Phenomenology and the Cognitive Sciences

, Volume 7, Issue 1, pp 51–66

Moral masquerades: Experimental exploration of the nature of moral motivation

Authors

Regular Article

DOI: 10.1007/s11097-007-9058-y

Cite this article as:
Batson, C.D. Phenom Cogn Sci (2008) 7: 51. doi:10.1007/s11097-007-9058-y

Abstract

Why do people act morally – when they do? Moral philosophers and psychologists often assume that acting morally in the absence of incentives or sanctions is a product of a desire to uphold one or another moral principle (e.g., fairness). This form of motivation might be called moral integrity because the goal is to actually be moral. In a series of experiments designed to explore the nature of moral motivation, colleagues and I have found little evidence of moral integrity. We have found considerable evidence of a different form of moral motivation, moral hypocrisy. The goal of moral hypocrisy is to appear moral yet, if possible, avoid the cost of being moral. To fully reach the goal of moral hypocrisy requires self-deception, and we have found evidence of that as well. Strengthening moral integrity is difficult. Even effects of moral perspective taking – imagining yourself in the place of the other (as recommended by the Golden Rule) – appear limited, further contributing to the moral masquerade.

Keywords

Moral hypocrisyMoral integrityMoral perspective takingSelf-deception

Moral principles motivate moral action. This truism undergirds much preaching, teaching, parenting, and politicking. But is it true? Is the person who endorses a moral principle more likely to act accordingly? Or, as astute observers of the human condition like Jane Austen (1811/1969), Charles Dickens (1843–1844/1982), and Mark Twain (1884/1959) have it, do morals often serve another master, providing convenient and high-sounding rationalizations for self-interest? Even though the term is harsh, I know no better name for this latter possibility than moral hypocrisy. Morality is extolled – even enacted – not with an eye to producing a good and right outcome but to appear moral (or at least not immoral) yet, if possible, still benefit oneself. (Webster’s Dictionary 1990, defines moral as “1. of or concerned with principles of right or wrong conduct. 2. being in accordance with such principles” [p. 589]; it defines hypocrisy as “a pretense of having desirable or publicly approved attitudes, beliefs, principles, etc., that one does not actually possess” [p. 444].)

Moral philosophers often challenge us to act in accord with some universal and impartial moral principle, such as justice (Rawls 1971). They have not, however, entirely ignored the question of whether acting with an ultimate goal of upholding a moral principle is really possible. Even Kant (1785/1898) admitted that action that appears prompted by duty to principle may actually be prompted by self-love:

We like to flatter ourselves by falsely taking credit for a more noble motive....A cool observer, one that does not mistake the wish for the good, however lively, for its reality, may sometimes doubt whether true virtue is actually found anywhere in the world....(Section 2, paragraph 2).

Kant quickly set this doubt aside. Given the current moral climate, I think we badly need to pick it up again, considering why and how people might be motivated to appear moral yet remain steadfastly focused on self-benefit.

Why be a moral hypocrite?

There are many possible reasons for wanting to appear moral both to others and to oneself. Most obviously, one can garner the social and self-rewards of being moral – such as feeling pride – as well as avoid the social and self-punishments for failing to be so – such as guilt. Freud (1930/1961) suggested that society inculcates moral principles in us when young in order to bridle our selfish impulses by making it in our best interest to act morally even when unobserved. We are constrained by conscience. But the bridle chafes; there is an unrelenting impulse to break free. One way out of this bind is a moral masquerade, to make a show of morality, but only a show.

How?

From the perspective of moral philosophy, the generality and abstractness of a universal moral principle like justice are major virtues. They expand our moral universe beyond narrow partialities of self-interest, kinship, friendship, and group interest. From the perspective of moral psychology, especially the psychology of moral motivation, generality and abstractness can be an Achilles’ heel. The more general and abstract a principle is, the more vulnerable it is to rationalization.

Most people are adept at moral rationalization, at justifying to themselves – if not to others – why a situation that benefits them or those they care about does not violate their principles. Why, in Jonathan Kozol’s (1991) apt phrase, the “savage inequalities” between public school systems of rich and poor communities in the USA are just. Why storing one’s nuclear waste in someone else’s backyard or using a disproportionate amount of the earth’s natural resources is fair. Why the killing of innocents by one’s own side is a regrettable necessity, whereas such killing by the other side is an atrocity. The abstractness of most moral principles makes such rationalization especially easy. Principles may be used more reactively than proactively, more to justify or condemn action than to motivate it.

