foundations of Moral Conviction
165
Skitka, L. J., Bauman, C. W., and Sargis, E. G. (2005). Moral conviction: Another contributor to attitude strength or something more? Journal of Personality and Social Psychology, 88, 895–917. Skitka, L. J., and Morgan, G. S. (2009). The double-edged sword of a moral state of mind. In D. Narvaez and D. K. Lapsley (eds), Moral Self, Identity, and Character: Prospects for New Field of Study. Cambridge, UK: Cambridge University Press, pp. 355–74. Skitka, L. J., and Wisneski, D. C. (2011). Moral conviction and emotion. Emotion Review, 3, 328–30. Smith, M. (1994). The Moral Problem. Oxford, England: Blackwell. Sturgeon, N. (1985). Moral explanations. In D. Copp and D. Zimmerman (eds), Morality, Reason, and Truth. Totowa, NJ: Rowman and Allanheld, pp. 49–78. Swink, N. (2011). Dogmatism and moral conviction in individuals: Injustice for all. (Doctoral dissertation). Wichita State University. Tetlock, P. E., Kirstel, O. V., Elson, S. B., Green, M. C., and Lerner, J. S. (2000). The psychology of the unthinkable: Taboo trade-offs, forbidden base rates, and heretical counterfactuals. Journal of Personality and Social Psychology, 78, 853–70. Turiel, E. (1978). Social regulations and domains of social concepts. In W. Damon (ed.), New Directions for Child Development. Vol. 1. Social Cognition. New York: Gardner, pp. 45–74. —(1983). The Development of Social Knowledge: Morality and Convention. New York: Cambridge University Press. —(1998). The development of morality. In W. Damon (Series ed.) and N. Eisenberg (Vol. ed.), Handbook of Child Psychology: Vol. 3. Social Emotional and Personality Development (5th ed.). New York: Academic Press, pp. 863–932. Uhlmann, E. L., Pizarro, D. A., Tannenbaum, D., and Ditto, P. H. (2009). The motivated use of moral principles. Judgment and Decision Making, 6, 476–91. Van Zomeron, M., Postmes, T., Spears, R., and Bettache, K. (2011). Can moral convictions motivate the advantaged to challenge social inequality?: Extending the social identity model of collective action. Group Processes and Intergroup Relations, 14, 735–53. Wisneski, D. C., Lytle, B. L., and Skitka, L. J. (2009). Gut reactions: Moral conviction, religiosity, and trust in authority. Psychological Science, 20, 1059–63. Wisneski, D. C., and Skitka, L. J. (2013). Flipping the “moralization switch:” Exploring possible routes to moral conviction. Emotion pre-conference, Society for Personality and Social Psychology, New Orleans, LA. Wright, J. C. (2012). Children’s and adolescents’ tolerance for divergent beliefs: Exploring the cognitive and affective dimensions of moral conviction in our youth. British Journal of Developmental Psychology, 30, 493–510.
166
Advances in Experimental Moral Psychology
Wright, J. C., Cullum, J., and Schwab, N. (2008). The cognitive and affective dimensions of moral conviction: Implications for attitudinal and behavioral measures of interpersonal tolerance. Personality and Social Psychology Bulletin, 34, 1461–76. Wright, J. C., Grandjean, P. T., and McWhite, C. B. (2012). The meta-ethical grounding of our moral beliefs: Evidence of meta-ethical pluralism. Philosophical Psychology, ifirst, 1–26. Zaal, M. P., Van Laar, C., Ståhl, T., Ellemers, N., and Derks, B. (2011). By any means necessary: The effects of regulatory focus and moral conviction on hostile and benevolent forms of collection action. British Journal of Social Psychology, 50, 670–89.
9
How Different Kinds of Disagreement Impact Folk Metaethical Judgments James R. Beebe*
Although the empirical study of folk metaethical judgments is still in its infancy, a variety of interesting and significant results have been obtained.1 Goodwin and Darley (2008), for example, report that individuals tend to regard ethical statements as more objective than conventional or taste claims and almost as objective as scientific claims, although there is considerable variation in metaethical intuitions across individuals and across different ethical issues. Goodwin and Darley (2012) also report (i) that participants treat statements condemning ethical wrongdoing as more objective than statements enjoining good or morally exemplary actions, (ii) that perceived consensus regarding an ethical statement positively influences ratings of metaethical objectivity, and (iii) that moral objectivism is associated with greater discomfort with and more pejorative attributions toward those with whom individuals disagreed. Beebe and Sackris (under review) found that folk metaethical commitments vary across different life stages, with decreased objectivism during the college years. Sarkissian et al. (2011) found that folk intuitions about metaethical objectivity vary as a function of cultural distance, with increased cultural distance between disagreeing parties leading to decreased attributions of metaethical objectivity. Wright et al. (2013) found that not only is there significant diversity among individuals with regard to the objectivity they attribute to ethical claims, there is also significant diversity of opinion with respect to whether individuals take certain issues such as abortion or anonymously donating money to charity to be ethical issues at all, despite the fact that philosophers overwhelmingly regard these issues as ethical.2 Wright et al. (2013) provide the following useful summary of the current set of findings on folk metaethical intuitions:
168
Advances in Experimental Moral Psychology
People do not appear to conceive of morality as a unified (metaethically speaking) domain, but rather as a domain whose normative mandates come in different shapes and sizes. They view the wrongness of some moral actions as clear and unquestionable, unaltered (and unalterable) by the feelings/beliefs/values of the individual or culture. They view the wrongness of other actions (though still genuinely moral in nature) as more sensitive to, and molded by, the feelings/beliefs/values of the actor and/or the people whose lives would be (or have been) affected by the action. This possibility is one we’ve not seen seriously considered in the metaethical literature—and perhaps it is time that it was.
The present chapter reports a series of experiments designed to extend the empirical investigation of folk metaethical intuitions by examining how different kinds of ethical disagreement can impact attributions of objectivity to ethical claims. Study 1 reports a replication of Beebe and Sackris’ work on metaethical intuitions, in order to establish a baseline of comparison for Studies 2 through 4. In Study 2, societal disagreement about ethical issues was made salient to participants before they answered metaethical questions about the objectivity of ethical claims, and this was found to decrease attributions of objectivity to those claims. In Studies 3 and 4, the parties with whom participants were asked to consider having an ethical disagreement were made more concrete than in Studies 1 and 2, using either verbal descriptions or facial pictures. This manipulation was found to increase attributions of metaethical objectivity. In a final study, metaethical judgments were shown to vary with the moral valence of the actions performed by the disagreeing party—in other words, a Knobe effect for metaethical judgments was found. These studies aim to increase our understanding of the complexity of the folk metaethical landscape.
Study 1 Method Participants Study 1 was an attempt to replicate Beebe and Sackris’ (under review) initial study with a population of participants that was limited to the same university
How Different Kinds of Disagreement Impact Folk Metaethical Judgments
169
student population from which participants for Studies 2 and 3 would be drawn. Participants were 192 undergraduate students (average age ⫽ 20, 53% female, 40% Anglo-American) from the University at Buffalo (a large, public university in the northeastern United States) in exchange for extra credit in an introductory course.
Materials Beebe and Sackris asked two and a half thousand participants between the ages 12 and 88 to indicate the degree to which they agreed or disagreed with the claims that appear in Table 9.1 and the extent to which they thought that “people in our society” disagreed about whether they are true. The same set of claims was used in Studies 1 through 3.
Procedure The items from Table 9.1 were divided into three questionnaire versions, and participants indicated their agreement or disagreement with them on a six-point scale, where “1” was anchored with “Strongly Disagree” and “6” with “Strongly Agree.” Participants rated the extent to which they thought people in our society disagreed about the various claims on a six-point scale anchored with “There is no disagreement at all” and “There is an extremely large amount of disagreement.” In order to capture one kind of objectivity that participants might attribute to the various claims in Table 9.1, participants were asked, “If someone disagrees with you about whether [one of these claims is true], is it possible for both of you to be correct or must one of you be mistaken?” The answer “At least one of you must be mistaken” was interpreted as an attribution of objectivity, and an answer of “It is possible for both of you to be correct” was taken to be a denial of objectivity.
Results As can be seen from Figure 9.1, the items in Table 9.1 are ordered within each subcategory in terms of increasing proportions of participants who attributed objectivity to them.
170
Advances in Experimental Moral Psychology
Table 9.1 Factual, ethical, and taste claims used in Beebe and Sackris (under review) and in Studies 1 through 4 Factual 1. Frequent exercise usually helps people to lose weight. 2. Global warming is due primarily to human activity (for example, the burning of fossil fuels). 3. Humans evolved from more primitive primate species. 4. There is an even number of stars in the universe. 5. Julius Caesar did not drink wine on his 21st birthday. 6. New York City is further north than Los Angeles. 7. The earth is only 6,000 years old. 8. Mars is the smallest planet in the solar system. Ethical 9. Assisting in the death of a friend who has a disease for which there is no known cure and who is in terrible pain and wants to die is morally permissible. 10. Before the third month of pregnancy, abortion for any reason is morally permissible. 11. Anonymously donating a significant portion of one’s income to charity is morally good. 12. Scientific research on human embryonic stem cells is morally wrong. 13. Lying on behalf of a friend who is accused of murder is morally permissible. 14. Cutting the American flag into pieces and using it to clean one’s bathroom is morally wrong. 15. Cheating on an exam that you have to pass in order to graduate is morally permissible. 16. Hitting someone just because you feel like it is wrong. 17. Robbing a bank in order to pay for an expensive vacation is morally bad. 18. Treating someone poorly on the basis of their race is morally wrong. Taste 19. Classical music is better than rock music. 20. Brad Pitt is better looking than Drew Carey. 21. McDonald’s hamburgers taste better than hamburgers made at home. 22. Gourmet meals from fancy Italian restaurants taste better than microwavable frozen dinners. 23. Barack Obama is a better public speaker than George W. Bush. 24. Beethoven was a better musician than Britney Spears is.
How Different Kinds of Disagreement Impact Folk Metaethical Judgments
Proportion of objectivity attributions
1.0
171
Question type Factual Ethical Taste
0.5
0.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Question number
Figure 9.1 Proportions of participants who attributed objectivity to the 24 items in Study 1. Error bars in all figures represent 95 percent confidence intervals.
Goodwin and Darley (2008) and Beebe and Sackris both found that more participants attributed objectivity to factual claims than to ethical or taste claims. In Study 1, a greater proportion of participants attributed objectivity to factual claims (0.64, averaged across all claims in the factual subcategory) than to ethical (0.34) or taste (0.10) claims. Chi-square tests of independence reveal that the difference between the factual and ethical proportions was significant, c2 (1, N ⫽ 926) ⫽ 80.523, p ⬍ 0.001, Cramér’s V ⫽ 0.30, and the difference between the ethical and taste proportions was significant as well, c2 (1, N ⫽ 826) ⫽ 61.483, p ⬍ 0.001, Cramér’s V ⫽ 0.27.3 Study 1 also replicates earlier findings that objectivity attributions are positively associated with strength of belief about an issue (c2 (2, N ⫽ 1,224) ⫽ 67.276, p ⬍ 0.001, Cramér’s V ⫽ 0.23) but negatively associated with the extent of perceived disagreement about the issue (c2 (5, N ⫽ 1,218) ⫽ 89.517, p ⬍ 0.001, Cramér’s V ⫽ 0.27). In other words, participants tended to attribute more objectivity to claims that they had stronger opinions about than to claims they had weaker
172
Advances in Experimental Moral Psychology
opinions about, but they tended to attribute less objectivity to claims they recognized were widely disputed in society. Somewhat surprisingly, higher ratings of perceived disagreement about an issue were positively associated with participants’ strength of opinion about the issue, c2 (10, N ⫽ 1,212) ⫽ 100.897, p ⬍ 0.001, Cramér’s V ⫽ 0.20.
Discussion Like Goodwin and Darley (2008) and Beebe and Sackris, Study 1 found that participants attribute more objectivity to some ethical claims than to some factual claims and that there is significant variation concerning the degree of objectivity attributed to different claims within each subcategory.4 Thus, Study 1 reinforces the conclusion already established by Goodwin and Darley (2008) and Beebe and Sackris that the question of whether ordinary individuals are moral objectivists is not going to have a simple “Yes” or “No” answer.
Study 2 Method Participants A total of 195 undergraduate students (average age ⫽ 19, 47% female, 69% Anglo-American) from the University at Buffalo participated in Study 2 in exchange for extra credit in an introductory course.
Materials and procedure The primary purpose of Study 1 was to construct a baseline of data with which the results of Studies 2 through 4 could be compared. These latter studies each introduce some kind of modification to the research materials used in Study 1 in order to see how folk metaethical judgments will be affected. The manipulation in Study 2 was simply a change in the order of the tasks participants were asked to complete. As noted above, Study 1 followed Beebe and Sackris in having participants perform the following tasks in the following order:
How Different Kinds of Disagreement Impact Folk Metaethical Judgments
173
Task 1: Participants indicated the extent to which they agreed or disagreed with a given claim. Task 2: Participants answered the question “If someone disagrees with you about whether [the claim in question is true], is it possible for both of you to be correct or must at least one of you be mistaken?.” Task 3: Participants rated the extent to which they thought people in our society disagreed about whether the claim in question is true.
Thus, the last thing participants were asked to do was to consider about the extent of societal disagreement with respect to the claims. Given the negative association between perceived disagreement and objectivity attributions, it was hypothesized that if participants were directed to think about societal disagreement before completing Task 2, their attributions of metaethical objectivity would decrease. Disagreement was not hypothesized to have a similar effect on factual and taste claims.
Results As expected, the overall proportion of objectivity attributions in the ethical subcategory was lower in Study 2 (0.29) than in Study 1 (0.34). This difference was significant, c2 (1, N ⫽ 1045) ⫽ 4.015, p ⬍ 0.05, Cramér’s V ⫽ 0.06. There were no significant differences in the factual and taste subcategories. Thus, it appears that making disagreement about ethical issues salient to participants can have a modest effect on the metaethical judgments they make. The fact that this result was obtained in the ethical domain but not in the factual domain is consistent with the widespread view among philosophers that ethical disagreement—because of its seemingly intractability—poses a significant challenge to the objectivity of ethical claims in a way that disagreement about factual matters fails to do for the objectivity of factual claims.5
Discussion The findings of Study 2 are consistent not only with the correlational data obtained by Goodwin and Darley (2008) and Beebe and Sackris but also with
174
Advances in Experimental Moral Psychology
the experimental data obtained by Goodwin and Darley (2012). The latter manipulated participants’ perceived consensus about ethical issues by giving them bogus information about the percentage of students from the same institution who agreed with them. Participants who were told that a majority of their peers agreed with them about some ethical statement were more likely to think there was a correct answer as to whether or not the statement was true than participants who were told that significantly fewer of their peers agreed with them. These studies show that perceived disagreement or consensus can be a causal and not a merely correlational factor in folk metaethical decisionmaking.
Study 3 Various studies of folk intuitions about moral responsibility have shown that individuals hold agents more responsible for their actions when the situations of those agents are described concretely than when they are described abstractly. Nichols and Knobe (2007), for example, obtained significantly higher ratings of moral responsibility for “Bill,” who was attracted to his secretary and killed his wife and three children in order to be with her, than for “a person” whose actions were left unspecified. Small and Loewenstein (2003, 2005) showed that the subtlest change in the concreteness of the representation of an individual can lead to surprising differences in judgments or decisions regarding them. When their participants were given the opportunity to punish randomly selected defectors in an economic game, participants selected significantly harsher punishments for anonymous defectors whose numbers had just been chosen than for anonymous defectors whose numbers were about to be chosen. Because increased concreteness appears to heighten or intensify the engagement of cognitive and affective processes associated with attributions of blame and responsibility and to lead participants to treat the actions of concrete individuals as more serious than abstractly represented ones,6 it was hypothesized that increasing the concreteness of those with whom participants were asked to imagine they disagreed would lead participants to take the disagreements more seriously and to increase attributions of metaethical objectivity.
How Different Kinds of Disagreement Impact Folk Metaethical Judgments
175
Method Participants A total of 108 undergraduate students (average age ⫽ 19, 59% female, 66% Anglo-American) from the University at Buffalo participated in Study 3 in exchange for extra credit in an introductory course.
Materials and procedure In Beebe and Sackris’ materials, which serve as the basis for Studies 1 and 2, each participant was asked “If someone disagrees with you about whether [one of these claims is true], is it possible for both of you to be correct or must one of you be mistaken?.” In Study 3, this unspecified “someone” was replaced with “Joelle P., a junior nursing major at UB,” “Mike G., a freshman computer science major at UB,” or some other student from the participant’s university, whose first name, last initial, class, and major were specified. In between completing Tasks 1 and 3 (which were described above) for 8 of the 24 claims found in Table 9.1, each participant completed a modified version of Task 2 such as the following: Madeline B., a senior biology major at UB, believes it is permissible to lie on behalf of a friend who is accused of murder. If you disagree with Madeline B., is it possible for both of you to be correct or must one of you be mistaken? It is possible for both of you to be correct. At least one of you must be mistaken. [If you agree with Madeline B., please skip to the next question.]
Results In accord with my expectations, having more concrete parties with which to disagree resulted in a significantly greater overall proportion of objectivity attributions to ethical claims in Study 3 (0.43) than in Study 1 (0.34), c2 (1, N ⫽ 826) ⫽ 5.399, p ⬍ 0.05, Cramér’s V ⫽ 0.08. The proportions were
176
Advances in Experimental Moral Psychology
numerically higher for eight of the ten ethical claims. Having more concrete parties in Study 3 did not, however, result in any significant difference in the objectivity attributed to factual or taste claims.
Discussion The results from Study 3 are consisted with those obtained by Sarkissian et al. (2011), who found that strong objectivity ratings were obtained when participants were asked to consider disagreeing with a concretely presented individual from their same culture (vs. a concretely presented individual from a different culture). The fact that the concreteness of the disagreeing parties used in Study 3 led to increased metaethical objectivity attributions may also explain why the objectivity ratings obtained in Study 1 fell below those obtained by Goodwin and Darley (2008), even though both used samples of university students. The Task 2 objectivity question in Study 1 asked participants to consider a situation of hypothetical disagreement (“If someone disagrees with you . . .”). Goodwin and Darley (2008, 1344), however, instructed participants, “We have done prior psychological testing with these statements, and we have a body of data concerning them. None of the statements have produced 100% agreement or disagreement.” Each of Goodwin and Darley’s objectivity questions then reiterated that some individuals who had been previously tested disagreed with participants about the relevant issue. Goodwin and Darley thus constructed situations of disagreement that were more concrete than those in Studies 1 and 2 by having (allegedly) actual rather than merely hypothetical individuals who disagreed with participants.
Study 4 Study 3 made the parties with whom experimental participants were asked to consider disagreeing concrete by providing them with given names, surname initials, academic classes, and majors. In Study 4, the disagreeing parties were made concrete by having pictures of their faces shown. Faces (and parts of faces) have been shown to have a variety of effects on morally relevant
How Different Kinds of Disagreement Impact Folk Metaethical Judgments
177
behavior. For example, Bateson et al. (2006) found that academics paid 276 percent more for the tea they took from a departmental tea station when an image of eyes was displayed by the station than when an image of flowers was displayed. Rezlescu, Duchaine, Olivola, and Chater (2012) found that unfakeable facial features associated with trustworthiness attracted 42 percent greater investment in an economic game that required trust.7
Method Participants A total of 360 participants (average age ⫽ 32, 38% female, 82% AngloAmerican) were recruited through Amazon’s Mechanical Turk (www. mturk.com) and were directed to complete online questionnaires hosted at vovici.com.8
Materials and procedure Combining behavioral studies and computer modeling, Oosterhof and Todorov (2008) found that individuals make surprisingly consistent judgments about socially relevant traits of individuals on the basis of differences in their facial characteristics. They claim that the two most important dimensions of face evaluation are trustworthiness/untrustworthiness and dominance/submissiveness. Judgments concerning the first dimension are reliably associated with judgments about whether an individual should be approached or avoided and with attributions of happiness or anger. Judgments concerning dominance or submissiveness were found to be reliably associated with judgments of the maturity, masculinity, and physical strength of an individual. Both untrustworthy and dominant faces were associated with potential threat.9 By exaggerating features specific to one of these evaluative dimensions, Oosterhof and Todorov (2008) created the set of faces represented in Table 9.2.10 Each of the non-neutral faces was plus or minus three standard deviations from the mean along the relevant dimension. The faces in Table 9.2 were used in Study 4, along with a control condition in which no face was displayed.
