Three Problems For Appreciation-Based Objective List Theory and Other Hybrid Views

Technically, these are universally applicable, but they are not worrisome for hedonists or desire theorists

Introduction

Why ya so afraid of pleasure? Do you hope to die on a hill?
You’ve been keeping up a ledger that you wrote while standing still

—The collection

My articles are notorious for stirring up the passions of the masses. After my articles come out, mass protests movements take to the streets shouting “no more troubling parochialism in our morality on account of the two intensions of moral terms differing.” My essays have been frequently analogized to Pericles’s speeches. Well, enough rabble-rousing—for a change of things, let’s talk about some narrow, technical objections to objective list theory. People who read my blog a lot will know that I very often provide objections to objective list theory. This is a sign of respect—I think desire theory is basically totally dead, so objective list theory is the only remotely plausible nonhedonist option. Unfortunately, even objective list theory has very severe problems that do not allow it to survive critical scrutiny. I think that these three arguments are the main reason, other than the fact that they’re clearly gerrymandered and consequently implausible, that the gerrymandered solutions I lay out at the end of my piece on lopsided lives are implausible. If you haven’t read that piece and you find this one confusing, probably check that one out!

1 EPTLW plus accessories

In his original article, about lopsided lives, Theron Pummer defends the following very intuitive 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, if you’re in intense agony each moment, your life can’t be made arbitrarily good by just adding other things like friends, knowledge, and achievements that bring you no happiness. Seems true! But Pummer shows that this principle leads to troubles. However, Pummer just shows how specific ways that we could accommodate the principle open up problems of their own. But showing that lots of specific ways to accommodate some principles incur problems doesn’t mean all ways do.

Fortunately, we can show that one cannot accept EPTLW and two other plausible principles. The following two principles are both intuitively very plausible:

1. Non relative hyperinsensitivity: If some action gives you 100,000,000 times more of an objective list good with 1,000,000,000 times the probability of another act, and has no other effects, it is better than the action that gives you 1/100,000,000th the amount of objective list goods with 1/100,000,000th the probability.

2. Objective good non-inferiority: Objective list goods are not lexically inferior to preventing extreme agony.

However, these three principles are inconsistent.

Good A is lexically inferior to good B if no amount of good A can ever be more valuable than some amount of good B. So one might think that ice cream is lexically inferior to preventing death, such that no amount of ice cream is as valuable as saving a life (this is false, but that’s a story for another day).

You might think that the objective list theorist can accept 2. However, if they accept 2, then they have to think that averting extreme agony is more important than any amount of objective list goods. This would mean that the harm of one person being tortured is greater than the non-instrumental harm would be of wiping out all knowledge in the world.

Given that objective list theorists generally hold that pleasure and objective list goods are comparable—otherwise they must think that a bit of pleasure is more valuable than any amount of objective list goods—this would also entail that life on Earth would be a great tragedy, even if only one being experienced extreme agony and everyone else was as well off as people can be. This is hard to believe.

Why must one give up one of the three principles? Consider the following argument:

1. If objective list goods are not lexically inferior to preventing extreme agony, then there are some circumstances where objective list goods can counterbalance extreme agony.

2. If EPTLW is true, then objective list goods are, in some circumstances, unable to counterbalance extreme agony.

3. If it is sometimes the case that an infinite amount of A can’t offset the harm of B, and other times a finite amount of A can offset the harm of B, then in some circumstances, A is infinitely more valuable than it is in other circumstances.

4. If something is infinitely more valuable than something else, then if something gives you 100,000,000 times more of the infinitely less valuable thing with 100,000,000 times the probability, versus another thing that gives you 100,000,000 times less of the infinitely more valuable thing with 100,000,000 times less probability, the second thing is better than the first.

5. Therefore, if some action gives you 100,000,000 times more of an objective list good with 1,000,000,000 times the probability of another act, and has no other effects, it is sometimes worse than the action that gives you 1/100,000,000th the amount of objective list goods with 1/100,000,000th the probability.

This shows that if EPTLW and the second principle are both true then the first must be false.

You might reject 4. This seems very implausible to me, especially because it runs into the challenges raised here. But if you reject 4, you should still accept

6. If something is infinitely more valuable than something else, then if something gives you 100,000,000 times more of the infinitely less valuable thing, versus another thing that gives you 100,000,000 times less of the infinitely more valuable thing, the second thing is better than the first.