Economist Robert Frank (1988), building on biologist Robert Trivers’s (1971) ideas about reciprocal altruism, presented a similar analysis of moral motivation, but with an interesting twist. Frank argued that people are motivated to present themselves as passionately committed to moral principles in order to gain the self-benefits that the ensuing trust provides. He also argued that, due to our well-developed ability to detect deception, the effort involved in shamming this commitment is so great that the more evolutionary stable strategy is genuine commitment. Our ancient ancestors may have taken up morality lightly as part of a masquerade, but over time, natural selection favored those whose appearance of morality was genuine. Only in them was hypocrisy likely to be undetected.

If Frank is right, then moral principles should motivate moral action. Trivers (1985, pp. 415–420; also see Alexander 1987, pp. 114–125) suggested another possibility. If I can convince myself that serving my own interests does not violate my principles, then I can honestly appear moral and so avoid detection without paying the price of actually upholding the principles. In the moral masquerade, self-deception may be an asset, making it easier to deceive others.

These observations about moral hypocrisy suggest a two-step process. The first step is perceptual. It involves seeing a choice or decision as moral, as an opportunity to uphold or violate some moral principle. The second step is motivational. It involves seeking to maximize personal gain by appearing moral while, if possible, not incurring the costs associated with actually being moral.

Moral integrity

Perhaps, however, the cynical claim that moral principles lack intrinsic motivating power is wrong. Even if principles are inculcated in childhood through appeals to reward and punishment, once they are internalized, upholding them may become an ultimate rather than an instrumental goal. One may be motivated not simply to appear moral but to actually be moral – or at least not immoral. Intrinsic moral motivation of this kind, if it exists, might be called moral integrity (McKinnon 1991; Williams 1976).

Intrinsic moral motivation – moral integrity – should not be equated or confused with altruistic motivation. The ultimate goal of the former is to act in accord with one or more moral principles; the ultimate goal of the latter is to increase another’s welfare (Batson 1994). Research has documented the difference between these two distinct forms of prosocial motivation. Batson et al. (1995) found evidence that empathy-induced altruism – much like self-interested egoism – can conflict with and, at times, overpower moral motivation.

Testing the nature of moral motivation: A moral dilemma in the lab

What is the nature of moral motivation – hypocrisy, integrity, both, or neither? To address this question empirically, one needs a research strategy that enables one to determine the ultimate goal motivating some moral act. Following the same logic used over the past several decades to empirically test for the existence of altruistic motivation (Batson 1991), two steps seem necessary.

First, we must elicit moral action. In so value-laden an area as morality, it is unwise to trust self-reports or judgments about what one would do in a hypothetical situation; we need to infer motivation from behavior. Moral action by itself, however, only tells us there is some motivation, not the nature of that motivation. Therefore, second, we must vary the circumstances under which the moral act occurs, varying them in a way that reveals whether the goal is to uphold moral principle (integrity) or to uphold the appearance of morality only to serve self-interest (hypocrisy).

To implement this strategy colleagues and I created a moral dilemma in the laboratory. Our dilemma was a simple, straightforward zero-sum conflict between self-interest and the interest of another person. We purposely wanted a dilemma that was real, not hypothetical, so we could observe actual behavior. We wanted a dilemma that was simple, not complex, so there would be no problem understanding what was at stake and so we could easily create variations to disentangle goals. We wanted a dilemma that pitted self-interest against the interest of another individual, not against the interest of a hallowed institution such as church or state, so responses would not be affected by institutional affiliation or allegiance. We wanted a dilemma that was mundane, not stereotypically moral (e.g., stealing or killing), so we would get less scripted responses. We wanted a dilemma where there would be broad consensus about the morally right course of action, not one where opinions would conflict, so we could know what action would be considered moral.

Drawing on the widely endorsed principle of procedural justice (“be fair”), we gave research participants the chance to assign themselves and another participant (actually fictitious) to tasks. There were two tasks: a positive consequences task, on which each correct response earned a raffle ticket for a $30.00 prize, and a neutral consequences task, on which each correct response earned nothing and which was described as rather dull and boring. One person had to be assigned to each task. Participants were told that the other person would not know that they were allowed to assign the tasks; the other would think the assignment was made by chance. Participants were left alone to make the assignment decision, simply indicating which task consequences each participant would receive by entering an S (for self) and an O (for other) in the two blanks on an assignment form. They knew they would never meet the other participant.