178
Advances in Experimental Moral Psychology
Table 9.2 Faces used in Study 4
Dominant
Trustworthy
Neutral
Untrustworthy
Submissive
Claims (12), (13), and (14) from Table 9.1—concerning embryonic stem cell research, lying for a friend accused of murder, and treating a national flag disrespectfully—were selected for use in Study 4. The degrees of objectivity attributed to them in Studies 1 through 3 fell in the middle range, suggesting that judgments about them could be more easily manipulated than judgments near the floor or ceiling. The first screen contained one of the pictures from Table 9.2, along with the following (Task 1) question:
How Different Kinds of Disagreement Impact Folk Metaethical Judgments
179
Mark (pictured above11) believes that [statement (12), (13), or (14) is true]. Please indicate whether you agree or disagree with Mark’s belief. Agree Disagree
If participants selected “Agree” in response to one Task 1 question, they would be directed to answer the Task 1 question for one of the other target claims. However, if participants selected “Disagree,” they were directed to answer the following (Task 2) metaethical question about their disagreement before moving on to the next Task 1 question: You disagree with Mark about whether [the target claim is true]. Is it possible for both of you to be correct about this issue or must at least one of you be mistaken? It is possible for both of you to be correct. At least one of you must be mistaken.
Each screen that presented the metaethical question included the same picture (if any) that participants saw at the top of their Task 1 question. Each participant was presented with claims (12), (13), and (14) in counterbalanced order. The same picture (if any) of Mark appeared above each of these questions. Thus, no participant saw more than one version of Mark’s face. It was hypothesized that the five facial conditions would engage online processes of social cognition to a greater degree than the control condition and that this would result in higher attributions of metaethical objectivity. On the basis of Oosterhof and Todorov’s (2008) finding that untrustworthy and dominant faces were associated with potential threat, it was also hypothesized that untrustworthy and dominant faces would elicit lower objectivity attributions than their dimensional pairs, since participants might be more tentative or anxious about disagreeing with potentially threatening interlocutors.
Results The proportion of objective attributions was significantly higher in the Neutral (0.65), Dominant (0.61), Submissive (0.60), Trustworthy (0.67),
180
Advances in Experimental Moral Psychology
and Untrustworthy (0.66) face conditions than it was in the No Face (0.46) condition. The proportions of objectivity attributions in the five face conditions did not differ significantly from each other.
Discussion Thus, it appears that having a face—any face, perhaps—makes the situation of moral judgment more concrete and engages moral cognitive processes in a way that increases attributions of objectivity. Because there were no significant differences between the Trustworthy and Untrustworthy conditions and the Dominant and Submissive face conditions, the second hypothesis concerning the effect of specific kinds of faces on folk metaethical intuitions failed to receive confirmation from Study 4.12
Study 5 Method Participants Using a between-subject design, 160 participants (average age ⫽ 34, 38% female, 80% Anglo-American) were recruited through Amazon’s Mechanical Turk and were directed to complete online questionnaires hosted at vovici. com.13
Materials and procedure A final study was constructed to see if the moral valence of the actions that disagreeing parties were described as performing would have an effect on folk metaethical judgments. Building upon work on the well-known Knobe effect in experimental philosophy,14 in which individuals’ folk psychological attributions have been shown to depend in surprising ways upon the goodness or badness of agents’ actions, the following four descriptions were constructed:
How Different Kinds of Disagreement Impact Folk Metaethical Judgments
181
1. The CEO of a company that helps and preserves the environment believes that it is morally wrong to harm the environment. 2. The CEO of a company that helps and preserves the environment believes that it is not morally wrong to harm the environment. 3. The CEO of a company that harms and pollutes the environment believes that it is morally wrong to harm the environment. 4. The CEO of a company that harms and pollutes the environment believes that it is not morally wrong to harm the environment. In (1) and (2), the CEO is depicted doing something morally good, namely, helping and preserving the environment, whereas the CEO’s actions in (3) and (4) are morally bad. In (1) and (3), the CEO is described as having a morally good belief about the environment, namely, that it should not be harmed; in (2) and (4), the CEO has the corresponding morally bad belief. The crossing of good and bad actions with good and bad beliefs results in the actions and beliefs of the CEO being congruent in (1) and (4) and incongruent in (2) and (3). Participants were first asked to indicate in a forced-choice format whether they agreed or disagreed with the CEO’s belief. They were then asked, “If someone disagreed with the CEO about whether it is morally wrong to harm the environment, would it be possible for both of them to be correct or must at least one of them be mistaken?.” Participants were then directed to choose between “It is possible for both of them to be correct” and “At least one of them must be mistaken.”
Results The results of Study 5 are summarized in Figure 9.2. Participants were more inclined to attribute objectivity to the ethical beliefs in question when the protagonist performed morally bad actions than when he performed morally good ones. This difference was significant, c2 (1, N ⫽ 160) ⫽ 5.013, p ⬍ 0.05, Cramér’s V ⫽ 0.18. Neither belief valence nor the congruence between action and belief significantly affected folk metaethical judgments. However, it is noteworthy that the highest proportion of objectivity attributions was obtained in the “double bad” (i.e., Bad Action/
182
Advances in Experimental Moral Psychology Judgment valence
Proportion of objectivity attributions
1.0
*
Good Bad
0.5
0.0 Good
Bad Action valence
Figure 9.2 Mean objectivity attributions in the Good Action/Good Belief (0.55), Good Action/Bad Belief (0.43), Bad Action/Good Belief (0.57), and Bad Action/Bad Belief (0.75) conditions of Study 5.
Bad Belief) condition, since it is badness (rather than goodness or neutrality) that has been shown to be the driving force behind the various forms of the Knobe effect.
Discussion As with other findings from the Knobe effect literature, the moral valence of a protagonist’s action significantly affected participants’ responses to probe questions. However, unlike other results in this literature, the responses in question were not folk psychological ascriptions. They were second-order attributions of objectivity to ethical beliefs held by the protagonist. These results provide further evidence that individuals’ assessments of metaethical disagreements are significantly affected by a variety of factors in the situation of disagreement.
How Different Kinds of Disagreement Impact Folk Metaethical Judgments
183
General discussion The foregoing studies show (i) that making disagreement salient to participants before asking them to make metaethical judgments can decrease objectivist responses, (ii) that increasing the concreteness of the situation of disagreement participants are directed to consider can increase objectivist responses, and (iii) that the moral valence of the actions performed by agents whose ethical beliefs participants are asked to consider affected attributions of objectivity to those beliefs. Because philosophical discussion—whether in the classroom or at professional conferences—often takes place in a somewhat rarefied atmosphere of abstractions, philosophers should be aware that intuitive agreement or disagreement with their metaethical claims can be affected by the very abstractness of those situations and that the amount of agreement or disagreement they encounter might be different in other situations. In spite of the fact that an increasing number of philosophers are familiar with the Knobe effect and its seemingly unlimited range of applicability, many philosophers continue to give little thought either to the moral valence of the actions depicted in their favored thought experiments and/or to the consequences this might have. An important question raised by the studies reported above concerns the coherence of folk metaethical commitments. Most philosophers assume that the correct semantics for ordinary ethical judgments must show them to be uniformly objective or subjective.15 Yet, Studies 2 through 5—in addition to work by Goodwin and Darley (2008), Beebe and Sackris (under review), and Sarkissian et al. (2011)—reveal that there are several kinds of variation in folk metaethical judgments. The lack of uniformity in the objectivity attributed to ethical claims might make us wonder how well ordinary individuals grasp the ideas of objectivism and subjectivism (and perhaps the related ideas of relativism and universalism). It might also lead us to question their reasoning abilities. Goodwin and Darley (2008, 1358, 1359), for example, suggest that “individuals were not particularly consistent in their meta-ethical positions about various ethical beliefs” and that “requirements of judgmental consistency across ethical scenarios are not considered.” However, this attribution of inconsistency seems both uncharitable and unwarranted. Why should we believe that the ordinary use of ethical terms requires a semantics that assumes uniform objectivity or subjectivity? Because
184
Advances in Experimental Moral Psychology
armchair philosophers who have gathered no empirical evidence about the actual practice of using ethical terms say so? It seems that the practice should dictate the semantics, and not the other way around. If we find variability in the practice, we should look for semantic theories that can accommodate such variation. Furthermore, a variety of semantic theories can do so. For example, in Beebe (2010) I offer a relevant alternatives account of ethical judgments that borrows heavily from the semantic machinery of the epistemic contextualists (e.g., Lewis 1996; DeRose 2011). I argue that treating ethical terms as context-sensitive yields a better interpretation of ordinary normative and metaethical judgments than interpretations that treat them as context-invariant. Without delving into the details of the view, the upshot for present purposes is that attributions of inconsistency or incoherence to folk metaethical practice are premature when there are more charitable interpretive options available. Another important issue raised by the above studies concerns my hypothesis that it is concreteness that is driving the effects observed in Studies 3 and 4. An alternative possibility is that when undergraduates at the University at Buffalo are told that Madeline B., a senior biology major at UB, believes that some action is morally permissible, it may be Madeline’s cultural proximity or group affiliation that leads participants to make more objectivist judgments. Signaling that someone from the same university believes that p may suggest to participants that they should believe it as well, if they are to remain members in good standing in the relevant group. And it is of course possible that some other kind of social influence might be operative as well. Further research is required to determine whether it is concreteness or other social factors that push individuals in the direction of greater objectivism.16 The studies reported above show that not only are there differences in folk metaethical judgments that track the content of ethical claims (Goodwin and Darley 2008; Beebe and Sackris, under review; Study 1), how contested they are (Goodwin and Darley 2012; Study 2), and the cultural distance between disagreeing parties (Sarkissian et al. 2011); there are also differences that track the goodness or badness of disagreeing parties (Study 5) and possibly their concreteness as well (Studies 3 and 4). It is hoped that the present research sheds useful light on the multidimensional variation that characterizes the folk metaethical landscape.
How Different Kinds of Disagreement Impact Folk Metaethical Judgments
185
Notes * James R. Beebe, Department of Philosophy, University at Buffalo. The author would like to thank Mike Giblin and Anthony Vu for assistance in data collection and Hagop Sarkissian and Jen Cole Wright for helpful comments on a previous version of this manuscript. Correspondence should be addressed to James R. Beebe, 135 Park Hall, Department of Philosophy, University at Buffalo, Buffalo, NY 14260. Email: [email protected]. 1 However, cf. Beebe and Sackris (under review, sec. 1) for critical discussion of many of the measures of metaethical commitments that are employed in the published literature. 2 Forty-one percent of participants in one study and 38 percent of participants in another classified first trimester abortion as a personal rather than a moral issue; 89 percent and 73 percent of participants did the same for anonymously donating money to charity. 3 All statistical tests reported in this chapter are chi-square tests of independence. On the conventional interpretation of Cramér’s V, an effect size of 0.1 to 0.29 counts as small, one 0.3 to 0.49 counts as medium, and one 0.5 or larger counts as large. 4 The only gender differences in the data were that females held slightly less strong opinions than males on factual matters (c2 (2, N ⫽ 303) ⫽ 6.124, p ⬍ 0.05, Cramér’s V ⫽ 0.14) and reported greater societal disagreement than males concerning matters of taste (c2 (2, N ⫽ 225) ⫽ 11.296, p ⬍ 0.05, Cramér’s V ⫽ 0.22). 5 Cf. Sidgwick (1907/1981, 342), Mackie (1977, 36–8), Wong (1984), and Tersman (2006). Because salient disagreement impacted participants’ second-order (metaethical) judgments in Study 2, a follow-up study was performed to see if salient disagreement might have a similar impact upon participants’ first-order judgments—that is, upon the degree of agreement they expressed in response to various ethical claims in Task 1. Participants were directed to complete Task 3 immediately before Task 1, and it was hypothesized that salient disagreement would result in less confident Task 1 judgments. However, this manipulation failed to have a significant impact on participants’ Task 1 judgments. 6 Nahmias et al. (2007) found that this was especially true if wrongdoing is involved. 7 Thanks to Mark Alfano for bringing this work to my attention. 8 Participants were required to reside in the United States and to have at least a 95 percent approval rating on more than 500 mturk tasks. Each participant was paid $.30.
186
Advances in Experimental Moral Psychology
9 Untrustworthy faces were associated with potentially harmful intentions, while dominant faces were associated with the capacity to cause harm. 10 Oosterhof and Todorov constructed the faces using FaceGen Modeller 3.2 (Singular Inversions 2007). 11 This phrase of course did not appear in the No Face condition. 12 It may be that how people respond to these kinds of faces depends upon whether they themselves are dominant, submissive, etc. The method of the present study did not allow this factor to be explored. Thanks to Hagop Sarkissian for raising this point. 13 Participants were required to reside in the United States and to have at least a 95 percent approval rating on more than 500 mturk tasks. Each participant was paid $.30. 14 See Alfano et al. (2012) for an overview of this literature. 15 See Sinnott-Armstrong (2009) for further discussion of this point. 16 Thanks to Hagop Sarkissian and Jen Cole Wright for pressing these points.
References Alfano, M. (2010). Social cues in the public good game. Presentation at KEEL 2010 Conference: How and Why Economists and Philosophers Do Experiments, Kyoto Sangyo University, Kyoto, Japan. Alfano, M., Beebe, J. R., and Robinson, B. (2012). The centrality of belief and reflection in Knobe effect cases: A unified account of the data. The Monist, 95, 264–89. Bateson, M., Nettle, D., and Roberts, G. (2006). Cues of being watched enhance cooperation in a real-world setting. Biology Letters, 12, 412–14. Beebe, J. R. (2010). Moral relativism in context. Noûs, 44, 691–724. Beebe, J. R., and Sackris, D. (Under review). Moral objectivism across the lifespan. DeRose, K. (2011). The Case for Contextualism: Knowledge, Skepticism, and Context, vol. 1. Oxford: Oxford University Press. Goodwin, G. P., and Darley, J. M. (2008). The psychology of meta-ethics: Exploring objectivism. Cognition, 106, 1339–66. —(2012). Why are some moral beliefs perceived to be more objective than others? Journal of Experimental Social Psychology, 48, 250–6. Lewis, D. K. (1996). Elusive knowledge. Australasian Journal of Philosophy, 74, 549–67. Nahmias, E., Coates, D., and Kvaran, T. (2007). Free will, moral responsibility, and mechanism: Experiments on folk intuitions. Midwest Studies in Philosophy, 31, 214–42.
How Different Kinds of Disagreement Impact Folk Metaethical Judgments
187
Nichols, S., and Knobe J. (2007). Moral responsibility and determinism: The cognitive science of folk intuitions. Noûs, 41, 663–85. Oosterhof, N., and Todorov, A. (2008). The functional basis of face evaluations. Proceedings of the National Academy of Sciences of the USA, 105, 11087–92. Rezlescu, C., Duchaine, B., Olivola, C. Y., and Chater, N. (2012). Unfakeable facial configurations affect strategic choices in trust games with or without information about past behavior. PLoS ONE, 7, e34293. Sarkissian, H., Parks, J., Tien, D., Wright, J. C., and Knobe. J. (2011). Folk moral relativism. Mind & Language, 26, 482–505. Sidgwick, H. (1907/1981). The Methods of Ethics. Indianapolis: Hackett. Sinnott-Armstrong, W. (2009). Mixed-up meta-ethics. Philosophical Issues, 19, 235–56. Small, D. A., and Loewenstein, G. (2003). Helping a victim or helping the victim: Altruism and identifiability. Journal of Risk and Uncertainty, 26, 5–16. —(2005). The devil you know: The effects of identifiability on punitiveness. Journal of Behavioral Decision Making, 18, 311–8. Tersman, F. (2006). Moral Disagreement. Cambridge: Cambridge University Press. Wong, D. B. (1984). Moral Relativity. Berkeley: University of California Press. Wright, J. C., Grandjean, P. T., and McWhite, C. B. (2013). The meta-ethical grounding of our moral beliefs: Evidence for meta-ethical pluralism. Philosophical Psychology, 26(3), 336–61.
10
Exploring Metaethical Commitments: Moral Objectivity and Moral Progress Kevin Uttich, George Tsai, and Tania Lombrozo*
People have beliefs not only about specific moral issues, such as the permissibility of slavery, but also about the nature of moral beliefs. These beliefs, or “metaethical commitments,” have been the subject of recent work in psychology and experimental philosophy. One issue of study has been whether people view moral beliefs in more objectivist or relativist terms (i.e., as more like factual beliefs or more like personal preferences). In this chapter, we briefly review previous research on folk moral objectivism. We then present the results of an experiment that compares two different approaches to measuring moral objectivism (those of Goodwin and Darley 2008, and Sarkissian et al. 2011) and consider the relationship between objectivism and two additional metaethical beliefs: belief in moral progress and belief in a just world. By examining the relationships between different metaethical commitments, we can better understand the extent to which such commitments are (or are not) systematic and coherent, shedding light on the psychological complexity of an important area of moral belief and experience. To preview our results, we find that different metaethical beliefs are reliably but weakly associated, with different measures of moral objectivism generating distinct patterns of association with belief in moral progress and belief in a just world. We highlight some of the challenges in reliably measuring metaethical commitments and suggest that the distinctions that have been useful in differentiating philosophical positions may be a poor guide to folk moral judgment.
