Useful Utility: Accounting for Taste (Part 2)
This is part two of a series defending utilitarianism. You can read Part 1 here:
From here on out, I’m assuming you’re all good, right-headed consequentialists. You try to take whatever action gives you the optimal outcome.
What, exactly, are you optimizing for?
The answer, I think, is just whatever best-satisfies your tastes.
This is an unsatisfying answer. Most philosophers expect more from you: they assume you want to optimize1 for “the right thing to do”. And by right, they mean the moral thing to do.
But when it comes to defining morality, the argument evaporates. In almost every case they make something up. They appeal to what somebody else defines as moral, or to some principle like “rationality” or “justice”, or simply to whatever intuitively feels “good”.
None of these are grounded in any base truth; it’s turtles all the way down. You cannot derive morality from first principles because there are no principles to start from. I mean, maybe you believe in some god, and you believe he handed down some principles on stone tablets that you can derive the rest of you ethics from. If that’s the case, there’s no use reading this – read your tablets instead.
Assuming you don’t have such a god, we can’t find a first principle to build from. There is no intrinsic moral quality to an abstract concept like “justice”. There is not an intrinsic Truth in the universe that defines morality; it is a nonesuch beast. There is no “right thing to do”, only the right thing to do to achieve a goal. And there are any number of goals to pursue. The goal you pick is really a matter of taste.
In a sense this is a colossal dodge – what’s the point of a system of ethics that just says, “to hell with it, do whatever feels right to you”? But that misses some subtlety. I’m not arguing that you should do whatever feels right to you. I’m arguing that you should do whatever best-satisfies your tastes. That’s not necessarily the same thing – as with the Monty Hall example in the last post, sometimes what feels right isn’t actually the best way to accomplish your goals. I won’t try and persuade you of what your tastes should be, but I think you can be persuaded to satisfy those tastes better than you have been.
To use an analogy, you might not have control over liking pizza, but you do have control over what restaurants to go to. Now you could attend random restaurants, look at the menu, and order whatever most closely resembles a pizza. Or you could read about the restaurants in your neighborhood beforehand, and go to the best pizza place. A good philosophy can help you find the best pizza place, but it can’t persuade you that pizza tastes good to begin with.
It can also help reconcile conflicting tastes. To stretch this metaphor to the breaking point, you might like pizza, but dislike spicy food. But you can eat spicy food over and over again until you acquire that taste. In a vacuum, is this a useful thing to do? Perhaps not. But if you really like pizza, it’s perhaps worth adjusting your other tastes such that you can enjoy the very best pizza available, even if it’s spicy. Or perhaps you just have a “taste” for eating with friends, and that matters more to you than the flavor profile of the food. You might want to acquire whatever taste your friends have, so that you can satisfy your primary taste (eating with friends), without disliking the food too much.
For the rest of this post, I’m going to refer to a lot of preferences as “tastes”, because I think “taste” has a wonderful capricious quality to it. You don’t really argue with somebody about their tastes. I might think some people have bad taste in music, and for the life of me I don’t know why so many people like Brussels sprouts,2 but I don’t really judge them for it in a serious sort of way. And that’s what I want to capture – the somewhat arbitrary nature of these first-order preferences, these tastes. Now I don’t think all preferences have this quality – some are more second-order. For instance, I might say I have a preference for living in the city because there are more people. That is a preference that I think can be debated as right or wrong. But the underlying first-order preference (wanting to be around more people) cannot be. It is as unassailable as any other taste; it is just something that is.
The misguided search for rationality
A huge number of arguments stem from conflating arguments about taste with arguments about reason.
Allow me to recount a story, to illustrate this point: A few weeks ago, it came to my attention that some people will not sit on a bed if they’re wearing pants they’ve worn outside. As a matter of fact, there are some people who will, immediately upon entering their home, change into their “inside pants”.
This is not a minority position. This is not one clean freak. This is, like, maybe a quarter to a third of the people I asked.3
Now there is something quite sensible, I think, about a rule like “don’t wear shoes in bed”. Shoe bottoms are gross. There is often dirt on them, sometimes mud. This isn’t true every time you go outside, but it is true enough that a general rule like “don’t wear shoes in bed” is a good idea, instead of evaluating your shoe soles every time you come inside. But pants? It’s ridiculous.
