The demandingness objection to consequentialism is frequently taken to be decisive.Footnote 1 The objection is a complaint leveled against consequentialism that it asks too much of us – or, more precisely, that it asks more of us than morality may ask of us. In the face of the demandingness objection, we are asked to either abandon consequentialism entirely, or else to soften the demands of the theory through some new formulation with appropriately modest expectations. This latter strategy has spawned a variety of clever versions of consequentialism, all sharing a healthy respect for the demandingness objection.Footnote 2 Other consequentialists have resisted the objection, embracing the radical demands entailed by their impartial and optimific theory, recognizing all the while how ‘deeply counterintuitive’Footnote 3 their theory is.
One consequentialist who has resisted the demandingness objection is David Sobel.Footnote 4 Sobel's elegant response to the demandingness objection demonstrates that, at its heart, the objection begs the question against consequentialism: it assumes a relevant moral distinction between requiring and permitting outcomes that therefore assumes the wrongness of consequentialism from the start.Footnote 5 This renders the demandingness objection impotent. To the extent that the objection seems compelling, it is because consequentialism fails to recognize the requiring/permitting distinction and not because the theory asks too much of us. Any emphasis on the demands themselves is therefore misplaced. This invites the consequentialist to follow Sobel's analysis one step further: for the critic of consequentialism to resuscitate the real criticism underlying the demandingness objection, the critic must recognize and provide a rationale for this distinction underlying the objection. It is not clear what this rationale might be, beyond the initial intuitive appeal of the distinction.
How might the critic respond to this challenge? One strategy would be to turn the tables on the consequentialist, raising a new kind of question-begging challenge. That is, just as the defender of the demandingness objection begs the question against the consequentialist by assuming the truth of a deeper distinction (regarding requiring and permitting) in offering her criticism, so does the consequentialist's response ultimately run afoul of the same sort of question-begging problem: it involves making a deeper assumption on which her response to the demandingness objection turns, and this assumption also stands in need of defense. The question is begged here at the methodological level, regarding the scrutiny we apply to our moral intuitions as we balance them against our moral reasoning in reflective equilibrium. On the face of it, it is not clear what kind of methodological authority intuitions should have. A compelling reason is needed in order to justify scrutinizing our moral intuitions to the extent that the consequentialist ultimately requires, in order to avoid begging the question against her critic and the demandingness objection. The critic, in other words, is questioning why she must provide the rationale demanded by Sobel for the intuitive distinction between requiring and permitting. This reason, I argue, is provided by the emerging neuroscientific work on intuitions.
Quite independently of the literature on the demandingness objection to consequentialism, a body of literature has begun to develop that connects moral theory with cognitive neuroscience. The Rawlsian device of reflective equilibrium is ubiquitous in contemporary moral methodology, where intuitions provide the raw data against which our moral theories and principles are balanced.Footnote 6 Yet these moral intuitions have lately been subjected to scrutiny by experimental philosophers for their biological, cultural and social origins.Footnote 7 Their work is in many respects in its nascent stages, and its implications for moral methodology are not yet settled. Its relevance here, however, comes clear when we examine the debate over the demandingness objection more closely. This work advises us to give our moral intuitions a modest (rather than robust) role in our moral methodology, and it is this exhortation of modesty that in turn recommends the careful examination of our moral intuitions that the consequentialist's response to the demandingness objection requires. This answers the critic's question-begging challenge, and revives the consequentialist's demand that the critic account for the intuitions propping up the demandingness objection.
The structure of this article is as follows. In section 1, I explain how Sobel's response to the demandingness objection renders the objection impotent by showing how it begs the question against the consequentialist. In section 2, I show how the consequentialist's response to the demandingness objection commits the same sort of sin that the objection itself commits, by smuggling in an undefended assumption about our intuitions, this time on the methodological level. Here I utilize two familiar formulations of consequentialism in order to bring to light the role of intuitions in consequentialist methodology, and in reflective equilibrium more generally: Shelly Kagan's defense of ‘extreme consequentialism’Footnote 8 and Brad Hooker's defense of rule-consequentialism. The methodologies employed by Kagan and Hooker are similar in many respects, yet they differ in the role they assign to our moral intuitions. The worry is that the extreme consequentialist needs some compelling reason for endorsing the methodological role she assigns to intuitions; absent this reason, she begs the question by assuming the correctness of her methodology over a methodology such as Hooker's that accepts the demandingness objection. In section 3, I find this compelling reason in the work of experimental philosophers. I turn to the neuroscientific literature on intuitions to recommend a modest role for intuitions in our moral methodology, thus rescuing the consequentialist's response to the demandingness objection. I also make the case that we need not take the drastic step of rejecting wholesale the device of reflective equilibrium in our moral methodology, as Peter Singer suggests.Footnote 9 In either case, however, the demandingness objection remains adequately answered.