This would entail that EPTLW and Non relative hyperinsensitivity are true, then so too is

7. Non relative hyperinsensitivity: If some action gives you 100,000,000 times more of an objective list good, and has no other effects, it is better than the action that gives you 1/100,000,000th the amount of objective list goods.

Thus, not only do objective list theorists have to think that infinite objective list goods are less valuable than a bit of pleasure sometimes (see section 4), they also have to think sometimes they’re less valuable than a bit of objective list goods. This is hard to believe. I think the least costly way to go is just to accept Non relative hyperinsensitivity, but that’s quite an unintuitive consequence.

2 Hypersensitivity plus bizarre insensitivity

Suppose you get an arbitrarily large amount of pleasure from an objective list good. Does the objective list good increase the value of each unit of pleasure by an amount that doesn’t approach zero, relative to a world where there is the same internal state of appreciation but minus the objective list good? For example, if some utility monster gets 100,000,000 units of joy from appreciating their children’s achievements, is the 100,000,000th unit of joy appreciably better on account of their children actually existing, relative to a world, for example, where they’re in the experience machine?

Suppose you say that the objective list good does increase the value of each unit of pleasure. Well then, you get hypersensitivity, wherein a very small change in some good makes an arbitrarily large difference to well-being, without having any other effects. To see this, note that there are a continuum of cases between there not being any child and there being the child that is appreciated. In the middle there might be a child who exists but is missing many of the properties of the child that is appreciated.

Suppose it doesn’t. Well, for appreciation to have intrinsic value, it must be that for the first some number of units of appreciation, their value is enhanced by a nonzero amount. But then if appreciation’s value approaches zero this implies a puzzling form of relative insensitivity—it is sometimes better to get 1 unit of happiness from appreciating a shitty thing that’s only barely on the objective list and 9,999 units from no source than to get an additional 10,000 units of happiness from appreciating something very good.

Why is this? Well, on this account, objective list goods cap out in the sense that as you appreciate them more, the marginal value of the nonhedonic element of the good approaches zero. Therefore, if one appreciates their child a lot, for instance, the nonhedonic value of increasing their appreciation of their child by 100,000 units of pleasure would produce a very small amount of nonhedonic value—say .1 units of well-being. But in contrast, if one appreciates the shittiest thing on the objective list just a little bit, that would bring them more than .1 units of well-being. Therefore, in this scenario, if we compare

1. 9,999 units of pleasure from no source + 1 unit of pleasure from appreciating the worst thing on the objective list, there would be 10,000 units of pleasure plus .2 objective list units of well-being. In contrast;

2. 10,000 additional units of pleasure from appreciating one’s child’s achievements, for instance, or what a great person they’ve become, you’d get 10,000 units of pleasure plus .1 objective list units of well-being.

Therefore, on this account, the first would be better than the second. But that’s wildly counterintuitive, on objective list theory.

Thus, one either has to accept hypersensitivity or that it’s sometimes better to get 9,999 units of pleasure from eating tasty food or masturbation, for instance, and 1 from appreciating something barely good than an additional 10,000 units of pleasure from appreciating something great (ignoring any effects of declining marginal utility of pleasure. If you want to have intuitions that ignore them, just assume we’re adding an additional 9,999 units of pleasure from masturbation or eating food, when there’s already lots of pleasure from those things. Assume that pleasure equals the amount of pleasure the person is already getting from appreciating the great thing).

3 The vagueness problem

I think value picks up on something objective in the universe. Pleasure is objectively better than pain. It was bad when the dinosaurs had their throats ripped out in agony, even if no human had ever had any attitude about them.

Value is also not vague. There is an amount of value that things bring people. This can be supported through many arguments—money pump arguments and considerations about gambles and more and even more than that. If value is an objective part of the world then it plausibly can’t be vague because what actually exists can’t be vague. What would it mean for something to be vague—to neither fully possess nor not possess a certain degree of a property? Vagueness is purely linguistic or perhaps conceptual, but certainly not metaphysical. You can’t have something with an indeterminate heat, for instance.

