1 Introduction

(Crosspost of this).  

A warning: This is going to get very complicated. The various different models of well-being end up producing weird and counterintuitive conclusions. This is probably the topic that I have spent the most time thinking about of any topic in my life. Very few people have spent much time thinking about lopsided lives, so the terrain is largely unexplored. There are maybe two people on the planet who have spent lots of time thinking about this argument, and I’m one of them. So hopefully this is interesting, but it will only be interesting if you’re a giant nerd.

It’s a shame that very often the best arguments for things are totally ignored by almost all people. The number of people who have read various papers which provide quite decisive cases against deontology is quite small—no one has heard of the suitcase cases or Chappell’s paradox of deontology or the paralysis argument. The best arguments are often churned out in the dusty halls of academia, relegated to some obscure journal, footnoted occasionally—never seriously explored. There are, as far as I can tell, upwards of 8 arguments against deontology that have been made at various points, that no one has ever published a response to. These arguments appeal to deeply intuitive, widely shared premises, showing that if you accept, for example, the principle that you should want others to act rightly, you should give up your deontological beliefs.

But my target here is not deontology. Instead, I am arguing for hedonism about well-being, roughly the idea that the only thing that makes a person’s life go well is happiness and the only thing that makes a person’s life go poorly is suffering. In other words, the only things that determine how well your life goes are your mental states—the ones that feel good are good for you, ones that feel bad are bad for you. How much you know, how many friends you have, and so on matter instrumentally, as friendship and knowledge tend to make a life happier than it would be otherwise. Still though, if two lives are equal in pleasure and pain, it is utterly irrelevant how much of the other things they each contain. Friendship, knowledge, and achievements are valuable as means towards an end—as ways to get pleasure for yourself and others—but are not ends in and of themselves.

This argument comes from one of my favorite philosophers—Theron Pummer. Pummer is quite smart—his stuff is very worth checking out. You can read his original paper on the subject. While Pummer is the progenitor of the lopsided lives challenge, many of the arguments that I advance here do not come from Pummer’s paper.

One brief note: when you first hear the challenge, in the section titled “The basic challenge,” it is going to seem trivial. Some idea will come into your mind about how to solve the puzzle and you will think that it is not a real challenge. I assure you, the full extent of the puzzle is not obvious at first, so do not just assume that the solution you hear when I present the basic idea of the challenge works. It will almost certainly not work.

One more other brief note (next brief note, same as the first, a little bit longer and a little bit worse): this argument is mostly targeted towards pluralism about well-being. Pluralism is the idea that there are lots of things that make your life go well—knowledge, friendship, achievement, etc. There are roughly three views about well-being: desire theory, which says that what makes your life go well is just getting what you want; hedonism, which says pleasure is what makes your life go well, and pluralism/objective list theory, which says that there are lots of things of intrinsic value. I think this argument does suffice to refute desire theory, but it’s mostly an argument against objective list theory. The good news, though, is that desire theory is obviously false—Ben Bradley argues for this convincingly in his book Well-being and Death, and this article also has various arguments against it (I think much of what I said here in this article is rubbish, but the desire theory stuff is mostly right).

Well, now that we have a list of caveats and clarifications long enough to resemble a typical academic paper—and to cause Steven Pinker to keel over and die—ladies and gentlemen, friends and foes, hedonists and objective list theorists, strap in! Things are about to get crazy.

2 The basic challenge


Suppose you’re a naïve objective list theorist. You think that pleasure is good, pain is bad, knowledge is good, achievements are good, friends are good, etc. So the more you have of these good things the better it is. If you have a lot of knowledge, that is very good, and if you have a lot of pleasure, that is also good. These good things contribute in a straightforward way—they just make your life better. No complex mathematics involved. Knowledge adds value the way whales add weight—no matter how much stuff there is working against gravity, if you’ve got enough whales, then you can overwhelm all of it.

On this simple account, let’s say you have some score that measures how well your life is going. Things like pleasure add to the score. Let’s say every unit of pleasure increases the score by 1 unit. Let’s say every unit of knowledge increases the score by 10. And let’s say every unit of agony decreases your score by 10.

