The ‘how’ and the ‘what’ of Effective Altruism (and of good decision making)

Good decision making is more about the “making” than about the “decision”

Altruism is an intriguing phenomenon. Many of us make material sacrifices in money, effort or time that benefit others, without a clear immediate material benefit to ourselves. We hold the door open for someone, or allow a driver to join a queue of traffic. We volunteer at the annual fête of our children’s school, or we stay late at work to help a colleague finish a report. And many of us give away actual money to good causes. On the face of it, such altruism goes against our self-interest. How could evolution favour such an organism?

Interestingly, much of this can be explained through evolution. Prosocial behaviour and reciprocity help us cooperate with others. And over many thousands of generations, populations that were better at working together were able to achieve more than those that were less cooperative, and hence prospered in comparison. Another angle is that altruistic behaviour patterns signal our character, and help us build a good reputation. This in turn may help us to find a mate, or secure our ongoing membership of a collaborative tribe. But while this may explain why altruism prevails at a community level, the benefits to an individual may be too vague, too distant or too conditional to really encourage the behaviours instrumentally. There must be something else that mediates it.

Choosing for altruism

When we behave in an altruistic way, we generally feel good — this is sometimes called the ‘warm glow’. And this would appear to have been developed this into something even bigger: our sense of morality, of good and evil, of right and wrong. We may not all agree on such matters as what is good and evil, or right and wrong, but almost everyone experiences positive feelings when they do what they believe is right and good (and negative feelings at doing things that are wrong and bad).

If our sense of morality drives us towards doing good, it is but a small step to the point where we seek to do “the most good” — the Benthamite philosophy of utilitarianism. The idea of doing the most good is embodied in the notion of ‘ Effective Altruism’ (EA), a philosophy (and a community gathered around it) focused on maximizing the good one can do. It emerged in the late 2000s, and has gathered some high-profile supporters including Bill and Melinda Gates, Elon Musk and Warren Buffett.

The tenets of EA are essentially the same as those of more general good decision making: both aim to use evidence and rational argument to determine what the best possible option is in a particular situation, taking into account the overall costs (in the widest possible sense) and the overall benefits (across all relevant stakeholders).

But not everyone is so positive, and EA has many critics. A recent, comprehensive and thoughtful critique by Michael Nielsen discusses several of the problems some people see with EA, especially with the social movement around it. It also challenges the basic principles of EA, and there I think it overeggs the omelette, mischaracterizing EA as a principle (and by extension the basis of ethical, evidence-based decision making).

To be fair, the wording used by EA proponents (“doing the most good”, “maximizing”) is an open goal for critics, in the same way that any definition of good decision making that incorporates the aim to maximize (or minimize) some quantity is. Decision making is not, and cannot be, about unconditionally maximizing anything. It is about making — and the same inevitably applies to altruistic decisions.

Don’t deny the trade-off

But a sizeable part of Nielsen’s post ignores this fundamental constraint, and focuses on what is called ‘strong’-EA, characterized by the explicit pursuit of maximizing doing good. Precisely because such a stance necessarily disregards the existence of trade-offs, it leads to trouble. The article cites Peter Singer (a philosopher who has been championing the idea of EA for over 40 years) recounting the story of a pioneering Effective Altruist who determined that it would be immoral for her to have children (as that would take away resources that would do more good in other ways). This made her miserable, and eventually she and her husband decided “they could afford to raise a child and still give plenty”. It also observes that strong-EA would require every effective altruist in possession of two healthy kidneys to donate one for transplantation. And it alludes to the fact that strong-EA adherents would really need to weigh up every bit of expenditure for personal enjoyment against the amount of good the money could do: how many malaria nets could the cost of cinema ticket, or a restaurant meal fund?

Even committed effective altruists are ultimately forced to set boundaries, and this makes them miserable. Die-hard effective altruists see this as a flaw of the people involved, but Nielsen implies that this is a flaw in EA. However, the need to recognize boundaries and make trade-offs is not a flaw of strong-EA, it is a fact of life.

Implicitly or explicitly, we set boundaries around many other decisions. We don’t seek to earn “the most” money (if that is what we really pursued, we would work more hours, have a second job, or do something a more dangerous or unpleasant that pays better, or even spend our spare time begging on the streets). We may want to spend more time with our family, but we don’t stay at home all day, every day. We may want to get fitter, but we don’t spend all our spare time exercising — you get the picture: we compromise all over the place, because we are not prepared to go indefinitely far in the pursuit of our goals.

So, I entirely agree with Nielsen that the fact that some people become miserable when they find it is impossible to live up to the demands of (strong-)EA is not their fault. But it is also not EA’s fault for demanding too much from those in its community. It is that it is setting literally impossibly high standards. It is always possible to earn more, spend more time with one’s family, and exercise more, but it is not possible to ‘earn the most’ etc. For the same reason it is always possible to do more good, but it is not possible to do ‘the most good’.

The fact that so many effective altruists seem to need to resort to “special pleading” (‘extension clauses’ about having children or setting aside an ice cream budget or a dinner budget) is invoked to support the claim that EA is “the wrong way to live”. This is, in a way, similar to claiming that human decision making is fundamentally irrational because we fail to make ‘optimum’ choices according to some narrow criterion.

In our world, resources are scarce. We have only 24 hours in a day, and the time we spend at work, we cannot spend with our families (and vice versa). $4,000 might buy 40 interventions to prevent trachoma, a significant public health problem in dozens of countries with over 130 million people at risk of permanent blindness. It might also buy 800 mosquito nets, which statistically prevents the death of one child. Which does the most good?

It ain’t what you do, it’s the way that you do it

Strong-EA cannot give an answer to such questions, and so it is, as Nielsen argues, a poor life philosophy. But that is not because there is something amiss with EA per se. Any ‘maximizing’ philosophy would be poor — maximizing wealth, maximizing having fun, maximizing healthy eating, the amount of exercise etc. The reason is that maximizing is inherently denying the existence of the need for balance. Economists circumvent this by positing an abstract concept, ‘utility’, that integrates all material and immaterial benefits — but nobody has ever figured out how anyone performs that integration. It is an interesting and useful theoretical notion, but of little practical value. Strong-EA is an ideal that is literally impossible to realize in a real world with scarce resources, where compromises and trade-offs are inevitable.

So, weak-EA is the only feasible version of EA and, in contrast, can, as Nielsen writes, be an inspiring philosophy. Like rational decision making, it is not about the overall maximization of some measure, quite the contrary. Both are about how best to spend a limited budget of resources, not about maximizing that budget or maximizing some outcome with no regard of the sacrifices that need to be made.

This requires the exercise of judgement based on evidence and rational argument — but also based on one’s preferences and values. Consequently, good decision making is more about the “making” and less about the “decision”. It is about how the evidence is gathered and evaluated and how robust the argument is, not about the eventual choice or the outcome. Different people, with different preferences and values making a decision in a rational manner may still end up with different choices.

This applies to EA just as well. It is not about finding that particular choice that maximizes the greater good, it is about putting in the effort, thinking and considering, the process of weighing up the alternatives.

Both for effective altruism, and for good decision making more generally, it is the how that matters, more than the what, just like the old song observes.

Originally published at http://koenfucius.wordpress.com on June 17, 2022.

Thanks for reading this article — I hope you enjoyed it. Feel free to share it (the ‘share’ button below or at the top has direct options for Twitter, Facebook and LinkedIn) or simply copy and paste this link. See all my other articles featuring observations of odd human behaviour (I have been publishing one every Friday since 2016) here. Thank you!