Page 0: Page 1: The Moral Behaviour of Ethicists: Peer Opinion Eric Schwitzgebel and Joshua Rust Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 If philosophical moral reflection tends to improve moral behaviour, one might expect that professional ethicists will, on average, behave morally better than non-ethicists. One potential source of insight into the moral behaviour of ethicists is philosophers’ opinions about ethicists’ behaviour. At the 2007 Pacific Division meeting of the American Philosophical Association, we used chocolate to entice 277 passers-by to complete anonymous questionnaires without their knowing the topic of those questionnaires in advance. Version I of the questionnaire asked respondents to compare, in general, the moral behaviour of ethicists to that of philosophers not specializing in ethics and to non-academics of similar social background. Version II asked respondents similar questions about the moral behaviour of the ethics specialist in their department whose name comes next in alphabetical order after their own. Both versions asked control questions about specialists in metaphysics and epistemology. The majority of respondents expressed the view that ethicists do not, on average, behave better than non-ethicists. Whereas ethicists tended to avoid saying that ethicists behave worse than non-ethicists, non-ethicists expressed that pessimistic view about as often as they expressed the view that ethicists behave better. 1. One might suppose that ethicists would behave with particular moral scruple. After all, they devote their careers to reflecting on and teaching about morality. Presumably, many of them care deeply about it. And if they care deeply about it, it is not unreasonable to expect them to act on it. Furthermore, many people might be willing to grant the following: moral reflection tends to promote moral behaviour, and professional ethicists are on average both more prone to and more skilled at moral reflection than non-ethicists. On the other hand, the connection between career and behaviour can be tenuous and complicated. Police officers commit crimes. Doctors smoke. Economists invest badly. Clergy flout the rules of their religion. Whether they do so any less than people of other professions, or any less than they would have had they chosen another Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 doi:10.1093/mind/fzp108 Advance Access publication 16 November 2009 Page 2: 1044 Eric Schwitzgebel and Joshua Rust career, can be difficult to assess.1 Likewise, Kantians lie, Confucians disrespect their elders, utilitarians buy expensive coffee. Whether they do so any less than others has never been systematically examined. We intend this essay as a preliminary investigation into this question. Since there are broad areas of agreement between mainstream ethical theories and everyday intuition, the question of whether ethicists behave better than non-ethicists by widely accepted moral standards is open to empirical investigation. The challenge, of course, is to obtain good data about ethicists’ moral behaviour. Suppose empirical research can establish that philosophical moral reflection (or a particular type of philosophical moral reflection) is, or is not, morally improving. Such results could profoundly affect both our self-conception as philosophers and our sense of the proper role of philosophical reflection in moral education and everyday life. Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 2. Obviously, no single study could resolve a question of this magnitude and complexity. We decided to begin simply by asking philosophers (both ethicists and non-ethicists) for their views on the moral behaviour of ethicists. We asked philosophers because, more than any other potential group of respondents, they have extensive interaction with a broad range of ethicists and otherwise socially comparable nonethicists. We are of course aware that responses are likely to be biased by a number of factors and at best represent beliefs based largely on behaviour as observed in professional contexts. However, even if peer 1 Doctors report smoking at rates substantially lower than do members of other professions. However, the data on nurses are mixed and the self-reports of doctors are probably compromised to some extent by embarrassment (Squier et al. 2006; Jiang et al. 2007; Lee et al. 2007; Sezer, Guler, and Sezer 2007; Smith and Leggat 2007). Studies of doctors’ general health practices are mixed but confounded by issues of convenience, embarrassment, and the temptation to self-diagnose and self-treat (Richards 1999; Kay, Mitchell, and Del Mar 2004). A Los Angeles Times reporter interviewed Nobel Prize winners in economics and says that many confess to having invested badly, especially too conservatively or passively (Gosselin 2005). On the other hand, Danish economists are more likely to hold stocks — and thus presumably not fall into the common error of excessive passivity or conservatism — than are comparably educated non-economists (Christiansen, Joensen, and Rangvid 2008). The relationship between religiosity and crime or social deviance has been extensively studied. The results here are also mixed (Hirschi and Stark 1969; Baier and Wright 2001; Eshuys and Smallbone 2006). Comparable philosophical examples might include whether decision theorists make decisions more in accord with the principles of decision theory, whether logicians commit fewer fallacies, and whether feminists are less sexist. Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 3: The Moral Behaviour of Ethicists 1045 opinion turns out only to be a mediocre indicator of the actual moral behaviour of ethicists, philosophical opinion on this issue merits study simply as a sociological or psychological fact in its own right, illuminating how optimistic or pessimistic we are, as a group, about the practical moral benefits of philosophical ethics as currently practised. In casual conversations over several years, we informally solicited the opinions of about 200 philosophers. Most of our interlocutors were sceptical of the practical value of philosophical ethics, describing it as behaviourally inert or even harmful. Many offered anecdotes about vicious ethicists (e.g. a historian of ethics repeatedly pursuing secret extramarital affairs). Only a few (mostly ethicists) stood by the idea that the serious study of philosophical ethics is, on average, morally edifying. Surprisingly to us, among philosophers expressing the view that the overall quality of ethicists’ moral behaviour varies according to their broad normative commitments (e.g. Kantianism, consequentialism, virtue ethics), nearly all said that Kantians behave on average less well than the others.2 To test opinion more formally, we set up a table in a high-traffic area outside the book display at the April 2007 American Philosophical Association Pacific Division meeting in San Francisco. The table bore a sign that said ‘Fill out a 5-minute philosophical-scientific questionnaire, get four Ghirardelli chocolate squares!’ Respondents generally sat in one of the two chairs next to the prominently displayed chocolates. Before handing them questionnaires, we assured them that their answers would be kept anonymous and we asked that they place the completed questionnaires in a ballot-style collection box. We did not reveal the contents of the questionnaire in advance. Respondents completed the questionnaire on the spot without consulting anyone else. When they had finished, we asked them orally and also in writing on a debriefing sheet not to discuss the contents of the questionnaire with other people at the meeting. Virtually everyone who received a questionnaire completed it. One respondent objected to the questionnaire on moral grounds. Over the course of three and a half days, we collected 277 questionnaires from approximately 1,500 conference attendees.3 We did not ask about this systematically. Rather our interlocutors sometimes raised the issue spontaneously on their own. We would estimate that it is about fifteen to zero so far for consequentialists and/or virtue ethicists over Kantians. If Kantians behave less well, this may harmonize with Greene 2007. 3 Although the near-100% rate of completion among those receiving the questionnaire encourages us to think that our respondents were not self-selected by attitudes toward the 2 Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 4: 1046 Eric Schwitzgebel and Joshua Rust A number of people stole candy without completing a questionnaire or took more than their share without permission. One eminent Kantian ethicist grabbed a single Ghirardelli square in passing and announced, ‘I’m being evil!’ Unfortunately, we were unable to study this behaviour systematically. 3. There were two versions of the questionnaire. Version I asked respondents to reflect on the behaviour of ethicists in general, while Version II asked respondents to reflect on the behaviour of a particular, arbitrarily selected ethicist. Each version was divided into two sub-versions (A and B) differing only in the order of the questions. Question (1) of Version I (Sub-Version A) was: (1) Take a moment to consider the various ethics professors you have known, both as colleagues and in the student–mentor relationship. As best you can determine from your own experience, do professors specializing in ethics tend, on average, to behave morally better, worse, or about the same as philosophers not specializing in ethics? (Please circle one number below.) Immediately below the question was a seven-point numerical scale, where 4 was marked ‘about the same’, 1 was ‘substantially morally better’, and 7 was ‘substantially morally worse’. The same sevenpoint scale was used in Questions (2)–(4). Question (2) asked: (2) As best you can determine from your own experience, do professors specializing in ethics tend, on average, to behave morally better, worse, or about the same as non-academics of similar social background? Questions (3) and (4) were essentially the same as Questions (1) and (2), except asking about ‘specialists in metaphysics and/or specific items on the questionnaire, we do acknowledge that under-represented in our sample were people in a hurry, people not tempted by chocolate, and people inclined to be suspicious of the intentions of two guys at a table handing out candy for completing a ‘philosophicalscientific questionnaire’. Although gender data were not recorded, we did have the impression that women responded at somewhat higher rates than men. Whether any of these factors are likely to interact with attitudes toward the moral behaviour of ethicists, we can only speculate. Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 5: The Moral Behaviour of Ethicists 1047 epistemology (including philosophy of mind)’ instead of ethicists. Sub-Version B was identical to Sub-Version A, except that the two metaphysics and epistemology specialist questions preceded the two ethicist questions. Questions (5)–(8) were demographic. Question (5) asked respondents their highest level of academic achievement (from undergraduate to distinguished professor). Question (6) asked respondents their level of professional involvement in ethics (response options being: specialist in ethics (‘area of specialization’); substantial secondary teaching or research interest in ethics (‘area of competence’); nonethicist philosopher; non-philosopher academic; academic publisher; and non-academic). Question (7) asked the type of institution at which the respondent has done most of her teaching, if she has taught at least three years beyond completing graduate study (from two-year college to university with a Ph.D. programme in philosophy). Question (8) (specially marked ‘optional’ and ‘do not answer this question if you are currently a graduate student’) asked the respondent at what institution she had done most of her graduate work. Questions (9)–(11) asked about prior knowledge of the questionnaire. Question (9) asked if the respondent had completed a similar questionnaire at the Eastern Division meeting in December 2006 (where we piloted this project). Question (10) asked if the respondent knew or suspected what the questionnaire would be about before taking it. Question (11) asked if she had heard or seen any discussion of it. A facsimile of all versions of the questionnaire is available as supplementary data at Mind online. Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 4. In all, 138 respondents completed Version I of the questionnaire. The order of the questions did not appear to make a difference.4 Nor did prior knowledge of the questionnaire, academic rank, institution type, or graduate institution.5 Results did vary by area of The mean response for each of the four main questions never differed by more than 0.24 between sub-versions, and none of the differences was statistically significant at an a-level of 0.05, using a two-tailed t-test (the lowest p-value was 0.18 [t(132) = 1.36]; SDs were 0.73–1.31). 5 Of the 277 respondents, 26 respondents revealed some prior knowledge of the questionnaire by answering ‘yes’ to at least one of the Questions (9)–(11). Their mean responses to the four main questions never differed by more than 0.25 from those answering ‘no’ to the 4 Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 6: 1048 Eric Schwitzgebel and Joshua Rust Table 1: Mean responses for Version I, Questions (1)–(4), by specialization Respondent’s specialization M&E Total Ethicists vs. Ethicists M&E vs. vs. nonvs. non- other respondents other philosophers academics philosophers academics 34 32 49 23 3.4à 3.7 4.0 3.8 3.1à 3.5 3.7 4.0 4.3à 4.3à 4.0 3.8 3.8 3.8 3.6à 3.7 Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 Ethics specialists Secondary interest in ethics Non-ethicist philosophers Non-philosophers Note: Here à indicates a statistically detectable difference from 4.0 (two-tailed t-test, p < 0.05). Pooled standard deviation: 1.07. 1 = ‘substantially morally better’, 4 = ‘about the same’, and 7 = ‘substantially morally worse’. specialization, however, as shown in Tables 1 and 2. The majority of philosophers — both 65% of ethics specialists and 68% of non-ethicist philosophers — expressed the view that ethicists do not behave better than other philosophers. However, few ethicists expressed the view that ethicists actually behave worse, while non-ethicists were about evenly divided between describing ethicists as better, worse, or about the same as other philosophers.6 Respondents with a secondary interest in ethics showed roughly intermediate results. A slender majority of ethicists (56%) expressed the view that ethicists behave better than non-academics of similar social background, while this was a minority opinion (41%) among non-ethicists. Implicit in these responses is a tendency for philosophers to think that philosophers behave morally better than non-academics. Philosophers ranked both ethicists and specialists in metaphysics knowledge questions and were never statistically significant at an a-level of 0.05 (two-tailed t-test, lowest p-value 0.55 [t(14) = 0.62]; SDs 0.24–1.53; pooling all respondents answering ‘yes’ to any one of the Questions (9)–(11) yields a minimum p of 0.40 [t(34) = 0.85] and SDs 0.23–1.27). To test for effects of academic rank and institution type we used ANOVAs with an a-level of 0.01 as a correction for multiple comparisons (academic rank: pooled SDs 0.74–1.23, lowest p-value 0.03 [F(6, 126) = 2.38, with full professors tending to rate M&E specialists better than did other ranks]; institution type: pooled SDs 0.70–1.29, lowest p-value 0.17 [F(4, 72) = 1.68]). Characteristics of graduate institution were evaluated only post hoc for obvious trends (e.g. prestige, location). 6 The difference in the rates at which ethicists and non-ethicists characterized ethicists as actually morally worse was marginally statistically significant (4/34 vs. 14/47, Fisher’s exact test, p = 0.06). It seems to be largely this difference driving the difference in the means displayed in Table 1. Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 7: The Moral Behaviour of Ethicists 1049 Table 2: Distribution of responses to Version I, ethicist questions, by specialization Better (1–3) N (%) Ethicists vs. other phil. Ethics specialist respondents Respondents with secondary interest in ethics Non-ethicist philosopher respondents Ethicists vs. non-acad. Ethic specialist respondents Respondents with secondary interest in ethics Non-ethicist philosopher respondents Same (4) N (%) Worse (5–7) N (%) 12 (35) 14 (44) 15 (32) 18 (53) 14 (44) 18 (38) 4 (12) 4 (13) Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 14 (30) 19 (56) 17 (55) 19 (41) 11 (32) 9 (29) 16 (35) 4 (12) 5 (16) 11 (24) Note: Percentages exclude respondents who left the question blank. 1 = ‘substantially morally better’, 4 = ‘about the same’, and 7 = ‘substantially morally worse’. and epistemology better in comparison with non-academics of similar social background than in comparison with other philosophers.7 (By analogy: If someone says Peter behaves better than Mark and also that he behaves the same as Tim, she implies that Tim behaves better than Mark.) Non-philosophers showed no such tendency. Indeed, among the eleven non-academic respondents, none expressed the view that ethicists behave better, on average, than non-academics, and five expressed the view that they behave worse.8 5. We hoped respondents would answer Version I of the questionnaire based on their experience of the actual behaviour of ethicists and metaphysics and epistemology specialists, as instructed in Questions (1) and (3). However, we recognized that many respondents might be driven by antecedent theoretical commitments, or by a tendency to Two-tailed paired t-test (on the mean of the ethicist and metaphysics and epistemology specialist ratings), difference in mean 0.37, p < 0.001, t(114) = 4.38, SD(diff) = 0.73. 8 Despite the tiny sample, this result is marginally statistically significant (two-tailed binomial test, p = 0.06). On average, non-academics rated the moral behavior of ethicists 3.91 compared to other philosophers and 4.45 compared to non-academics. Our impression is that most of the non-academic respondents were philosophers’ spouses. 7 Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 8: 1050 Eric Schwitzgebel and Joshua Rust overstate the value of the projects to which they are committed, or by an appreciation of irony. We were also concerned that vicious ethicists might come more readily or vividly to respondents’ minds than virtuous or ordinary ones and, so, disproportionately influence their reflections. Version II of the questionnaire was intended to avoid or minimize these potential difficulties by asking respondents to concentrate on a single, arbitrarily (alphabetically) chosen ethicist and metaphysics and epistemology specialist. In Version II (Sub-Version A) of the questionnaire, Question (1) was prefaced by the following: Think of the ethics specialist in your department whose name comes soonest after yours in alphabetical order (wrapping around from Z back to A if necessary). (If your department has no ethics specialist or you are the only one, consider the philosophy department at the institution where you received your highest degree.) Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 Question (1) was: (1) As best you can determine from your own experience, does this person tend, on average, to behave morally better, worse, or about the same as non-ethicists in your department? (The question is not about whether you enjoy this person’s company but rather, to the extent this is separable, about the moral qualities of her or his behavior — honesty, treatment of students and staff, etc.) (Please circle one number below.) Immediately below this question was the same seven-point scale as in Version I, from 1 (‘substantially morally better’) to 4 (‘about the same’) to 7 (‘substantially morally worse’). Question (2) asked: (2) As best you can determine from your own experience, does this person tend, on average, to behave morally better, worse, or about the same as non-academics of similar social background? (Please circle one number below.) The same seven-point scale followed. Question (3) asked respondents to ‘list two or three aspects of this person’s behavior most central to your assessment’.9 Questions (4)–(6) were essentially the same as questions (1)–(3), except asking about ‘M&E specialists’ (compared to ‘non-M&E specialists in your department’ and to ‘non-academics 9 We thank Jonathan Ichikawa for the suggestion to ask about the next ethicist in alphabetical order in one’s department as a means to select an arbitrary ethicist, and we thank Dale Jamieson for the suggestion to ask respondents the bases of their assessments. Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 9: The Moral Behaviour of Ethicists 1051 of similar social background’). A parenthetical remark instructed respondents to ‘interpret “M&E” in this case to refer to metaphysics and/or epistemology, including philosophy of mind’. The questionnaire concluded with the same demographic and prior knowledge questions as in Version I. Sub-Version B was identical to Sub-Version A, except that the ‘M&E’ questions preceded the ethicist questions. Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 6. The results of Version II largely mirrored those of Version I — though we discarded non-philosophers’ responses (13 out of 139 total responses) because it was unclear how they would interpret the phrase ‘the ethicist in your department’. As in Version I, no differences were evident between the sub-versions, or between naive respondents and those indicating some prior knowledge of the questionnaire, or on the basis of academic rank, institution type, or graduate school.10 Tables 3 and 4 present the results by area of specialization. Ethicists again tended, on average, to rate ethicists more favourably than they rated non-ethicists, while non-ethicists saw ethicists as no different and those with a secondary interest in ethics showed intermediate results.11 In Version II, as in Version I, the effect appears to be largely There was somewhat more variance in Version II, that is, more extreme responses than in Version I (pooled SD: Version I, 1.07; Version II, 1.41). The means on the four main questions never differed by >0.35 between sub-versions (two-tailed t-test, lowest p = 0.14 [t(134) = 1.47; SDs 1.29–1.56). Between naive respondents and the 30 respondents who indicated some prior knowledge the means never differed by more than 0.43 (treating the three prior knowledge questions separately: two-tailed t-test, lowest p = 0.09 [t(33) = 1.76]; SDs 1.05–3.06 [the last with an n of only three people who said they had taken the questionnaire at the Eastern APA]; looking at the group answering ‘yes’ to any one of the prior knowledge questions: minimum p = 0.07 [t(41) = 1.85], SDs 1.24–1.55). We again used an a of 0.01 for the demographic analyses due to multiple comparisons. There were nearly-significant trends (0.01 p < 0.05) for opinions about the moral behaviour of ethicists to worsen with rank and for professors at MA-granting institutions to view their selected colleagues (both ethicists and metaphysics and epistemology specialists) more negatively than those at other institutions (pooled SDs: academic rank 1.34–1.47, institution type 1.36–1.50). However, the Version I and Eastern APA pilot data do not confirm these trends. 11 Although the ethicists’ preference for ethicists shows in the t-tests vs. 4.0 in Table 3 and the binomial test in Table 4, two-tailed paired t-tests eth-dept vs. ME-dept and eth-nonac vs. ME-nonac are marginally significant to non-significant (respectively, diff = –0.60, p = 0.07 [t(39) = 1.89], SD(diff) = 2.01; diff = –0.33, p = 0.25 [t(38) = 1.17], SD(diff) = 1.78). We attribute the failure of significance on the paired t-tests to the high variance in the data and the relatively small sample size. The consistency of the trends among the subgroups (specialists 10 Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 10: 1052 Eric Schwitzgebel and Joshua Rust Table 3: Mean responses for Version II, Questions (1) and (2) and (4) and (5), by specialization Respondent’s specialization Total respondents Ethicist vs. others in dept. 3.4à 3.7 3.6 Ethicist vs. nonacademics 3.3à 3.5 3.4à M&E vs. others in dept. 4.0 3.9 3.5à M&E vs. nonacademics 3.6 3.7 Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 Ethics specialists Secondary interest in ethics Non-ethicist philosophers 40 36 50 3.4à Note: Here à indicates a statistically detectable difference from 4.0 (two-tailed t-test, p < 0.05). Pooled standard deviation: 1.41. 1 = ‘substantially morally better’, 4 = ‘about the same’, and 7 = ‘substantially morally worse’. Table 4: Responses to Version II, by specialization, rating of selected ethicist compared to rating of selected M&E specialist (averaging the vs. department and vs.-non-academics ratings) Respondent’s specialization Selected ethicist Same rating rated better than N (%) selected M&E specialist N (%) 10 (25) 11 (31) 12 (24) Selected M&E specialist rated better than selected ethicist N (%) 9 (23) 9 (26) 19 (38) Ethics specialist respondents 21 (53)à Respondents with secondary 15 (43) interest in ethics Non-ethicist philosopher 19 (38) respondents Note: Here à indicates a statistically significant tendency to rate the ethicist as better (two-tailed binomial test, ethicist better vs. M&E better, p < 0.05). However, a two proportion z-test of 21/30 vs. 19/38 is only marginally statistically significant (p = 0.09). driven by ethicists rarely describing the moral behaviour of ethicists as worse than the comparison groups.12 and secondary) and between Version I and Version II suggests against a purely sampling-error explanation of the difference in means. 12 In Version II, only 4 of 40 (10%) of ethics specialists expressed the view that the selected ethicist behaved morally worse than the non-ethicists in her department and only 3 of 40 (8%) expressed the view that the selected ethicist behaved morally worse than non-academics of similar social background. Non-ethicists, in comparison, ranked the selected ethicist morally Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 11: The Moral Behaviour of Ethicists 1053 As in Version I, philosophers tended implicitly to express the view that philosophers behave morally better than non-academics of similar social background by rating philosophers a bit more favourably in comparison with non-academics than in comparison with other philosophers.13 Also, overall, respondents tended to rate the arbitrarily selected individuals as better than others in their departments.14 The latter may reflect a general bias to regard individuals more favourably