As you may have anticipated from my earlier description of this as naïve objective list theory, this view does not work. It has a fatal problem, namely, that it produces totally incorrect descriptions of how well one’s life is going when their life is lopsided. Suppose that a person is in extreme agony every moment of every day. Their life is as miserable as anyone’s life could be imagined to be. Every moment of every day, he experiences more suffering than the sum total of all suffering that has ever been experienced in human history.

Hell creatures

I think it’s safe to say that this guy is not living a good life. No matter what else is going on, if he has no pleasure but unfathomable amounts of agony, no matter how many friends or achievements he has, if he is experiencing a torture 10 billion times worse than the torture of any human every single second of every day, his life is not going well, obviously.

But naïve objective list theory has to deny that. If each unit of knowledge gives him 10 units of well-being, then even if he has 10^100 units of agony, that would be offset by 10^99 units of knowledge. In other words, if pain is bad and knowledge is good, on this simple account, then no matter how much pain you have, it could be offset by enough knowledge. The same is true if, rather than knowledge being the thing of value, instead achievements or friendships are—it doesn’t matter how many friends you have or how many times you achieved your goal of winning the science fair—if you are being unfathomably tortured all the time, your life is not good.

So basic, vanilla objective list theory is out. In its place are a whole host of other things you can believe. But as we’ll see, none of them are plausible either.

3 What if the other stuff only counts if you’re happy


“Before focusing on the guy next to you in an airplane, put on your own mask,” has been said in every xenophobic ham-fisted political message, about why, before we help starving African children or cease our policy of making it illegal for people to move over a border in ways that would dramatically improve their lives and also benefit us, we really need to focus on the current conditions of veterans and unemployed construction workers who earn more than almost all people in world history. But you might think that the basic principle can be applied here—the other stuff like knowledge, friendship, and so on only matters when you’re not in constant, extreme agony. Here’s one way of articulating this—maybe the other stuff of value only starts to matter when you have a positive pleasure score. So only when you have more happiness than misery do your achievements, friends, and so on start to be good for you intrinsically.

But here we come to the heart of the problem: this view implies hypersensitivity, wherein some extreme amount of value supervenes on a very small amount of some pleasure. In this case, a slight amount of pleasure has the potential to make a person better off to an arbitrarily large degree.

Imagine two people; one has 100000000000000000000 units of knowledge, 10000000000000000000000000 friends, and 1000000000000000000 achievements (many people have said I have these things). However, their hedonic score—their pleasure minus pain—is .01. On this account, this person is living a super excellent life. But compare this person’s life to another person whose hedonic score is -.01, but who also has 100000000000000000000 units of knowledge, 10000000000000000000000000 friends, and 1000000000000000000 achievements. This second person is living a bad life on this account—because their hedonic score is negative, none of the knowledge, friendship, or achievements actually positively contributes to their well-being. Thus, the extra .02 units of pleasure—which we can suppose is the amount of pleasure one gets from eating a popsicle—makes one’s better to an enormous degree. But this is implausible—eating a popsicle doesn’t make a person’s life enormously, mind-bogglingly, massively better. In fact, as the amount of knowledge, friendship, and achievements approach infinity, the benefits of eating the popsicle also approach infinity. But this is crazy! Eating a popsicle cannot have an arbitrarily large amount of value—its value can’t approach infinity. We’re assuming here that it produces no other effects.

This, in short, is the challenge (though we’ll see it getting even more forceful in the next section). Avoiding the first conclusion requires accepting the following principle:

Enough Pain at Each Time, Limited Well-Being (EPTLW): Any life that contains no pleasure and at least finite amount of pain P at each time cannot have an overall well-being score that exceeds finite limit L, no matter how much nonhedonic goodness it contains.

Basically, EPTLW says that there’s some amount of pain such that if you’re in that amount of pain every second and you experience no pleasure, there’s a limit to how well your life can go. So it’s not the case that, without reducing your agony at all, we can just increase your knowledge a lot, and your well-being keeps increasing all the way to infinity. This seems true.