Documenting a clear preference for the positive consequences task, Batson et al. (1997b, Study 1) found that most research participants faced with this simple dilemma assigned themselves the positive consequences task (80% – 16 of 20), even though when asked later on an open-ended questionnaire item (“In your opinion, what was the most morally right way to assign the task consequences?”), very few said that assigning oneself to the positive consequences was the most moral way to assign the consequences (5% – 1 of 20). Moreover, when asked later in the questionnaire to rate the morality of their own action (“Do you think the way you made the task assignment was morally right?” 1 = not at all; 9 = yes, totally), the 80% who assigned themselves to the positive consequences rated the morality of their assignment relatively low (M = 4.38) compared to the 20% who assigned the other person to the positive consequences (M = 8.25; see Row 1 of Table 1). Thus, the action of the majority of participants failed to match their retrospective judgments about the most moral thing to do and about the morality of what they had done. This behavior-standard discrepancy was not, however, evidence of moral hypocrisy. It could simply reflect moral oversight. Participants may not have thought about the moral implications of their decision until it was too late.
Table 1

Percentage assigning self to positive consequences, percentage perceiving assigning self to positive as most moral, and rated morality of the way one assigned the tasks

Task assignment condition

Assign self to positive consequences (%)

Perceive self to positive as most moral (%)

Rated morality of the way one assigned the tasks

By those who assigned self to positive

By those who assigned other to positive

1. Moral standard not salient; no coin to flipa

80

5

4.38

8.25

2. Moral standard salient; coin to flip. Participant did not flip coinb

90

10

3.56

8.00

3. Moral standard salient; coin to flip. Participant flipped coinb

90

0

7.11

9.00

4. Moral standard salient; labeled coin. Participant did not flip coinc

83

17

3.90

8.50

5. Moral standard salient; labeled coin. Participant flipped coinc

86

0

7.42

9.00

6. Moral standard salient; colored labels on coin. Participant did not flip coind

75

0

3.89

8.00

7. Moral standard salient; colored labels on coin. Participant flipped coind

86

0

6.33

8.80

8. Moral standard salient; coin; not facing mirror. Participant did not flip coine

85

8

4.00

8.00

9. Moral standard salient; coin; not facing mirror. Participant flipped coine

85

0

7.82

9.00

10. Moral standard salient; coin; facing mirror. Participant did not flip coine

62

0

3.50

8.00

11. Moral standard salient; coin; facing mirror. Participant flipped coine

50

0

8.80

8.00

aBatson et al. (1997b, Study 1). Replicated, Batson et al. (1999, Study 3).

bBatson et al. (1997b, Study 2). Replicated, Batson et al. (1999, Study 2).

cBatson et al. (1999, Study 1).

dBatson et al. (2002, Study 2).

eBatson et al. (1999, Study 2).

Two procedural additions were necessary to provide a test of the nature of moral motivation. First, the moral relevance of the decision needed to be made salient. Second, there needed to be sufficient “wiggle room” or ambiguity so that it would be possible to appear moral without actually having to be moral. If participants are motivated to actually be moral (moral integrity), those for whom the relevance of fairness to the task assignment is made salient should not favor themselves – even when provided with enough wiggle room that they can appear fair without having to be fair. If participants are motivated to appear moral yet, if possible, avoid the cost of being moral (moral hypocrisy), those provided with such wiggle room should favor themselves. Only when there is no wiggle room – when the appearance–reality link is unambiguous – should moral hypocrisy produce a moral result. Batson et al. (1997b) used two techniques to introduce wiggle room: They allowed participants to flip a coin (in private) to assign the tasks (Study 2) or to accept a “random” assignment that they knew gave them the positive task (Study 3). Results of each technique led to the same conclusion, so I shall focus only on the first.

Initial tests of the nature of moral motivation

To make the moral relevance of the task assignment salient, Batson et al. (1997b, Study 2) included the following sentence in the task-assignment instructions: “Most participants feel that giving both people an equal chance – by, for example, flipping a coin – is the fairest way to assign themselves and the other participant to the tasks.” To create wiggle room, Batson et al. provided a coin for participants to flip if they wished (a quarter in a clear plastic pouch taped to the inside of the folder containing the task-assignment instructions). Under these conditions, virtually all participants said in retrospect that using a fair method such as the coin flip was the most moral way to assign the consequences. Yet only about half chose to flip the coin.

