Commentary on Rachlin, 2002
Ainslie, G. & Haslam, N, (2002) Altruism is a primary impulse, not a discipline. Behavioral and Brain Sciences 25, 251.




Altruism is a primary impulse, not a discipline

Commentary on Rachlin, H. (2002). Altruism and selfishness.
Behavioral and Brain Sciences, 25(2), 239-250

George Ainslie
Veterans Affairs Medical Center, Coatesville PA, USA
University of Cape Town, South Africa

Nick Haslam
Department of Psychology
University of Melbourne
Parkville VIC 3010


Published in Behavioral and Brain Sciences 25, 251, 2002

This material is the result of work supported with resources and the use of facilities at the Department of Veterans Affairs Medical Center, Coatesville, PA, USA. The opinions expressed are not those of the Department of Veterans Affairs of the US Government.


Intertemporal bargaining theory based on the hyperbolic discounting of expected rewards accounts for how choosing in categories increases self- control, without postulatin, as Rachlin does, the additional rewardingness of patterns per se. However, altruism does not seem to be based on self-control, but on the primary rewardingness of vicarious experience. We describe a mechanism that integrates vicarious experience with other goods of limited availability.



Utility theory is frequently read as declaring altruism to be irrational. Rachlin offers one of many current rebuttals of this counterintuitive conclusion (see, e.g., Batson & Shaw, 1991; Field, 2001). His argument is that altruism is a kind of self-control, overriding one's current impulse for the sake of a longer-range good. He depicts the mechanism as learning to see choices not as isolated instances but as part of overall patterns. We agree that such overall interpretation is a key mechanism of self-control; but Rachlin's mechanism entails the unnecessary assumption that the reward for a pattern of self-control is greater than the sum of rewards for the choices that make up the pattern. Furthermore, we do not agree that altruism is motivated mainly by the incentives that exist for self-control.

Rachlin confronts two conceptual problems, both of which can be solved more specifically by assuming only hyperbolic discounting of delayed rewards, the additivity of discounted sequential rewards, and the dependence of total available appetite on rate of satiation. The first problem is how patterns of choices come to be rewarded preferred oppositely from individual members of the pattern. The second is how an organism comes to be rewarded by another organism's experience.

1. Rachlin's note 4 concisely characterizes the differences in between his and Ainslie's theories of "complex ambivalence." In Rachlin's model, the value of a pattern of being sober, say, or being altruistic, is greatly reduced by being intoxicated or selfish just today, just as the value of a symphony plummets if notes are taken out; but the lapse does not make the remaining pattern unavailable. In Ainslie's intertemporal bargaining model the lapse is attractive but is avoided, when it is avoided, because of the risk that it will break off the pattern. The advantage of the bargaining theory is that it accommodates the widely reported urge to duck out of a pattern as well as the urge to maintain it, without postulating more than the rewards literally available in each choice: When a person sees each choice in a category as a test case for her continuing cooperation in an intertemporal prisoner's dilemma, she will face both an incentive to distinguish the present choice from the pattern, i.e. that is, to rationalize a defection vis-a-vis her future selves, and a growing incentive to preserve her expectation of future prudence, i.e. that is, to cooperate with these selves. Rachlin's alcoholic is less vulnerable to temptation after refusing 100 drinks than after refusing one, not because he sees a pattern of sobriety as any more desirable than before, but because it has become more believable that he will attain it if he does not slip this time. To get the pattern-keeping effect, he must still see each refusal of alcohol as necessary to this believability; if someone offers him a drink under circumstances that he does not expect to reduce this believability—a really good rationalization, say, like coercion or a rare occasion—he is apt to welcome it.

Hyperbolic discounting and its implication of intertemporal bargaining would be a good theory of the benefits of incentives for pattern-following even without independent evidence that it exists. In fact, there is not only overwhelming evidence that all organisms discount rewards in single-shot choices hyperbolically (e.g., Kirby, 1997), but good evidence that they reverse preferences from smaller-earlier to larger-later rewards when they choose between a whole sequence of pairs at once, a phenomenon that would not occur with conventional (exponential) discounting (Kirby & Guastello, 2001; Ainslie & Monterosso, 2003; Kirby & Guastello 2001). Softer evidence that willpower is based on the perception of current choices as test cases predicting such sequences comes from thought experiments (Ainslie, 2001, pp. 117-–138) and experimental bargaining analogs (Ainslie et. al., 2002, Monterosso, 2002).

2. Certainly choosing according to principle increases self-control. The assertion that the same mechanism overcomes selfish motives by creating altruistic ones is more tenuous.

Rachlin does not make any attempt to say why altruistic patterns of choice should be rewarding when their components are not, merely assuming that "habitual altruism is a happier mode of existence than habitual selfishness." But the same appeal to common experience argues that the vicarious feeling of other people's emotions is a primary good.