1. THE DISTINCTION UNDERLYING THE DEMANDINGNESS OBJECTION
Sobel's response to the demandingness objection begins with a thought-experiment.Footnote 10 Joe has two good kidneys; Sally needs one of them, or she will die. Consequentialism, in its standard impartial and optimific form, requires Joe to sacrifice one kidney to Sally at the cost of a decent but reduced life. The demandingness objection tells us that morality cannot impose such a high cost on Joe, and so consequentialism must be a defective moral theory. The implications of consequentialism are simply too demanding for Joe. Yet, as Sobel notes, defenders of this objection have failed to notice the way that it can be turned on its head. From Joe's perspective, yes, consequentialism is a demanding theory, for the way that it imposes a significant cost on him: he must endure the surgical removal of a kidney, and whatever reductions in his lifestyle follow from that. These are the demands that drive the demandingness objection. But from Sally's perspective, whatever non-consequentialist theory we are subsequently favoring is very demanding of her: she will now die, because the theory allows others to refrain from helping her when they could do so. In fact, this alternate theory (whatever it is) is even more demanding than consequentialism, given that death is a higher cost than suffering a permanent but only marginally lower standard of living.
Surely the defender of the demandingness objection would be suspicious of this response on Sally's behalf. But why? According to Sobel, the impotence of the demandingness objection is revealed when we understand the critics’ distinction between consequentialism's demands on Joe and non-consequentialism's demands on Sally. The ‘obvious answer’, as he puts it, is as follows:
[T]he costs of what a moral theory requires are more demanding than the costs of what a theory permits to befall the unaided, size of cost held constant. The moral significance of the distinction between costs a moral theory requires and costs it permits must already be in place before the Objection gets its grip. But this is for the decisive break with Consequentialism to have already happened before we feel the pull of the Demandingness intuitions.Footnote 11
The demandingness objection is impotent as a fundamental criticism of consequentialism, then, because it is only forceful given this distinction between requiring and permitting. If the requiring/permitting distinction is legitimate, then it is not the demands of consequentialism, but rather its failure to recognize this distinction, that is its downfall; conversely, if the distinction is illegitimate, then the demandingness objection is groundless. It is in this way that the demandingness objection begs the question against the consequentialist: where the distinction between requiring and permitting is assumed rather than shown, the deck is stacked against consequentialism from the start.
2. ASSUMPTIONS AT TWO LEVELS: BEGGING THE QUESTION BOTH WAYS
Sobel's analysis demonstrates that the demandingness objection, to be compelling, must smuggle in a deeper assumption regarding the asymmetry between requiring and permitting. As Sobel notes, this suggests a clear next step in the debate emerging from the demandingness objection: ‘[w]e should focus on reasons to think there is a serious moral difference between costs a moral theory requires and costs it permits, or between causing and allowing, rather than focusing on our feeling that sometimes Consequentialist morality asks too much.’Footnote 12 But just as the non-consequentialist critic resisted Sally's complaint against the demands of non-consequentialism in her thought-experiment, there are grounds for the critic to resist the consequentialist's call here to examine the reasons underlying the requiring/permitting distinction at the heart of the demandingness objection. The non-consequentialist critic, as Sobel notes, smuggles in an assumption about the requiring/permitting distinction in offering the demandingness objection. Yet the consequentialist smuggles in a deeper assumption on the methodological level, and so the seemingly innocent request made here by Sobel to examine the rationale begs the question against the non-consequentialist critic, one level down from the initial question-begging charge. This assumption concerns the role we give to intuitions in our moral methodology.
The methodological distinction at issue can be seen clearly by looking at two very different formulations of consequentialism: Shelly Kagan's defense of extreme consequentialism and Brad Hooker's defense of rule-consequentialism. While Hooker is transparently not a non-consequentialist, his importance here lies in the stance he adopts towards the demandingness objection. Hooker accepts the demandingness objection and modifies his theory accordingly; Kagan rejects the objection. These differing consequentialist theories are useful for two reasons: first, both philosophers are explicit about their methodologies, and second, while they both utilize the standard methodology of reflective equilibrium, they adopt very different positions on how intuitions are to be employed in balancing them against our moral reasoning in reflective equilibrium. Ultimately, in order to reject the demandingness objection in a non-question-begging way, the extreme consequentialist must find a compelling reason to regard intuitions as she does, rather than as Hooker and others who accept the demandingness objection do.