Of those who chose not to flip, 90% assigned themselves to the positive consequences task (see Row 2 of Table 1), similar to what participants without the sentence about fairness and with no coin had done (Row 1). More interesting and revealing, the same was true among those who flipped the coin; 90% assigned themselves the positive-consequences task (Row 3 of Table 1). Clearly, either the coin was very charitable or some who flipped it failed to abide by the outcome. The appearance of morality – flipping the coin – while avoiding the cost of being moral – in spite of the result of the flip assigning self the positive consequences – provided evidence of moral hypocrisy. Had participants been motivated by moral integrity, assignment after the coin flip would have been unbiased.

In spite of the same pattern of assignment, there was one notable difference between participants who assigned themselves to the positive consequences after flipping the coin and those who made this assignment without flipping. Those who flipped the coin rated the morality of the way they assigned the tasks considerably higher (M = 7.11) than those who did not flip the coin (M = 3.56). Even though many did not abide by the outcome of the flip, they managed to see themselves as moral (or to say they were).

The pattern of results reflected in Rows 2 and 3 of Table 1 has proved very robust. In study after study, the percentage of participants assigning themselves the positive consequences after flipping the coin (always in private) has been significantly greater than the 50% that would be expected from an unbiased coin flip, providing evidence of motivation to appear moral yet avoid the cost of being moral (i.e., moral hypocrisy). It was found even in a study in which the less desirable consequences were more obviously negative – uncomfortable electric shocks (Batson et al. 2001). This variation led a smaller percentage of participants to flip the coin (30%) but did not reduce the percentage of those who assigned themselves to the positive consequences after flipping (100%). It was also found in a study in which participants were told that the other participant knew they were assigning the tasks (Batson et al. 2002, Study 1). This variation led a larger percentage of participants to flip the coin (90%) but did not noticeably reduce the percentage of those who assigned themselves to the positive consequences after flipping (89%). It was found in a study in which participants indicated the importance of their making the assignment in a fair way (and whether they flipped the coin) at the top of the form on which they made the task assignment (Batson et al. 2005). Again, this variation led more participants to flip the coin (83%) but did not noticeably reduce the number who assigned themselves to the positive consequences after flipping (93%).

Nor has the evidence of moral hypocrisy been limited to those scoring relatively low on measures of personal morality; quite the contrary. Batson et al. (1997b, Study 2) found that an index of personal moral responsibility correlated positively with choosing to flip the coin (r = 0.40). Among participants who flipped, however, those who scored higher on this index were no less likely to assign themselves to the positive consequences task than those who scored low. Thus, those with greater self-reported moral responsibility did not show signs of greater moral integrity; they showed signs of greater hypocrisy. They were more likely to appear moral (flip the coin) but no more likely to be moral (allow the coin flip to determine the task assignment). Batson et al. (2002, Study 2) replicated this finding.

Taken together, the results of these studies seem to provide considerable evidence of moral hypocrisy and very little evidence of moral integrity. The results consistently conform to the pattern we would expect if the goal of those who flipped the coin was to appear moral yet, if possible, avoid the cost of being moral. These results beg for a clearer understanding of the psychology underlying the moral hypocrisy effect.

Exploring possible explanations for the moral hypocrisy effect: The role of self-deception

How did participants in these studies manage to negotiate the task assignment dilemma so that they could appear moral yet still unfairly favor themselves? They do not seem to have distorted or disengaged their moral standards. When asked later about the most morally right way to assign the tasks, by far the most common response was (a) use a random method (e.g., flip a coin), followed by (b) give the other participant the positive consequences. Moral standards were still there, and for many so was the appearance of morality (flipping the coin), but real morality was rare (the outcome of the flip was biased).

More is at stake here than appearing moral to others. If one is to gain the self-rewards for being moral, and more importantly perhaps, to avoid the self-punishments for being a hypocrite, then one must appear moral to oneself. How is this possible when violating one’s moral standards to serve self-interest?

As noted earlier, one way is to engage in self-deception, as suggested by Trivers (1985). Self-deception is a concept that has been used widely and with varying meanings in both philosophy and psychology, so it is important to consider (a) what form of self-deception is implied by the motive of moral hypocrisy and (b) what strategies might serve this motive. Sometimes self-deception has been thought to require that a person simultaneously hold two contradictory beliefs, being unaware of one (e.g., Gur and Sackheim 1979), or that a person believe what at some level he or she knows is not so (e.g., Demos 1960). Such paradoxical forms of self-deception are not required for moral hypocrisy. To reach the goal of appearing moral to myself, it is sufficient that I engage in what Mele (1987) has called “ordinary self-deception,” or “desire-influenced manipulation of data” (p. 126). The goal of moral hypocrisy can be reached if I can manipulate the data so that I avoid confronting the discrepancy between my self-serving behavior and my moral standards.