Exploring Metaethical Commitments
189
Recent work on metaethical beliefs Moral objectivity Moral objectivity is a complex idea with multiple variants and diverse proponents (for useful discussions see Goodwin and Darley 2010; Knobe et al. 2012; Sinnott-Armstrong 2009). For our purposes, to accept moral objectivism is to believe that some moral claims are true in a way that does not depend on people’s decisions, feelings, beliefs, or practices. Thus, to reject the objectivity of moral claims one can either deny that moral claims have a truth value or allow that moral claims can be true, but in a way that does depend on decisions, feelings, beliefs, or practices (e.g., Harman 1975; Sinnott-Armstrong 2009). Non-cognitivism is typically an instance of the former position, and cultural or moral relativism of the latter. Recently, there have been a few attempts to examine empirically what people believe about moral objectivity (Goodwin and Darley 2008, 2010; Forsyth 1980; Nichols 2004; Sarkissian et al. 2011; see Knobe et al. 2012, for review). Goodwin and Darley (2008) asked participants to rate their agreement with statements that were factual, ethical, social-conventional, or about personal taste, and then asked them whether these statements were true, false, or “an opinion or attitude.” For example, one of the ethical statements was “Robbing a bank in order to pay for an expensive holiday is a morally bad action,” while one of the social-conventional statements was that “Wearing pajamas and bath robe to a seminar meeting is wrong behavior.” Responding that these were either true or false was considered a more objectivist response than selecting “an opinion or attitude.” Participants were later asked whether the fact that someone disagreed with them about a given statement meant that the other person was wrong, that neither person was wrong, that they themselves were wrong, or something else entirely. On this measure, responding that one of the two people must be wrong was taken as a more objectivist response. Using a composite of these two measures, Goodwin and Darley found evidence that people treat statements of ethical beliefs as more objective than either social conventions or taste. They also found a great deal of variation in objectivism across both ethical statements and individuals. Strongly held ethical beliefs were seen as more objective than beliefs that people did not hold
190
Advances in Experimental Moral Psychology
as strongly, and those who said they grounded their ethical beliefs in religion, moral self-identity, or the pragmatic consequences of failing to observe norms were more likely to be objectivist about ethical statements. Subsequent work has suggested that variation in objectivist beliefs is not an artifact of variation concerning which issues participants themselves take to be moral, nor of misunderstanding moral objectivism (Wright et al. 2012). More recently, Sarkissian et al. (2011) have argued that relativist beliefs are more prevalent than suggested by Goodwin and colleagues, but that these beliefs are only observed when participants are comparing judgments made by agents who differ from each other in important ways. In their studies, participants were presented with two agents who disagreed about a moral claim and were asked whether one of them must be wrong. For example, participants were asked to imagine a race of extraterrestrial beings called “Pentars” who “have a very different sort of psychology from human beings.” Participants were then presented with a hypothetical case in which a classmate and a Pentar had differing views on a moral case, and were asked to rate their agreement with the statement that “at least one of them must be wrong.” Participants provided more objectivist answers (“one of them must be wrong”) when comparing judgments made by agents from the same culture, but more relativist answers (denying that “at least one of them must be wrong”) when comparing judgments made by agents from different planets (i.e., a human and a Pentar). Sarkissian et al. argue that engaging with radically different perspectives leads people to moral relativism. What are the implications of this research? On the one hand, the findings from Goodwin and Darley (2008) and Sarkissian et al. (2011) suggest that metaethical beliefs are not particularly developed or unquestionably coherent. They certainly challenge the idea that those without philosophical expertise can be neatly classified as “moral objectivists” versus “moral relativists.” Instead, judgments vary considerably depending on the moral claim in question and the way in which objectivism is assessed—in particular, whether a case of disagreement involves similar or dissimilar agents. On the other hand, a growing body of research suggests that moral objectivism is systematically related to aspects of cognition and behavior that go beyond metaethical beliefs. For example, Goodwin and Darley (2012) found that moral claims were judged more objective when there was greater
Exploring Metaethical Commitments
191
perceived consensus. They also found that participants judged those who held opposing beliefs as less moral and harder to imagine interacting with when disagreement concerned a claim that was considered objective (see also Wright et al. in press). Finally, Young and Durwin (2013) found that participants primed to think in more objective terms were more likely to give to charity. These findings, among others, suggest that despite intra- and interpersonal variation in judgments, moral objectivism relates to factual beliefs (e.g., about consensus), attitudes (e.g., tolerance of others), and decisions (e.g., about whether to give to charity). We aim here to better understand the ways in which metaethical beliefs are and are not systematic and coherent by considering the relationship between three different metaethical beliefs: belief in moral objectivism, belief in moral progress, and belief in a just world.
Moral progress A belief in moral progress is a commitment to the idea that history tends toward moral improvement over time. This notion, which postulates a certain directionality in human history, can be contrasted with the notion of mere moral change. Although moral progress has been defended by philosophers in the history of philosophy, notably Marx and Hegel, the notion also finds expression in people’s ordinary thinking. For example, Martin Luther King famously proclaimed, “the arc of the moral universe is long but it bends towards justice” (King 1986). It is worth noting that a belief in a historical tendency toward moral progress can be consistently held while maintaining that moral progress can be imperceptible, occurring over long stretches of time. Sometimes moral improvement can be dramatic and rapid, while at other times not. Thus, belief in a tendency toward moral progress does not require committing to a particular rate of moral progress. Additionally, to hold that there is a basic tendency toward moral progress in human history is also compatible with allowing that these tendencies do not inevitably or necessarily prevail. Believing in some tendency need not require belief in inevitability. For example, one could believe that 6-year-old children tend to grow physically larger (e.g., that a child at 14 years of age will be larger than that very same child at age 6) without claiming that they inevitably or necessarily get physically larger (serious illness
192
Advances in Experimental Moral Psychology
or death could prevent their continuing to grow in size). Likewise, in the case of moral progress, one could still allow that there could be exogenous forces such as environmental and biological catastrophes or foreign invasions that prevent the historical development toward moral progress. One reason to focus on moral progress is that the notion is commonly invoked, reflecting ideas in the broader culture. There is therefore reason to suspect that people have commitments concerning its truth, and it is natural to ask, with philosopher Joshua Cohen (1997), “Do [ideas of moral progress] withstand reflective examination, or are they simply collages of empirical rumination and reified hope, held together by rhetorical flourish?” (p. 93). In particular, we might ask whether moral progress typically involves a commitment to moral objectivism, as objective norms might be thought to causally contribute to progress or simply provide a metric against which progress can be assessed. It is also important to note that the notion of moral progress does not merely contain metaethical content but also a kind of descriptive content: to believe in moral progress involves believing something about the nature of human history and the character of the social world. This suggests that our metaethical beliefs, including beliefs about moral objectivity, do not stand alone, compartmentalized from other classes of beliefs. Not only might they be rationally and causally related to each other, in some cases these beliefs are inseparable, expressing a union between the ethical and the descriptive. Thus, a second reason for our interest in considering moral progress in tandem with moral objectivity is that it may reveal important connections between different types of metaethical beliefs as well as connections between metaethical beliefs and other beliefs (such as descriptive beliefs about consensus, or explanatory beliefs about social phenomena).
Belief in a just world While previous research has not (to our knowledge) investigated beliefs about moral progress directly, there is a large body of research on a related but distinct metaethical belief, “belief in a just world” (e.g., Lerner 1980; Furnham 2003). Belief in a just world refers to the idea that good things happen to good people while bad things happen to bad people. The belief that people experience consequences that correspond to the moral nature of their actions or character
Exploring Metaethical Commitments
193
is potentially consistent with belief in moral progress, although the relationship is complex. For example, it is not obvious that the world is morally improved when a criminal experiences a string of bad luck, unless retribution or the deterrence of future criminal activity is itself the moral payoff. Nonetheless, we focus on belief in a just world as a third metaethical belief for two reasons. First, doing so allows us to examine empirically whether belief in a just world is in fact related to belief in moral progress, and thus relate our novel measures to existing research. Second, investigating a third metaethical commitment can help us differentiate two possibilities: that relationships between metaethical commitments are relatively selective, such that (for example) moral objectivity and moral progress might be related but have no association with belief in a just world, or alternatively, that the relationship reflects a single and more general tendency, such that individuals with strong metaethical commitments of one kind will typically have strong metaethical commitments of all kinds.
Method We present a subset of results from a larger experiment investigating people’s beliefs about moral objectivity using modified versions of both the Goodwin and Darley’s (2008) and Sarkissian et al.’s (2011) measures, as well as people’s beliefs about moral progress and belief in a just world. We also solicited explanations for social changes to investigate the relationship between metaethical beliefs and ethical explanations. In the present chapter, we focus on the relationships between different metaethical beliefs. In ongoing work, we consider the relationship between these beliefs and explanations (Uttich et al. in prep).
Participants Three hundred and eighty-four participants (223 female; mean age ⫽ 33) were recruited from Amazon Mechanical Turk, an online crowd-sourcing platform. Participants received a small payment for their participation. All participants identified themselves as being from the United States and as fluent speakers of English.
194
Advances in Experimental Moral Psychology
Materials and procedure We report a subset of a larger set of experiments concerning the relationship between metaethical beliefs and the use of explanations that cite ethical norms. The full experiment consisted of four main parts: (1) explanation solicitation, (2) moral objectivity measures, (3) moral progress measures and general belief in a just world measure (GBJW), and (4) baseline check on beliefs about the morality of social changes. The ordering of the parts was partially counterbalanced, as detailed below.
Explanation solicitation In the full experiment, participants were presented with a description of a social change and asked to explain it in a few sentences (e.g., “Why was slavery abolished? ”). The changes included the abolition of slavery, women’s suffrage, and the potential legalization of same-sex marriage. Given our present focus on the relationship between different metaethical beliefs, we do not report findings concerning explanation here (see Uttich et al. in prep).
Moral objectivity: Disagreement measure Participants’ views concerning moral objectivity were examined in two different ways. The first involved an adaptation of the “disagreement” method used by both Goodwin and Darley (2008) and Sarkissian et al. (2011). Participants read vignettes where either a person similar to themselves (i.e., from their same time and culture) or a person from another time period (e.g., the eighteenth century) disagreed with an imagined friend of the participant about whether a social fact was morally problematic. The relevant social fact was always matched with that for which participants had been asked to provide an explanation. An example from the slavery condition involving the current time and place is presented below: Imagine a person named Allison, a fairly ordinary student from your town who enjoys watching sports and hanging out with friends. Consider Allison’s views concerning the moral status of the following social institution: Slavery. Allison thinks that slavery is not morally wrong.
Exploring Metaethical Commitments
195
This scenario was matched with one involving a judgment from a different time or place: Imagine the social world of the United States in the eighteenth century. Most people in this time and place view slavery as morally acceptable. The existence of slavery is seen by many as part of the natural social order, slavery is permitted by the law and the slave trade is at its peak, and someone who owns many slaves is esteemed as admirable. An individual, Jessica, from this society (eighteenth-century United States), regards slavery as not morally wrong.1
In both cases, participants were then presented with a friend who disagreed: Imagine that one of your friends thinks that slavery is morally wrong. Given that these individuals (Allison [Jessica] and your friend) have different judgments about this case, we would like to know whether you think at least one of them must be wrong, or whether you think both of them could actually be correct. In other words, to what extent would you agree or disagree with the following statement concerning such a case? “Since your friend and Allison [Jessica] have different judgments about this case, at least one of them must be wrong.”
Participants rated their agreement with this statement on a 1–7-point scale with 1 corresponding to “definitely disagree,” 7 to “definitely agree,” and 4 to “neither agree nor disagree.” Each participant saw one current and one historical vignette, with order counterbalanced across participants.
Moral objectivity: Truth-value measure Participants’ beliefs about moral objectivity were also examined using a method adapted from Goodwin and Darley (2008). Participants were asked whether statements about the moral permissibility of the social facts are true, false, or an opinion. The question prompt was adjusted from the original multiple-choice format used by Goodwin and Darley to a 7-point Likert scale to make it more comparable to the disagreement measure. Thus, participants rated their agreement with statements concerning the moral permissibility of a social practice (e.g., “slavery is not morally wrong”) on a 1–7-point scale with 1 being “is best described as true or false,” 7 “is best described as an opinion,” and 4 “is equally well described as either true/false or as an
196
Advances in Experimental Moral Psychology
opinion.” Participants answered questions concerning moral beliefs for all three historical facts (demise of slavery, women’s suffrage, legalization of same-sex marriage), with the historical fact for which they rated explanations presented first.
Moral progress and belief in a just world measures Participants rated their agreement with eighteen statements intended to measure their metaethical commitments concerning moral progress and belief in a just world. Twelve items were constructed to measure participants’ beliefs in moral progress. The statements examined two dimensions of this belief: whether they concerned something concrete (i.e., moral progress with respect to a particular social practice or area of social life) or abstract (i.e., moral progress in general), and whether progress was described as a tendency or as inevitable. There were three questions for each of the four possible combinations (e.g., three concrete questions about tendency, three abstract questions about tendency, and so on). Participants also evaluated six statements concerning belief in a just world, taken from the GBJW (Dalbert et al. 1987). All eighteen statements are included in Table 10.1. Participants rated the statements on a 1–7-point scale with 1 being “definitely disagree,” 7 “definitely agree,” and 4 “neither agree nor disagree.” The order of all moral progress and GBJW statements was randomized.
Baseline check Participants were also asked for their personal views on whether the three social changes were good or bad. For example, for the slavery fact participants were presented with the following statement: The demise of slavery was a good thing. Participants rated their agreement with this statement on a 1–7-point scale with 1 being “definitely disagree,” 7 “definitely agree,” and 4 “neither agree nor disagree.” All three social facts were rated. The social fact related to the explanation for which each participant had been prompted was always presented first.
Table 10.1 Means and factor loadings for statements of moral progress and belief in a just world. Items with an asterisk were reverse coded.
Mean (SD)
Factor 1 (“Abstract progress”)
Factor 2 (“GBJW”)
Factor 3 (“Concrete inevitability”)
Human history tends to move in the direction of a more perfect moral world.
3.84 (1.60)
0.757
0.212
⫺ 0.001
0.189
As time goes on, humanity does NOT generally become more morally advanced.∗
4.21 (1.68)
0.707
0.011
0.061
0.077
Over time we will move toward a more morally just world.
4.22 (1.60)
0.782
0.184
0.093
0.272
The moral advancement of humanity is NOT inevitable.∗
3.66 (1.76)
0.686
0.238
0.227
⫺ 0.068
It is inevitable that on average our morals will be better than those of our distant ancestors.
3.94 (1.77)
0.743
0.173
0.274
0.091
An increase in moral justice in the world is inevitable.
4.05 (1.69)
0.729
0.340
0.227
0.072
Over time there is moral progress concerning slavery.
5.54 (1.45)
0.159
0.146
0.179
0.813
Over time there is moral progress concerning voting rights.
5.41 (1.35)
0.250
0.066
0.249
0.733
Over time there is moral progress concerning marriage rights.
4.79 (1.60)
0.586
⫺ 0.001
0.290
0.354
Statement
Factor 4 (“Concrete progress”)
Abstract Tendency Exploring Metaethical Commitments
Abstract Inevitability
Concrete Tendency
197
(Continued)
198
Table 10.1 (Continued)
Factor 2 (“GBJW”)
The demise of slavery was inevitable.
5.14 (1.78)
0.262
0.093
0.652
0.162
The extension of the right to vote to women was inevitable.
5.43 (1.66)
0.097
0.186
0.776
0.258
The legalization of same-sex marriage is inevitable.
5.14 (1.82)
0.189
0.008
0.774
0.059
I think basically the world is a just place.
3.77 (1.62)
0.170
0.563
0.116
⫺ 0.109
I believe that, by and large, people get what they deserve.
4.16 (1.62)
0.093
0.705
⫺ 0.017
0.037
I am confident that justice always prevails over injustice.
3.64 (1.75)
0.293
0.723
0.121
⫺ 0.067
I am convinced that in the long run people will be compensated for injustices.
3.95 (1.75)
0.036
0.775
0.108
0.128
I firmly believe that injustices in all areas of life (e.g., professional, family, politic) are the exception rather than the rule.
3.82 (1.52)
0.082
0.620
0.055
0.122
I think people try to be fair when making important decisions.
4.69 (1.46)
0.159
0.607
⫺ 0.006
0.271
Statement
Factor 3 (“Concrete inevitability”)
Factor 4 (“Concrete progress”)
Concrete Inevitability
General Belief in a Just World
Advances in Experimental Moral Psychology
Mean (SD)
Factor 1 (“Abstract progress”)
Exploring Metaethical Commitments
199
Counterbalancing Participants either provided an explanation first (part 1) and then completed the two moral objectivity measures (part 2) and the moral progress measures and GBJW measures (part 3), with the order of parts 2 and 3 counterbalanced, or they first completed the moral objectivity measures (part 2) and the moral progress and GBJW measures (part 3), with order counterbalanced, followed by explanations (part 1). Participants always completed the baseline check on social facts (part 4) last.
Results We begin by reporting the data for each set of questions individually, and then consider the relationship between different metaethical commitments.
Individual measures Baseline check measures The baseline check confirmed our assumptions about participants’ own attitudes toward the moral claims in question. The average ratings were 6.70 of 7 (SD ⫽ 0.95) for the demise of slavery, 6.63 (SD ⫽ 1.00) for women’s suffrage, and 5.15 (SD ⫽ 2.20) for same-sex marriage.
Moral objectivism: Disagreement The first measure of objectivism concerned participants’ responses to disagreement between a friend and an individual in a current or historical period. Overall, participants provided higher ratings for the current scenario (M ⫽ 4.84, SD ⫽ 1.95) than for the historical scenario (M ⫽ 4.62, SD ⫽ 1.97), indicating greater objectivism in the latter case and consistent with Sarkissian et al.’s findings. To analyze the data statistically, we performed a repeatedmeasures ANOVA with time period (current vs. historical) as a withinsubjects factor and social fact (slavery, women’s suffrage, same-sex marriage) as a between-subject factor. This revealed two significant effects: a main effect of time period, F(1,381) ⫽ 13.17, p ⬍ 0.01, with more objectivist responses for the current vignette than for the historical vignette, and a main effect of social
200
Advances in Experimental Moral Psychology
fact, F(2,381) ⫽ 36.35, p ⬍ 0.01, with responses that were more objectivist for slavery (M ⫽ 4.99, SD ⫽ 1.90) and women’s suffrage (M ⫽ 4.90, SD ⫽ 1.69) than for same-sex marriage (M ⫽ 4.30, SD ⫽ 1.95). Because the correlation between participants’ current (C) and historical (H) ratings was very high (r ⫽ 0.817, p ⬍ 0.01), we consider the average rating (C⫹H) for each participant (M ⫽ 4.72, SD ⫽ 1.87) in most subsequent analyses.
Moral objectivism: Truth value Our second measure of moral objectivism was the “true, false, or opinion?” (TFO) measure adapted from Goodwin and Darley (2008). The average rating for the TFO measure was 4.31 (SD ⫽ 2.15), with lower scores indicating greater moral objectivism. This measure varied as a function of social fact, F(2,382) ⫽ 53.65, p ⬍ 0.01, with the most objectivist responses for slavery (M ⫽ 3.71, SD ⫽ 2.58), followed by women’s suffrage (M ⫽ 4.31, SD ⫽ 2.51) and same-sex marriage (M ⫽ 4.91, SD ⫽ 2.30).2 Nonetheless, participants’ ratings across the three social facts were highly related (α ⫽ 0.82). In subsequent analyses, we therefore focus on a given participant’s average TFO rating across the three social facts. To facilitate the interpretation of correlations across measures, we report a reversed average (8 minus each participant’s score) such that higher numbers correspond to greater objectivism, as in the C⫹H measure.
Moral progress and belief in a just world measures We analyzed ratings across the 18 statements with a factor analysis employing principal components analysis as an extraction method and a varimax rotation. This analysis resulted in four factors with eigenvalues over one, accounting for a total of 59.4 percent of the variance. Table 10.1 reports the average rating for each statement as well as the factor loadings for each statement in the rotated component matrix, and suggests that the four factors can be characterized as follows: abstract progress (34.2% of variance), GBJW (11.5%), concrete inevitability (8.1%), and concrete tendency (5.6%). It’s worth noting that beliefs about moral progress were indeed differentiated from GBJW, and that the dimension of abstract versus concrete appeared to be psychologically meaningful while the distinction between tendency and
Exploring Metaethical Commitments
201
inevitability emerged only for the concrete items, where participants may have been able to look back over time at the specific issues we considered to identify both general trends and temporary setbacks.3 In subsequent analyses we examine correlations between these four factors and our two measures of moral objectivism.
Relationships between metaethical commitment measures Table 10.2 reports the correlations between our two measures of moral objectivism (C⫹H and TFO) as well as the four factors extracted from the factor analysis on moral progress items (abstract progress, GBJW, concrete inevitability, concrete tendency). There are several notable results. First, while the correlation between the C⫹H ratings and the TFO ratings was significant (r ⫽ 0.271, p ⬍ 0.01), it was low enough to suggest that each Table 10.2 Correlations between metaethical measures C⫹H
Current
Historical
C-H
Avg TFO
⫺0.020
0.271**
C⫹H: Current/ Historical disagreement
1
0.953**
0.954**
Current
0.953**
1
0.817**
Historical
0.954**
0.817**
1
C-H: Difference score
⫺0.020
0.285**
⫺0.320**
0.285**
0.258**
⫺0.320**
0.258**
1
⫺0.005
Avg TFO: True, False, or Opinion?