After explicitly endorsing the process of reflective equilibrium and enumerating a set of criteria that any moral theory must adequately fulfill, Kagan includes a specific proviso against ‘dangling distinctions’.Footnote 13 Though our intuitions comprise the data against which we test our moral principles, we have to double back and scrutinize those intuitions, examining the key distinctions underlying them; without the critical support that this scrutiny provides, we have left our distinctions dangling, and a moral theory must not do this. This methodology fits well with the sort of methodology that supports Sobel's challenge to the demandingness objection: when the consequentialist pushes back against her critic and demands a rationale for the requiring/permitting distinction underlying the demandingness objection, the consequentialist is essentially challenging the critic to show that the distinction doesn't dangle. And the reasoning behind this proviso against dangling distinctions seems clear enough: we ought not to draw lines that we cannot support.
Hooker, however, provides a counterpoint.Footnote 14 Hooker's methodology is in many respects similar to Kagan's, including his explicit endorsement of reflective equilibrium, but the two diverge in a striking way in their discussion of the intuitions that provide the data for settling on our theory in reflective equilibrium. For Hooker, our starting point in moral reasoning is with intuitive convictions that come with ‘independent credibility’: these intuitions bring their credibility with them, prior to any reasoned theorizing. While independently credible intuitions may turn out to be mistaken, Hooker analogizes them to self-evident propositions, in that they are evident without any need of further proof. These independently credible intuitions are the building blocks of our theory: without them, the project is futile from the start. And like Kagan's proviso against dangling distinctions, the reasoning behind this emphasis on independent credibility also seems clear enough: after all, as Kagan and Hooker both acknowledge, our explanations must end somewhere. At some point, in other words, our distinctions must dangle. Following Bernard Williams, Hooker notes, ‘[i]f we take up a point of view stripped of all evaluative conviction, we have no basis for evaluation.’Footnote 15
At the methodological level, then, we seem to face a kind of stalemate in how we understand the role of our intuitions. Either we need to double back and scrutinize them, or else we do not. Returning to the demandingness objection, this dispute about the role of intuitions reveals how the consequentialist's seemingly-innocent request to provide a rationale for the requiring/permitting distinction actually begs the question in return against the critic of consequentialism. The critic begged the question in the first place against the consequentialist by assuming the relevance of the requiring/permitting distinction in asserting the demandingness objection; the consequentialist, by requesting a defense of that distinction, now returns the favor. The question begged here is at a deeper methodological level. The intuitions that make plausible the requiring/permitting distinction either stand in need of further defense, or else they do not; assuming the former begs this very question. The door remains open to the critic of consequentialism to defend herself as follows: she may acknowledge Sobel's observation that the demandingness objection really turns on a distinction deeper than the superficial demands of consequentialism, and still hold fast to the belief that the objection, in its revised form, is in the end correct. Her rationale is that the demandingness objection reveals a deeper distinction between requiring and permitting, a distinction whose intuitive force is independently credible and so immune from the kind of scrutiny that Kagan recommends in his proviso against dangling distinctions. Demanding some further account of those intuitions as Sobel does is therefore to assume, without argument, that a methodology that finds intuitions to be independently credible is mistaken. But why privilege the extreme consequentialist's modest understanding of intuitions over a theoretical standpoint that accords our intuitions a more robust methodological role? Given this apparent methodological stalemate regarding the role of intuitions in reflective equilibrium, the critic of consequentialism is not obliged to acquiesce to the consequentialist's demand for further defense.Footnote 16
3. BREAKING THE STALEMATE: THE COGNITIVE NEUROSCIENCE OF INTUITIONS
At base, the methodological stalemate at hand turns on the fundamental way we regard our moral intuitions.Footnote 17 Assigning our intuitions a methodologically modest role favors a methodology like Kagan's, where dangling distinctions are forbidden and the deeper distinctions underlying the demandingness objection stand in need of defense. Conversely, assigning our intuitions a methodologically robust role favors a methodology like Hooker's, where intuitions have independent credibility and the deep distinctions underlying the demandingness objection need no defense – by their very nature, they defend themselves. This revives the demandingness objection. Can this stalemate be broken? One sort of strategy for doing so seeks an independent reason to be wary of employing our intuitions foundationally, absent some serious scrutiny. Such a reason would, presumably, tilt the playing field toward the more modest strategy, thus putting the critic on her heels in her attempt to resuscitate the demandingness objection. Here, the cognitive neuroscience of intuitions is illuminating.
Trolley problems have long provided an interesting set of cases for intuition-pumping in moral philosophy.Footnote 18 In one version, we stand at the track-switch as an out-of-control trolley barrels toward five innocent victims, and at the flick of the switch, we can divert the trolley toward one innocent victim; in another version, we stand on a footbridge next to a very fat man as the same trolley speeds along, where a well-timed nudge will save the five at the unsuspecting fat man's expense. These cases are clearly similar in many relevant ways, yet the differences in our intuitive responses to them have long been a source of puzzlement. Why is it that respondents seem to have reservations about shoving the fat man, but not about flipping the track-switch?
