by Richard Robb
But just as I was about to send the usual it’s-the-same-for-everyone email, I switched her grade to a B. I asked her not to tell anyone, and as far as I know she didn’t.
Why did I give her a break? I had nothing to gain. I’d never see her again and didn’t care about her personally because I didn’t know her. Nor could I justify it in terms of some moral precept to act for the common good. Presumably, if this young woman lost her fellowship, the money would go to a student more deserving of Singapore’s assistance, one who legitimately earned the required grades.
In terms of identifiable interests, however, I did have something to lose. If word got out that I was arbitrarily changing grades, this one act of mercy would have opened the floodgates to similar petitions. For obvious reasons, I didn’t want to set myself up as the judge of each student’s circumstances. It could have become a slippery slope, quickly reducing my grading process from order to chaos. Now that five years have passed, I can safely say that I didn’t slip down that slope: I’ve never bestowed a similar act of mercy on another student.
Given that I had no incentive to favor her plea over all the others and good reasons to dismiss it, did I behave irrationally? Was I losing my grip? I don’t think so, even though this act of altruism can’t be explained by purposeful choice. It was a spontaneous, random, one-time gesture of mercy. I am not proud of helping her, nor do I regret it. I can only say that I did it. It was out of character and for-itself and cannot be explained by an appeal to any motive beyond acting with mercy.
Love Altruism Redux
Let’s revisit the woman from the beginning of this book who jumps into the river to save her drowning husband. We’ll add some details to their story.
The husband hardly ever walks their dog because the wife bought it and he never wanted it in the first place. Sometimes the wife has to leave her office early to walk the dog, even though the husband is already home. Left to himself, he doesn’t mind getting a little exercise, but he won’t walk the dog because he sees it as his wife’s responsibility.
When the wife returns home, she scolds her husband for his selfishness. Scolding is unpleasant for him, but he doesn’t change his behavior. The wife likes venting, but not as much as her husband dislikes being on the receiving end.
This suggests that the two don’t care about each other in the sense that the wife’s well-being enters into her husband’s utility or vice versa. Of course, if the dog were the only contentious issue, we could overlook it. But to prove there is one more category of for-itself altruism, let’s continue with our example. Let’s assume there are many such cases where one spouse fails to expend a minor effort that would deliver a larger benefit to the other. This is not to say they have a bad marriage—they just don’t attribute value to each other’s well-being. They may regularly help each other, exemplifying two varieties of purposeful altruism: selfish altruism (that is, trading favors) and social norms. They may enjoy each other’s company and mean it when they say that their marriage is a happy one. But they don’t care directly about each other’s well-being.15
Nonetheless, the wife can love her husband. To be precise: when he’s drowning in the river, she suspends the entire schema that defines her purposeful actions. In this instance, love is not a desire and saving her beloved is not a preference. The commitment to her many preferences is no greater than her love of her husband. It’s not mercy; she might jump into the river every time the same situation arises. But she can’t supply a reason that would mean much. A reason for jumping would be in reference to her other choices, and this one doesn’t fit. Its authority is bigger than any means, ends, or trade-offs.
It would be wrong to try to cast the woman’s choice as unobserved care that rises to the surface when the husband’s life is in danger. To make the point logically, we can embellish the example further by supposing that as soon as her husband is out of the river, the wife resumes her nagging. Or we can provide for other conduct that substantially diminishes her husband’s quality of life. Still, when the time comes, she jumps.
We’ve mentioned a few extraordinary for-itself acts of mercy, for example, the senior monk who carries the woman and the saving of Captain Locher, but those are logically different from love altruism. They are sporadic, whereas the woman might try to save her husband every time. Nor are those acts of mercy driven by strong bonds with the recipient. Most importantly, we don’t have to say that the senior monk and the American general were drawn to these acts by forces greater than all their other concerns. I am not denying that there are borderline cases—perhaps the end of Casablanca—where we could argue for mercy or love altruism. But either way, it’s for-itself.
Malevolence
Selfish altruism has a counterpart in strategic competition. In a game of chess, we’re happy to lose a piece as long as our opponent loses one of greater value. This behavior is clearly purposeful. But just as a self-interested performance of favors lacks the goodwill associated with altruism, strategic competition lacks the vindictiveness and smoldering ill will that evoke acts of malevolence.
Strategic competition aside, modeling malevolence isn’t as simple as applying our taxonomy of prosocial altruism in reverse. Consider the manners or ethical code branch: while some societies care a great deal about vengeance, most moral codes stress loving your neighbor rather than harming your enemy.
Nor is there a clear analog for the observed care altruism modeled with the Rotten Kid Theorem. People rarely expend resources up to the point where they equate the loss to themselves to the loss to their enemy. A struggle so bitter that one person hates another as passionately as a mother loves her child is inherently unstable. Very soon, the malevolent person would destroy her enemy. Even unobserved care malevolence, caring about my enemy’s distress enough to trade my own distress for it at some ratio, seems implausible. Suppose I was walking down an isolated street and came across my enemy’s parked car. It would only take a moment to throw a rock through his windshield and no one would ever suspect that I was the culprit. Yet I can’t conceive of doing something so deranged.