0.271**
0.258**
0.258**
⫺0.005
1
Factor 1: “Abstract progress”
0.063
0.059
0.061
⫺0.003
0.127*
Factor 2: “GBJW”
0.042
0.018
0.061
⫺0.072
⫺0.116*
⫺0.040
⫺0.009
⫺0.066
0.094
⫺0.063
0.093
⫺0.056
Factor 3: “Concrete inevitability” Factor 4: “Concrete tendency” *⬍0.05, ** ⬍0.01.
0.239**
0.258**
0.199**
202
Advances in Experimental Moral Psychology
measure captures some important and unique variance in beliefs about moral objectivism, perhaps roughly capturing relativism and non-cognitivism, respectively. To further investigate the possible relationships between measures, we also considered whether TFO might be related to the difference between C and H ratings (C-H), which can be conceptualized as a measure of the extent to which a participant is influenced by sociocultural factors in evaluating the truth of a moral claim. One might therefore expect a significant negative correlation between TFO and C-H, but in fact the relationship was very close to zero. Coupled with the high correlation between judgments on the C and H questions, and the fact that C and H had very similar relationships to other variables, this suggests that varying the sociocultural context for a belief can indeed affect judgments concerning disagreement, but that the effect is more like a shift in the absolute value of participants’ judgments than the recruitment or application of different moral commitments. Second, while both the C⫹H and TFO ratings were related to moral progress and GBJW, they had unique profiles in terms of the specific factors with which they correlated. The C⫹H measure was correlated with the concrete tendency factor (r ⫽ 0.239, p ⬍ 0.01), while the TFO measure was positively correlated with the abstract progress factor (r ⫽ 0.127, p ⬍ 0.05) and negatively correlated with the GBJW factor (r ⫽ ⫺ 0.116, p ⬍ 0.05). Although these correlations were small, they suggest systematic relationships between measures, and more surprisingly, non-overlapping relationships, providing further evidence that judgments of disagreement (C⫹H) and judgments concerning whether moral claims have a truth value (TFO) reflect different facets of folk metaethical commitments. Finally, it’s worth considering why C⫹H and TFO had these distinct profiles. We speculate that the dimension of concrete versus abstract evaluation can partially explain these results. Specifically, C⫹H and the concrete tendency factor were positively associated and involved particular moral claims (e.g., about slavery) rather than abstract claims, while TFO and the abstract progress factor were positively associated and involved judgments that were more explicitly metaethical in that they concerned the status of particular moral ideas (i.e., whether there is moral progress in general and whether particular claims have a truth value). However, this speculation does not explain why the C⫹H measure was not also associated with the concrete tendency factor, nor does it explain the negative association between TFO and the GBJW factor.
Exploring Metaethical Commitments
203
General discussion Our results suggest that metaethical beliefs are varied and complex, with significant but modest relationships across different sets of beliefs. Our results also reinforce some of the conclusions from prior research. Like Goodwin and Darley (2008, 2012), we find significant variation in objectivism across individuals, and also that judgments reflect greater objectivism for some social facts (slavery) than for others (same-sex marriage), perhaps echoing their findings on the role of consensus, and also consistent with the strength of participants’ attitudes concerning each social fact. Like Sarkissian et al. (2011), we find evidence that measures that highlight different perspectives seem to increase non-objectivist responses, as our “historical” vignette generated less objectivist responses than the matched “current” vignette, although the responses were strongly correlated. Our findings therefore support the need to consider the characteristics of both participants and measures in drawing conclusions about metaethical beliefs. Beyond illuminating variation between individuals, our findings shed light on the coherence and variability of metaethical beliefs within individuals. Correlations between our measures of metaethical beliefs suggest two conclusions: that the metaethical concepts we investigate have some common elements, but also that there is only partial coherence in the corresponding beliefs. Our two separate measures of moral objectivity (C⫹H and TFO) were significantly correlated, but only weakly so. The correlation was weak despite modifications from Goodwin and Darley (2008) and Sarkissian et al. (2011) to make the measures more comparable: both involved judgments on 7-point scales and referred to the same moral claims. Analyses of the relationship between these two measures and the four factors concerning moral progress and GBJW suggest that moral objectivism is related to these ideas, but the two measures of objectivism had unique patterns of association. If participants have strong, stable, and consistent metaethical commitments, why might responses to metaethical questions be so weakly related? We first consider methodological and conceptual answers to this question. One possibility is that we observe weak associations between metaethical commitments as an artifact of our methods of measurement. This idea is consistent with a suggestion by Sarkissian et al. (2011), who argue that when
204
Advances in Experimental Moral Psychology
forced to address radically different perspectives, people who appeared to have completely objectivist commitments reveal some underlying, relativist intuitions. It follows that methods for soliciting commitments might themselves account for substantial variance in responses. We suspect there is some truth to this idea, and our particular measures certainly have limitations. Nonetheless, it’s worth repeating that while disagreements in current and historical contexts (C and H) involved different absolute ratings, they were very highly correlated and had matching patterns of association with our other measures. Moreover, the difference between these ratings (C-H)—that is the extent to which context shifted judgments—was not reliably associated with any measures. One interpretation is that people’s absolute judgments may be quite easy to manipulate, but that the relationships between metaethical commitments, while weak, may be more stable. Another possibility is that the metaethical commitments we investigated do not in fact correspond to coherent and unified sets of beliefs. Thus, participants’ judgments could be inconsistent across measures because the philosophical constructs we aim to assess are themselves diverse or incoherent. For example, we expected a stronger relationship between moral objectivism and belief in moral progress, but such a relationship is not logically required—one can, for example, be a relativist and endorse moral progress, or an objectivist and deny it. We also expected our two measures of moral objectivism to be more strongly associated given their association within the philosophical literature and the fact that prior research has simply combined both measures (Goodwin and Darley 2008; Wright et al. in press), but it is logically possible, if unusual, to be (for example) a non-cognitivist universalist (e.g., Hare 1952). While we suspect that conceptual dissociations between metaethical commitments partially explain our results, and that the findings are doubtless influenced by our particular methods of measurement, our results also point to three possible (and mutually consistent) proposals concerning the psychology of metaethical belief. First, as suggested by Wright et al. (in press), it could be that objectivism in the moral domain is tempered by the need to tolerate and effectively interact with others who hold divergent beliefs. On this view, the apparent incoherence in participants’ metaethical commitments serves an important psychosocial function, and we would expect the observed relationship
Exploring Metaethical Commitments
205
between the prevalence of a particular moral belief and an objectivist stance toward it. Second, it could be that people do not hold a single “intuitive theory” of metaethics, but instead hold multiple theories with some moral content. For example, intuitive theories could be organized around general principles (such as fairness vs. justice), moral patients (such as humans vs. non-humans), or particular practices (such as slavery vs. marriage). This idea can help make sense of reliable relationships between metaethical commitments and other beliefs (e.g., descriptive beliefs about consensus, explanatory beliefs), attitudes (e.g., tolerance), and behaviors (e.g., charitable giving) despite only modest associations across different metaethical beliefs. This proposal builds on prior research positing “intuitive theories” across a wide range of domains, where such theories embody somewhat coherent but not full articulated bodies of belief (e.g., Carey 1985; Shtulman 2010; Thagard 1989). In the moral domain, for example, Lombrozo (2009) investigated the relationship between deontological versus consequentialist commitments and found evidence of a systematic but imperfect correspondence across more abstract and explicit versus scenariobased measures. With explicit articulation and examination, as typically occurs with philosophical training, different metaethical commitments could potentially become more reliably associated. Finally, it could be that categories that make sense a priori philosophically play a relatively minor role in driving peoples’ responses, with a much greater role for (arguably philosophically irrelevant) properties, such as whether the question prompts are abstract or concrete. Both our factor analysis—which suggested that the dimension of “abstract” versus “concrete” was more psychologically significant than that between tendency and inevitability—and the patterns of correlations across measures support the importance of this dimension. Along these lines, Nichols and Knobe (2007) found that concrete vignettes about free will elicited compatibilist responses, while abstract vignettes elicited incompatibilist responses. More generally, research on Construal Level Theory suggests that level of abstraction can have important consequences for cognition (Trope and Liberman 2010). This final point should give pause to the assumption that folk morality will have any clean correspondence to extant philosophical categories. Instead, a more bottom-up, data-driven approach to understand folk moral commitments may be more successful.
206
Advances in Experimental Moral Psychology
Further research aimed directly at measuring the nature of metaethical commitments will aid in distinguishing these possibilities and further clarify the status and coherence of folk metaethical commitments. If such commitments don’t correspond to philosophical distinctions that can be motivated a priori, which dimensions of moral experience do they track, and why? These are important questions for future research.
Notes * Author’s Note: Kevin Uttich, University of California—Berkeley, George Tsai, University of California—Berkeley and University of Hawaii, and Tania Lombrozo, University of California—Berkeley. Corresponding author: Kevin Uttich, Email: [email protected], 3210 Tolman Hall, Berkeley, CA, 94720. We thank Nicholas Gwynne, Michael Pacer, Kathie Pham, Jennifer Cole Wright, Hagop Sarkissian and the Moral Psychology group at Berkeley and the Concept and Cognition lab for feedback and data collection assistance. This research was supported by research funds from the McDonnell Foundation. 1 Information on the norms of the time period was added to the historical scenario to ensure that participants were aware of the relevant norms and understood that the scenario takes place before the change in the social fact. 2 We obtained similar results in a separate experiment which used Goodwin and Darley’s original multiple-choice format rather than a Likert scale: 18 “true” responses (6%), 102 “false” responses (35%), and 162 “opinion” responses (56%) out of 288 total responses (96 participants ⫻ 3 social facts). 3 We thank Jennifer Cole Wright for suggesting this interpretation for why the concrete items may have shown a differentiation between “tendency” and “inevitability” while the abstract items did not.
References Carey, S. (1985). Conceptual Change in Childhood. Cambridge, MA: MIT Press. Cohen, J. (1997). The arc of the moral universe. Philosophy and Public Affairs, 26(2), 91–134.
Exploring Metaethical Commitments
207
Dalbert, C., Montada, L., and Schmitt, M. (1987). Glaube an eine ge- rechte welt als motiv: Vali-dierungskorrelate zweier Skalen [Belief in a just world as motive: Validity correlates of two scales]. Psychologische Beitrage, 29, 596–615. Forsyth, D. R. (1980). A taxonomy of ethical ideologies. Journal of Personality and Social Psychology, 39, 175–84. Furnham, A. (2003). Belief in a just world: Research progress over the past decade. Personality and Individual Differences, 34, 795–817. Goodwin, G. P., and Darley, J. M. (2008). The psychology of meta-ethics: Exploring objectivism. Cognition, 106, 1339–66. —(2010). The perceived objectivity of ethical beliefs: Psychological findings and implications for public policy. Review of Philosophy and Psychology, 1, 161–88. —(2012). Why are some moral beliefs seen as more objective than others? Journal of Experimental Social Psychology, 48, 250–6. Hare, R. M. (1952). The Language of Morals. New York: Oxford University Press. Harman, G. (1975). Moral relativism defended. Philosophical Review, 84, 3–22. King, M. L., Jr. (1986). A Testament of Hope: The Essential Writings of Martin Luther King, Jr. In J. Washington (ed.). San Francisco: Harper and Row. Knobe, J., Buckwalter, W., Nichols, S., Robbins, P., Sarkissian, H., and Sommers, T. (2012). Experimental philosophy. Annual Review of Psychology, 63, 81–99. Lerner, M. (1980). The Belief in a just World: A Fundamental Delusion. New York: Plenum. Lombrozo, T. (2009). The role of moral commitments in moral judgment. Cognitive Science, 33, 273–86. Nichols, S. (2004). After objectivity: An empirical study of moral judgment. Philosophical Psychology, 17, 3–26. Nichols, S., and Knobe, J. (2007). Moral responsibility and determinism: The cognitive science of folk intuitions. Nous, 41, 663–85. Sarkissian, H., Park, J., Tien, D., Wright, J. C., and Knobe, J. (2011). Folk moral relativism. Mind & Language, 26, 482–505. Shtulman, A. (2010). Theories of god: Explanatory coherence in a non-scientific domain. In S. Ohlsson and R. Catrambone (eds), Proceedings of the 32nd Annual Conference of the Cognitive Science Society. Austin, TX: Cognitive Science Society, pp. 1295–300. Sinnott-Armstrong, W., Moral skepticism. The Stanford Encyclopedia of Philosophy (Summer 2009 Edition), Edward N. Zalta (ed.), URL ⫽ http://plato.stanford.edu/ archives/sum2009/entries/skepticism-moral/. Thagard, P. (1989). Explanatory coherence. Behavioral and Brain Sciences, 12, 435–502.
208
Advances in Experimental Moral Psychology
Trope, Y., and Liberman, N. (2010). Construal-level theory of psychological distance. Psychological review, 117(2), 440. Uttich, K., Tsai G., and Lombrozo, T. (2013). Ethical Explanations, Moral Objectivism, and Moral Progress. Manuscript submitted for publication. Wright, J. C., Grandjean, P. T., and McWhite, C. B. (2013). The meta-ethical grounding of our moral beliefs: Evidence for meta-ethical pluralism, Philosophical Psychology, 26(3), 336–61. Wright, J. C., McWhite, C. B., and Grandjean, P. T. (in press). The cognitive mechanisms of intolerance: Do our meta-ethical commitments matter? In J. Knobe, T. Lombrozo, and S. Nichols (eds), Oxford Studies in Experimental Philosophy, Vol. 1, New York, NY: Oxford University Press. Young, L., and Durwin, A. (2013). Moral realism as moral motivation: The impact of meta-ethics on everyday decision-making. Journal of Experimental Social Psychology, 49(2013), 302–6. doi:10.1016/j.jesp.2012.11.013
11
Agent Versus Appraiser Moral Relativism: An Exploratory Study Katinka J. P. Quintelier, Delphine De Smet, and Daniel M. T. Fessler*
Theories of moral relativism do not always fit well with common intuitions. In the Theaetetus, Plato ridiculed the relativist teachings of Protagoras (Plato 1921), and Bernard Williams dubbed moral relativism “possibly the most absurd view to have been advanced even in moral philosophy” (Williams 1972, p. 20). Nonetheless, even though some moral philosophers oppose theories of moral relativism due to its counterintuitive implications (e.g., Streiffer 1999), other philosophers defend it by referring to common intuitions, lay people’s speech acts, or common understandings of certain moral terms (e.g., Brogaard 2008; Harman 1975; Prinz 2007). These intuitions have been investigated empirically: On the one hand, empirical studies suggest that survey respondents can make relativist moral judgments (Goodwin and Darley 2008; Sarkissian et al. 2012; Wright et al. 2012; Wright et al. in press). On the other hand, the folk’s moral relativist intuitions might be self-contradictory (cf. Beebe 2010), and this can be used as an argument against relativist moral theories (Williams 1972). If the prevalence and coherence of folk moral relativism are to play a role in arguments regarding the philosophical merits of moral relativism, then we need to know what the folk actually adhere to. In this regard, for several reasons, it is important to take into account that there are different kinds of moral relativism (Quintelier and Fessler 2012). First, failure to do so may lead to an underestimation of the prevalence of folk moral relativism, as respondents may employ relativist intuitions of a kind other than that being
210
Advances in Experimental Moral Psychology
measured. Second, some kinds of moral relativism might be more coherent than others (Quintelier and Fessler 2012). An important distinction that has received attention in recent philosophical work is agent versus appraiser relativism (Beebe 2010; Lyons 1976/2001; Prinz 2007). As we further explain in the next two sections, this distinction refers to the individual or group of individuals toward whom a moral statement is relativized. Agent moral relativism states that the appropriate frame of reference is the moral framework of the person who performs the act, or of the cultural group to which this person belongs. Appraiser moral relativists state that the appropriate frame of reference is the moral framework of the person who makes the moral judgment, or of the cultural group to which this person belongs. Contemporary empirical work on moral relativism has largely failed to investigate both (a) this critical distinction between agent and appraiser moral relativism, and (b) the corresponding comparative intuitive appeal of each type of moral relativism. Here, we explore the existence of both agent and appraiser moral relativist intuitions in lay people. Below, we briefly define moral relativism, after which we explain in more detail the difference between agent and appraiser moral relativism. In the main section, we describe our study and report our findings. Finally, we discuss the possible implications of our findings.
Moral relativism We construe the notion of moral relativism as consisting of three necessary components: First, X is relative to Y; second, X is an aspect of the moral phenomenon; and third, there is variation in Y that cannot be eliminated, as illustrated below (cf. Quintelier and Fessler 2012). First, we take moral relativism to mean that some aspect of moral statements (e.g., their truth, their referent) or morally relevant acts (e.g., their moral rightness) is relative to a moral framework (cf. Harman and Thompson 1996). By a moral framework, we mean a set of moral values, principles, or sentiments that play a justifying role in one’s moral reasoning (e.g., one justifies moral judgments by referring to this framework). Consider the following example, inspired by Lyons (1976/2001): Assume that pro-choice activists endorse a moral framework that prioritizes the value
Agent Versus Appraiser Moral Relativism: An Exploratory Study
211
of personal choice over the value of the unborn life. According to some kinds of moral relativism, a pro-choice activist—say, Christine—can truthfully judge that abortion is permissible because it is in accordance with her moral framework. Nonetheless, if a pro-life activist—say, Lisa—abhors abortion, Lisa’s statement regarding the impermissibility of abortion is also true because it is in accordance with Lisa’s moral framework that prioritizes the value of the unborn life over personal choice. In this example, the truth of moral statements thus depends on the moral framework of the person uttering a moral statement. Second, moral relativism holds that there is variation between these moral frameworks. In our example, some people are pro-choice and others are prolife. People’s moral judgments will therefore sometimes differ because their respective moral frameworks differ. Finally, moral relativism rests on philosophical assumptions, such that this variation in moral frameworks cannot be eliminated. For instance, one can hold that both frameworks are equally true, that there is no truth about the matter, or that they are equally practical, etc. If moral relativism would allow that all variation in moral frameworks could be eliminated, moral relativism would be compatible with (most forms of) moral universalism. This meaning of moral relativism would be too broad for our purposes.
