Abstract Benevolence as a Virtue

Better approximate your ideal self

Many people find effective altruism emotionally unappealing. They find hands-on, locally-oriented volunteering more fulfilling, and may resent the suggestion that it would be even better to do things that help others more at the cost of being less personally satisfying. (For the record, I’m happy for people to do imperfectly good things, but I think they should acknowledge the truth that more good is better, and give some thought to where they could get the most positive impact for marginal increases in moral effort or self-sacrifice.) Some seem to assume that their current sympathies mark the limits of what’s humanly possible, and thus criticize EA as an essentially “cold” or “unemotional” approach to ethics. I think that’s a mistake. What they’re picking up on is rather that EA won’t automatically validate your emotional inclinations. It may uphold a different emotional response as more ideal. But it’s surely right to do so: our actual inclinations are not guaranteed to be ideal.

So I have two main responses to those who—like Curtis Yarvin—oppose efforts to extend one’s circle of moral concern beyond those to whom one (currently) has “real sympathy”: (1) Broader “abstract sympathy” is both possible and, when it occurs, more virtuous; and (2) When our actual motivations fall short of full virtue, it’s better to at least be strong-willed (or “continent”, as Aristotle would say) and muster the motivation to simulate virtue rather than to wholly neglect what we know to be true and good.

1. Abstract Sympathy

As Bertrand Russell wrote in Education and the Good Life:

There is a purely physical sympathy: a very young child will cry because a brother or sister is crying. This, I suppose, affords the basis for the further developments. The two enlargements that are needed are: first, to feel sympathy even when the sufferer is not an object of special affection; secondly, to feel it when the suffering is merely known to be occurring, not sensibly present. The second of these enlargements depends largely upon intelligence. It may only go so far as sympathy with suffering which is portrayed vividly and touchingly, as in a good novel; it may, on the other hand, go so far as to enable a man to be moved emotionally by statistics. This capacity for abstract sympathy is as rare as it is important.

Abstract sympathy: almost as rare as the ability to read upside-down.

In my 2019 paper ‘Overriding Virtue’, I expand upon Russell’s idea:

The possibility of such abstract sympathy undermines the charge that it is necessarily “callous” to maximize net welfare at the cost of more proximate interests. On the contrary, such impartially benevolent preferences may instead reveal a deeper wellspring of emotional concern for others than is found in the merely ordinarily (concretely) sympathetic. Indeed, we should surely expect that the compassion of the ideally virtuous agent would extend more broadly than our own, flawed and imperfect compassion manages to do. Insofar as moral perfection is thought to involve a kind of universal love, it is very natural to conceive of the ideally virtuous agent as one who would feel the moral-emotional pull of others’ needs just as strongly even when they are distant from the agent herself. And it would certainly not be callous or lacking in compassion when such an ideally virtuous agent acted upon her expansive sense of compassion to protect a greater number of people despite their lack of proximity (just as there is nothing callous about saving the nearby many over the nearby few).

See also Beneficentric Virtue Ethics: “Prioritizing the homeless person before your eyes over distant children dying of malaria is not virtuous… It rather reflects a failure of empathy: you feel for those you see (good so far!), but not those you don’t (which is obviously less than morally ideal).” Similar things can be said of the common tendency to prioritize cute animals over equally intelligent ugly ones, etc. The positive concern is good as far as it goes, but it doesn’t go as far as it should (or ideally would).

2. Abstract Benevolence as Backup Motivation

My paper continues:

Of course, we are not ideally virtuous agents. Even many of us who are moved by moral reasons to prefer saving the greater number may nonetheless find that this verdict conflicts with our strongest sympathetic impulses (which remain tethered to more proximate, salient needs). This raises interesting questions about how to evaluate our characters when we choose to save the distant many over the nearby few. Is this a virtuous choice, since it is done for good moral reasons and in recognition that this is what the ideally virtuous agent would do? Or is it disreputably callous, as we are in fact overriding our strongest sympathetic impulses, and letting harm come to those we see most vividly, merely for the sake of some “greater good” that we do not fully (i.e., emotionally) comprehend? To answer this question, consider the following character trait:

Abstract benevolence: The disposition to allow abstract, globally-oriented moral reasons to override or redirect one’s natural inclination to prioritize the most salient needs one faces, when this is necessary to address more objectively pressing needs.