And the good news is, there are lots of ways to accommodate EPTLW. The bad news: they generally imply hypersensitivity, wherein one small change makes a huge difference to the goodness of a state of affairs even though it doesn’t affect anything else. So as we saw of the view that other stuff only matters if you’re happy, this implies that any arbitrarily large amount of well-being can depend entirely on whether you get the tiny amount of pleasure from eating a lollipop. This is obviously false. So this is the challenge: avoiding hypersensitivity, and accepting EPTLW, while not having to say other crazy implausible things. Of course, as we’ll see in the next section, there’s a simple proof that one will have to say something implausible, even if they accommodate EPTLW without implying hypersensitivity.

4 Insensitivity


Here’s something we can prove: if you accept a few modest premises, including EPTLW, it follows that it is better to get a small amount of pleasure, in lots of cases, than any amount of any of the other things on the objective list. This is hard to believe if we’re objective list theorists. But as we’ll see, it’s true. Note that this is sort of orthogonal to the original lopsided lives challenge, but it is still quite forceful. This is not the main argument of this article, but it’s nonetheless convincing.

Let’s start by granting the assumption that EPTLW is true. In addition, it seems that pleasure can offset pain. Thus, if you’re super miserable, and you get enough pleasure, your life will be good overall. Pleasure and pain are opposites, so if you get enough pleasure when you’re in constant agony, your life will be good overall. Thus, we should accept:

Pleasure Offsetting: No matter how much agony one is in, if their pleasure is increased sufficiently, their well-being will be positive.

Pleasure Offsetting seems very intuitive. But there is an additional reason to believe it. It seems that when some experience is experienced is irrelevant to how good the experience is, assuming it does not affect the qualitative nature of the experience. So if, for example, instead of some amount of pleasure being experienced in a single moment, it’s experienced in lots of moments, this doesn’t affect how good it is. But this means that a sufficiently large amount of pleasure can be divided up into entire years or decades of pleasure without affecting their goodness. But if one experienced a decade of mindbogglingly intense ecstasy, that then contained a short period of intense pain—much less intense than even a single second of the pleasure—that would be obviously a worthwhile decade. This is especially plausible given that it seems like pain can be divided up in the same way. Intense agony is no worse than lots of slightly irritating dust specks; but intense ecstasy can obviously offset the badness of lots of dust specks, so by transitivity, intense ecstasy offset any amount of suffering.

In addition, as a consequence of accepting EPTLW, we should accept:

Objective List Non-Offsetting: There is some amount of agony such that no non-pleasure good on the objective list can offset it, no matter how much of it there is.

But this means that at various levels of suffering one is experiencing, a slight amount of pleasure is more valuable than any amount of objective list theory. If we accept that if one is in extreme agony no amount of non-pleasure objective list goods can offset their misery, and some amount of pleasure can offset their misery, and that any amount of extreme pleasure can be decomposed into lots of small instances of pleasure, and that any very extreme amount of objective list goods can be decomposed into lots of very great but less extreme amounts of objective list goods, then some of the slight increments of pleasure would have to be more valuable than any amount of objective list goods. To illustrate this, suppose that one is experiencing 100,000,000 units of pain. If they experienced 100,000,000,000 units of pleasure their life would be great. In contrast, if they got 100,000,000,000 x 100,000,000,000,000 units of knowledge, their life would still suck. This means that getting a single unit of pleasure over and over again would be less valuable than getting 100,000,000,000,000 units of knowledge over and over again. But if that’s true, then it must be the case that, at least some of the time, slight amounts of pleasure are more valuable than any amount of objective list goods. This is hard to believe if we accept objective list theory, where knowledge is valuable. It’s hard to believe that an arbitrarily large amount of one of the intrinsically valuable things is less valuable than a minuscule amount of one of the other intrinsically valuable things.

The objective list theorist might object that when talking about trading off pleasure against pain, in order to make it analogous to the knowledge case, we have to assume that they’re happening at the same time. If a person experiences 100,000 years of immense knowledge, before experiencing tons of suffering, it’s not at all obvious that this is bad for them. And this is analogous to the pleasure case.

A few points are in order. First, if the pleasure and pain are experienced simultaneously, if the pleasure swamps the pain, despite them being extreme, that seems obviously good. Think about a time when you’re extremely happy—say, you’re in love—but you also have a painful cut on your knee. That seems good. If we keep making the experience more intense—doubling the pleasure and pain—it seems like it only gets better and better.