As Greene and others answer the puzzle, ‘the crucial difference between the trolley dilemma and the footbridge dilemma lies in the latter's tendency to engage people's emotions in a way that the former does not’.Footnote 19 This conclusion was generated from a battery of tests employing functional magnetic resonance imaging (fMRI), where brain images reveal that the different cases engage different areas of the brain. The thought of pushing someone off a bridge engages the parts of the brain concerned with emotional processing; this generates the differential response from the track-switch case, where the details of the case are less psychologically vivid, in spite of the fact that the outcomes (in terms of innocent lives saved and lost) are identical. These neuroimaging studies are supplemented by studies of individuals exhibiting aberrant moral behavior, all to the conclusion that at the heart of our intuitive moral responses, ‘automatic emotional processes tend to dominate’.Footnote 20 What cognitive neuroscience tells us, in other words, is that the more we understand our intuitions, the more reason we have to attribute them to the residue of our biological, social and cultural heredity. When we intuitively favor the switch of the track but reject the nudge of the fat man, we do so not because of some important intuitive insight into morality, but rather because of the biological constitution of our brains. Some situations engage one part of our brain, some situations another, and our intuitive response to any particular case is essentially captive to this evolutionary development.
Armed with this neuroscientific insight, we return once more to the demandingness objection and the distinction that underlies it. Is the intuition favoring this distinction the sort of thing that brings with it independent credibility, or is it the sort of thing that we ought to double back and reconsider more carefully before settling on our endorsement of it? If the theoretical investigation of reflective equilibrium leaves us in a stalemate on this question, an empirical investigation of the neuroscience of moral intuitions tilts toward assigning the more modest role to intuitions that forbids dangling distinctions. Once we recognize that our intuitions may be little more than evolutionary residue, it is hard to see why they merit the robust role in our moral reasoning that methodologies like Hooker's recommend.Footnote 21 This in turn bolsters the consequentialist's response to the demandingness objection: at the heart of the objection lies a deeper distinction, and this distinction stands in need of defense.
Peter Singer has a similar respect for this empirical investigation of intuitions, claiming that ‘[t]he single most important advantage we have over the great moral philosophers of the past is our understanding of evolution and its application to ethics’.Footnote 22 Yet Singer uses these neuroscientific insights to reach a more radical conclusion: he lays bare his long-standing suspicion of reflective equilibrium and challenges the role of reflective equilibrium itself in moral reasoning. Methodologically, reflective equilibrium requires the use of intuitions as data; if the data are fundamentally flawed, then the method that relies on them must be fundamentally flawed as well. It is not immediately clear that Singer needs to commit himself to such a radical conclusion; he admits that ‘it is possible to interpret the model of reflective equilibrium so that it takes into account any grounds for objecting to our intuitions’.Footnote 23 Such a version of reflective equilibrium would be unobjectionable, though he worries that making reflective equilibrium ‘all-encompassing’ in this way comes ‘only at the cost of making it close to vacuous’.Footnote 24 Reflective equilibrium requires intuitions, as data, at the very foundation: weaken the foundation too much, and it is unclear how reliable the resulting edifice can be.
Yet the inescapability of the need to start somewhere in our moral explanations builds a powerful case for continuing to endorse reflective equilibrium in our moral methodology – indeed, it was just this sentiment that justified the robust role given to intuitions by Hooker in the first place. But perhaps more importantly, the sort of model that Singer finds unobjectionable is precisely the sort of model that we find recommended by Kagan. Kagan's proviso against dangling distinctions seems to account for the objections that Singer has in mind, by forcing us to double back and scrutinize our intuitions. And while it is certainly the case that this kind of methodology radically curtails the weight of our intuitions in reflective equilibrium, it is not clear that it makes the model ‘close to vacuous’ – though it is also not clear exactly what that criticism amounts to. We face a choice in light of the analysis of intuitions provided by experimental philosophers: either we ought to give our intuitions a more modest role in reflective equilibrium, or else we should abandon the model of reflective equilibrium entirely. While we may in the end agree with Singer that we ought to dispose of reflective equilibrium, it is not yet obvious that such a radical step is needed, when a more measured step that preserves the methodological advantages of reflective equilibrium remains viable.
In the end, then, Singer's closing recommendation is consonant with the wide version of reflective equilibrium endorsed by Kagan. Singer suggests that we ‘attempt the ambitious task of separating those moral judgments that we owe to our evolutionary and cultural history, from those that have a rational basis’.Footnote 25 As a stand-alone project, this might be immense; we would be better served to scrutinize each intuition as it arises, to make sure that none of our distinctions dangle, on biological grounds or otherwise. In just this way, the methodological stalemate over our moral intuitions is broken. Such a procedure equips the consequentialist with a compelling last response to the demandingness objection: the objection rests on a deeper distinction, and it simply will not do to rest that distinction on the supposed independent credibility of the intuition on which it depends.