I propose that abstract benevolence is a neglected virtue, specific to imperfect agents like ourselves, that serves to moderate the biasing effect of ordinary sympathy…

While it would be most ideal to be directly sympathetically moved by all needs in proportion to their objective merit (which is unaffected by factors like distance and cuteness), the next-best option is to possess sufficient abstract benevolence to do the ideal thing even in the absence of ideal feelings. Blindly indulging your actual feelings, regardless of how badly distorted they are, must surely come in a distant third.

Yarvin asserts that exercising abstract benevolence “wrecks” one’s “capacity for genuine sympathy,” but he offers no (non-fictional) evidence that this is true.¹ Scott Alexander, in Against The Concept Of Telescopic Altruism, argues more convincingly for correlated altruism: “People who are nice to a far-off group are more likely to be nice to a nearby group, because all forms of compassion come from the same place.”

Fixing Incentives

I’m no fan of pretending that subjectively-fulfilling helping is guaranteed to be the most objectively helpful form of helping. I’m also not impressed by egoistic views that claim we have more reason to prioritize personal fulfilment over helping others more. I think those views are wrong.

But I agree it can be more challenging to motivate ourselves (and others) to act in ways that lack subjective appeal. Rather than conveniently revising our beliefs about morality to match our personal interests, I think we do better to try to reshape our incentives so that it becomes easier to act well. Taking abstract benevolence as a second-best virtue available to imperfect agents, the practical question becomes how to scaffold it. Examples might include:

(1) Taking the 🔸10% Pledge (previously recommended here) and publicizing this fact, as a precommitment device and costly signal.

(2) Take time to vividly imagine and reflect on the good done (in expectation) by your exercises of abstract benevolence.

(3) Surround yourself with people who have good values and who will reward you with social esteem for exercising abstract benevolence.

(4) Appreciate being part of a community with all the right enemies, spanning the political spectrum.

(5) Find meaning and satisfaction in all of the above.

This seems better than trying to directly optimize our “moral” actions for meaning and satisfaction. Especially when there are gadflies like me about, pointing out that a preference for hands-on helping is not really supportable by moral reasons… You may actually find that the most sustainable route to moral fulfilment is to first just focus on doing good efficiently, and then separately figure out how to find fulfilment in this.²

Conclusion

Critics imagine impartiality as “leveling down” our concern for those near, until we care as little for them as we do for the distant. The more ideal response, of course, is the opposite: “leveling up” impartiality. We’re absolutely right to care so much about our loved ones. Where we go wrong is in our deficient treatment of others:

We glimpse but a glimmer of the world’s true value. It’s enough to turn our heads, and rightly so. If we could but see all that’s glimpsed by various others, in all its richness, depth, and importance, we would better understand what’s truly warranted. But even from our limited personal perspectives, we may at least come to understand that there is such value in everyone, even if we cannot always grasp it directly. And if we strive to let that knowledge guide our most important choices, our actions will be more in line with the reasons that exist—reasons we know we would endorse, if only we could see them as clearly as we do the ones in our more personal vicinity.

That ideal of fully seeing (and feeling) others’ value is beyond most of us. But we can at least get closer to following the guidance of abstract sympathy when we muster abstract benevolence as a backup motivation. Refusing to act on concern that exceeds what you currently feel leaves your ethics vulnerable to emotional dictatorship. When we know there are good reasons our emotions are overlooking, we need to find other ways to ensure that our behavior is suitably reasons-responsive.

1. Yarvin also invokes a lazily indiscriminate form of aid skepticism, apparently unaware of the lack of controversy over well-targeted aid.

2. The separation there is a bit exaggerated: in reality, personal tastes and such will influence how “costly” different options are for you, and hence which option gives you the most altruistic bang for your buck. Smaller costs remain preferable all else equal, after all. And for major life choices, like career decisions, it would seem very inadvisable to try to force yourself in a direction that you find personally repulsive or deeply unappealing—it’s hard to see that working out successfully.