In addition, in the case where pleasure is dispersed, it is totally clear that the pleasure experienced is valuable enough to offset any amount of pain. As I’ve argued elsewhere, duration and intensity can be traded off, such that any amount of pleasure experienced in some amount of time is just as valuable as a much greater amount of pleasure experienced in some much smaller amount of time, and comparably, that any immense amount of pain is less bad than a large amount of very mild pain. Suppose—as we should—that we accept that. Well then, for any amount of pain, there’s some amount of pleasure such that if one experienced both the pleasure and the pain, that would be just as good as experiencing a very lengthy and intense amount of pleasure interspersed by occasional mild pain. That would clearly be good, so as a consequence, very extreme pain can be offset by sufficiently intense pleasure. So there are good reasons to think that, whether one experiences the pleasure and pain simultaneously or not, any amount of pain can be offset by some amount of pleasure. In contrast, there is a very good reason to accept that if one experiences non-hedonic objective list goods and pain simultaneously, no amount of objective list goods can offset the badness of the pain, and no reason that does not already assume the truth of objective list theory can be given to think that when the objective list goods are spread out they can offset any amount of pain.

Suppose you accept the reasoning up until this point. There are two things you should think that the objective list theorist will not want to accept.

Sometimes, mild pleasure is more valuable than an infinite amount of every objective list good.

A life that contains short periods of mild suffering, interspersed with tons of non-hedonically beneficial experience of objective list goods, is very bad. So if one experiences 100,000 years of immense knowledge and then experiences mild pain—equivalent to stubbing their toe—and then repeats, this would be a bad life, rather than a good one.

To make the reasoning behind each of them explicit, the reasoning for the first claim is as follows:

Pleasure Offsetting + Objective List Non-Offsetting + Extreme pleasure can be broken down into lots of experiences of mild pleasure + Sufficiently extreme non-hedonic objective list goods —> Lots of mild increases in pleasure add up to being more valuable than lots of immense increases in non-hedonic objective list goods —> Some experience of mild pleasure are more valuable than immense increases in non-hedonic objective list goods.

To make the reasoning behind two explicit:

A life that contains constant intense suffering and lots of objective list goods is bad + any experience of extreme pain is as bad as lots of mild pains being spread out across time + getting some sufficiently extreme amounts of lots of objective list goods is as good as getting lots of objective list goods spread out across a very long period of time —> a life containing some very large amount of mild suffering, containing a much greater amount of objective list goods, such that one has 10,000 years of extreme non-hedonic objective list goods followed by one second of mild suffering is not good at all.

In other words, if you’re in extreme agony all the time, your life can’t be good, no matter how much objective list goods you have. However, separating the experiences so they’re not experienced simultaneously doesn’t affect their goodness. Therefore, if you have an experience where you have lots of objective list goods for a while, intense agony for a bit, and then repeat, that is not a good life. But given that that life is better than one where, instead of just intense agony, you have mild agony for a long period of time—equivalent to a mild toe-stub—and then a long period of objective list good fulfillment, that life is good. But given that when events occur doesn’t matter, this life could be scrambled, so that instead one experiences slight pain for one second and then days worth of objective list goods. But, by the minimal premises described here, this wouldn’t be good at all.

Both of these are very counterintuitive and only naturally explained by hedonism. But we have pretty much knock-down arguments for both of them. This makes a very strong case for hedonism—claims that objective list theorists would vehemently reject have quite potent justifications.

Suppose you reject that some amount of mild pain is as bad as extreme torture. Well, you still get the conclusion that if one experienced, for example, 100,000,000,000,000,000,000 years of extreme non-hedonic goods, followed by one second of intense suffering, and then repeated, that would be very bad. However, if the objective list goods were pleasure instead, that would be great. This is hard to believe if we’re objective list theorists, while it’s trivially entailed by hedonism.

So this is the first big problem for objective list theory. We can prove from fairly minimal premises both that slight pleasure is sometimes more valuable than tons of every non-hedonic objective list good and that a life that contains only very mild pain and lots of non-hedonic goods is not worth living. But the challenge gets much thornier.

We can give another argument for 2. Suppose that we accept the following premise, given by Pummer:

Enough Pain, Limited Well-Being (EPLW): Any life that contains no pleasure and at least finite amount of pain P cannot have an overall well-being score that exceeds finite limit L, no matter how much nonhedonic goodness it contains.