How can I manipulate the data to avoid this discrepancy? Assuming that (a) I have behaved in a way that violates my moral standards and (b) responsibility for the behavior cannot be denied or justified, ordinary self-deception strategies that would serve moral hypocrisy can be classed into two types. First, I could perceive my behavior as moral (i.e., as being in line with my moral standards) even though it actually is not. Second, I could avoid comparing my behavior to my moral standards. The first of these two strategies, if available, seems preferable because the second leaves me vulnerable to anything that might focus my attention on the behavior-standard discrepancy. If I can convince myself that I have acted morally, then my behavior can be scrutinized from the perspective of my moral standards with impunity, even pride.

Might participants in the studies providing evidence for moral hypocrisy have been able to perceive their self-serving behavior as in line with their standards? Might they, for example, have been able to perceive their coin flip as fair even when it was biased in their favor? It seems possible. Participants were left alone to decide whether to flip the coin and, if so, how to interpret the outcome. They may not have decided the meaning of heads and tails in advance. Rather, once they saw the result of the flip was heads (or tails), they may have “remembered” that this result assigned them the positive consequences task: “It’s heads. Let’s see, that means I get the positive task.” “It’s tails. Let’s see, that means the other participant gets the neutral task.” In this way, they could enjoy both the positive consequences and the knowledge that they had assigned the tasks fairly because they abided by the result of the coin flip.

Labeling the coin to eliminate the first type of self-deception

To the degree that participants relied on this first type of self-deception, it should be relatively easy to diminish if not eliminate the moral hypocrisy effect. All that is required is to clearly identify who gets which task if the coin comes up heads or tails. Accordingly, Batson et al. (1999, Study 1) placed participants in the task assignment dilemma, making the moral standard of fairness salient and providing a coin. This time, however, there was a sticker on each side of the coin; one said “SELF to POS” and the other said “OTHER to POS.” If, using this labeled coin, the hypocrisy effect disappeared (i.e., there was no longer a significant deviation from 50 percent in the direction of assigning self the positive consequences), this would provide evidence the hypocrisy effect is a product of the first self-deception strategy. If the effect remained, it would be necessary to look elsewhere for an explanation.

As can be seen in Rows 4 and 5 of Table 1, adding the labels did nothing to reduce the moral hypocrisy effect. Once again, the percentage of those assigning themselves to the positive consequences task after flipping the coin (86%) was much like the percentage doing so without flipping the coin (83%), and was significantly greater than the 50% one would expect from a fair coin flip. And once again, those who assigned themselves to the positive consequences task after flipping the coin rated the morality of the way they assigned the tasks higher (M = 7.42) than did those who assigned themselves to the positive consequences task without flipping the coin (M = 3.90).

Participants said the labels were important. When asked afterward whether it would make any difference if the coin had no labels, most participants said that without labels it would be easier to cheat (e.g., “It would be easier to pick me if the coin was not labeled. You could fool yourself and say, ‘No, I meant that side.’ ” “Without the labels you’d be more likely to change your mind after the flip.” “People would probably cheat their way out and take the positive for themselves.” “It’s not as concrete; you could fudge it.” “You could play mind games until you came to the conclusion that you get the positive.”). Still, labeling the coin did not eliminate–or even noticeably reduce–the hypocrisy effect.

After running this study, we realized that labels on the coin could provide even more useful information if the two labels were of different colors. The small window in the door of the research cubicle in which participants sat to make the task assignment decision was covered with paper. By peeking through a tiny opening in the covering, it would be possible to see not only whether participants flipped the coin but also which color label came up. This way, we could know which participants won the flip fair and square and which did not. Comparing the ratings of the morality for each of these groups with the ratings of those who assigned themselves to the positive consequences without flipping the coin would shed light on an ambiguity in the results of previous studies. We could learn whether the higher rating among those who flipped the coin was simply a product of those who flipped, won fairly, and assigned themselves to the positive consequences or whether the charade of a dishonest flip was sufficient to produce higher ratings of morality.