than groups. In light of this tendency, we view the differences between the means (which express implicit comparisons between ethicists and metaphysics and epistemology specialists and between the respondents’ colleagues and non-philosophers) as a more telling measure of opinion than the absolute means. We did not detect any systematic differences in respondents’ descriptions of the bases of their opinions. The most common responses were ‘conscientious’, ‘fair’, ‘generous’, ‘honest’/‘dishonest’, ‘integrity’, ‘kind’, ‘selfish’/‘self-centred’, and ‘thoughtful’. Ethicists and non-ethicists were about equally likely to receive each of these approbations and disapprobations. Since mainstream ethical theories broadly agree about the general content and valence of such attributions in ordinary life, we see these data as supporting our expectation that respondents’ judgements would not be grounded narrowly in standards specific to particular moral theories. The only multiply cited basis that seemed to us contentious was ‘vegetarian’ or ‘vegan’, cited in 6 of the 585 total attributions. Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 7. Our results suggest that non-ethicist philosophers do not tend to see ethicists, in general, as particularly well behaved. Indeed, a substantial minority of non-ethicists asserted that ethicists on average behave morally worse than non-ethicists. The same mediocre view of ethicists emerges when non-ethicist philosophers are asked to rate the behaviour of particular arbitrarily selected ethicists and metaphysics and epistemology specialists in their department. Across the data, respondents tended to rate their own groups a bit more favourably, on average, than other groups. Ethicists tended worse at rates of 15 out of 49 (31%) and 14 out of 49 (29%) respectively (Fisher’s exact test, p = 0.02 in both cases). 13 14 Difference in mean 0.26, two-tailed paired t-test, p < 0.001 (t(125) = 3.83), SD (diff) = 0.76. Mean 3.7, two-tailed t-test vs. 4.0, p < 0.001 (t(240) = 3.76), SD = 1.38. Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 12: 1054 Eric Schwitzgebel and Joshua Rust to rate ethicists better than they rated metaphysics and epistemology specialists; philosophers tended to rate philosophers better than nonphilosophers. By some measures non-ethicist philosophers (a substantial proportion of which must have viewed themselves as metaphysics and epistemology specialists) rated metaphysics and epistemology specialists slightly better than ethicists; and indeed the small sample of non-academics tended to rate non-academics better than philosophers (though this finding did not approach statistical significance15). Simple in-group/out-group bias may be driving these differences; or arguably one group or another could have a more accurate perception of ethicists’ and non-ethicists’ behaviour (we could see the argument going either way). The overall pattern in both versions of the questionnaire is this: ethicists rarely rated ethicists as morally worse than either of the two comparison groups and tended to be about equally divided between rating ethicists as morally better and rating them as about the same; non-ethicists were about equally divided between rating ethicists’ behaviour as morally better, the same, or worse on average than non-ethicists’; and those with a secondary interest in ethics tended to show intermediate results. Overall, the majority of philosophers expressed the view that ethicists behave no better than non-ethicists. They expressed that view directly in Version I, and they expressed it indirectly as a group in Version II, where the majority ranked the arbitrarily selected ethicists no better than the arbitrarily selected metaphysics and epistemology specialists. Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 8. Do moral reflection and philosophical ethical inquiry help us to become better people? Socrates thought so — and so did Mencius, Kant, and Mill.16 We the authors also find this view attractive. If we suppose that professional ethicists are more inclined to or skilled at such reflection than non-ethicists (especially non-academics), and if there is no reason to suspect that ethicists enter the field with a prior Combining Versions I and II, non-academics rated ethicists or metaphysics and epistemology specialists worse in comparison to non-academics than in comparison to other philosophers six times, and better three times, out of seventeen respondents. 16 For example, Plato’s (4th c. BCE/1961) Apology and Protagoras (though the end of the Meno jars a bit); Mencius 3rd c. BCE/1970; Kant 1785/1998; Mill 1859/2003. Among contemporary philosophers, see Moody-Adams 1997 and Nussbaum 1997 and 2007. 