Agent versus appraiser moral relativism The above picture leaves out an important component of moral relativism, namely, whose moral framework matters in deciding whether a moral statement is true or not: Does it matter who is evaluating the act, does it matter who is doing the act, or both? Another example can illustrate this: Both Carol, a prochoice activist, and Laura, a pro-life activist, are having an abortion. They are agents performing an act with moral relevance. Lisa (the other pro-life activist) and Christine (pro-choice activist) again contemplate these actions and utter their respective judgments: Lisa says neither abortion is permissible, while Christine says both abortions are permissible. They are appraisers evaluating the act. Which moral statement is true now? Should we assign truth values based on the moral frameworks of the agents performing the act—this being
212
Advances in Experimental Moral Psychology
Carol and Laura—or based on the moral frameworks of the appraisers judging the act—this being Lisa and Christine? Or could any moral framework be an appropriate frame of reference? According to agent moral relativism, the appropriate frame of reference is the agent’s moral framework. In this example, it would be permissible for Carol (the pro-choice agent) to have an abortion, but it would not be permissible for Laura (the pro-life agent) to have an abortion. Viewed from the perspective of agent moral relativism, Christine’s evaluative statement that both abortions are permissible is false, even though this statement is in accordance with her own moral framework. In contrast, for an agent moral relativist, it would be correct for an appraiser, such as Christine, to disapprove of Laura’s abortion (as inconsistent with Laura’s own moral perspective) and to permit Carol’s abortion (as consistent with Carol’s own moral perspective). In contrast, according to appraiser relativism, the moral frameworks of the agents (Laura and Carol) are irrelevant for a moral judgment to be true or false. What matters instead are the moral frameworks of the appraisers, Christine, and Lisa. Viewed from the perspective of appraiser moral relativism, Christine’s evaluative statement that both abortions are permissible is correct, even though abortion is against Laura’s (the agent’s) framework. In what follows, we consider appraisers as only those who evaluate a moral act without being involved in the act. We consider agents as only those doing the act without uttering a statement about the act. Thus, considering the act of lying, when A lies to B, A and B are not appraisers. Of course, in reality, agents can appraise their own actions. Moreover, when appraisers are uttering a moral statement—for example, C says to D that lying is wrong—they might in the first place have themselves as agents in mind; thus, appraisers can also be agents. However, simplifying matters this way will make it easier to investigate whether lay people indeed draw a distinction between agents and appraisers when assessing the status of moral statements and behavior.
Importance of the agent-appraiser distinction The distinction between agent moral relativism and appraiser moral relativism is important when evaluating moral theories. One possible argument against
Agent Versus Appraiser Moral Relativism: An Exploratory Study
213
moral relativism is that it has counterintuitive implications (e.g., Williams 1972). Moral relativism is often taken to imply that at least some moral statements are true or false depending on the appraiser. In the above example, this would mean that it is true (for Christine) that Carol’s abortion is permissible, while it is true (for Lisa) that Carol’s abortion is not permissible. As a consequence, conflicting moral statement can both be true at the same time, which is hard to reconcile with common intuitions. Moreover, according to appraiser moral relativism, Christine cannot reasonably say that Lisa is wrong. However, most people do admonish others when they utter apparently conflicting moral statements. Thus, the moral speech acts of most people are not in line with moral relativism. While this argument against moral relativism holds for appraiser relativism, it does not necessarily hold for agent relativism. According to agent moral relativism, each moral statement about a specific act performed by a specific agent is either true or false, irrespective of who is appraising the act. In the above example, Carol’s abortion is permissible, irrespective of who is judging Carol. As a consequence, conflicting moral statements are not both true at the same time, and it is not unreasonable for discussants to admonish those who utter conflicting moral statements. This is easier to reconcile with common intuitions. Also, agent moral relativism is easier to reconcile with certain existing social practices than appraiser moral relativism. According to agent group moral relativism, the appropriate frame of reference is the moral framework of the group the agent belongs to. This is akin to cultural relativism: an act is right or wrong depending on the moral values that prevail in the culture in which the act takes place. Cultural relativism has been vehemently defended in the past, and moderate forms of cultural relativism—where the wrongness of at least some, but not all, acts depends on the culture in which the act takes place—are still defended and applied in public policy. For instance, in Belgium, it is illegal to kill animals without previously anesthetizing them. However, the same does not hold for religious groups when ritually slaughtering animals. Thus, whether an act is legally right or wrong depends on the group performing the act. Such policies are put in practice for at least some moral issues, and people seem to be able to relativize their practical judgments to the moral frameworks of agents.
214
Advances in Experimental Moral Psychology
In contrast, according to appraiser group moral relativism, the appropriate frame of reference is the moral framework of the group to which the appraiser belongs. In the case of slaughtering animals, everyone would judge the ritual and non-ritual practices according to their own moral framework, and all these conflicting judgments would be true. This is hard to reconcile with the observation that, in fact, the relevant conflicting judgments were discussed by politicians and pundits in Belgium until a consensus was reached, and an agent group moral relativist solution was adopted. Thus, those who reject moral relativism because of its counterintuitive implications should clarify what kind of moral relativism they have in mind: appraiser moral relativism might well be counterintuitive in ways that agent moral relativism is not, and, of course, agent moral relativism might be counterintuitive in ways that appraiser moral relativism is not.
Previous evidence for folk moral relativism Existing studies about folk moral relativism most often vary only the appraisers. To date, investigators have yet to examine whether participants also reveal agent relativist intuitions in experimental studies. Goodwin and Darley (2008) and Wright et al. (2012; in press) both report the existence of relativist moral positions. In these studies, participants are presented with statements such as “Before the 3rd month of pregnancy, abortion for any reason (of the mother’s) is acceptable.” Some participants indicated that the statement was true (or false) but that a person who disagrees with them about the statement need not be mistaken. Hence, in these studies, participants allowed the truth value of a moral statement to vary when the appraiser varied. We do not know if participants would also allow the truth of a moral statement, or the rightness of an act, to vary when the agent would vary. Sarkissian et al. (2011) were able to guide participants’ intuitions in the direction of moral relativism by varying the cultural background of the appraisers. They also varied the cultural background of the agents, but this did not have an effect on participants’ intuitions. However, this apparent null result is subject to the methodological limitation that the cultural backgrounds of the hypothetical agents were much more similar to each other
Agent Versus Appraiser Moral Relativism: An Exploratory Study
215
(an American vs. an Algerian agent) than were the cultural backgrounds of the hypothetical appraisers (a classmate vs. an appraiser from a fictitious primitive society, or vs. an extraterrestrial). Because the above studies do not allow us to conclude whether the folk show agent relativist moral speech acts, we developed scenarios in which we explicitly varied the moral frameworks of both agents and appraisers.
Method We tested whether manipulating the moral frameworks of agents would have an effect on lay people’s moral speech acts. We asked participants about their own moral judgments about moral acts performed by agents holding various moral frameworks. We then also tested whether manipulating the moral frameworks of agents and appraisers would make a difference. To do so, we asked the same participants to assess the truth of others’ moral judgments about moral scenarios. These moral statements were uttered by appraisers who held different moral frameworks. Moreover, these statements evaluated the acts of moral agents that also held different moral frameworks.
Participants From April to June 2013, we recruited participants using Amazon.com’s Mechanical Turk web-based employment system (hereafter MTurk). This is a crowdsourcing website that allows people to perform short tasks, including surveys, for small amounts of money. Anyone over 18 could participate. We analyzed data from 381 participants, who were mostly from the United States (234) and India (118).
Materials and design We developed two questionnaires featuring agents and appraisers. All participants completed only one of the two questionnaires. The first questionnaire featured employees in a firm where, as a punishment, reducing the time allowed for lunch was either against, or in accordance with, the employees’
216
Advances in Experimental Moral Psychology
moral framework. The second questionnaire featured sailors on a ship in international waters, where whipping as a punishment was either against, or in accordance with, the sailors’ moral frameworks. The “sailors” questionnaire was a modified version of a questionnaire previously employed in related research; see Quintelier et al. (2012) for a description of this instrument. In this chapter, we therefore give a detailed overview of the “employees” questionnaire first, followed by only a short description of the “sailors” questionnaire. The full text of both questionnaires is available upon request. In order to investigate whether participants’ moral intuitions vary depending on the agents’ moral frameworks, participants were first presented with two scenarios describing the same act, done by different agents. In one scenario, the act was concordant with the agents’ own moral framework and in the other scenario, the act was discordant with the agents’ own moral framework. After each scenario, we asked participants to morally judge this act. In order to check whether they had understood the particular vignette, we also asked them to complete two comprehension questions. Because the order of presentation of the two scenarios might unintentionally shape responses due to priming or similar psychological effects that are orthogonal to the phenomena in which we are interested here, the order of presentation was randomized across participants. This generated relatively equal subsamples that differed by order of presentation, allowing us to control for order effects in the final analysis. The “employees” questionnaire consisted of the following scenarios:
Scenario 1 Mr Jay is the boss of family business J in a small town in the Midwestern United States. In this company, when employees are late for work, their wages are reduced by a proportionate amount. As a consequence, everyone in this company has come to think that a proportionate wage reduction is a morally right punishment for being late for work. They think reducing lunch breaks as a punishment is morally wrong because this is never done and they value their lunch breaks. One day, John is late for work. This day, his boss is not in the mood to deal with administrative issues such as adjusting John’s wages. Instead, he punishes John by shortening his lunch break, even though Mr Jay himself, John, and all the other employees in this company think this is morally wrong.
Agent Versus Appraiser Moral Relativism: An Exploratory Study
217
Because this punishment is discordant with the agents’ moral framework, we refer to this scenario as “AD.” Participants then answered the following judgment question on a 5-point Likert scale: “To what extent do you think Mr Jay’s behavior is morally wrong?” (1 ⫽ certainly morally wrong; 5 ⫽ certainly not morally wrong). The higher the participants’ scores, the more their judgment was discordant with the agent’s moral framework. Participants then answered two comprehension questions to check if they had read and understood the scenario.
Scenario 2 Mr May is the boss of another family business M in the same small town in the Midwestern United States. In this company, when employees are late for work, their lunch break is proportionately shortened. As a consequence, everyone in this company has come to think that a proportionately shorter lunch break is a morally right punishment for being late for work. They think that reducing wages as a punishment is morally wrong because this is never done and they value their income. One day, Michael is late for work. His boss punishes Michael by shortening his lunch break. Mr May himself, Michael, and all the other employees in this company think that this is morally right. Because this punishment is concordant with the agents’ moral framework, we refer to this scenario as “AC.” Participants then answered the following judgment question on a 5-point Likert scale: “To what extent do you think Mr May’s behavior is morally wrong?” (1 ⫽ certainly morally wrong; 5 ⫽ certainly not morally wrong). Thus, the higher the participants’ scores, the more their judgment was concordant with the agents’ moral frameworks. Participants again answered two comprehension questions. In order to test whether participants’ moral judgments depended on the agents’ moral frameworks, we used AC and AD as within-subject levels of the variable AGENT. In order to test whether participants’ moral intuitions varied depending on the appraisers’ and the agents’ moral frameworks, participants were presented with two additional scenarios, presented in randomized order, that extend
218
Advances in Experimental Moral Psychology
the previous scenarios through the addition of appraisers who utter a moral statement about the act.
Scenario 3 James and Jared are employees in Mr Jay’s company. They both know that in Mr May’s company, everyone thinks shortening lunch breaks is morally right. Of course, in their own company, it is just the other way around: Everybody in Mr Jay’s company, including James and Jared, think that shorter breaks are a morally wrong punishment, and that wage reduction is a morally right punishment. James and Jared have heard that Mr May shortened Michael’s lunch break. James says to Jared: “What Mr May did was morally wrong.” This statement is discordant with the agents’ moral framework and concordant with the appraisers’ moral framework. We therefore label this scenario “AGDAPC.” Participants answered the following question: “To what extent do you think that what James says is true or false?” (1 ⫽ certainly true; 5 ⫽ certainly false). Thus, the higher the participants’ scores, the more that their truth evaluation was concordant with the agents’ moral frameworks but discordant with the appraisers’ moral frameworks. Since this is at odds with the scenario label, we reverse coded this item. Thus, for the final variable that was used in the analysis, higher scores indicate that the response was more discordant with the agents’ moral frameworks and more concordant with the appraisers’ moral frameworks. Participants answered one comprehension question. Participants were then led to the following text: Now Jared replies to James: “No, what Mr. May did was not morally wrong.” This statement is concordant with the agents’ moral framework and discordant with the appraisers’ moral framework. We therefore label this scenario “AGCAPD.” Participants answered the following question: “To what extent do you think that what Jared says is true or false?” (1 ⫽ certainly true; 5 ⫽ certainly false). Again, we reverse coded this item. For the final variable that was used in the analysis, higher scores indicate that the response was more concordant with the agents’ moral frameworks and more discordant with the appraisers’ moral frameworks, in line with the label for this scenario. Participants answered one comprehension question.
Agent Versus Appraiser Moral Relativism: An Exploratory Study
219
Scenario 4 Mark and Matthew are employees in Mr May’s company. They both know that in their own company, everybody, just like Mark and Matthew themselves, thinks that reducing wages is a morally wrong punishment, and that shortening lunch breaks is a morally right punishment. Mark and Matthew have heard that Mr May shortened Michael’s lunch break. Mark says to Matthew: “What Mr. May did was morally wrong.” This statement is discordant with the agents’ moral framework and discordant with the appraisers’ moral framework. We therefore label this scenario “AGDAPD.” Participants answered the following question: “To what extent do you think that what Mark says is true or false?” (1 ⫽ certainly true; 5 ⫽ certainly false). Higher scores on this statement indicate that the participant’s truth evaluation was more concordant with both the appraisers’ and the agents’ moral frameworks. We reverse coded this item. For the final variable that was used in the analysis, higher scores indicate that the response was more discordant with the agents’ moral frameworks and more discordant with the appraisers’ moral frameworks, in line with the label for this scenario. Participants answered one comprehension question. Participants were then led to the following text: “Now Matthew replies to Mark: No, what Mr. May did was not morally wrong.” This statement is concordant with the agents’ moral framework and concordant with the appraisers’ moral framework. We therefore label this scenario “AGCAPC.” Participants answered the following question: “To what extent do you think that what Matthew says is true or false?” (1 ⫽ certainly true; 5 ⫽ certainly false). We reverse coded this item, such that higher scores indicate that the response was more concordant with the agents’ moral frameworks and more concordant with the appraisers’ moral frameworks, in line with the label for this scenario. Participants again answered one comprehension question. Participants thus had to indicate the truth of four moral statements. The variable AGENT TRUTH consists of the following two within-subject levels: AGCAPC⫹AGCAPD and AGDAPC⫹AGDAPD. The variable APPRAISER TRUTH consists of the following two levels: AGCAPC⫹AGDAPC and AGCAPD⫹AGDAPD. The “sailors” questionnaire featured the following scenario:
220
Advances in Experimental Moral Psychology
Scenario 1 Mr Johnson is an officer on a cargo ship in 2010, carrying goods along the Atlantic coastline. All the crew members are American but the ship is mostly in international waters. When a ship is in international waters, it has to follow the law of the state whose flag it sails under and each ship can sail under only one flag. This ship does not sail under the US flag. The law of this ship’s flag state allows both whipping and food deprivation as a punishment. On this ship, food deprivation is always used to discipline sailors who disobey orders or who are drunk on duty; as a consequence, everyone on this ship, Mr Johnson as well as all the sailors, has come to think that food deprivation is a morally permissible punishment. Whipping, however, is never used to discipline sailors and everyone on this ship. Mr Johnson, as well as all the sailors, thinks whipping is a morally wrong punishment. One night, while the ship is in international waters, Mr Johnson finds a sailor drunk at a time when the sailor should have been on watch. After the sailor sobers up, Mr Johnson punishes the sailor by giving him 5 lashes with a whip. This does not go against the law of the flag state. Subsequent scenarios, experimental and comprehension questions were analogous to the “employees” questionnaire: As in the “employees” questionnaire, there were eight comprehension questions and six experimental questions.
Results In order to ensure that participants read and understood the scenarios, we only retained those participants that answered all eight comprehension questions correctly. We analyzed the data from the two questionnaires separately. We analyzed data from 272 participants (50.4% women) for “employees” and 109 participants (51.4% women) for “sailors.” For some analyses, the total number of participants was lower due to missing values. For “employees,” mean age was 34.92 years (SD ⫽ 12.42), ranging from 19 to 75 years old. For “sailors,” mean age was 35.63 years (SD ⫽ 12.11) ranging from 20 to 68. Participants were mostly from the United States (58.1% and 69.7%) and India (34.2% and 22.9%) for “employees” and “sailors,” respectively.
Agent Versus Appraiser Moral Relativism: An Exploratory Study
221
To determine whether participants considered punishment as less morally wrong when it was in accordance with the agents’ frameworks, we used a mixed design ANOVA with AC and AD as the two within-subject levels of the variable AGENT, and order of presentation as the between-subject variable. We found a significant main effect of AGENT on the extent to which participants thought that the punishment was morally wrong (employees: F(1,270) ⫽ 223.9, p ⬍ 0.001; sailors: F(1,107) ⫽ 43.2, p ⬍ 0.001). Specifically, participants thought that the “agent concordant” punishment was more morally permissible (less morally wrong) (see Figure 11.1; employees: M ⫽ 3.94, SD ⫽ 0.07; sailors: M ⫽ 2.86, SD ⫽ 0.14) than the “agent discordant” punishment (employees: M ⫽ 2.55, SD ⫽ 0.08; sailors: M ⫽ 2.05, SD ⫽ 0.11). We found no significant order effect (employees: F(1,270) ⫽ 0.05, p ⫽ 0.868; sailors: F(1,107) ⫽ 1.97, p ⫽ 0.164) and no interaction effect between AGENT and order (employees: F(1,270) ⫽ 0.406, p ⫽ 0.525; sailors: F(1,107) ⫽ 2.47, p ⫽ 0.119). To determine whether there was an effect of the agents’ and appraisers’ moral frameworks on participants’ truth evaluation of a moral statement, we conducted a mixed design ANOVA with AGENT TRUTH and APPRAISER TRUTH as within-subject variables and order as between-subject variable.
Mean moral permissibility of behavior
5.00
4.00
3.00
2.00
1.00
0.00
Employees QUESTIONNAIRE
Sailors Error bars: 95% Cl
SCENARIO Agent CONCORDANT behavior
Agent DISCORDANT behavior
Figure 11.1 Moral permissibility of behavior as a function of concordance of the behavior with the agents’ moral frameworks, for two different questionnaire vignettes.
222
Advances in Experimental Moral Psychology
We found that the agents’ moral frameworks (AGENT TRUTH) had an effect on whether participants thought that the moral statement was true or not (employees: F(1,270) ⫽ 76.3, p ⬍ 0.001; sailors: F(1,107) ⫽ 53.9, p ⬍ 0.001). Specifically, participants thought that the statement was more likely to be true when it was in accordance with the agents’ moral frameworks (see Figure 11.2; employees: M ⫽ 3.46, SD ⫽ 0.053; sailors: M ⫽ 3.62, SD ⫽ 0.089) than when it was not in accordance with the agents’ moral frameworks (employees: M ⫽ 2.61, SD ⫽ 0.053; sailors: M ⫽ 2.61, SD ⫽ 0.096). We found that the appraisers’ moral frameworks (APPRAISER TRUTH) also had a significant effect on whether participants thought that the moral statement was true or not (employees: F(1,270) ⫽ 2496, p ⬍ 0.001; sailors: F(1,107) ⫽ 33.3, p ⬍ 0.001). Specifically, participants thought that the moral statement was more likely to be true when it was in accordance with the appraisers’ moral frameworks (see Figure 11.3; employees: M ⫽ 3.75, SD ⫽ 0.051; sailors: M ⫽ 3.71 SD ⫽ 0.081) than when it was not in accordance with the appraisers’ moral frameworks (employees: M ⫽ 2.32, SD ⫽ 0.050; sailors: M ⫽ 2.51, SD ⫽ 0.092). We did not find a main effect of order (employees: F(1,270) ⫽ 0.318, p ⫽ 0.573; sailors: F(1,107) ⫽ 0.067, p ⫽ 0.797).
Mean truth of statement
4.00
3.00
2.00
1.00
0.00
Employees QUESTIONNAIRE
Sailors Error bars: 95% Cl
SCENARIO Agent CONCORDANT statement
Agent DISCORDANT statement
Figure 11.2 Truth assessment of a moral statement as a function of concordance of the statement with the agents’ moral frameworks, for two different questionnaire vignettes.
Agent Versus Appraiser Moral Relativism: An Exploratory Study
223
Mean truth of statement
4.00
3.00
2.00
1.00
0.00 Employees QUESTIONNAIRE
Sailors Error bars: 95% Cl
SCENARIO Appraiser CONCORDANT statement Appraiser DISCORDANT statement
Figure 11.3 Truth assessment of a moral statement as a function of concordance of the statement with the appraisers’ moral frameworks, for two different questionnaire vignettes.