Accordingly, Batson et al. (2002, Study 2) placed a yellow “SELF to POS” sticker on one side of the coin and a blue “OTHER to POS” sticker on the other side. As can be seen in Rows 6 and 7 of Table 1, responses were much the same as in the previous labeled-coin study and in the studies in which the coin was not labeled. However, the surreptitious observation permitted classification of task assignment behavior into four categories: (a) assign other to positive task (8 participants did this – 3 without flipping the coin, 5 with); (b) flip the coin, get SELF to POS, assign self to positive task (11 participants did this); (c) not flip the coin, assign self to positive task (9 participants did this); and (d) flip the coin, get OTHER to POS or otherwise “fiddle” the coin flip, assign self to positive task (16 participants did this). Of the 16 participants in the final category, five participants flipped the coin once, got OTHER to POS, yet assigned themselves to the positive consequences task. Seven used the coin, but rigged the flip so that they “won.” (For example, several participants who lost the initial flip, flipped the coin again until it came up SELF to POS.) Four did not flip the coin at all, but reported on the task-assignment reaction questionnaire and in a post-decision interview that they did.

Table 2 presents – separately for each of the four task-assignment-behavior categories – participants’ mean ratings of the morality of the way they made the task assignment decision. As can be seen, those in Category 1, who assigned the other to the positive consequences task, rated the morality of the way they made the decision quite high (M = 8.50). So did those in Category 2, who flipped the coin, won, and assigned themselves to the positive consequences task (M = 7.45). In contrast, participants in Category 3, who did not flip the coin and assigned themselves to the positive consequences task, rated the morality of the way they made the task assignment decision relatively low (M = 3.89). Most important, participants in Category 4, who fiddled the coin flip and assigned themselves to the positive consequences task, rated the morality of the way they made the decision moderately high (M = 5.56), significantly higher than participants in Category 3. Nor was the higher rating among Category 4 participants a function of only one type of fiddler (e.g., those who rigged the flip); a comparison of rated morality among the three types revealed no statistically reliable differences.
Table 2

Mean rated morality of the way they assigned the tasks by participants in four task-assignment-behavior categories

Task-assignment-behavior category

N

M

SD

Assign other to positive task

8

8.50

0.76

Flip the coin, win, and assign self to positive task

11

7.45

1.37

Not flip the coin, assign self to positive task

9

3.89

1.45

Fiddle the coin flip, assign self to positive task

16

5.56

2.37

Adapted from Batson et al. (2002).

In sum, even though the coin had no more effect on their decision than it had on the decision of those who did not claim to use the coin at all, the fiddlers still said they thought the way they made the task assignment decision was more moral. Their sham use of the coin seems to have provided sufficient appearance of morality that they could claim to have acted, if not totally morally, at least moderately so. And by fiddling the flip, they made sure that they got the more desirable task. They were able to appear moral – or at least not immoral – without incurring the cost of being moral. This pattern of responses could not be explained by the first type of self-deception. However, it seemed quite consistent with the second type.

Using a mirror to eliminate the second type of self-deception

How might the second type of self-deception – avoid comparing one’s immoral behavior (unfairly assigning oneself the positive consequences task) with one’s moral standards (be fair) – operate in the task-assignment situation? Consistent with Baumeister and Newman’s (1994) analysis of the role of strategically interrupted thought in self-deception, those who fiddled the flip may have reasoned to the point, “I flipped the coin, which is fair,” and stopped without adding, “But I ignored the result, which is not.” In this way, they may have managed to avoid assessing their behavior in light of the salient moral standard.

Self-awareness manipulations, such as looking at one’s face in a mirror, have been found to heighten awareness of discrepancies between behavior and salient personal standards, creating pressure to act in accord with standards (Wicklund 1975). In the moral domain, for example, Diener and Wallbom (1976) induced self-awareness in research participants who had the opportunity to cheat on a test. They found that those who were self-aware cheated significantly less often than those who were not self-aware.

If people who are motivated to appear moral yet avoid the cost of actually being moral rely on not comparing their self-serving behavior to their moral standards, then introducing self-awareness should reduce or eliminate the hypocrisy effect. It should make the behavior-standard discrepancy salient, creating pressure to act in accord with the standard. Pursuing this logic, Batson et al. (1999, Study 2) placed participants in the task assignment dilemma, gave them the statement about fairness, and provided a coin (unlabeled). A mirror was also present. Participants in the high self-awareness condition were seated facing the mirror while deciding how to assign the tasks; those in the low self-awareness condition were also seated facing the mirror, but it was turned to the wall so they saw only its back. A sign in the corner of the mirror said: “ANDERSON STUDY – DON’T TOUCH PLEASE,” and participants were asked not to move the mirror. None did.