In other words, there’s some amount of pain such that if you’re in this much pain, there is a limit to how good your life can be if you just increase your non-hedonic objective list goods sufficiently. This seems true. If you’re in intense agony for ten billion years and are never happy, your life seems bad. But given the earlier reasoning, this life is as bad as a life containing some huge number of years worth of time experiencing mild suffering. And so if the first life is not good, then a life where the non-hedonic objective list goods are spread out—such that one has 100^100 years worth of mild pain but 100^100^100 years worth of intense objective list goods, that would be as good as the not good world, which would thus be bad.

5 The multiplicative view


Okay, back to the basic lopsided lives challenge. Here’s a view you might have. Perhaps to ascertain the value of pleasure, you must consider both the amount of pleasure and the source. Thus, various goods on the objective list—relationships, knowledge, and so on—serve to multiply the value of pleasure. As a consequence, Enough Pain, Limited Well-Being is true, as is Enough Pain at Each Time, Limited Well-Being. If non-hedonic objective list goods just enhance the value of pleasure, then if there’s no pleasure, there’s nothing to be enhanced. As a consequence, if one has no pleasure and has lots of misery, their life couldn’t be very good.

Note, there’s a more sophisticated version of this that I will talk about later that has been proposed by Richard Chappell.

This view has the problems described in the previous section. In addition, it implies hypersensitivity.

Suppose you think that knowledge is on the objective list. A person thinks that he knows some list of 100,000 facts, and derives 100^100^100 units of pleasure from these facts. Suppose he is justified in believing all of them, but only 50,000 of them are true. Some action causes him to experience more suffering than has ever existed in all of human history, but makes one of those facts true. Is this good? No, obviously. However, on this account, if it multiplies the value of the pleasure, then if the pleasure is sufficiently great, this would be good, despite not giving him any pleasure. This is obviously false.

The basic point is that if the objective list goods multiply the value of pleasure, then slightly increasing the objective list good would increase the multiplier, which tends towards infinite value as the level of pleasure approaches infinity.

6 The lifetime caps view


Maybe the most obvious proposal for avoiding the lopsided lives challenge involves supposing that there is a cap on the amount of well-being one can get from objective list goods. For example, perhaps as one’s amount of knowledge approaches infinity, the amount of well-being this brings them approaches some finite cap.

If there’s a cap on how much well-being one can derive from objective list sources over the course of their life, this cap is probably pretty huge. One can presumably eke out lots of non-hedonic value from knowledge, friendship, and such. But this implies, implausibly, that if one lived a short time, experienced intense agony, acquired lots of knowledge and friendship, and then died, their life would be great.

One might object and say that death is a significant harm—therefore, the badness of death is greater than the value of the non-hedonic goods. But this poses two additional problems:

It just doesn’t jive with objective list theorist intuitions. Intuitively, it seems that the non-hedonic value one gets from being in love over the course of their life is greater than the non-deprivational badness of death.

It implies that a life where one has intense agony, no pleasure, lives a short time, gets lots of knowledge or friendship, and then dies is better than one where a person has great pleasure instead, but a life of considerable length.

Another problem for this view relates to memory loss. In order for the total value of non-hedonic goods to approach some non-infinite amount as the amount of non-hedonic goods approaches infinity, the marginal value of extra non-hedonic goods has to approach zero as the amount of non-hedonic goods one has already had approaches infinity.

Now suppose that one had lots of knowledge, friendship, and so on, but they lost their memory of it. After that, they get more knowledge and more friends. Does this have much non-hedonic value? More specifically, is the value of the extra knowledge affected by the non-hedonic value they had before, which they have no memory of?

Suppose that it is. Well then, this would mean that if we actually existed prior to birth for a very long time, and got lots of knowledge and friends, we should be hedonists, because this would reduce non-hedonic value down to nothing. If aliens flooded us with tons of knowledge, before erasing our memories of it, that would make no future knowledge intrinsically valuable. That’s hard to believe—forgotten things shouldn’t affect our core normative considerations. If Huemer turns out to be right, and we have infinitely old souls that have acquired infinite friends and knowledge, then it would not be the case that we should act as hedonists because of that, on account of the infinite age wiping out all marginal non-hedonic value. Well, actually it is the case that we should act as hedonists, because hedonism is true, but whether we should act as hedonists shouldn’t depend on Huemer’s hypothesis.