For employees, but not for sailors, we found a significant two-way interaction between AGENT TRUTH and APPRAISER TRUTH (see Figure 11.4; employees: F(1,270) ⫽ 7.58, p ⫽ 0.006; sailors: F(1,107) ⫽ 0.199, p ⫽ 0.657). Examining Figure 11.4, although this interaction was significant, the effect of agents’ (or appraisers’) moral frameworks was in the same direction in both conditions. Interestingly, in the “employees” questionnaire, the statement was perceived to be more true when it was concordant with the appraisers’ moral framework and discordant with the agents’ moral framework (M ⫽ 3.40, SD ⫽ 1.39), than when it was concordant with the agents’ moral framework and discordant with the appraisers’ moral framework (M ⫽ 2.81, SD ⫽ 1.42). In the sailors “questionnaire” though, the truth values of these statements were similar (M ⫽ 3.21, SD ⫽ 1.47, M ⫽ 3.04, SD ⫽ 1.47). This suggests that, in the “employees” questionnaire, the appraisers’ moral framework was more important than the agents’ moral frameworks when there was some discordance, while in the “sailors” questionnaire, the appraisers’ and agents’ moral frameworks were almost equally important when there was some discordance.
224
Advances in Experimental Moral Psychology
5.00
Mean truth of statement
4.00
3.00
2.00
1.00
0.00 I...Appraiser CONCORDANT statement...II...Appraiser DISCORDANT statement...I SCENARIO Error bars: 95% Cl SCENARIO Agent CONCORDANT statement
Agent DISCORDANT statement
Figure 11.4 Truth assessment of a moral statement as a function of concordance of the statement with the agents’ and with the appraisers’ moral frameworks, for the “employees”’ questionnaire only.
Because these results suggests that agents’ and appraisers’ moral frameworks independently matter for people’s evaluations of the truth of moral statements, it might be the case that some people are predominantly and consistently agent relativists while others are predominantly and consistently appraiser relativists. In order to explore this possibility, we calculated three new variables: AGENT DEGREE (AC-AD) as the degree to which participants relativized the permissibility of behavior according to the agents’ moral frameworks, AGENT TRUTH DEGREE (AGCAPC⫹AGCAPD-AGDAPC-AGDAPD) as the degree to which participants relativized the truth of moral statements according to the agents’ moral frameworks, and APPRAISER TRUTH DEGREE (AGCAPC⫹AGDAPC-AGCAPD-AGDAPD) as the degree to which participants relativized the truth of moral statements according to the appraisers’ moral frameworks. For “sailors,” we found that AGENT DEGREE was positively and significantly related to AGENT TRUTH DEGREE (F(1,108) ⫽ 11.0, p ⫽ 0.001) but not to APPRAISER TRUTH DEGREE (F(1,108) ⫽ 0.000, p ⫽ 0.989). This suggests that participants who were relativists with regard to moral permissibility were more likely to be agent relativists with regard to moral truth. Thus, they might
Agent Versus Appraiser Moral Relativism: An Exploratory Study
225
have been agent relativists with regard to moral permissibility and with regard to moral truth, and therefore quite consistent in their relativist intuitions. However, for employees, it was just the other way around: AGENT DEGREE was positively and significantly related to APPRAISER TRUTH DEGREE (F(1,271) ⫽ 5.30, p ⫽ 0.022) but not to AGENT TRUTH DEGREE (F(1,271) ⫽ 0.141, p ⫽ 0.708). In this scenario, participants might have been inconsistent in their relativist intuitions, alternating between agent relativist speech acts with regard to moral permissibility and appraiser relativist speech acts with regard to moral truth. Alternatively, participants might have interpreted the actors in the moral permissibility scenario as appraisers instead of agents—as explained in the introduction. Thus, they might have been appraiser relativists with regard to moral permissibility and with regard to moral truth. Finally, for employees, but not for sailors, we found a significant interaction effect between APPRAISER TRUTH and order of presentation (see Figure 11.5; F(1,270) ⫽ 26.76, p ⬍ 0.001). Examining Figure 11.5 though, we see that the
5.00
Mean truth of statement
4.00
3.00
2.00
1.00
0.00
Appraiser CONCORDANT FIRST
Appraiser CONCORDANT SECOND
ORDER
Error bars: 95% Cl
SCENARIO Appraiser CONCORDANT statement Appraiser DISCORDANT statement
Figure 11.5 Truth assessment of a moral statement as a function of concordance of the statement with the appraisers’ moral frameworks, for two different orders of presentation, for the “employees’ ” questionnaire only.
226
Advances in Experimental Moral Psychology
effect of appraisers’ moral frameworks was again in the same direction in both orders. Thus, the folk seem to be appraiser relativists regardless of order of presentation or variation in the appraisers’ moral frameworks. There were no interaction effects for AGENT TRUTH and order of presentation.
Discussion We investigated whether lay people’s moral evaluations were in accordance with agent moral relativism. We tested this in two ways. First, we asked participants to make a moral judgment about an act while manipulating the moral frameworks of the agents. We found that participants were more likely to consider the act as morally permissible when it was in accordance with the agents’ moral frameworks than when it was not in accordance with the agents’ moral frameworks. This suggests that agents’ moral frameworks have an effect on lay people’s moral speech acts about the moral wrongness or permissibility of behavior: People are more likely to say that a behavior is morally permissible when it is in accordance with the agents’ moral frameworks compared to when it is not in accordance with the agents’ moral frameworks. Second, we asked participants to assess the truth of a moral statement while manipulating the moral frameworks of the agents and of the appraisers. We found that participants were more likely to answer that a moral statement is true when its message was in accordance with the agents’ moral frameworks. This suggests that agents’ moral frameworks have an effect on lay people’s moral speech acts about the truth of moral statements: People are more likely to say that a moral statement is true when the message is in line with the agents’ moral frameworks compared to when the message is not in line with the agents’ moral frameworks. However, we also found that participants were more likely to answer that a moral statement is true when its message was in accordance with the appraisers’ moral frameworks. This suggests that appraisers’ moral frameworks also have an effect on lay people’s moral speech acts about the truth of moral statements.
Agent Versus Appraiser Moral Relativism: An Exploratory Study
227
For “employees,” but not for sailors, we found two interaction effects. We found a significant two-way interaction between AGENT TRUTH and APPRAISER TRUTH and between APPRAISER TRUTH and order of presentation. However, the effects were always in the same direction, meaning that our second conclusion is upheld: individuals take both agents’ and appraisers’ moral frameworks into account when assessing the truth of moral statements. Further research may reveal whether these interaction effects are a consistent pattern in folk moral relativism, or whether they were an artifact of the “employees” scenario. Finally, we explored the possibility that some people are predominantly and consistently agent relativists while others are predominantly and consistently appraiser relativists. Our results are not conclusive. Whether people are predominantly agent moral relativists or appraiser moral relativists might vary depending on the scenario or depending on the moral aspect (truth vs. permissibility) that is to be evaluated. Our results are not definitive. Notwithstanding the fact that we excluded all participants who did not fill out all comprehension questions correctly, given the complexity of our scenarios and questions, future investigations would benefit from simpler materials. Also, we examined assessments of only two acts, namely reduction in lunch time, and whipping, both as a punishment. The extent of lay people’s moral relativism may depend on the kind of act or the modality of the moral statement. In addition, it remains to be seen whether agent relativism and appraiser relativism are stable intuitions or vary across a range of situations. These and other possibilities warrant future research, some of which has already been undertaken by the present authors (Quintelier et al. 2013). With the above caveats in mind, our study reveals that there is interindividual as well as intra-individual variation in whether individuals relativize moral speech acts to agents or to appraisers. Such variation in types of moral intuitions is in line with previous suggestions (e.g., Gill 2009; SinnottArmstrong 2009) that different individuals employ quite divergent moral language. The variation that we have documented thus supports Gill’s position that philosophical theories that appeal to lay people’s speech acts cannot rely on a “handful of commonsense judgments,” (2009, p. 217), as the philosopher’s commonsense judgment will often fail to reflect the actual distribution of
228
Advances in Experimental Moral Psychology
moral reasoning among the folk. Moreover, that people may employ divergent relativist forms of language indicates that researchers of moral relativism cannot make claims regarding moral relativism without first specifying the type of relativism at issue, nor can they attend only to appraiser relativism. Methodologically, researchers must take care in designing stimuli and queries in order to minimize ambiguity as to which type of relativism is made salient. Whether they be empiricists or theorists, researchers of moral relativism must take seriously the existence of agent moral relativism, and must consider the differences between it and appraiser moral relativism.
Note * Author’s note: Katinka J. P. Quintelier, Amsterdam Business School, University of Amsterdam, Plantage Muidergracht 12, 1080TV Amsterdam, The Netherlands. Delphine De Smet, Department of Legal Theory and Legal History and Research Unit “The Moral Brain,” Ghent University, Universiteitstraat 4, B-9000 Ghent, Belgium. Daniel M. T. Fessler, Department of Anthropology and Center for Behavior, Evolution, & Culture, 341 Haines Hall, University of California, Los Angeles, Los Angeles, 375 Portola Plaza, Los Angeles, CA 90095-1553, USA. Correspondence should be addressed to Katinka J. P. Quintelier, Amsterdam Business School, University of Amsterdam, Plantage Muidergracht 12, 1080 TV Amsterdam, The Netherlands. E-mail: [email protected]. Acknowledgments: The authors thank the participants of the first workshop of the Experimental Philosophy Group UK, and Hagop Sarkissian and Jen Cole Wright for their valuable feedback. The first author received funding from the Flemish Fund for Scientific Research (FWO) and from the Konrad Lorenz Institute for Evolution and Cognition Research while working on this chapter.
References Beebe, J. (2010). Moral relativism in context. Noûs, 44(4), 691–724. doi:10.1111/ j.1468-0068.2010.00763.x Brogaard, B. (2008). Moral contextualism and moral relativism. Philosophical Quarterly, 58(232), 385–409. doi:10.1111/j.1467-9213.2007.543.x
Agent Versus Appraiser Moral Relativism: An Exploratory Study
229
Gill, M. (2009). Indeterminacy and variability in meta-ethics. Philosophical Studies, 145(2), 215–34. doi:10.1007/s11098-008-9220-6 Goodwin, G. P., and Darley, J. M. (2008). The psychology of meta-ethics: Exploring objectivism. Cognition, 106(3), 1339–66. doi:10.1016/j. cognition.2007.06.007 —(2010). The perceived objectivity of ethical beliefs: Psychological findings and implications for public policy. Review of Philosophy and Psychology, 1(2), 161–88. doi: 10.1007/s13164-009-0013-4 Harman, G. (1975). Moral relativism defended. The Philosophical Review, 84(1), 3–22. Stable URL: http://links.jstor.org/sici?sici⫽00318108%28197501%2984%3A 1%3C3%3AMRD%3E2.0.CO%3B2-%23 Harman, G., and Thompson, J. J. (1996). Moral Relativism and Moral Objectivity. Blackwell. Lyons, D. (1976/2001). Ethical relativism and the problem of incoherence. In P. K. Moser and T. L. Carson (eds), Moral relativism - A reader. New York/Oxford: Oxford University Press, pp. 127–41. Plato (1921). Theaetetus (H. N. Fowler, Trans. Vol. 12). Cambridge, MA/London: Harvard University Press/William Heinemann. Prinz, J. J. (2007). The Emotional Construction of Morals. Oxford: University Press. Quintelier, K., and Fessler, D. (2012). Varying versions of moral relativism: The philosophy and psychology of normative relativism. Biology and Philosophy, 27(1), 95–113. doi:10.1007/s10539-011-9270-6 Quintelier, K., Fessler, D. M. T., and De Smet, D. (2012). The case of the drunken sailor: On the generalizable wrongness of harmful transgressions. Thinking & Reasoning. doi:10.1080/13546783.2012.669738 —(2013). The moral universalism-relativism debate. Klesis Revue Philosophique, 27, 211–62. Richard, M. (2004). Contextualism and relativism. Philosophical Studies, 119(1), 215–42. doi:10.1023/B:PHIL.0000029358.77417.df Sarkissian, H., Park, J., Tien, D., Wright, J., and Knobe, J. (2012). Folk moral relativism. Mind and Language, 26(4), 482–505. doi:10.1111/j.14680017.2011.01428.x Sinnott-Armstrong, W. (2009). Mixed-up meta-ethics. Philosophical Issues, 19(1), 235–56. doi:10.1111/j.1533-6077.2009.00168.x Streiffer, R. (1999). Moral Relativism and Reasons for Action. Doctor of Philosophy, Massachusetts Institute of Technology, Cambridge. Williams, B. (1972). Morality: An Introduction to Ethics. Cambridge: Cambridge University Press.
230
Advances in Experimental Moral Psychology
Wright, J. C., Grandjean, P., and McWhite, C. (2012). The meta-ethical grounding of our moral beliefs: Evidence for meta-ethical pluralism. Philosophical Psychology, 26(3), 1–26. doi:10.1080/09515089.2011.633751 Wright, J. C., McWhite, C., and Grandjean, P. T. (in press). The cognitive mechanisms of intolerance: Do our meta-ethical commitments matter? In T. Lombrozo, S. Nichols and J. Knobe (eds), Oxford Studies in Experimental Philosophy, Vol. 1. Oxford: Oxford University Press.
Part Three
Measuring Morality
12
Know Thy Participant: The Trouble with Nomothetic Assumptions in Moral Psychology Peter Meindl and Jesse Graham*
Imagine a world in which researchers measure morality by determining how often people eat peanut butter sandwiches (considered morally good behavior) versus how often they eat jelly sandwiches (considered morally bad behavior). Researchers in this world investigate factors that influence morality by assessing the relationships that exist between situational variables, individual differences, and the eating of peanut butter and jelly sandwiches. Also imagine that in this world there exist reasonable philosophical arguments for why peanut butter sandwich eating is morally good and jelly sandwich eating is morally bad, but there also exist well-worn ethical theories that reject the moral significance of peanut butter and jelly sandwich eating altogether. Additionally, although a majority of this world’s people consider peanut butter sandwich eating to be highly morally good and jelly sandwich eating to be highly morally bad, a sizeable portion of the population also considers peanut butter and jelly sandwich eating morally irrelevant. Further, a large percentage of people in this world actually hold views that are diametrically opposed to the implicit assumptions of researchers—according to these people, it is morally bad to eat peanut butter sandwiches and it is morally good to eat jelly sandwiches. The field of peanut-butter-and-jelly moral psychology is in crisis. In this chapter, we argue that researchers in our own world currently study morality and its contributing factors in much the same way as researchers in the moral PB&J world described above. Moral psychologists tend to deem certain patterns of thoughts and behaviors “moral” even though neither
234
Advances in Experimental Moral Psychology
scholars nor laypeople agree about the moral significance of these thoughts and behaviors. As a consequence, moral psychologists often investigate the causes and consequences of behaviors that either have little impact on the world or are of little moral interest to scholars and laypeople. We also explain how a small change in how morality is typically assessed could significantly increase the scope and importance of morality research. Finally, we present an empirically informed list of traits and behaviors that we consider useful proxies of morality. In this chapter, we will focus primarily on the construct of morality; however, most of the concerns we raise and the solutions we offer are relevant to other normative concepts that are likely of interest to readers of this volume, such as prosociality and selfishness. Although we think moral psychologists’ current nomothetic bent poses a problem for many areas of psychology (e.g., judgment and decision-making, motivation), here we focus on moral behavior, the area of research where we think this bias is most prevalent and problematic.
Operationalizing “Morality”—An introduction to the problem Traditionally, researchers have taken one of two approaches to operationalizing the concept “morality.” Following Frimer and Walker (2008), we refer to these as the third-person and first-person approaches. Like in the moral PB&J world, researchers in our world tend to take a third-person, normative, non-neutral approach to operationalizing the concept of morality, in which what is moral is determined by the researchers, who deem one set of principles or actions morally good and another set morally bad. The principles that researchers use as proxies for morality can be based on normative ethical theory, religious prescriptions, cultural values, lay definitions of morality, or the idiosyncratic predilections of individual researchers. This is called the “3rd-person” approach because the participant’s perspective of what is moral does not influence how the researcher defines morality. The third-person approach is widely used in moral psychology. As Frimer and Walker (2008) noted, researchers investigating moral behavior have used bravery (Walker and Frimer 2007), extraordinary care (Matsuba and Walker
The Trouble with Nomothetic Assumptions in Moral Psychology
235
2004), social activism (Colby and Damon 1992), honesty (Derryberry and Thoma 2005), environmentally friendly behavior (Kaiser and Wilson 2000), and community service (Hart et al. 2006) to operationalize morality. To this list, we add volunteerism (Aquino and Reed 2002), honesty (Teper et al. 2011), and cooperation (Crockett et al. 2010), to name a few. The traditional alternative to the third-person approach is the first-person, value-neutral, descriptive approach (Frimer and Walker 2008). In contrast to the third-person approach, the first-person approach assesses morality according to what the participant herself considers moral. The impartial researcher deems each individual’s set of principles or actions moral, and deems failing to follow or enact those principles or actions not moral. Although no less a figure than Gordon Allport proposed that a first-person approach is the only valid means of assessing moral behavior (Allport 1937), and researchers have long warned of the flaws inherent in third-person morality research (e.g., Pittel and Mendelsohn 1966), moral behavior has rarely been assessed using the first-person approach. Psychologists have occasionally taken participants’ idiosyncratic moral beliefs into account when determining which variables to include in their analyses, but have taken this step almost exclusively when studying moral cognition (e.g., Goodwin and Darley 2008; Wright 2008; Wright 2010), but not when studying moral behavior. That said, the first-person approach is very similar to the approach taken by advocates of social-cognitive process models of general personality, such as Cognitive Affective Personality System (Mischel and Shoda 1995, 1999) and Knowledge-and-Appraisal Personality Architecture (Cervone 2004). These models were developed largely as a response to calls that behavior was not consistent across situations. The creators and advocates of these models suggested that past research showed that behaviors associated with traits such as agreeableness and conscientiousness were not meaningfully consistent, because participants’ perceptions of situations were not taken into account. When researchers assessed behavior across situations that were psychologically similar according to the participants (not just nominally similar to an outsider’s perspective), researchers discovered that cross-situational consistency was high (Mischel and Shoda 1998). The idea behind the first-person approach to operationalizing morality is that something similar might be true for moral behavior: people’s behavior
236
Advances in Experimental Moral Psychology
might not fall in line with what researchers deem moral (i.e., what is nominally moral), but their behavior might fall in line with what they personally consider moral (i.e., what is psychologically moral).
Comparing the two approaches The first- and third-person approaches both have a number of unique advantages and disadvantages, but both have their own main advantage and disadvantage. Because the most important advantage of one approach naturally relates to the most important disadvantage of the other approach, here we will discuss the main disadvantage of each approach.