And finally, the reverse of effective altruism would be monstrous: except in cases of mental illness, no one has a general desire to harm strangers. A disturbed person who feels wronged by life may enjoy inflicting pain from time to time, but to qualify as a reverse effective altruist, she would have to devote herself to inflicting the maximum possible pain on humanity as a whole. While a terrorist might want to wreak damage on a particular population, he acts out of selfish malevolence to achieve a political agenda, albeit mixed with pleasure from harming his enemies.
Love altruism has a straightforward malevolent counterpart in crimes of passion. There must also be a malevolent counterpart to mercy, since neither the pleasures of malevolence nor the sweetness of revenge can be denied. In for-itself malevolence—let’s call it spite—it’s the act that matters. The opportunity for this type of malevolence must arise in a natural context, and the punishment ideally preys on the enemy’s flaws. Poetic justice is sweet in real life as well as in fiction.
One of literature’s most artful acts of vengeance occurs in Edgar Allan Poe’s “The Cask of Amontillado.” After a chance encounter during a carnival, Montresor exploits his enemy Fortunato’s vanity to lure him to his doom in the catacombs. This murder is premeditated—Montresor has simply been waiting for the right opportunity. Spontaneity plays a less crucial role in spite than in mercy, since the most inventive punishments often require a degree of planning. And while mercy can benefit a stranger, spite involves strong animosity toward a particular individual. As in all well-constructed malevolent acts, the victim here recognizes the author of his downfall.
There are a few other key differences between mercy and spite. While we can only pursue a small fraction of the abundant chances that life affords us to act altruistically, opportunities to enjoy the pleasures of spite are rare. (Given that the malevolent act need only seem artful to the actor, people with low standards migh
t find more opportunities for for-itself malevolence.) And despite philosopher Jeremy Bentham’s claim that malevolence is one of life’s basic pleasures, it may come with inner conflict.16 Acts of malevolence, especially ones that violate our ethical principles, can generate nagging feelings of guilt and self-doubt. Fifty years later, Montresor was still rationalizing his crime.
Regret versus Remorse
Regret and remorse are different classes of feeling, but if we view all behavior as purposeful, we risk confusing them.
The anodyne case of regret occurs when we make an error while attempting to optimize. When people say they feel regret, they might simply mean that with hindsight, they recognize that they would be better off had they chosen differently. Or maybe they mean that they miscalculated: if they had thought more carefully, they would have acted differently. Regret belongs squarely in the purposeful realm. We should be able to learn from mistakes and let regret fade into the past.
Remorse is not simply a stronger version of regret—it belongs to a different category. It would be unremarkable to say that I feel regret for not buying gold in 1986, since its price increased by a factor of three over the next thirty-three years. It would sound strange, however, to say that I feel remorse for not buying Microsoft stock at its 1986 initial public offering, even though it increased by a factor of five hundred over the same period. I just feel more regret about Microsoft stock than I do about gold.
If a person makes a profound error in judgment or commits an immoral act, the remedy (if there is one) is to repent. That repentance, like the original act, stands for itself. Repentance can’t be outsourced. Scapegoats, sin eaters, and the buying of indulgences have all passed into history. Charitable giving (which is not exempt from purposeful calculations) may help relieve a general sense of guilt, but it cannot assuage remorse over a particular act. If that act was not part of an optimization problem, then the subsequent contrition cannot be either.
Jean-Paul Sartre tells the story of a student whose brother was killed in the war in 1940. The student is torn: he feels compelled to join the Free French Forces but also to stay at home and care for his mother, who would suffer without him.17 This is more than just a difficult decision. He can weigh the chance that he would affect the outcome of the war against the likely impact of his absence on his mother, and either way, he could argue afterward that he chose wisely within the context of his moral precepts. Nonetheless, his decision cannot possibly be cast in the framework of purposeful choice. He can’t quantify and compare the benefits of fighting against the costs of abandoning his mother—he’s grappling with love altruism for his mother, his country, and his brother’s memory, rendering the decision for-itself. Whatever he chooses, he is destined to feel remorse. But that does not mean he’ll necessarily feel regret; he can simultaneously feel remorse over his decision and believe it was the right one. If he could do it over, he’d make the same choice.
Back to the Good Samaritan
The Good Samaritan did nothing noteworthy by stopping to check on the beaten man. That could have been predicted: it’s likely he would have stopped every time he found himself in a similar situation. He was merely exercising the values instilled by his upbringing. Since adherence to principles or social mores is purposeful, we can model his response with rational choice.
But taking the beaten man to the inn, caring for him, and paying the innkeeper to continue caring for him lifts the Samaritan’s behavior into a different category. If he’d lavished that degree of attention on every needy person he met, he’d never have made his way out of Jerusalem. This altruism was unpredictable and so qualifies as for-itself mercy.