From early childhood we spontaneously put ourselves in other people's shoes, a phenomenon that has been demonstrated in nonhuman animals and for which neurophysiological substrates have been found (Preston & de Waal, 2002).

The vicarious reward that empathy supplies is often so compelling that people make efforts to discipline it—to rein in altruistic tendencies, for example, suppressing sympathy towards those whom one expects to request costly help later (Shaw et al., 1994). It is quite believable that the woman who saves the stranger's child from a burning building may do so impulsively, in violation of a perceived duty to her own family, because she cannot tolerate imagining a burning baby or an anguished mother. Furthermore, if altruism were dependent on farsightedness learned, like civility, over the person’s lifetime, then altruism would be unknown among young children and would increase with age. The evidence contradicts both of these implications (e.g., Fiske, 1991; Frank et al., 1993). Empathy is a robust and early-developing process that underpins prosocial behavior (altruism) as well as antisocial behavior (retribution, gloating) behavior.

Two issues have apparently kept utility theorists from accepting vicarious experience as a primary good: the lack of a physical need for stimuli from other people in order to have a positive emotional response to them, and, conversely, the difficulty of avoiding negative emotional responses to information about other people's painful experiences.  Conventional theory does not tell us why we want to know that others are happy, or why we allow ourselves to be moved when they are not.  Folk psychology depicts our responses as unmotivated, perhaps classically conditioned.  Rejoicing in someone else's joy seems like an arbitrary decision, not consumption of an external good; and sharing someone else's anguish seems to violate maximization of utility. However, the fact of hyperbolic discounting predicts that I will be impatient in consuming reward that is at my free disposal, as emotional reward apparently is. The consequent premature satiation should reduce emotional patterns to the quality of daydreams, unless I learn to cue my emotional behaviors with occasions that are outside of my control and adequately surprising. Vicarious experience represents a rich source of such occasions, which may thus come to govern my emotions almost as if they were stimuli for reflexes. The same hyperbolic curves may also cause vivid aversive experiences to seduce my attention, just as, over a slower time course, these curves may cause addictive substances to lure me into unrewarding choices. Thus even anguish need not be seen as either imposed by a process like classical conditioning or accepted through the kind of discipline Rachlin proposes, but rather as a good that can compete in the free market of choice (see Ainslie, 1995; and 2001, pp. 161-186).




Ainslie, George. (1995). A utility-maximizing mechanism for vicarious reward. Rationality and Society, 7, 393-–403.

Ainslie, George. (2001). Breakdown of will. Cambridge: Cambridge University Press.

Ainslie, George and & Monterosso, John (2003). Building blocks of self-control: Increased tolerance for delay with bundled rewards. Journal of the Experimental Analysis of Behavior, 79, 37-48.

Ainslie, George, Monterosso, John, Toppi-Mullen, P.amela and & Gault, Barbara (2002) The fragility of cooperation: An empirical study employing false feedback in a sequential iterated prisoner's dilemma. Journal of Economic Psychology, 23, 437-448.

Batson, C. Daniel and & Shaw, Laura L. (1991) Evidence for altruism: Toward a pluralism of prosocial motives. Psychological Inquiry 2, 159-168.

Field, Alexander (2001) Altruistically Inclined? The Behavioral Sciences Evolutionary Theory and the Origins of Reciprocity U Michigan.Press.

Fiske, Alan P. (1991). The cultural relativity of selfish individualism: Anthropological evidence that humans are inherently sociable. In: M. Clark (Ed.), Review of Personality and Social Psychology, vol. 12, Altruism and prosocial behavior, ed.M. Clark. (pp. 176-214). Beverley Hills, CA: Sage.

Frank, Robert. H., Gilovich, Thomas. & Regan, Dennis. (1993). Does studying economics inhibit cooperation?. Journal of Economic Perspectives, 7, 159-171.

Kirby, Kris N. (1997) Bidding on the future: Evidence against normative discounting of delayed rewards. Journal of Experimental Psychology: General 126, 54-70.

Kirby, Kris N., and & Guastello, Barbarose (2001) Making choices in anticipation of similar future choices can increase self-control. Journal of Experimental Psychology: Applied 7, 154-164.

Monterosso, J., Ainslie, G., Toppi-Mullen, P. & Gault, B. (2002) The fragility of cooperation: A false feedback study of a sequential iterated prisoner's dilemma. Journal of Economic Psychology 23, 437-448.

Preston, Stephanie D. and & de Waal, Frans B. M. (2002 Empathy: Its ultimate and proximate bases. Behavioral and Brain Sciences 25(1), 1-70.

Shaw, Laura L., Batson, C. Daniel, & Todd, R. Matthew (1994). Empathy avoidance: Forestalling feeling for another in order to escape the motivational consequences. Journal of Personality and Social Psychology, 67, 879-–887.