Third-person disadvantage: Poor construct validity We begin by discussing the main disadvantage associated with the third-person approach—it results in poor construct validity. The third-person approach’s poor construct validity stems from the fact that, though objectively moral facts might exist, there is certainly no consensus about whether moral objectivity is possible, let alone which theories or principles are the most objectively correct (Shafer-Landau 1994). As Kohlberg and Mayer (1972, p. 479) noted, “one person’s ‘integrity’ is another person’s ‘stubbornness,’ [and one person’s] ‘honesty in expressing your true feelings’ is another person’s ‘insensitivity to the feelings of others.’” Hence, though some people might have reason to consider a behavior virtuous, someone else might have reason to consider it virtue-neutral or even vicious. It was due in part to the apparent arbitrariness of lists of moral traits and behaviors (pejoratively dubbed the “bag of virtues” problem; Kohlberg 1970) that Kohlberg avoided research on moral behavior and instead focused on moral reasoning (Narvaez and Lapsley 2009), and it is in part for this same reason that it might be wise for researchers to avoid the third-person approach to defining morality. Not only do scholars of ethics disagree about what is morally good and morally bad, but laypeople do as well. Here we call this lay disagreement. There is evidence of at least three important types of lay moral disagreement: (a)
The Trouble with Nomothetic Assumptions in Moral Psychology
237
variability in the degree to which laypeople consider actions and traits morally relevant, (b) disagreement about the moral valence of certain actions and behaviors (e.g., some see obedience to authority figures as good, some see it as bad), and, perhaps most problematic for moral psychology, (c) disagreement between researchers and laypeople about the moral relevance of several behaviors and traits. The first two types of lay disagreement demand little attention here, as intra- and intercultural variations in moral concerns and judgments have recently received much attention in the psychology literature (for a review, see Graham et al. 2013). Consequently, we will focus on the discrepancy between the amount of moral relevance that researchers place on behaviors and traits themselves (or assume that their participants place on them) and the moral relevance that their participants actually place on these behaviors and traits (if any at all).
Empirical evidence To begin to assess this discrepancy, we instructed two groups of participants to rate either the moral importance of a list of traits and behaviors (“How important it is to have each of the following traits in order to be a moral person?”; Table 12.1) or the moral valence of traits and behaviors (“How morally good or morally bad it is to possess or perform each of the following traits or behaviors?”; Tables 12.2 and 12.3). The moral importance of each trait was rated by 905 participants on YourMorals.org (YM), and the moral valence of each behavior or trait was rated by 125 participants on Amazon’s Mechanical Turk. Both of these qualify as Western, Educated, Industrialized, Rich, and Democratic (Henrich et al. 2010) samples, so it is likely that the researcher-participant discrepancies suggested here underestimate the true discrepancies. Included in our trait list were all traits that have been previously included in morally relevant trait lists that we are aware of (e.g., Aquino and Reed 2002; Lapsley and Lasky 2001; Smith et al. 2007; Walker and Pitts 1998), as well as traits listed by participants in our own pretests who completed open-ended questions such as “In order to be moral, what traits are important for people to possess?” The list of behaviors consisted of (a) actions that psychologists often
238
Advances in Experimental Moral Psychology
Table 12.1 Moral importance survey
Behavior
Mean score
SD
Behavior
Mean score
SD
Honest
4.39
0.90
Wise
3.00
1.34
Just
4.20
Compassionate
4.04
0.96
Controls thoughts
2.98
1.34
1.10
Straightforward
2.95
1.22
Treats people equally
3.91
1.20
Courageous
2.87
1.28
Genuine
3.86
1.07
Hardworking
2.83
1.24
Kind
3.83
1.08
Environmentally friendly
2.76
1.19
Honorable
3.81
1.16
Purposeful
2.73
1.19
Tolerant
3.80
1.12
Perseverant
2.71
1.18
Responsible
3.74
1.07
Controls emotions
2.68
1.16
Merciful
3.68
1.16
Modest
2.65
1.17
Humane toward animals
3.61
1.19
Friendly
2.61
1.17
Forgiving
3.59
1.19
Brave
2.61
1.24
Respectful
3.56
1.19
Determined
2.56
1.22
Conscientious
3.51
1.09
Non-materialistic
2.48
1.23
Helpful
3.44
1.00
Resourceful
2.22
1.20
Nonjudgmental
3.34
1.32
Optimistic
2.18
1.22
Loyal
3.34
1.19
Spends money wisely
2.12
1.12
Giving
3.31
1.08
Spiritual
1.88
1.20
Rational
3.29
1.28
Obedient
1.80
1.02
Self-controlled
3.28
1.18
Is patriotic
1.59
0.97
Generous
3.24
1.14
Supportive
3.22
1.08
Selfless
3.19
1.26
Patient
3.10
1.15
Cooperative
3.01
1.15
Note: 5 ⫽ “It is extremely important that a person possesses this characteristic” (in order to be moral). 1 ⫽ “It is not important that a person possess this characteristic” (in order to be moral). Sample sizes range from 867 to 905 raters for each item.
The Trouble with Nomothetic Assumptions in Moral Psychology
239
Table 12.2 Moral valence survey (behaviors)
Behavior
Mean score
SD
Kicking a dog in the head, hard. (Graham et al. 2009)
8.54
0.99
Steal a cellphone.
7.96
1.67
A psychologist tells his or her participants that their behavior is anonymous when it is not.
7.72
1.65
Cheat on trivia game (for money). (DeAndrea et al. 2009)
7.27
1.50
Lie about predicting the outcome of a coin toss (for money). (Greene and Paxton 2009)
7.17
1.51
Gossiping.
6.89
1.33
Failing to pay for a subway ticket which costs somewhere between $1.50 and $5.00.
6.72
1.24
Not stopping at a stop sign.
6.60
1.43
Not recycling.
6.08
1.26
Being impatient with people.
6.07
1.33
Illegally watching movies online.
6.06
1.27
Looking at pornography.
6.05
1.42
Keeping the majority of $20 in an ultimatum game.
5.97
1.38
Failing to flip a coin to decide whether to place self or other participant in a “positive consequence” condition (and choosing to place self in positive condition). (Batson et al. 1999)
5.86
1.37
Showing up late for something.
5.85
1.07
Not holding a door open for someone behind you.
5.84
1.02
Taking a pen that is not yours. (Mullen and Nadler 2008)
5.76
1.09
Illegally walking across a street.
5.57
0.83
Having the opportunity to keep $50 for oneself or keeping $25 and giving $25 to charity, and deciding to keep all $50 for oneself.
5.56
1.26
Not cooperating in a one shot public goods game.
5.48
1.37
Lying in order to avoid hurting someone’s feelings.
5.21
1.52
Eating pickles.
4.97
0.48
Defecting in a one shot prisoner’s dilemma game.
4.94
1.26
Cooperating in a one shot prisoner’s dilemma game.
3.77
1.62
Giving the majority of $20 away in a one shot Dictator Game.
3.61
1.50
(Continued)
240
Advances in Experimental Moral Psychology
Table 12.2 (Continued) Mean score
SD
Agree to talk to xenophobic inmates on the virtues of immigration. (Kayser et al. 2010)
3.21
1.75
Helping an experimenter pick up and organize dropped papers. (van Rompay et al. 2009)
3.01
1.32
Having the opportunity to keep $50 for oneself or keeping $25 and giving $25 to charity, and deciding to give $25 to charity and keeping $25 for oneself.
2.57
1.35
Behavior
Note: 1 ⫽ “This is an extremely morally good behavior/trait”; 5 ⫽ “This is neither a morally good nor a morally bad behavior/trait”; 9 ⫽ “This is an extremely morally bad behavior/trait.” Sample sizes range from 94 to 124 raters for each item.
use as proxies of morality and (b) behaviors we thought laypersons would consider morally relevant and should be relatively easy to assess. In both surveys we also included presumably morally neutral traits and behaviors (e.g., eating pickles) that we used as points of comparison. People rated the moral importance and moral goodness of many of the traits and behaviors on our list as might be expected, but other traits and behaviors—including those that have been used as proxies of morality and prosociality—were rated differently than researchers might expect. For instance, though researchers have often used people’s choices in prisoner’s dilemma games (PDGs) and public goods games as proxies of moral or prosocial behavior (e.g., Batson and Moran 1999; Cohen et al. 2006; Rand et al. 2012; Twenge et al. 2007), our findings suggest that laypeople do not consider such cooperation in general to be highly morally relevant or morally good. In fact, participants considered non-cooperation in these games to be almost entirely morally neutral. For instance, non-cooperation in a classic PDG was rated as no more morally bad than eating pickles (on average, both were rated as “This is neither a morally good nor a morally bad behavior”; see Table 12.2). Similarly, participants rated behaviors such as helping an experimenter pick up dropped papers—a behavior that has often been used as a proxy for prosocial behavior (e.g., Isen and Levin 1972)—as only slightly morally good. As a point of comparison, the use of deception in psychology research—something common in experimental moral psychology—tended to be rated as extremely morally bad. Overall, these results suggest that laypersons and researchers
The Trouble with Nomothetic Assumptions in Moral Psychology
241
Table 12.3 Moral valence survey (traits)
Behavior
Mean score
SD
Behavior
Mean score
SD
Honest
1.62
1.05
Patient
2.72
1.29
Treats people equally Compassionate
1.82
1.02
Open-minded
2.90
1.45
1.95
1.14
Friendly
2.97
1.36
Humane toward animals
2.03
1.33
Self-controlled
2.99
1.38
Honorable
2.05
1.24
Non-materialistic
3.02
1.41
Charitable
2.06
1.07
Cooperative
3.05
1.36
Giving
2.13
1.24
Conscientious
3.07
1.43
Respectful
2.14
1.22
Modest
3.16
1.36
Faithful
2.15
1.43
Brave
3.42
1.48
Helpful
2.22
1.16
Wise
3.46
1.68
Generous
2.24
1.23
Perseverant
3.57
1.47
Is Fair
2.25
1.28
Patriotic
3.59
1.50
Hardworking
2.45
1.25
Independent
3.93
1.38
Loyal
2.48
1.46
Obedient
3.94
1.59
Empathetic
2.49
1.25
Sociable
4.00
1.37
Dependable
2.53
1.28
Intelligent
4.09
1.46
Humble
2.55
1.36
Lively
4.13
1.34
Polite
2.55
1.40
Bold
4.30
1.19
Selfless
2.59
1.70
Creative
4.32
1.18
Tolerant
2.61
1.27
Perfectionist
4.67
0.92
Nonjudgmental
2.65
1.51
Genuine
2.67
1.33
Environmentally friendly
2.69
1.26
Note: 1 ⫽ “This is an extremely morally good behavior/trait”; 5 ⫽ “This is neither a morally good nor a morally bad behavior/trait”; 9 ⫽ “This is an extremely morally bad behavior/trait.” Sample sizes range from 95 to 125 raters for each item.
disagree about the morality of traits and behaviors. Consequently, the behaviors researchers use as indicators of morality may not be optimal proxies of moral behavior—a potential pitfall that seems to inherently accompany the third-person approach.
242
Advances in Experimental Moral Psychology
Assessing a different construct Due to the normative and descriptive disagreement described above, any time a researcher conceptualizes morality in its broadest sense (i.e., in relation to ethics in general), the third-person approach is likely to result in poor construct validity. Due to ongoing normative and descriptive moral disagreement, the third-person approach can only test the extent to which people’s behavior is in line with the tenets of one particular set of ethical principles. Thus, this approach may produce information on the degree to which people behave in accord with particular moral principles, but assessing moral behaviors that are in line with only one particular set of moral beliefs does not result in information about morality in general (i.e., “morality” in its most overarching sense), which appears to often be the goal of researchers who investigate moral behavior (e.g., Aquino et al. 2009; Crockett et al. 2010; Gu et al. 2013; Jordan et al. 2011; Kouchaki 2011; Perugini and Leone 2009; Reynold et al. 2010; Sachdeva et al. 2009; Teper et al. 2011). A person can consistently behave contrary to one code of ethics while simultaneously adhering to a different code of ethics (e.g., people involved in organized crime who adhere to strict norms of loyalty and respect). Researchers who use a third-person approach may thus be assessing a construct (behavior that is in line with one particular moral viewpoint or belief) that is different from the construct they intend to assess (morality in general). Some psychologists may not be interested in morality in general; instead they might be interested only in certain behaviors that they personally consider moral or immoral. For such researchers, the concerns mentioned above still deserve consideration for at least two reasons. First, we believe that if psychologists were to only focus on a short list of moral behaviors, they would be ignoring much of what makes morality and moral life interesting and important (including the very arguments about what constitutes morality that make universally accepted objective criteria impossible). Furthermore, even when researchers are only interested in particular types of morally relevant behaviors, taking a strict third-person approach is likely to produce many of the same shortfalls described above. For instance, if a psychologist is only interested in assessing cheating behavior it would be unwise to use a measure of cheating that they only assume their participants
The Trouble with Nomothetic Assumptions in Moral Psychology
243
consider meaningfully bad. As the results of our surveys suggest (Table 12.1), laypersons consider behaviors that are used as standard measures of cheating to be only mildly wrong. When psychologists make assumptions about the moral importance of their dependent variables, they run the risk of assessing behaviors that are so mundane that they are not assessing something of real interest or importance. Consequently, in many cases using these measures as dependent variables may provide researchers with relatively little information about the relationship between cheating and other variables of interest.
First-person disadvantage: Complexity and impracticality Recall that the first-person approach operationalizes morality according to whatever each person considers moral (Frimer and Walker 2008). Thus, “1stperson” moral research is immune to the problems with the third-person approach that were outlined above. That said, an important disadvantage of the first-person approach is its complexity and relative impracticality. This approach requires that researchers accurately assess a person’s moral values; unfortunately, social desirability and self-deception are likely to make it extremely difficult for researchers to do so. It is likely that some moral beliefs are more socially desirable than others, and thus it is likely that people will sometimes incorrectly report their moral beliefs in order to optimize appearances. For instance, conservatives participating in a study run by a presumably liberal researcher may not be forthcoming about their true moral beliefs, even if their anonymity is ensured. As for self-deception, research shows that self-enhancement causes people to label a trait as good if they think they possess that trait (Dunning et al. 1991); thus, in a study setting people might define what is moral according to how they actually act. As a result, the moral values people list in a first-person study might not necessarily be those which they consider moral, but instead they might simply be descriptions of how they typically act.1 None of this is to say that third-person assessments of morality are immune to the effects of social desirability and self-deception; surely third-person self-report measures of morality are highly susceptible to these biases. However, first-person approaches require two steps that may
244
Advances in Experimental Moral Psychology
be influenced by these biases (value assessment and behavior assessment) whereas the third-person approach is vulnerable to these biases only once (during behavioral assessment). The first-person approach’s second source of complexity is that a strict first-person approach would probably require scholars to assess behaviors, thoughts, and emotions that are in line with an impractically large number of moral principles. Past research suggests that people attach high levels of moral significance to hundreds of different traits and actions (Aquino and Reed 2002; Lapsley and Lasky 2001; Smith et al. 2007; Walker and Pitts 1998). Even if scholars assessed moral behavior only according to those moral principles people self-described as their most valued, it is likely that subjects would list a very large amount of principles. For instance, in a recent study we asked participants to list only their most important moral values; even after combining similar answers we were left with more than 70 different values. And these are only consciously endorsed moral values; if one were to try to include moral concerns or intuitions of which participants are unaware, this would add a huge new layer of complexity to the first-person approach. In contrast, the third-person approach is relatively simple and convenient: researchers simply pick a behavior or idea they deem as moral, and then assess participants accordingly.
A combined first- and third-person operationalization The advantages of the two approaches complement each other, and their disadvantages mostly negate each other. For this reason, we suggest that a synthesis of the two approaches can lead to the best operationalizations of morality. We call this the “mixed approach.” As is the case with the thirdperson approach, we suggest that scholars should assess morality according to predetermined moral principles, but in line with the first-person approach we suggest that scholars should also examine which principles each participant (consciously) values. Thus, in effect we are suggesting that researchers interested in moral behavior should assess moral hypocrisy, which is often conceptualized as the degree to which people go against what they consider morally right—and thus involves both first-person
The Trouble with Nomothetic Assumptions in Moral Psychology
245
assessments of individual moral values and third-person assessments of moral behavior. The advantages of the two approaches can be combined by assessing morality using traits and behaviors that prior research suggests people (either in general or in a particular sample) consider highly morally relevant. For instance, a researcher using the mixed approach might assess morality by measuring a person’s honesty, because research suggests that people in general consider honesty to be highly morally important (Aquino and Reed 2002; Smith et al. 2007; see also Tables 12.1 and 12.3). Surprisingly, researchers interested in assessing morality very rarely do this (or if they do, they fail to mention that this logic underlies their choice of moral proxy). Alternately, a researcher could choose to operationalize morality as, say, bravery because pretests with their particular sample suggest that their participants generally consider bravery to be the most morally important trait. Furthermore, a researcher who is interested in only one particular type of morally relevant behavior (e.g., cheating) may also use the mixed approach; this type of researcher could first determine what type of cheating behavior their participants are likely to consider highly morally bad (but are still relatively likely to perform). All that the mixed approach requires is information on the extent to which participants consider traits and/or behaviors morally good or morally bad. The results of our recent surveys provide insight into which traits and behaviors researchers can use in order to successfully measure moral behavior and which traits and behaviors are not good candidates for this purpose. Some of these suggestions are underwhelming: as most readers would probably expect, diverse groups of people—liberals and conservatives, men and women, theists and atheists—tend to rate traits such as honesty and fairness as highly morally important (Table 12.1). More surprisingly, people in general do not rate traits such as cooperative, helpful, or charitable as highly morally relevant. Researchers should be very cautious when making assumptions about normative concepts such as morality, selfishness, and prosociality. To reiterate, researchers could choose to measure morality according to that which previous research (such as our own research summarized above) suggests people in general consider highly morally important, but
246
Advances in Experimental Moral Psychology
they could also determine what their own particular participants consider most morally relevant by performing pretests with the particular population or even sample that they will use in their research. As long as a group of participants’ (probable) moral beliefs are somehow being taken into account, this approach will likely produce more accurate results than the third-person research, and at the same time require far less time and effort than a purely first-person approach.
Examples of the mixed approach A mixed approach is not the norm in moral psychology, but it has occasionally been used. For example, research on the effect that moral convictions have on behavior tends to involve assessments of each participant’s personal moral convictions in regard to the target behavior (Skitka 2010; Skitka and Baumann 2008). And in their investigation of the neural correlates of admiration and moral elevation, Immordino-Yang et al. (2009) first determined for each individual participant where the emotional high point of the eliciting stimuli were, aiding their ability to assess moral emotions in the scanner by taking individual variation in the emotional reaction into account. Given that the mixed approach is a means of assessing moral hypocrisy, it should not be surprising that in the past moral hypocrisy has been assessed by way of the mixed approach. For instance, in order to investigate the situational factors that cause people to act unfairly, Batson et al. (1999) ensured they were using an appropriate proxy for unfairness by asking a separate sample of participants whether they considered their behavior of interest—giving oneself a chance to win money rather than giving someone else a chance to win money—to be unfair. Other examples of the mixed approach in moral psychology go beyond moral hypocrisy. For instance, prior to performing research designed to test whether various presumed moral behaviors—helping behavior, moral courage, and heroism—are associated with people’s moral prototypes, Osswald, Greitemeyer, Fischer, and Frey (2010) first determined whether people actually considered these three behaviors to be distinct moral behaviors. In order to do this, Osswald et al. (2010) simply performed a
The Trouble with Nomothetic Assumptions in Moral Psychology
247
pretest with participants who were demographically similar to the people who would participate in their main studies. These pretest participants were asked to rate the morality of different scenarios in which helping behavior, moral courage, or heroism were depicted, and by way of this simple pretest it was determined that people indeed considered these behaviors to be distinct moral behaviors. Had these authors taken a third-person approach, the results of these studies may have been less valid and their implications would have been less profound (for the reasons described previously), and had either Batson et al. (1999) or Osswald et al. (2010) taken a first-person approach, their projects would have become much more complicated. Thus, by taking a mixed approach, these authors struck an important balance between methodological rigor and practicality. These extra steps required relatively little effort, but in both cases the extra step taken provided strong support for the assumptions that lay at the foundation of their respective research projects.
