Sidetracked by trolleys: Why sacrificial moral dilemmas tell us little (or nothing) about utilitarian judgment

Abstract

Research into moral decision-making has been dominated by sacrificial dilemmas where, in order to save several lives, it is necessary to sacrifice the life of another person. It is widely assumed that these dilemmas draw a sharp contrast between utilitarian and deontological approaches to morality, and thereby enable us to study the psychological and neural basis of utilitarian judgment. However, it has been previously shown that some sacrificial dilemmas fail to present a genuine contrast between utilitarian and deontological options. Here, I raise deeper problems for this research paradigm. Even when sacrificial dilemmas present a contrast between utilitarian and deontological options at a philosophical level, it is misleading to interpret the responses of ordinary folk in these terms. What is currently classified as “utilitarian judgment” does not in fact share essential features of a genuine utilitarian outlook, and is better explained in terms of commonsensical moral notions. When subjects deliberate about such dilemmas, they are not deciding between opposing utilitarian and deontological solutions, but engaging in a richer process of weighing opposing moral reasons. Sacrificial dilemmas therefore tell us little about utilitarian decision-making. An alternative approach to studying proto-utilitarian tendencies in everyday moral thinking is proposed.

Keywords:

• Morality
• Moral decision-making
• Sacrificial dilemmas
• Utilitarian judgment

Notes

¹ For a battery of validated vignettes involving everyday moral situations, see Knutson et al. (2010).

² For example, Christensen and Gomila (2012) write that sacrificial dilemmas “were instrumental in arguing for the inconsistency of utilitarianism (or Consequentialism, in general) as an ethical theory” (p. 1251). However, the Foot and Thompson articles that first introduced these trolley cases are not at all concerned with this issue; they proceed on the assumption that utilitarianism is false.

³ The scope of the problem should not be overly exaggerated. Some studies have used only variants of the original trolley problems (see, e.g., Greene et al., 2008) or better controlled stimuli (e.g. Moore, Clark, & Kane, 2008).

⁴ Since commonsense morality is not an explicit theory, the implicit rules that make it up need not be accessible to introspection or easy to articulate—just as our concepts of knowledge or causality clearly have a complex structure that is nevertheless difficult to spell out, even after centuries of philosophical reflection.

⁵ Commonsense morality further distinguishes more stringent duties to prevent suffering and harm from weaker duties to confer benefits and promote happiness—a familiar moral distinction that utilitarians reject.

⁶ Studies such as Greene et al. (2001) that ask participants whether the sacrificial act is “acceptable” cannot distinguish between the utilitarian view that this act is required and the much weaker view that it is permissible both to commit this act and to refuse to do so (Kahane & Shackel, 2010). But this is a critical distinction. In fact, Royzman et al. (2015) found that a greater tendency to greater reflection was associated only with judgments of permissibility.

⁷ An alternative explanation is that such effort is needed to overcome a persistent emotion or intuition that subjects take to be morally spurious; but see Kahane (2012) to see why this is an implausible explanation of the majority of cases. Notice that in a given context, some moral considerations or rules may be more salient than others, and thus have greater intuitive force, thereby either blocking further deliberation or dominating (or biasing) such deliberation. It does not follow from this, however, that the opposing moral considerations must therefore have a qualitatively different psychological character or source.

⁸ It is sometimes claimed that brain areas implicated in deliberative processing, such as the Dorsolateral Prefrontal Cortex (DLPFC), are involved in using “utilitarian” cost–benefit analysis to override, not only pre-potent “deontological” intuitions, but also pre-potent selfish impulses. But different kinds of responses will be pre-potent/intuitive in different contexts and populations. In some contexts, such as the Ultimatum Game, deliberative processing is needed to override a self-interested impulse and reject a beneficial yet unfair offer (Knoch, Pascual-Leone, Meyer, Treyer, & Fehr, 2006)—which is arguably a deontological response (Kahane & Shackel, 2010). And in some contexts, deliberative processing is needed to override pre-potent cooperative impulse, in order to arrive at a (counterintuitive) selfish decision (Rand, Greene, & Nowak, 2012; Suzuki, Niki, Fujisaki, & Akiyama, 2011). But such a general tie between deliberative processing and counterintuitive judgments is utterly unsurprising, and tells us nothing about utilitarian judgment per se; it just so happens that utilitarianism is associated with many counterintuitive moral conclusions. For further discussion, see Kahane et al. (2012), Kahane (2014).

⁹ A common criticism of sacrificial dilemmas is that they are unrealistic (Bauman, McGraw, Bartels, & Warren, 2014). This is not the worry I have been raising: lack of realism is both a disadvantage and an advantage (because, e.g., far-fetched examples allow us to better isolate distinct moral variables that would often be entangled in more realistic cases, or because they allow us to investigate judgments that do not merely reflect social convention). The issues I have been raising would remain in force even if we were to devise highly realistic instances of sacrificial dilemmas.

¹⁰ For suggestive evidence supporting this hypothesis see Kahane et al. (2015).

¹¹ Similar points apply to investigating the way an agent’s guilt, or otherwise their being a threat to others, can make that agent less immune to harm. It seems plausible, however, that standard sacrificial dilemmas can be useful for studying the way the inevitability of harm affects judgments of permissible harming.

Additional information

Funding

This work was supported by the Wellcome Trust [grant WT087208MF].