As can be seen in Rows 8 and 9 of Table 1, results in the low self-awareness condition of this study were much as in previous studies. Even among those who flipped the coin, well over half (85%) assigned themselves to the positive consequences task. However, results in the high self-awareness condition were dramatically different. Although participants who did not flip the coin were somewhat more likely to assign themselves to the positive consequences task (62%), those who flipped the coin while facing themselves in the mirror showed no bias (50% – see Rows 10 and 11 of Table 1). Reflecting the fairness of the assignment, these participants rated the morality of the way they assigned the tasks very high on the 1–9 scale (M = 8.80). The difference across the self-awareness conditions was precisely the pattern expected if making the behavior-standard discrepancy salient rendered the second type of self-deception ineffective, and forced participants to actually be moral in order to appear moral. It suggests that the second type of self-deception plays an important role in moral hypocrisy.

Perspective taking: A stimulus to moral integrity?

Having found evidence of the prevalence of moral hypocrisy and having some idea of the underlying psychology, but finding little evidence of moral integrity, one begins to think about possible ways to stimulate moral integrity. One strategy that has been suggested by religious teachers, moral philosophers, and moral psychologists is perspective taking.

Perhaps the most universal religious prescription for morality is the Golden Rule, “Do unto others as you would have them do unto you” (e.g., Matthew 7:12). This rule implies an act of perspective taking in which you mentally place yourself in the other’s situation. Presumably, to imagine how you would like to be treated provides the standard for how you should treat the other, leading you to consider the other’s interests as well as your own.

Mark Johnson (1993) makes the moral significance of perspective taking explicit in his analysis of moral imagination. He argues that moral insight and sensitivity require the ability to imagine ourselves in the other’s place:

Unless we can put ourselves in the place of another, unless we can enlarge our own perspective through an imaginative encounter with the experience of others, unless we can let our own values and ideals be called into question from various points of view, we cannot be morally sensitive....It is not sufficient merely to manipulate a cool, detached “objective” reason toward the situation of others. We must, instead, go out toward people to inhabit their worlds. (Johnson 1993, p. 199–200)

The idea seems to be that if individuals can be induced to take the perspective of another with whom their own interests conflict, they will be more inclined to move beyond narrow self-interest to consider and give weight to the interests and desires of the other. As a result, they will be more motivated to uphold moral principles that require consideration of the interests of others, such as principles of fairness and justice – i.e., they will display moral integrity.

Two different perspectives on another’s situation; two different motives

In his classic early research on empathy, Ezra Stotland (1969) identified two different forms of perspective taking. Using both self-report and physiological measures, Stotland found that (a) imagining what one’s own thoughts and feelings would be if one were in the situation of a person in need (an imagine-self perspective) and (b) imagining the thoughts and feelings of the person in need (an imagine-other perspective) both led to increased emotional arousal compared to an objective perspective condition. However, the emotions aroused by these two perspectives were not the same. An imagine-self perspective appeared to produce a mix of self-oriented distress feelings (tense, upset, etc.) and other-oriented empathic feelings (sympathetic, compassionate, etc.), whereas an imagine-other perspective produced relatively pure empathic feelings. (For further evidence of this difference in emotions produced by these two imagine perspectives, see Batson et al. 1997a.)

The Golden Rule and Johnson seem to agree that an imagine-self perspective is the one that should stimulate moral integrity. To actually be moral, a person should first imagine him or herself in the other’s place. This reasoning suggests that if research participants in the task-assignment paradigm are induced to imagine themselves in the other participant’s situation prior to making the task assignment, it will stimulate moral integrity. These participants should be more likely to flip the coin, and among those who flip, the outcome should be more fair.

Results from a number of experiments have indicated that an imagine-other perspective evokes empathic emotion, which in turn leads to increased altruistic motivation, not to increased moral motivation (Batson 1991; Batson et al. 1995). This research suggests that in the task-assignment paradigm, participants induced to imagine the other’s feelings prior to making the assignment will not be more fair. Instead, motivated at least in part to increase the other’s welfare, they will be more likely to assign the other participant to the positive-consequences task directly, without flipping the coin.

Effects of perspective taking on task assignment

To test these different predictions, Batson et al. (2003) had some participants perform a brief imagination exercise before making the task assignment decision, whereas others did not. Among the former group, half were asked to “imagine yourself in the place of the other participant” for one minute, then write down what they imagined (imagine-self condition); the other half were asked to “imagine how the other participant likely feels,” then write (imagine-other condition).