Conclusion Throughout the history of the field of psychology only a handful of researchers have extolled the virtues of a first-person approach to moral research (e.g., Allport 1937; Blasi 1990; Colvin and Bagley 1930). In contrast, contemporary moral psychologists usually assess moral behavior using a third-person approach. At this point, however, it should be clear that both approaches have disadvantages. Considering the third-person approach’s potentially negative impact on construct validity and the inherent complexity of effectively assessing morality using the first-person approach, one might argue that—as Allport (1937) contended long ago—morality is simply not a topic for psychological inquiry. However, in psychology and perhaps especially in moral psychology, error and inaccuracy will inevitably exist; this does not suggest that an entire field of study should be ignored. Nor does this mean we should settle for research with mediocre validity. Instead, it necessitates that the weaknesses that produce this error are minimized by operationalizing morality in the least problematic way possible.
248
Advances in Experimental Moral Psychology
By asking large groups of laypeople what traits and behaviors they consider moral, immoral, or morally irrelevant, we found evidence of problematic discrepancies between researchers’ nomothetic assumptions and participants’ views. For instance, measures that many researchers use as proxies for morality (e.g., cooperation or defection in PDG, picking up papers in the lab) are seen as about as morally irrelevant as eating pickles, while the actual behavior of moral psychology researchers (deceiving participants) is seen as highly morally bad. To help address this discrepancy we suggest that researchers use a mixed approach combining the strengths of the first- and third-person perspectives. This mixed approach still has disadvantages. For instance, even if the mixed approach were used, values will sometimes clash, and it is unclear how a person’s morality should be assessed in such cases. However, we believe that even with these disadvantages the mixed approach can help usher in a more interesting and significant era of morality research.
Notes * Author’s Note: Peter Meindl and Jesse Graham, University of Southern California. Address correspondence to: Peter Meindl, Department of Psychology, University of Southern California, 3620 S. McClintock Ave., SGM 501, Los Angeles, CA 90089, Email: [email protected]. This work was supported by Templeton Foundation Grant 53-4873-5200. 1 See also Frimer, this volume, on the differences between values people explicitly express on surveys and those they implicitly express in their everyday lives.
References Allport, G. W. (1937). Personality: A Psychological Interpretation. Oxford, England: Holt. Aquino, K., and Reed, A. (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83, 1423–40. Batson, C. D., and Moran, T. (1999). Empathy-induced altruism in a prisoner’s dilemma. European Journal of Social Psychology, 29, 909–24. Batson, C., Thompson, E. R., Seuferling, G., Whitney, H., and Strongman, J. A. (1999). Moral hypocrisy: Appearing moral to oneself without being so. Journal of Personality and Social Psychology, 77, 525–37.
The Trouble with Nomothetic Assumptions in Moral Psychology
249
Blasi, A. (1990). Kohlberg’s theory and moral motivation. New Directions for Child and Adolescent Development, 1990, 51–7. Cervone, D. (2004). The architecture of personality. Psychological Review, 111, 183–204. Cohen, T. R., Montoya, R. M., and Insko, C. A. (2006). Group morality and intergroup relations: Cross-cultural and experimental evidence. Personality and Social Psychology Bulletin, 32, 1559–72. Colby, A., and Damon, W. (1992). Some do Care: Contemporary Lives of Moral Commitment. New York: The Free Press. Colvin, S., and Bagley, W. (1930). Character and behavior. In Colvin, Stephen S. Bagley, William C. MacDonald, Marion E. (eds), Human Behavior: A First Book in Psychology for Teachers (2nd ed. Rev.). New York, NY: MacMillan Co, pp. 292–322. Crockett, M. J., Clark, L., Hauser, M., and Robbins, T. (2010). Serotonin selectively influences moral judgment and behavior through effects on harm aversion. Proceedings of the National Academy of Sciences of the United States of America, 107, 17433–8. DeAndrea, D. C., Carpenter, C., Shulman, H., and Levine, T. R. (2009). The relationship between cheating behavior and sensation-seeking. Personality and Individual Differences, 47, 944–7. Derryberry, W. P., and Thoma, S. J. (2005). Functional differences: Comparing moral judgment developmental phases of consolidation and transition. Journal of Moral Education, 34, 89–106. Doris, J. M. (2002). Lack of character: Personality and moral behavior. Cambridge: Cambridge University Press. —, ed. (2010). The Moral Psychology Handbook. Oxford, UK: Oxford University Press. Dunning, D., Perie, M., and Story, A. L. (1991). Self-serving prototypes of social categories. Journal of Personality and Social Psychology, 61, 957–68. Frimer, J. A., and Walker, L. J. (2008). Towards a new paradigm of moral personhood. Journal of Moral Education, 37, 333–56. Goodwin, G. P., and Darley, J. M. (2008). The psychology of meta-ethics: Exploring objectivism. Cognition, 106, 1339–66. Graham, J., Haidt, J., Koleva, S., Motyl, M., Iyer, R., Wojcik, S., and Ditto, P. H. (2013). Moral Foundations Theory: The pragmatic validity of moral pluralism. Advances in Experimental Social Psychology, 47, 55–130. Graham, J., Haidt, J., and Nosek, B. A. (2009). Liberals and conservatives rely on different sets of moral foundations. Journal of Personality and Social Psychology, 96, 1029–46. Greene, J. D., and Paxton, J. M. (2009). Patterns of neural activity associated with honest and dishonest moral decisions. Proceedings of the National Academy of Sciences, 106, 12506–11.
250
Advances in Experimental Moral Psychology
Gu, J., Zhong, C., and Page-Gould, E. (2013). Listen to your heart: When false somatic feedback shapes moral behavior. Journal of Experimental Psychology, 142, 307. Hart, D., Atkins, R., and Donnelly, T. M. (2006). Community service and moral development. In M. Killen and J. Smetana (eds), Handbook of Moral Development. Hillsdale, NJ: Lawrence Erlbaum, pp. 633–56. Immordino-Yang, M. H., McColl, A., Damasio, H., and Damasio, A. (2009). Neural correlates of admiration and compassion. Proceedings of the National Academy of Sciences, 106, 8021–6. Isen, A. M., and Levin, P. F. (1972). Effect of feeling good on helping: Cookies and kindness. Journal of Personality and Social Psychology, 21, 384–8. Jordan, J., Mullen, E., and Murnighan, J. K. (2011). Striving for the moral self: The effects of recalling past moral actions on future moral behavior. Personality and Social Psychology Bulletin, 37, 701–13. Kaiser, F. G., and Wilson, M. (2000). Assessing people’s general ecological behavior: A cross-cultural measure. Journal of Applied Social Psychology, 30, 952–78. Kayser, D., Greitemeyer, T., Fischer, P., and Frey, D. (2010). Why mood affects help giving, but not moral courage: Comparing two types of prosocial behavior. European Journal of Social Psychology, 40, 1136–57. Kohlberg, L. (1970). Education for justice: A modern statement of the Platonic view. In N. Sizer and T. Sizer (eds), Moral Education: Five Lectures. Cambridge, MA: Harvard University Press, pp. 56–83. Kohlberg, L., and Mayer, R. (1972). Development as the Aim of Education. Harvard Educational Review, 42(4). Kouchaki, M. (2011). Vicarious moral licensing: The influence of others’ past moral actions on moral behavior. Journal of personality and social psychology, 101, 702. Lapsley, D. K., and Lasky, B. (2001). Prototypic moral character. Identity, 1, 345–63. Matsuba, M. K., and Walker, L. J. (2004). Extraordinary moral commitment: Young adults involved in social organizations. Journal of Personality, 72, 413–36. McAdams, D. P. (2009). The moral personality. In D. Narvaez and D. K. Lapsley (eds), Personality, Identity, and Character: Explorations in Moral Psychology. New York: Cambridge University Press, pp. 11–29. Mischel, W., and Shoda, Y. (1995). A cognitive-affective system theory of personality: Reconceptualizing situations, dispositions, dynamics, and invariance in personality structure. Psychological Review, 102, 246–68. —(1998). Reconciling processing dynamics and personality dispositions. Annual review of psychology, 49, 229–58.
The Trouble with Nomothetic Assumptions in Moral Psychology
251
—(1999). Integrating dispositions and processing dynamics within a unified theory of personality: The Cognitive Affective Personality System (CAPS). In L. A. Pervin and O. John (eds), Handbook of Personality: Theory and Research. New York: Guilford, pp. 197–218. Mullen, E., and Nadler, J. (2008). Moral spillovers: The effect of moral violations on deviant behavior. Journal of Experimental Social Psychology, 44, 1239–45. Narvaez, D., and Lapsley, D. K. (2009). Personality, Identity, and Character. Cambridge: Cambridge University Press. Narvaez, D., Lapsley, D. K., Hagele, S., and Lasky, B. (2006). Moral chronicity and social information processing: Tests of a social cognitive approach to the moral personality. Journal of Research in Personality, 40, 966–85. Osswald, S., Greitemeyer, T., Fischer, P., and Frey, D. (2010). Moral prototypes and moral behavior: Specific effects on emotional precursors of moral behavior and on moral behavior by the activation of moral prototypes. European Journal of Social Psychology, 40, 1078–94. Perugini, M., and Leone, L. (2009). Implicit self-concept and moral action. Journal of Research in Personality, 43, 747–54. Pittel, S. M., and Mendelsohn, G. A. (1966). Measurement of moral values: A review and critique. Psychological Bulletin, 66, 22. Rand, D. G., Greene, J. D., and Nowak, M. A. (2012). Spontaneous giving and calculated greed. Nature, 489, 427–30. Sachdeva, S., Iliev, R., and Medin, D. L. (2009). Sinning saints and saintly sinners: The paradox of moral self-regulation. Psychological Science, 20, 523–8. Shafer-Landau, R. (1994). Ethical disagreement, ethical objectivism and moral indeterminacy. Philosophy and Phenomenological Research, 54(2), 331–44. Skitka, L. J., and Bauman, C. W. (2008). Moral conviction and political engagement. Political Psychology, 29, 29–54. Smith, K. D., Smith, S. T., and Christopher, J. C. (2007). What defines the good person? Cross-cultural comparisons of experts’ models with lay prototypes. Journal of Cross-Cultural Psychology, 38, 333–60. Teper, R., Inzlicht, M., and Page-Gould, E. (2011). Are we more moral than we think?: Exploring the role of affect in moral behavior and moral forecasting. Psychological Science, 22, 553–8. Turiel, E. (1983). The Development of Social Knowledge: Morality and Convention. Cambridge: Cambridge University Press. Twenge, J. M., Baumeister, R. F., DeWall, C. N., Ciarocco, N. J., and Bartels, J. M. (2007). Social exclusion decreases prosocial behavior. Journal of Personality and Social Psychology, 92, 56–66.
252
Advances in Experimental Moral Psychology
Valdesolo, P., and DeSteno, D. (2008). The duality of virtue: Deconstructing the moral hypocrite. Journal of Experimental Social Psychology, 44, 1334–8. Van Rompay, T., Vonk, D., and Fransen, M. (2009). The eye of the camera: Effects of security cameras on prosocial behavior. Environment and Behavior, 41, 60–74. Walker, L. J., and Pitts, R. C. (1998). Naturalistic conceptions of moral maturity. Developmental Psychology, 34, 403–19. Wright, J. C. (2010). On intuitional stability: The clear, the strong, and the paradigmatic. Cognition, 115, 491–503. Wright, J. C., Cullum, J., and Schwab, N. (2008). The cognitive and affective dimensions of moral conviction: Implications for attitudinal and behavioral measures of interpersonal tolerance. Personality and Social Psychology Bulletin, 34, 1461–76.
Index abortion 132, 149, 153, 157, 167, 211–14 agent moral relativism 210, 212–13, 226 agents 5, 7, 32, 73, 115, 174, 180, 183, 190, 211–16, 226–7 Algoe, S. 65 Allport, G. W. 42, 235, 247 ambient sensibilia 78 American Philosophical Association (APA) 92, 94, 98, 106n. 1, 131 Appiah, K. A. 131–2, 141 appraiser moral relativism 210 Aristotle 5 authority independence hypothesis 158–9 Batson, C. D. 39–40, 51, 246–7 behavioral immune system 10, 115–16, 121, 125 moral judgment 118–21 social attitudes 116–18 belief in a just world 192–3, 196, 200 Blass, T. 80 bystander apathy 80 Cannon, P. R. 113 charitable donation 97–8 Chomsky, N. 131 Cognitive Affective Personality System 235 cognitive science 130–1, 141 coin-flipping experiments 39–40 comparative disposition 76 competence 22–3 character judgments 30, 32 emotional responses 25 moral cognition 32–3 moral identity 30 motivations 28 status 24 traits 28–30 competitiveness 24, 27, 58, 60, 65–6
complex scoring systems 43 Co-opt thesis 114, 134 Darley, J. M. 167, 171–4, 176, 183, 189–90, 193–5, 200, 203, 214 Dasgupta, N. 116 death penalty 148, 153 debunking argument 132–3, 135–6, 138 derogation 51, 57–8, 60 disagreement folk metaethical intuitions 172–82 moral objectivism 194–5, 199–200 disgust amplifier view 112–13 consequence view 112–13 ethnic cleansing/child abuse 120 foreign outgroups 117 homosexuals 116–17 moral cognition 133 moralizer 112–13 and moral judgment 111 moral violations 119–21 permissible actions 124 repugnant foods, consumption of 119 sexual practices 119 dispositions 74 psychology of 77–80 disrupters 75, 81 domain theory of attitudes authority independence hypothesis 158–9 emotion 155–8 motivating behavior 154–5 schematic representation 150 universalism and objectivism 152–3 Durwin, A. 191 elevation 6, 64–5, 246 prosociality 61–3 email responsiveness 95 emotion 155–8
254
Index
Engel, C. 37 Entanglement thesis 134 ethicists 8 metaethicists 12 virtue ethicist 23, 33, 91–106 expectation confirmation 79 folk metaethical intuitions Beebe and Sackris’ work 168–72 disagreement studies 172–6 face evaluation studies 176–80 Knobe effect 180–2 folk psychology 131, 136, 139 foreign outgroups 117 general belief in a just world measure (GBJW) 197–8, 202–3 generosity 4, 40, 57, 61, 65, 67, 76, 82 Giner-Sorolla, R. S. 119 Goodwin, G. P. 167, 171–4, 176, 183, 189–90, 193–5, 200, 203, 214 Greene, J. 130, 132, 141 Gutierrez, R. 119 Haidt, J. 8, 60, 65, 91, 99, 113, 118–19, 121 Harman, G. 139 Helion, C. 121 Helzer, E. G. 122 Herrmann, B. 59 Hersch, M. 119 high-fidelity virtues 76–7, 81 homosexuals 116–17, 119 honesty 23, 76, 81–2, 84, 96, 245 Horberg, E. J. 119, 122 Hornsey, M. J. 151, 160 hypocrisy 39–40, 51, 244, 246 Immordino-Yang, M. H. 246 intuitive theory of metaethics 205 Knobe, J. 174, 205 Knowledge-and-Appraisal Personality Architecture 235 Kohlberg, L. 111, 158, 236 Koleva, S. P. 113 Lewis, D. K. 74, 76 Linguistic Inquiry and Word Count (LIWC) 50
Loewenstein, G. 174 low-fidelity virtues 76–7, 81 Lyons, D. 210 Lytle, B. L. 153 Marquez, M. J. 58 Mayer, R. 236 McAdams, D. P. 43 meat consumption 59, 96–7 metaethical beliefs belief in a just world 192–3 moral objectivism 190–1 moral progress 191–2 metaethical commitments 15, 183, 188, 193, 196, 199–206 intuitions 167–8, 180 objectivity 168, 173–9 Milgram paradigm 79–80 mind perception 31–2 Minson, J. A. 59 Monin, B. 58–9 mood modulators 78–9 (moral) character 1, 6–7, 21–32, 43, 49, 52, 73, 82, 86, 139, 192 moral cognition 11, 15–16, 21, 235 competence 29 disgust 133 mind perception and morality 32 person perception 29 moral conviction abortion 149 authority independence hypothesis 158–9 face valid items 150–1 motivational impact 155 nonconformity hypothesis 159–60 strong attitudes 148 moral exemplars 5–6, 9, 29, 44, 52, 167 vs. ordinary adults 45–9 moral judgment act-centered models 32 behavioral immune system 118–21 and disgust 111–14, 118, 121–4 emotions 156 mind perception 32–3 moral relativism 209–12, 215, 217 neuropsychology of 130 person-centered models 32
Index moral motivation calculation of 50 projective measure hypotheses 50–2 moral objectivism 188–90 disagreement measure 194–5, 199–200 just world measures 196, 200–1 truth-value measure 195–6, 200 moral progress 191–3, 196, 200–1 moral relativism agent moral relativism 210, 212–13, 226 appraiser moral relativism 212–14 kinds of 209–10 Morgan, G. S. 151, 153 motivating behavior 154–5 Mullen, E. 159 Nadler, J. 159 Narrow Principles model 100–2 Nichols, S. 174, 205 nonconformity hypothesis 159–60 objectivism 152–3 see also moral objectivism objectivist beliefs 190 Oosterhof, N. 177, 179 operationalizing morality first-person approach 234–5, 243–4 mixed approach 244–6 third-person approach 234–7, 242–3 outgroup bias 79 Parks, C. D. 60 patients 32, 205 peer evaluations 16, 84, 92–3, 160, 174 personal preference 149 person perception competence 26 evaluation, dimensions of 3 moral character 22–4 moral cognition 29 Plato 209 Power of Reason view 91, 100, 102–3 Pritchard, D. 83 professional ethicists, moral behavior of charitable donation 97–8 deficient intuitions, compensation for 103–4 email responsiveness 95
255
meat consumption 96–7 Narrow Principles model 100–2 Power of Reason view 102–3 rationally driven moral improvement 104–6 rational tail view 99–100 voting 92 prosociality 47, 56–7 elevation 61–3 punishment 32, 57, 60, 174, 215–21, 227 racial cognition 139–40 Rational Tail view 91, 99–100 Rawls, J. 99 relativist beliefs 190 repugnant foods, consumption of reputational status 9, 52, 56–60 Rosenberg, S. 22 Royzman, E. B. 119 Rozin, P. 114, 119
119
Sackris, D. 167–9, 171–3, 175, 183 Sawyer, P. J. 58 selective debunking argument see debunking argument self-affirmation 62 self-report measures 16, 37, 44–5, 47–8, 51 charitable donation 97–8 disgust 120 expedience 42 limitations 41–2 meat consumption 96–7 objectivity 42 selfish 38 Small, D. A. 174 Smith, A. 27–8 social attitudes 112 behavioral immune system 116–18 social domain theory 149 see also domain theory of attitudes specialization 27–8, 93 spoken words 43, 47–9 status 4, 9, 14, 24, 44, 56–7, 63–4, 66, 101, 135, 152 stereotype content model 25 Stohr, K. 94 Stone, A. B. 60 strong attitudes 148, 161
256
Index
subjunctive conditional analysis 74–5, 77–8, 81–2 Terrizzi, J. A. 117 Thematic Apperception Test 43 Todorov, A. 177, 179 Tybur, J. M. 114 universalism
152–3, 211
virtues ambient sensibilia 78 asocial situational influences 78–9 high-fidelity virtues 76–7, 81 low-fidelity virtues 76–7, 81 mood modulators 78–9 social influences 79–80
Walker, L. J. 44–5, 234 warmth 22 character judgments 30, 32 competition and status 24 elevation 61 emotional responses 25 perceptions 29 traits 23, 28, 31 weak disposition 76 Wheatley, T. 113 Williams, B. 209 Young, L. 191 yuck factor 132–3, 136 Zhong, C. B. 123