So I found a friend who was one of these no-jeans-in-bed people and told them so. And I proceeded to argue with them for like, half an hour. I thought I had all sorts of clever points. “What if”, I told them, “I could prove that you cannot visually distinguish between a sheet that’s been sat on by outside jeans vs inside sweatpants?”. Well perhaps there are microbes on the jeans, they said. So I proceeded to talk through elaborate thought experiments involving how many microbes are on the average pair of jeans vs sweatpants, and asked them how often they washed the sheets, or left the windows open, and all kinds of other hypotheticals that I thought were awfully shrewd and they thought were awfully tedious. Needless to say, they were not persuaded.
The mistake we both made was trying to rationally argue about this in the first place. I assumed there was some state-of-the-bedding they were trying to avoid. And indeed, they articulated such things – concerns about outside dirt, or bacteria, or some such. But at the end of the day, that wasn’t what it was about; they just didn’t like the idea that the pants that touched a public bus seat would touch their bed as well. And even if their preference really was to avoid dirt on the bed…so what? Not wanting dirt on the bed is a taste too. You could try and show that that’s just a mechanism to avoid bacteria, which makes you healthier, but honestly…it’s tenuous.4 And that’s fine. We don’t need to rationally justify liking pizza, and we don’t need to rationally justify not wanting “outside clothes” on the bed.
What we should do, though, is be clear-eyed about what’s a taste and what’s not. Banning jeans on the bed in service of a taste for not being sick is a bit silly, and if that was the motivator then we’d have something to argue about.
Back to moral philosophy…
This digression is relevant because it is a (hopefully) more obvious example of a mistake we make all the time. A lot of utilitarian conclusions are dumb (more on that in the next post), but even the smart ones tend to imply things like “it is better to donate to a cause that saves children you have never met than it is to donate to your local theater”. And then a bunch of non-utilitarians get angry and say things like “but what about community”, or “you’re completely discounting culture”. And the utilitarians say “but but but LIVES! Don’t you care about LIVES?” And they shout back and forth, and the first group decides the utilitarians are too bloodless and theoretical and the second group decides the non-utilitarians are too stupid, and the shouting dies down and they ignore each other again.
And the thing is, they’re just arguing about taste. If your tastes tell you that saving lives is more valuable than having more plays be performed, utilitarianism can help. I think this is a fairly common taste, and I think one of Effective Altruism’s greatest impacts is finding people with that taste, and helping them satisfy it better. But that is not everybody! Some people really, really like the opera. And I can’t really begrudge them that. It sounds strange, that somebody may prefer to live in a world with more opera than a world where more lives are saved, but it’s really not so weird. Most of us are pretty selfish. Some of us are more selfish than others, of course, but even the most altruistic people tend to spend a good amount of their resources on themselves. Because even those people have a taste for more than just altruism; they have a taste for material comfort, and self-actualization, and so on and so forth. Maybe the donation to the local theater comes from a non-altruistic place – the donor has a taste to style themself as a patron of the arts, that competes with their taste to be altruistic. And I can’t fault them for that taste, any more than I can fault them for liking pizza.
What I do fault people for is not accounting for them. Not everything is motivated by a desire to be moral, and tastes conflict. That’s ok. But when the opera donors complain that you can’t just compare everything to saving lives, or do some hand-waving about diminishing returns and investing in both…well, they’re not thinking clearly. They should decide if, to them, saving one life is better or worse than allowing hundreds to attend an opera. They should decide how much of their resources they want to spend on their “doing good” tastes, and how much they should spend on their more selfish ones. They should compare outcomes and decide what they prefer, whenever a significant decision arises. Utilitarianism rests on two pillars. One is consequentialism, and the other is this: any two states of the world can be compared, and one will always be preferable to the other.
A defensible utilitarianism isn’t one that prescribes maximizing the number of lives saved, or net happiness, or some other measure like that. It is one that prescribes being aware of all of your tastes, weighing them in some way, and choosing the course of action that best-satisfies them. The methods to do that will be discussed in the next post.
Terms like “optimize” get associated with utilitarianism in particular. But I will reiterate: every ethical theory involves optimizing for whatever that theory’s conception of “good” is. And no theory, including this one, provides a mathematical precision that you can apply in your day-to-day life. “The optimal choice” is merely a useful concept.
So we’re clear: Brussels sprouts taste bad. Full stop. It is true that one thing can always taste better or worse than something else; the boiled sprouts I ate as a kid tasted much worse than the roasted sprouts a modern-day San Franciscan will make, covered with balsamic vinegar and salt. But the fact remains that what tastes good about the dish is the balsamic vinegar and the salt, and it would in fact taste much better if the technique were applied to something that wasn’t a Brussels sprout.
Yes, this is entirely unscientific. No, that does not affect my point in the slightest.
And even if it weren’t tenuous, then you just have a taste for not being sick. Or a taste for not being bedridden. It is tastes all the way down.