7:04 AM :: Stephen (contact)

Reflections on the Duty to Vote, Part IV: The duty to vote (discussed in Part I, Part II, and Part III) is tricky in part because voting is fundamentally a group practice. Most moral theories are designed for individual agents, telling a single person how to live and act. But when they are applied to questions that require group coordination, they can produce some strange results.

For instance, according to the theory of "act consequentialism" (AC), an action is right if it produces the best (or equal best) consequences of any action available to the agent. This doctrine is relatively popular among philosophers for its simplicity; yet for a doctrine focused on the consequences, AC itself has consequences that are very odd.

For instance, suppose that a patient Joe approaches his doctor with a mild skin complaint, which she can treat with one of three drugs. Drug A would relieve Joe's condition but not cure it entirely; drugs B and C would each cure the condition completely for one-half of the population but instantly kill the other half--and no one can predict which will happen to Joe. Most people think the right thing to do would be to prescribe drug A; the marginal improvement in the skin condition simply isn't worth the risk of killing him. But in fact, this act could never be right according to AC, since it is guaranteed to produce less-than-optimal consequences. The best consequences are, in this case, achieved by prescribing drug B, which just happens to be the drug that will cure Joe entirely. The fact that the doctor didn't know this is irrelevant to AC's assessment of the consequences. Prescribing drug B was an available act, and had the doctor prescribed drug B on a hunch, her action would have right according to AC. In short, to an act-consequentialist, it can never be wrong to guess right.

One might be forgiven for thinking that there is something deeply wrong with a consequentialism which is so exacting. Indeed, such a doctrine can be shown to be flawed on its own terms. Consider the following case. I am a resident of the flooded city referred to in Part II, and when the call goes up for volunteers, no one leaves their homes to help sandbag. How will AC judge my decision to stay home? Well, given that no one else is helping, I am clearly doing the right thing. Of the actions available to me, staying at home is the best thing I could do; my own attempts at sandbagging would be woefully insufficient to save the city, and I can at least get some enjoyment out of the evening. (Assume that there are no other projects I could work on from home, like reconnecting lost puppies with their owners over the Internet, that would be a better use of my time.) Moreover, since the same argument could be made from anyone's perspective, we are all acting rightly according to AC. And the city drowns.

(Perhaps the consequentialist could object that we are not acting rightly on this very basis, as the best consequences would actually be produced by a large group of volunteers. Yet this would be to forget that act-consequentialism judges actions from the individual's point of view. Whether an action is "available" to me must take the simultaneous actions of others as given. Consider Bernard Williams' example of the traveller Jim, who is about to witness the execution of 20 innocent men, and is given the opportunity to save them by executing one innocent man himself. To the utilitarian, Jim is obliged to take the opportunity and save 19 lives. It would be no objection that the best "available" action would be for all the executioners to repent and change their ways; the crucial question is what actions would be available to Jim. Returning to the flood case, if I am unable to communicate with the thousands of needed volunteers, I cannot regard such coordinated action as "available" in any meaningful sense.)

Now consider a second case, where every resident of the flooded city turns out to help sandbag. In this case, my assistance is entirely unnecessary. The city will certainly be saved, and my efforts achieve no further benefit. I might as well go home and take the night off; doing so will make the world no worse a place, and in fact (by my enjoying a pleasant evening) will make it better. Thus, of the actions available to me, the best consequences would be produced by my staying home. And again, since the same argument could be made from anyone's perspective, we are all acting wrongly according to AC. Yet the city is saved.

Thus, a world in which AC was universally satisfied (the world where everyone stays home) could have worse consequences than a world in which AC were universally violated (and everyone helps sandbag). Universal attention to the consequences would not, even in theory, always produce better consequences, and may in fact produce inferior ones. Yet if our accepting a doctrine such as AC was motivated by the love of good consequences, why should we retain it when the world is thereby worsened? For those interested in pursuing this argument in greater detail, in my essay "Is Objective Consequentialism Self-Defeating?" I present what I believe to be a proof that every "objective" consequentialist theory -- one in which the rightness of an act does not depend on the agent's own beliefs, thoughts, or motivations -- will be flawed in this way. Consequentialism simply cannot capture everything we want to say about the consequences. And if such moral accuracy is unattainable, not merely in practice, but even in theory, then we might well wonder whether objective consequentialism is an appropriate theory to accept.

Moving beyond the issues addressed in that essay, even a subjective consequentialism -- demanding that we act in whatever way we subjectively expect to produce the best consequences -- will often, through its universal adherence, make the consequences worse. Due to my ignorance of the physics of floods and of the likely response by other volunteers, my expected benefit from sandbagging is negligible, below that of almost any alternative act. In that case, "doing my part" to save the city is not merely morally superogatory, but in fact morally perverse. I must stop sandbagging, if I am to act morally, and instead turn my attention to other affairs. Similarly, if one believes on consequentialist grounds that the duty to vote is so weak as to be permanently outweighed by other concerns--if one could always find something better to do than voting--then it is difficult to avoid the conclusion that one almost always has a duty not to vote. Even if the election is in fact going to be split, if I am unaware of this fact in advance, I am under a duty not to cast the deciding vote. Everyone who participates in an election, then, no matter how crucial the decision or how monstrous the opposing candidate, has acted wrongly. How can this be the case? How can it be that a morally required outcome for the group could only be achieved by morally forbidden individual acts?

For many moral theories, this won't pose a problem. A Kantian might well approve of the decision not to violate someone's rights, even where doing so would have resulted in a smaller number of rights violations overall. Yet for a theory that finds its motivations solely in the desire to maximize good consequences, we are unable to ignore the fact that our doctrines have made the world a worse place to live. Perhaps the case of voting is telling us something fundamental about ethics: that we must look beyond mere consequences to find where the aims of morality lie.