As we’ve discussed, several or all five categories of altruism can be in play at the same time. The Samaritan could have gone the extra mile partly out of selfish altruism if he planned to bolster his reputation by telling the story afterward. Or maybe he was an extreme effective altruist and felt deeply about the welfare of the beaten man and everyone else besides. But none of this explains the extravagance of his gesture, all the time and money spent on one person.
7
Public Policy
I’ve presented the argument that much of altruistic behavior is purposeful, and some is for-itself. In light of this distinction, let’s look at policies adopted by individuals and particularly by governments to advance collective welfare. Many policies rely in straightforward ways on purposeful choice economics. (We’ll use the term “purposeful” here rather than “rational” to allow for policies that exploit behavioral biases.) These policies can be boiled down to a fixed set of rules that maximize a well-defined concept of the public good. But other policy problems cannot be solved by even the most sophisticated application of purposeful choice and instead require tough, for-itself decisions. This discussion will not resolve intractable policy problems or moral dilemmas, but we can gain insight by assigning them to their proper realms.
Pareto Efficiency and Purposeful Public Policy
One simple rule, Pareto efficiency, can be used to determine whether a policy can be evaluated in terms of purposeful choice. In purposeful choice, one policy is superior to another if and only if it is Pareto efficient, that is, the superior policy helps someone and hurts no one, or at least makes the winners better off by a sufficiently large margin that they could theoretically compensate the losers and no one would be worse off. This approach assumes that policymakers can determine the preferences of all their constituents, including their constituents’ ethical principles and the weight given to those principles. It’s not practical, but that’s a technical issue which doesn’t affect whether a policy problem is purposeful or for-itself.1
Suppose a factory emits pollution that imposes health costs of $1 million on people who live nearby. It would be Pareto efficient to force the factory to pay $300,000 to install equipment that would reduce the impact of the pollution to $100,000. The factory owner would then be $300,000 worse off, but if those who most benefited from the reduced pollution chipped in to compensate him, everyone would win. If profits before installing the equipment were less than $300,000, another Pareto improvement could be achieved by shutting down the factory.
To measure the health costs ($1 million in this example), we’d estimate the total amount that each person afflicted would be willing to pay to reduce the factory’s pollution to zero, assuming everyone had perfect information. An even better estimate would tack on social costs, such as health care paid by the government.
Some decisions, however, must be made without the guidance of Pareto efficiency. Political scientist Michael Taylor provides an example in the story of Arizona’s Yavapai Indians, who would refuse “all the money in the world” from a federal bureau intent on building a dam on their ancestral lands.2 They could not be adequately compensated by any material good for the loss of what they have. It simply might not be possible to relocate the native people, give them a monetary side payment, and make everyone better off. Society, in deciding whether to let the government build a dam, would have to make a for-itself decision about whether the benefit of the electricity exceeds its costs to an unwilling few.
Two Moral Dilemmas
The distinction between purposeful and for-itself decision-making can be demonstrated by looking at two enduring moral puzzles: the merchant’s choice posed by Cicero in 44 BCE and the trolley problem posed by Philippa Foot in 1967.3 The merchant’s choice belongs in the purposeful category, where options can be evaluated, ranked, and traded. Choices in the trolley problem, however, depend ultimately on impulse—attempts to calculate the trade-offs are swamped by an individual exercise of will. Action (or inaction) is for-itself.
In Cicero’s story, Rhodes is suffering a famine when a merchant arrives at the port with a ship full of grain. The merchant knows that other ships carrying grain are en route. If, as Cicero tells us, the merchant is virtuous, must he inform the starving buyers that further supply is on the way, thereby reducing the price he can charge? Two cynics, Antipate
r and Diogenes, debate the matter.
Antipater and Diogenes agree that disclosure would be nice, all else being equal. But Diogenes argues that the merchant is not obligated to reveal this information because there is a significant difference between not revealing and actively concealing. Antipater favors disclosure on the basis that “it is your duty to consider the interests of your fellow-men and to serve society.”4
The primary impact of disclosure would be on the market-clearing price and hence on the merchant’s profit. Either way, the people of Rhodes, taken as a whole, will end up with the same amount of grain on the same schedule.5 Thus the question is really, how much money should the Rhodians transfer to the merchant? And how important is the principle that honesty is the best policy? Omitting material facts is not as bad as lying, but it still wouldn’t be the merchant’s finest moment. The merchant must consider how his actions will affect his reputation if he plans to do business in Rhodes again or if word of his conduct spreads beyond Rhodes. He’ll necessarily evaluate all these factors in the context of his wealth—if he’s struggling, he’ll tilt the scales in favor of not disclosing. None of the values in play are absolute.
The merchant can weigh the social and personal implications of this decision, then decide whether the potential collective gains from disclosing outweigh the private benefits of keeping quiet. Likewise, Antipater and Diogenes can deliberate, hone their views, and factor in self-interest. Still, their behavior is predictable: Antipater would always favor disclosing. Diogenes would be less inclined to disclose unless the damage from concealing was very great. Every aspect of this debate fits into the purposeful choice framework. It’s possible to do the necessary calculations to rationally balance concern for public welfare, private gain, and ethical principles.