In addition, this view implies that sometimes a person should pass up a thousand-fold increase in objective list goods with much greater probability. Suppose someone offers one the following deal:

Become omniscient, but only if they’re infinitely old (suppose they give 99.9% credence to being infinitely old, on account of Huemer’s hypothesis. Note, they only have memories from the last 50 years or so).

Learn one fact only if they’re not.

The first deal seems better, especially if objective list theory is true. But this account requires we deny that principle.

Suppose one instead says that only objective list goods that one has a memory of moves them closer to the cap. This has three problems.

It implies that sometimes a person becomes better off by losing memories of events, even if that doesn’t increase their pursuit of any of the things on the objective list. For example, this would imply that if you had a friend a long time ago, forgetting about them would be intrinsically valuable—it would make new friendships more valuable.

This implies that forgetting memories can be valuable to an arbitrarily large degree. If a person will learn one fact a year for infinite years, losing their memories of the previous facts would be arbitrarily valuable.

This implies that if a person is in constant misery all the time, their life could be very good, even if they never are happy. Suppose a person repeatedly gets tons of knowledge and friendship—tons of it every ten minutes. After ten minutes, their memories are erased and they repeat. They’re in constant extreme agony. On this account, their life would be great. Presumably, the benefit that a person can get over the course of their life from objective list goods is pretty significant—a person with a perfect memory would probably get lots of value from non-hedonic goods if objective list theory is true. But if a significant chunk of one’s lifetime value comes from objective list goods, then if all that value were condensed into ten minutes, that would be very valuable—more than enough to offset extreme agony. But a person who is constantly waterboarded, while getting all the knowledge on Wikipedia, would not be well off.

In addition, this view posits a weird asymmetry. Clearly, there is not a cap on the value that pleasure can give a person. Why would this be different for objective list goods.

For these reasons, I conclude that the cap view, while solving the lopsided lives problem without implying hypersensitivity, has enormous problems.

7 The momentary caps view


You might have the view that there is a limit on the amount of value one can get from objective list goods per moment. On this account, because you can only get up to, for example, 100 units of well-being per moment, if you’re at above 100 units of suffering, non-hedonic objective list goods can’t make you well off.

This view has two big problems.

It implies hypersensitivity. Suppose that one says that you can only get a fixed amount of well-being over the course of a second. Suppose someone repeatedly has a slight bit of knowledge at the beginning of the second and another jolt of knowledge either right before the end of the second or right after the end of the second, where those two possible events would only be separated by a plank instant. Suppose that after this, there’s a second of nothing else, and then it repeats. This goes on forever. On this account, their actions happening an instant before—such that it’s within the second rather than outside the second—is bad to an arbitrarily large degree. Thus, changing the time when people have experiences by a total of an arbitrarily small amount can be good to an arbitrarily large degree.

It implies that if someone is horrifically tortured for 100,000 years and then has a lot of seconds of having no joy but just lots of knowledge and friendship and such, their life could be very good.

Both of these are hard pills to swallow, especially the first. When an event happens doesn’t matter intrinsically, and certainly doesn’t make an arbitrarily small change to when events happen can’t be good to an arbitrarily large degree.

8 Richard’s solution


Here’s Richard’s summary of the solution:

I wonder whether a multiplicative approach might prove more promising. Here’s a super-rough toy model. Take value to come in two dimensions, ‘hedonic’ and ‘non-hedonic’. Map both dimensions onto a scale from 0 - 1, where 0 = maximum disvalue, 1/2 = neutral, and 1 = maximum value. Multiply both dimensions together to yield a composite score on a scale from 0 - 1 (area on the graph below), where neutrality now lies at 0.25 (so unit intervals are not assumed to be constant in value). Any composite score < 0.25 corresponds to negative value, whereas a composite score > 0.25 corresponds to positive value.


Problems with Richard’s solution:

It cannot work if the value of pleasure is unbounded. If there is no maximum to the value from pleasure—which there is not—then there is no model. If you extend hedonic value to infinity, you get hypersensitivity.

It cannot work if the value from non-hedonic goods is unbounded. Earlier in this article, I’ve argued that the value from non-hedonic goods is unbounded. If that is unbounded, then it implies that as long as one is slightly above the neutral point in terms of pleasure, as their non-hedonic value approaches infinity, total value approaches infinity. If this is true, then if a person gets tons of non-hedonic value and is very close to maximum misery, then their life can be arbitrarily good.

Most worryingly, it can’t work if hedonic disvalue is unbounded. Imagine the graph that goes below zero. This would imply hypersensitivity for very bad lives—where slight increases in non-hedonic value would be arbitrarily valuable.

I won’t rehash this here in any detail, but it runs head-on into the agony challenge that I raise in this article.

It implies that sometimes one life that has more pleasure and more of everything else on the objective list is worse than another life—because distribution matters. This is at least a bit counterintuitive.

It has the same problem as every other version of objective list theory does, raised in section 4.

There are two more specific versions of Richard’s view—one of which says that these scores are calculated over the course of a moment and the other of which says they’re calculated over the course of a lifetime. So the first view says that one calculates the value of one’s lifetime by looking at one’s pleasure score over the course of their lifetime and multiplying it by their objective list score, while the other says that one calculates the value of each moment by multiplying the pleasure in that moment by the objective list score. One of these views is lousy, and the other is even worse!!

Start with the lifetime view. This implies:

Something very anti-egalitarian and temporally odd. Suppose that we’re deciding whether to throw a birthday party for little Johnny. Not only does this view imply that the value of this will depend on how many friends Johnny will have when he’s 85—because those multiply his lifetime value of pleasure—it also implies that you should give pleasure to people who have higher objective list scores, rather than lower objective list scores. So you should help out people with lots of friends instead of people with fewer friends, even if all else is equal, precisely because they score higher on the objective list.

It requires one to think there’s a cap on the value that pleasure can give over the course of one’s life. That’s implausible for all the reasons discussed above (including here).

The momentary view has similar implausible implications. Specifically, it implies that there’s a maximum value one can get over the course of a moment. This again implies hypersensitivity.

Ultimately, Richard’s proposal is clever and it’s hard to get your head around at first. But when you work out the math, it’s just a mix of the cap-on-value view and the multiplicative view.

9 The asymptotic paired cap view


I think maybe the least bad view is the following. On this view, objective list goods by themselves aren’t valuable. However, when they’re paired with pleasure, they acquire intrinsic value proportional to the level of pleasure. As the level of pleasure one gets from some objective list source approaches infinity, the value of the objective list source approaches some finite amount.

What do I mean by objective list goods being paired with pleasure? An example of this would be when one appreciates their friends or spouse and has a good time. The pleasure derives from the objective list source and thus it gives value to the objective list source. A graph might look like the following:


The Y axis is the intrinsic value from the objective list source and the X axis is the amount of paired pleasure. This would be a function of the value of some constant amount of some objective list source—for example, of learning 100,000 civil war facts. It has to asymptotically approach a threshold to avoid the agony challenge.

The view will also hold that there is a cap on the value of some amount of knowledge. So if one has 100 units of pleasure and they derive that from some amount of objective list good sources, as the objective list goods approach infinity, the intrinsic value of that would approach a finite threshold.

I think this is the best way to go if you’re an objective list theorist. That said, it’s terribly unintuitive in the following ways.

Because it implies that there’s a cap on the value of pleasure one can get, it implies that utility monsters who can experience both knowledge, friendship, happiness, and so on 100,000,000 times more intensely than us humans should basically act as hedonists. Objective list theorists shouldn’t accept that.

It implies hyperinsensitivity, where sometimes increasing objective list goods by an arbitrarily large amount produces an arbitrarily small amount of extra value. Thus, sometimes you should take a bit of objective list goods over a huge amount of objective list goods later.

It implies, implausibly, that moving when one experiences some piece of knowledge or friendship to a time when they are experiencing more sublime joy from experiencing it to one where they experience less joy, could bring an improvement.  This is because the marginal effect of the objective list goods is lower when one has both more pleasure and objective list goods.  In fact, this seems to imply the same type of hypersensitivity as is implied by momentary caps—if we imagine over time one gains more and more pleasure that is appropriately tied to objective list goods, by shifting when some marginal quantity of objective list goods occur to an earlier moment with less pleasure, we could arbitrarily increase their welfare, even if this doesn’t affect the intensity of teh appreciation.  The hypersensitivity is odd, as is the conclusion that it’s sometimes better to have some objective list goods at a time when one experiences them less rather than more. If you could either derive 100 units of joy from appreciating your spouse, that would seem to better than deriving 100 units of joy from appreciating a mild acquaintance, no matter when those are experienced, but this view requires denying that.

This view is so weird and contrived, it just strikes me as adding epicycles to wriggle out of the plainly absurd conclusions of objective list theory. If you start having to have multiple thresholds depending on each other to meet common-sense, something has gone deeply wrong.

Because it implies distribution matters, it entails that one life can have more pleasure and more of every objective list good than another life, and yet the second can be better than the first.

It has all the problems discussed in section 4.

10 Conclusion


The lopsided lives challenge is the most pressing challenge for objective list theory. The basic version of the theory seems to imply something crazy. There are various proposals for adding epicycles to the theory to avoid this frightening conclusion. None of them are sufficient to rescue the theory. Only hedonism can accurately explain our verdicts about lopsided lives.



New Comment
8 comments, sorted by Click to highlight new comments since: Today at 9:30 PM

In other words, the only things that determine how well your life goes are your mental states—the ones that feel good are good for you, ones that feel bad are bad for you.

I predict from this that you would wirehead and urge wireheading for everyone else, given the possibility, and you would choose the Matrix over the real world. Is my prediction correct?

For some reason the words "flagrantly, confidently, and egregiously wrong" come to mind.

Good one!  

Though is there a reason? 

Let's say, fundamental differences in worldview.

I judge wireheading to be a step short of suicide, simulations to be no more than places that may be worth visiting on occasion, and most talk of "happiness" to be a category error.

And the more zeros in an argument, the less seriously I am inclined to take it.

I've no strong background in philosophy, but I am not convinced by your rebuttal of naïve objective list theory. As long as we concede that the same thing can be more or less good depending on context, I still think that the naïve vision has a point. Simple example: a billion dollars in cash usually translates to a lot of positive value, but not if you are stranded on a desert island with no way to spend the money.

Arguments of the form "super-torture trumps every positive thing" work only because we intuitively associate "extremly pain torture" with images of sadistic jailers preventing you from doing pretty much anything. Of course friendship is of no value if your jailer doesn't let you see your friends in the first place, and of course knowledge is of no value if you suffer so much that you can't think about anything else but your pain. I don't intuitively consider this as a flaw in the naïve theory, it kinda feels like putting the cart before the horse. A more accurate example than literal torture would be suffering from some terrible illness while still being able to enjoy things. Suppose that the Devil offers to make you the next Stephen Hawking, inflicting some horrible permanent disease upon you in exchange from wondrous scientific achievements and planetary fame. I predict that a nonzero number of people would gladly accept such a deal. By contrast, a deal such as "you'll get infinite knowledge but you won't ever be able to use it because of the constant pain" sounds very stupid from the start.

We're asking what's good for the person, not what deal they'd accept.  If we ask whether the person who is constantly tortured is well off, the answer is obviously no!  If naive OLT is true, then they would be well off.  It doesn't matter if they can ever use the knowledge. 

Is the naïve OLT so naïve that it always assign the same fixed amount of Value to the same bit of knowledge no matter what?

Anyway, I'm still not convinced that a person in constant pain should automatically be not well off. Who is better off, a world-famous scientist billionaire with a terrible illness causing constant pain, or a beggar without terrible illnesses living a miserable life in some third-world slum?

I've some trouble figuring out a similar scenario of well-off people involving literal torture, but that's because, as I said earlier, the very concept of "torture" involves jailers deliberately inflicting harm to segregated people. You say that OTL fails just because you can't imagine any realistic counterbalance to the torture itself. But since we are already in the realm of hypotheses, consider a fantasy setting where the demon-king routinely torture his generals, each of whom rules a whole realm anyway.