As can be seen in Table 3, the imagine-self perspective had little effect on the fairness of the task assignment. It produced a small and not statistically reliable reduction in the percentage of participants assigning themselves to the positive consequences task after flipping the coin (67%) compared to participants not asked to do an imagination exercise (85%). It produced a small and not statistically reliable increase in the percentage assigning themselves to the positive consequences without flipping the coin (89% vs. 64%). Nor did the imagine-other perspective have much effect on the fairness of the task assignment after flipping the coin (again, 67% assigned themselves to the positive consequences task). What is striking is the percentage of participants in the imagine-other condition who assigned themselves to the positive consequences task without flipping the coin (27%). It was by far the lowest percentage in any study using the task assignment procedure. After imagining the other’s feelings, almost three-fourths assigned the other to the positive consequences. Consistent with the idea that the imagine-other perspective produced empathy-induced altruistic motivation, the likelihood of participants in this condition assigning the other to the positive consequences task without flipping was significantly positively correlated with reported empathy felt for the other.
Table 3

Percentage of participants assigning self to positive-consequences task in each perspective condition

Did participant flip coin?

Perspective condition

No perspective (%)

Imagine self (%)

Imagine other (%)

No

64

89

27

Yes

85

67

67

Adapted from Batson et al. (2003).

Symmetrical and asymmetrical moral dilemmas

Why did the imagine-self perspective, widely touted as a stimulus to moral integrity, have such a small effect? Is it possible that those extolling the virtues of this form of perspective taking are simply wrong? I suspect more. The moral dilemma faced by participants in making the task assignment is of a particular type. The pre-assignment plight of both participants is exactly the same; each faces the prospect of being assigned either to a more desirable or to a less desirable task. In a symmetrical dilemma like this, to imagine oneself in the place of the other participant may not lead one to give more weight to the other participant’s interests, stimulating moral integrity. Instead, it may lead one to focus on one’s own interests.

If the inability of the imagine-self perspective to stimulate moral integrity was due to the symmetrical nature of the dilemma, then an imagine-self perspective should be more effective when one’s own and the other’s initial situations are not the same. When each is in the same initial situation, imagining oneself in the other’s place may lead one to focus on one’s own interests, not the other’s. When the other’s need is clearly greater than one’s own, imagining oneself in the other’s place may provide insight into what it is like to be in the other’s position of disadvantage and, as a result, lead to a more productive focus on the other’s interests, stimulating moral integrity. To illustrate: When considering whether to vote for an increase in one’s own taxes in order to fund a job-training program for the unemployed, imagining oneself in the place of someone in need of a job may stimulate moral action.

Pursuing this logic, Batson et al. (2003) ran a second study in which participants were told that they had been initially assigned to an asymmetrical condition in which they would receive two raffle tickets for each correct response on their task and the other participant would receive zero. However, if they wished, they could switch to a symmetrical condition in which each participant would receive one raffle ticket for each correct response. Participants who made the decision about switching after imagining themselves in the place of the other participant were far more likely to make the switch (83%) than were participants who engaged in no imagination exercise (38%—an imagine-other perspective condition was not included in this study). In this case, an imagine-self perspective did seem to stimulate moral integrity.

Results of these two studies suggest that an imagine-self perspective may have a limited, but still quite important, role in stimulating moral integrity. Imagining oneself in the other’s place may provide a corrective lens for the specific moral myopia to which a position of advantage is prone. The myopia of the advantaged is legendary. Those who, like Candide (Voltaire 1759/1930), live in the best of all possible worlds are not likely to trouble themselves thinking about the worlds in which others live. Those innocently born with a silver spoon in their mouth are not likely to ask whether it is morally right to keep it there. If introducing an imagine-self perspective can effectively stimulate the moral sensitivity of persons of privilege, then it has done important work.

Another mask?

This very effectiveness may, however, lead to a less salutary consequence. Persons of privilege, aware of the potential power of imagining themselves in the place of the less advantaged, may not simply neglect to adopt this perspective. They may actively resist it. If so, admonition or instruction to imagine oneself in the other’s place is likely to fall on deaf ears. This possibility, which raises the specter of motivation to avoid imagining oneself in the place of the less fortunate in order to avoid stimulating moral integrity, seems worth pursuing in future research. Distinct from moral hypocrisy, it may be another important reason for the lack of moral integrity. It is a mask that blinds one to unfairness. Moral hypocrisy is a mask that allows one to appear fair – even to oneself – yet still exploit advantage. Each mask promotes the moral masquerade.

Copyright information

© Springer Science+Business Media B.V. 2007