But this poses a problem for objective list theory—it sure seems that many of the things on the objective list are vague. Take friendship—is it really plausible that there’s an objective formula for determining how good a friend is? Or how significant an achievement is? It seems clear intuitively that the answer is no.

This can be seen more clearly intuitively by considering some theories of mereology. If you look at the three most popular views of mereology, they are:

1. Mereological restrictionism, according to which only some collections constitute an object (for example, there is such a thing as my laptop but not as the laptop Eiffel tower—a single object containing all the atoms in my laptop and also all the atoms in the Eiffel tower). But if you look at the more specific versions of restrictivism, they depend on linguistic facts, for instance, or facts about which atoms are joined together. None of these allow objective facts about friendship.

2. Mereological universalism, according to which all collections constitute an object. On this account, there is a single object composed of my hand, your mother, and 7 drops of rain in Spain that fall mainly in the plain. On this account, there are not privileged facts about friendship either, if friendship is a composite object. None of the compositions are privileged according to universalism.

3. Mereological nihilism according to which no collections constitute objects. Thus, only the most fundamental physical stuff—and perhaps also mental stuff—really exists. On this account, it seems like friends wouldn’t even exist, much less possess objective facts about them.

But if there are no precise, objective facts about relationships or friendship or knowledge or achievements, then it seems like OLT—and any view that implies that the facts aren’t just experiential—imply vagueness about value. But value isn’t vague, so those views are false.

Contrast this with hedonic value which doesn’t seem vague. Experiences seem precise. There is a fact about how intense a pleasure is, as is broadly true of consciousness facts—consciousness is not vague.

I think this core vagueness poses a series of more specific problems:

1. There’s the basic version of the vagueness problem discussed above. If there are no precise facts about any of the objective list goods, then value is vague. But it ain’t.

2. There’s the problem of arbitrariness. To avoid vagueness, you might just have stipulations about which things count as a friendship, for instance. But these things will be objectionably arbitrary—it doesn’t seem like there’s an objective fact about whether a person who bakes you cake or takes you bowling is a better friend, for instance.

3. There’s the problem of parsimony, as I’ve pointed out before. In order to specify what counts as friendship or achievements or relationships, one would have to have an absurdly long, complex, and specific formula. But that’s much too complex to be built into the fundamentals of morality, just as it would be a decisive objection to a theory of physics if it had to include some complex measure of friendship in its fundamental laws, that ended up being more complex than the other physical laws.

4. There’s another arbitrariness problem at the level of adding up goods. It seems there’s a very natural way to add up the value of pleasure. Plausibly, pleasure is defined as good mental states—as I’ve argued elsewhere. If so, then there’s a very natural way to determine the goodness of pleasure—you just add up the goodness of the mental states. In contrast, with other objective list goods, they don’t intrinsically contain goodness in them, so there must be some arbitrary seeming formula that converts friendship or achievements into goodness. For example, you’d measure the value of friendship and multiply it by 1.8 before adding it to the well-being score. But this is arbitrary. Why 1.8 times? Why not 1.7? If we’re modal rationalists—which we should be—then this will seem especially incredible—is it really plausible that an ideal being could deduce this formula a priori. Random arbitrary contingency and unexplainable values occurs in physical laws, but its wildly implausible that it occurs in noncontingent things. This strikes me as equally odd to claiming that it’s necessary that the universe has 6.02214076 × 10^23 atoms. Necessary things should not have brute, inexplicable, contingencies. But morality is clearly necessary.

Note that these concerns don’t apply to hedonism. The amount of pleasure one experiences is not vague and is consequently nonarbitrary, it’s very parsimonious—you just add up the amount of pleasure to measure the goodness of a state of affairs, which also addresses the fourth problem. In fact, I don’t think these apply to desire theory—just to OLT.

These are somewhat technical objections, invoking theoretical virtues, for instance. Here, let me just try to articulate the fundamental intuition here (while standing on one foot):

The world has lots of precise physical facts at the lowest level. Particles follow precise equations. Then, at the higher levels, these lower-level phenomena lead to various imprecise facts—where we arbitrarily order certain arrangements of fundamental stuff. There are also fundamental consciousness facts. Objective list theory requires there be precise, objective facts about the higher level phenomena, which there aren’t, so it’s false. This manifests itself in more specific ways, but that is the basic problem.