15 Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 13: The Moral Behaviour of Ethicists 1055 inclination towards delinquency, then it seems to follow that ethicists will tend to behave morally better than non-ethicists. But about twothirds of the non-ethicists and about half of the ethicists surveyed did not endorse this conclusion. Perhaps this scepticism betrays some disillusionment with the Socratic and Enlightenment ideals that many of us are otherwise so eager to share with our students. The expressed attitudes of 277 attendees at an APA meeting do not, of course, strictly imply anything either about ethicists’ behaviour or about the relationship, in general, between moral reflection and moral behaviour. Two gaps hamper the inference from our survey results to conclusions about ethicists’ behaviour: this survey is only an imperfect measure of opinion; and opinion is only an imperfect index of behaviour. Even if our respondents are a representative sample of philosophers, with stable opinions accurately expressible on a sevenpoint scale, responding to a survey of this sort at the APA is also a public act (though anonymous), and recognition of the possible consequences for (and of) a journal article like the one you are now reading must surely affect the responses. And although we think peer opinion is as good a tractable measure of moral behaviour as any other — what are we going to do, invent a moralometer? invest in a spy network? — peer opinion is of course apt to be distorted by the respondents’ limited exposure to ethicists’ behaviour, primarily in professional contexts, by shortcomings in the raters’ own visions of the moral good (especially the non-ethicists’, one might argue), by the flattering lens of friendship, by preference for one’s own group, by the salience of particular examples, etc. Because of these shortcomings in opinion surveys as a measure of behaviour, it would be desirable before reaching any sweeping conclusions to supplement our survey results with more direct measures of moral behaviour. In fact, we have already begun that project: in one study (Schwitzgebel forthcoming), Schwitzgebel examined the rate at which relatively obscure ethics books — the kind most likely to be borrowed exclusively by professors and advanced students in ethics — were missing from academic libraries compared to similar non-ethics philosophy books. The ethics books, it turns out, were somewhat more likely to be missing than the non-ethics books. In another study (on the assumption, controversial we know, that voting is a civic duty), we examined the rate at which ethicists, including political philosophers as a subgroup, voted in public elections (national, state, and local), compared to non-ethicists in philosophy, political scientists, and a group of professors in other fields Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 Page 14: 1056 Eric Schwitzgebel and Joshua Rust (Schwitzgebel and Rust forthcoming). All groups voted at about the same rate, except for the political scientists who voted about 10–15% more frequently. Other studies are under way. So far, we see no general pattern of better moral behaviour among ethicists, though we regard the question as empirically open. No one — not even Socrates we suspect — would argue that ordinary philosophical moral reflection is a panacea. Aristotle famously doubted that theoretical reflection alone could bring about moral change in those not already brought up well enough to have good habits as a foundation (e.g. fourth-century BCE/1962, p. 1095b, 1105b). Still, Aristotle’s own aim (or that for his students) in studying ethics was not just theoretical knowledge but actually ‘to become good’ (p. 1103b), so Aristotle must have thought it at least possible for philosophical inquiry to contribute to the improvement of moral character. There is of course no conflict between these two strands in Aristotle. To say that theoretical moral reflection is not by itself sufficient to produce virtuous behaviour is very different from saying that it does not on average have a good effect; an analogous point can be made of an athlete’s pre-game strategizing or weight room training and her athletic performance. Philosophical moral reflection may improve moral behaviour even if ethicists behave about the same, on average, as socially similar non-ethicists. It may be that ethicists are no more likely to engage in moral reflection than are non-ethicists (at least concerning their daily lives); or ethicists may start out morally worse and improve to average through explicit reflection; or a little reflection may be good but a lot problematic; or moral reflection may be bivalent, sometimes morally improving but just as often harmful. Of course, if further investigation comes to substantiate the opinion of the minority of philosophers who believe that ethicists actually behave worse, more sceptical explanations are possible. Bernard Williams has emphasized ways in which ethical reflection can hamper morality, for example, by undermining the use of traditional moral concepts, by introducing uncertainty, and — as in the case of the man who needs to apply a moral calculus before saving his wife from peril — by sometimes encouraging ‘one thought too many’ (1981, p. 18; 1985). Maybe explicit reflection crowds out other forms of moral responsiveness that are even better; or maybe reflection on philosophical examples eviscerates the intuitions on which we must depend; or maybe moral reflection is mostly just self-serving rationalization, at which ethicists are particularly talented. Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 Page 15: The Moral Behaviour of Ethicists 1057 Moral reflection and philosophical ethics may be inherently valuable, independent of their impact on behaviour — as is, perhaps, the study of metaphysics or of the early history of the universe. Perhaps, also, advocating moral views, for example, on environmentalism or social justice, can benefit the public sphere even if the philosophers advancing such views do not behave especially well. We the authors, however, hope for more from philosophical ethics and moral reflection than abstract knowledge and contributions to public discourse. We would like to think that, in addition, moral reflection and philosophical ethics, done well, can positively affect one’s own behaviour, and can be valuable for their tendency to point the person who reflects towards the good. If empirical inquiry eventually reveals, instead, that philosophical moral reflection is personally inert or even harmful, many of us will have to rethink our assumptions about moral psychology, moral education, and the role of reflection in the morally good life.17 Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 Department of Philosophy University of California at Riverside Riverside, CA 92521-0201 USA eschwitz@ucr.edu Department of Philosophy Stetson University Unit 8250, 421 North Woodland Boulevard DeLand, FL 32723 USA jrust@stetson.edu ERIC SCHWITZGEBEL JOSHUA RUST Supplementary data Supplementary data are available at Mind online. 17 We give our thanks to Anita Silvers and Richard Bett for permission to conduct our study at the 2006 Eastern and 2007 Pacific APA meetings; to Jeremy Wisnewski for help in conducting the survey at the Pacific; and to the many people with whom we have conversed about this issue over the years, both in person and on Eric Schwitzgebel’s blog, The Splintered Mind. Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 16: 1058 Eric Schwitzgebel and Joshua Rust References Aristotle 4th c. BCE/1962: Nicomachean Ethics. Trans. M. Oswald. New York: Macmillan. Baier, Colin J. and Bradley R. E. Wright 2001: ‘“If you love me, keep my commandments”: A meta-analysis of the effect of religion on crime’. Journal of Research in Crime and Delinquency, 38, pp. 3–21. Christiansen, Charlotte, Juanna S. Joensen, and Jesper Rangvid 2008: ‘Are Economists More Likely to Hold Stocks?’ Review of Finance, 12, pp. 465–96. Eshuys, Donna and Stephen Smallbone 2006: ‘Religious Affiliations Among Adult Sexual Offenders’. Sexual Abuse, 18, pp. 279–88. Gosselin, Peter G. 2005: ‘Experts Are at a Loss on Investing’. Los Angeles Times, 11 May 2005, A.1 ff. Greene, Joshua D. 2007: ‘The Secret Joke of Kant’s Soul’. In W. Sinnott-Armstrong (ed.) 2007, pp. 35–79. Hirschi, Travis and Rodney Stark 1969: ‘Hellfire and Delinquency’. Social Problems, 17, pp. 202–13. Jiang, Yuan, Michael K. Ong, Elisa K. Tong, Yan Yang, Yi Nan, Quan Gan, and Teh-wei Hu 2007: ‘Chinese Physicians and Their Smoking Knowledge, Attitudes, and Practices’. American Journal of Preventive Medicine, 33, pp. 15–22. Kant, Immanuel 1785/1998: Groundwork of the Metaphysics of Morals. Trans. M. Gregor. Cambridge: Cambridge University Press. Kay, Margaret P., Geoffrey K. Mitchell, and Christopher B. Del Mar 2004: ‘Doctors Do Not Adequately Look After Their Own Physical Health’. Medical Journal of Australia, 181, pp. 368–70. Lee, David J. et al. 2007: ‘Smoking Rate Trends in U.S. Occupational Groups: The 1987 to 2004 National Health Interview Survey’. Journal of Occupational and Environmental Medicine, 49, pp. 75–81. Mencius 3rd c. BCE/1970: Mencius. Trans. D. C. Lau. London: Penguin. Mill, John S. 1859/2003: On Liberty. Rpnt in M. Warnock (ed.), Utilitarianism and On Liberty, 2nd edn. Malden, MA: Blackwell, 2003, pp. 88–180. Moody-Adams, Michele M. 1997: Fieldwork in Familiar Places. Cambridge, MA: Harvard. Nussbaum, Martha C. 1997: Cultivating Humanity. Cambridge, MA: Harvard. Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 Page 17: The Moral Behaviour of Ethicists 1059 —— 2007: ‘On Moral Progress: A Response to Richard Rorty’. University of Chicago Law Review, 74, pp. 939–60. Plato 4th c. BCE/1961: The Collected Dialogues, ed. E. Hamilton and H. Cairns. Princeton, NJ: Princeton. Richards, J. G. 1999: ‘The Health and Health Practices of Doctors and Their Families’. New Zealand Medical Journal, 112, pp. 96–9. Schwitzgebel, Eric forthcoming: ‘Do Ethicists Steal More Books?’ Forthcoming in Philosophical Psychology. —— forthcoming: ‘Do Ethicists and Political Philosophers Vote More Often Than Other Professors?’ Forthcoming in Review of Philosophy and Psychology. Sezer, Hafize, Nuran Guler, and R. Erol Sezer 2007: ‘Smoking Among Nurses in Turkey: Comparison with Other Countries’. Journal of Health, Population, and Nutrition, 25, pp. 107–11. Sinnott-Armstrong, W. (ed.) 2007: Moral Psychology, Vol. 3: The Neuroscience of Morality. Cambridge, MA: MIT Press. Smith, Derek R. and Peter A. Leggat 2007: ‘Tobacco Smoking by Occupation in Australia: Results from the 2004 to 2005 National Health Survey’. Journal of Occupational and Environmental Medicine, 49, pp. 437–45. Squier, Christopher, Vicki Hesli, John Lowe, Victor Ponamorenko, and Natalia Medvedovskaya 2006: ‘Tobacco Use, Cessation Advice to Patients and Attitudes to Tobacco Control Among Physicians in Ukraine’. European Journal of Cancer Prevention, 15, pp. 548–63. Warnock, M. (ed.) 2003: Utilitarianism and on Liberty, 2nd edition. Malden, MA: Blackwell. Williams, Bernard 1981: Moral Luck. Cambridge: Cambridge University Press. —— 1985: Ethics and the Limits of Philosophy. Cambridge, MA: Harvard University Press. Downloaded from mind.oxfordjournals.org at Eccles Health Sci Lib-Serials on January 3, 2011 Mind, Vol. 118 . 472 . October 2009 ß Schwitzgebel and Rust 2009 Page 18: