Can philosophical accounts of altruism accommodate experimental data on helping behaviour?

Australasian Journal of Philosophy

Volumn 77, Issue 1, 1999, Pages 26-45

  Campbell, John ^a  

^a LA TROBE UNIVERSITY   (Australia)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0007269080     PISSN: 00048402     EISSN: None     Source Type: Journal    
DOI: 10.1080/00048409912348791     Document Type: Article

References (27)

1. Lee Ross and Richard E. Nisbett, The Person and the Situation (New York: McGraw-Hill, 1991), p. 79
2. John M. Darley and C. Daniel Batson, '"From Jerusalem to Jericho": A Study of Situational and Dispositional Variables in Helping Behavior', Journal of Personality and Social Psychology 27 (1973), pp. 100-108
3. John McDowell, 'Virtue and Reason', The Monist 62 (1979), pp. 331-350
4. Thomas Nagel, The Possibility of Altruism (London: Oxford University Press, 1970)
5. Lawrence Blum, 'Compassion' in Amelie Oksenberg Rorty (ed.), Explaining Emotions (Berkeley: University of California Press, 1980), p. 513
6. Joel Feinberg, 'The Moral and Legal Responsibility of the Bad Samaritan' in Freedom and Fulfillment (Princeton: Princeton University Press, 1992), p. 178
7. Darley and Batson, '"From Jerusalem to Jericho": A Study of Situational and Dispositional Variables in Helping Behavior', p. 101
8. Are these findings surprising? When presented with the data about hurriedness, etc., some people respond by saying that it is just what one would expect. However, in one study subjects were presented with the plan of the Darley and Batson experiment. The subjects were asked to predict the effect of hurriedness, etc., on helping behaviour. Subjects predicted that hurriedness would not make a difference, and also predicted that factors which in fact made no difference would make a pronounced difference. So if Darley and Batson's findings are 'just common sense', they may be so only for those who are wise after the fact. Indeed, even that may not be wholly true. Some subjects were informed of the results of the Darley and Batson experiment. This knowledge had only a modest effect on their predictions about helping situations that were new but very similar to the Darley and Batson situation. These subjects thought that hurriedness would make some difference but far less than that in the Darley and Batson experiment, and they thought that some other factors would make a significant difference even though they had made no difference in the Darley and Batson experiment. So when informed of counterintuitive results, subjects seem to find it difficult to fully take them to heart and to use them as a basis for making predictions about very similar situations. See Paula R. Pietromonaco and Richard E. Nisbett, 'Swimming Upstream Against the Fundamental Attribution Error: Subjects' Weak Generalizations from the Darley and Batson Study', Social Behavior and Personality 10 (1982), pp. 1-4. For our purposes it does not matter whether the Darley and Batson findings are surprising since we are interested in the implications of the findings whether surprising or not
9. Immanuel Kant, Foundations of the Metaphysics of Morals, trans, by Lewis White Beck (New York: Bobbs-Merrill, 1959), p. 14
10. Nagel, The Possibility of Altruism, p. 3
11. Lawrence Blum, Friendship, Altruism, and Morality (London: Routledge & Kegan Paul, 1980)
12. Not only can one interpret reliability in more than one way but one need not interpret the question whether one motive is more reliable than another as an empirical one. Barbara Herman has argued that Kant regarded the motive of duty as more reliable than motives based in compassion but on her account, reliability is not an empirical matter. Briefly, the idea seems to be this. Consider two agents, one dutiful and one compassionate, who each succeed in achieving what they set out to do. Then the dutiful agent will do what (truly) is dutiful and right. And the compassionate agent will do what (truly) is compassionate. But the compassionate agent may fail to do what is right, since the compassionate act is not always the right act. It is in that sense that the motive of duty is more reliable than compassion. So construed, reliability is not an empirical issue. If that is Kant's view, it is compatible with the claim that compassionate agents are more 'empirically reliable' than dutiful agents, i.e., with the claim that, as a matter of empirical fact, persons motivated by compassion are more likely to do the right thing than persons motivated by duty. But such 'empirical reliability' would not, on Kant's view, be relevant to assessing moral worth. See Barbara Herman, 'On the Value of Acting from the Motive of Duty' in The Practice of Moral Judgment (Cambridge, Mass.: Harvard University Press, 1993), pp. 1-22
13. Blum, Friendship, Altruism, and Morality, p. 136
14. Ross and Nisbett, The Person and the Situation, p. 49
15. Regarding these hurried subjects who perceived the victim's distress but decided not to stop, Darley and Batson suggest that they were in conflict between helping the victim and helping the experimenter and that it was this conflict that explained their failure to stop. This may be part of the explanation but it does not explain (a) why the hurried subjects' conflict was resolved in favour of helping the experimenter rather than in favour of helping the victim, nor does it explain (b) why the hurried subjects did not even mention to the persons in the second building that there was a distressed person that sort of minimal help would not have conflicted with helping the experimenter.To explain (a), Richard Holton has suggested to me that we invoke Michael Bratman's idea that prior intentions have a characteristic inertia, a 'causal tendency' to persist and 'to resist reconsideration' (Intention, Plans, and Practical Reason (Cambridge, Mass.: Harvard University Press, 1987), p. 60). Holton's suggestion is that this idea would explain why the hurried subjects' conflict was resolved in favour of helping the experimenter, since the prior intention was to help the experimenter and that intention would therefore resist reconsideration. This suggestion seems promising but may need supplementation. Note that the unhurried subjects were only a few minutes early and they offered some forms of help (e.g., to take the victim to the infirmary) that could only have been offered after reconsidering their prior intention to help the experimenter. If resistance to reconsidering prior intention explains why the hurried subjects failed to offer help that conflicted with their prior intention, why didn't such resistance also prevent the unhurried subjects from offering help that conflicted with their prior intention? Bratman's idea on its own cannot account for this. However, suppose one supplements Bratman's idea with the idea that being unhurried plays a key role in breaking down resistance to reconsideration of prior intention (e.g., being unhurried gives subjects some time to think and reflect). That is why various unhurried subjects did offer help that conflicted with their prior intention, while hurried subjects did not being hurried impeded the breaking down of hurried subjects' resistance to reconsidering their prior intention. So we may now have an explanation for (a). However, this still leaves (b) unexplained why didn't hurried subjects offer forms of help to the victim that were consonant with their prior intention to help the experimenter and which would not have required reconsideration of this prior intention?
16. This is how I understand Christine Korsgaard, 'Skepticism about Practical Reason', The Journal of Philosophy 83 (1986), pp. 5-25. Korsgaard maintains that metaphysical (nonempirical) arguments establish what a rational person would be moved to do and that although empirical data can tell us what considerations move humans, these data are irrelevant to determining what considerations move a rational person. She leaves it open (in this article) whether there are persuasive metaphysical arguments to show that considerations about the welfare of others have rational authority. But she maintains that if there are such arguments, then obviously any rational person will be motivated by considerations about the welfare of others and act accordingly. And in that case, if empirical data showed that we humans are not motivated to act by such considerations, that would only show that we humans are not rational. Korsgaard's approach to rationality can be useiully compared to Kant's approach to moral worth. Kant's view is that metaphysical arguments establish what makes actions morally worthy. In footnote 25, we noted a metaphysical (nonempirical) argument to show that only the motive of duty is reliable. That argument, in conjunction with the (nonempirical) claim that a morally worthy motive must be reliable, implies that only the motive of duty confers moral worth. If this metaphysical argument is persuasive, then if empirical data showed that humans are not ever motivated by duty, that would only show that humans are not ever morally worthy
17. Although being 'fully informed' may strike one as evidently an empirical idea (albeit one that is realisable in practice only to a limited extent possessing all pieces of information is a bit like possessing all pieces of gold), there are in fact great difficulties in giving clear empirical content to this idea. Some of these difficulties are discussed in J. David Velleman, 'Brandt's Definition of "Good"', The Philosophical Review 97 (1988), pp. 353-71
18. and Connie S. Rosati, 'Persons, Perspectives, and Full Information Accounts of the Good', Ethics 105 (1995), pp. 296-325. For an attempt to characterise ideal conditions in empirical terms without relying on the idea of being fully informed, and to then attempt to acquire some empirical data about people's choice of principles of justice when choosing in such conditions
19. see Norman Frohlich and Joe A. Oppenheimer, Choosing Justice (Berkeley: University of California Press, 1992)
20. Owen Flanagan, Varieties of Moral Personality (Cambridge, Mass.: Harvard University Press, 1991), p. 302
21. Since the publication of Carol Gilligan's In a Different Voice (Cambridge, Mass.: Harvard University Press, 1982), there has been debate about whether the moral perspectives of men and women differ. Even if they do differ in perspective, however, there is a further question whether the difference in perspective translates into differences in behaviour. Would the helping rates for hurried and unhurried men differ from those for hurried and unhurried women? Perhaps not, even if there is a difference in perspective. In the Milgram experiments, for example, the obedience rates for men and women were virtually identical, even though women generally appeared to experience more conflict and to characterise the conflict differently in the post-experiment interviews
22. (see Stanley Milgram, Obedience to Authority, (New York: Harper & Row, 1974), pp. 62-3)
23. Roger Brown, Social Psychology, 2nd edition (New York: Free Press, 1986), p. 4
24. Cf. Peter Railton's comment about a very similar case: 'Suppose that I, rushing down the sidewalk like the others around me, hurry past a child who is wandering toward a busy street, noticing but doing nothing. To scoop her up, return her to her yard, and shut the gate would take an insignificant amount of time and effort on my part. Would a moral evaluation that I have acted badly be inapplicable simply because my plans for that afternoon did not include any rescues and because I just couldn't be bothered? Hardly.' (Peter Railton, 'Some Questions about the Justification of Morality', in James E. Tomberlin (ed.), Philosophical Perspectives, 6 Ethics, 1992, (Atascadero, California: Ridgeview Publishing Company, 1992) p. 38). Note also that this is not just a spectator's judgement about these subjects. It seems likely that (most) subjects' sincerely professed values would rate running a bit late to record such a talk as inconsequential compared to taking a moment to check on someone in severe distress. The subjects' sincerely professed values do not explain their behaviour here
25. Alan C. Elms and Stanley Milgram did conduct a structurally similar study with some of the Milgram subjects. That is, disobedient subjects from variants of the experiment that had the highest obedience rates were compared with obedient subjects from variants of the experiment that had the highest disobedience rates. They tentatively concluded that there were some significant personality differences between those two groups. However, they were cautious about interpreting some of these differences, e.g., some of the differences might have been the effect of subjects' behaviour in the experiment rather than the cause of it (or causation may have gone in both directions). See Alan C. Elms and Stanley Milgram, 'Personality Characteristics Associated with Obedience and Defiance toward Authoritative Command', Journal of Experimental Research in Personality 1 (1966), pp. 282-9. My thanks to Neil Thomason for drawing my attention to this study, which refutes the common view that no personality differences were ever detected between obedient and disobedient subjects
26. There have been many studies on this. One is John M. Darley and Bibb Latane, 'Bystander Intervention in Emergencies: Diffusion of Responsibility', Journal of Personality and Social Psychology 8 (1968), pp. 377-83. This study examined the effect of subjects' thinking that there were other potential helpers, in a situation where the subjects did not know whether those other potential helpers were calling for help. In that respect the study resembled the infamous case where numerous New York apartment dwellers witnessed a long and vicious attack on Kitty Genovese but each dweller did not know whether anyone else was alerting police. In this study, subjects' knowledge that there were other potential helpers, even though they did not know whether the other potential helpers were calling for help, had the result that subjects were much less likely to call for help, and that they were much less likely to call quickly if they did call. And the greater the number of other potential helpers, the less likely these things became. This result seems congruent with the Kitty Genovese case, where there was a large number of other potential helpers and no one even telephoned the police. Just as subjects' degree of hurriedness is seemingly a better predictor of what they will do than facts about their character, so is whether subjects think that there are other potential helpers (and if so, how many) this despite the fact that subjects subsequently reported that they felt that their awareness of other people 'made no difference to their own behaviour' (p. 381). So what explains the behaviour of the New York apartment dwellers may be their awareness of other potential helpers rather than facts about their character it may be wrong to conclude that they were unusually callous and indifferent people or that they had been 'dehumanised' by modern urban living
27. For a discussion of much of this evidence, see Ross and Nisbett, The Person and the Situation. Their discussion covers not just helping behaviour but behaviour in general. Indeed, even self-interested behaviour seems heavily influenced by such factors (see, e.g., the study discussed on p. 10 of The Person and the Situation). It would be a mistake to interpret the data on helping behaviour as supporting a selfish interpretation of human behaviour, e.g., some version of psychological egoism