I think this is basically right (I don't think the upshot is that incomparability implies nihilism, but rather the moral irrelevance of most choices). I don't really understand why this is a reason to reject incomparability. If values are incomparable, it turns out that the moral implications are quite different from what we thought. Why change your values rather than your downstream beliefs about morally appropriate action?
I think it's a bad result of a view if it implies that no actions we perform are good or bad. Intuitively it doesn't seem like all chaotic actions are neutral.
Hi Matthew.
Our actions have lots of unpredictable effects. If you drive to the store, you will delay everyone behind you in traffic. This will change when they next have sex, thus completely changing the identity of their future child. A different sperm and egg will fuse. This new child will go on to take a staggeringly large number of actions, each of which will change the identity of still more people. For this reason, because of your decision to drive to the store, the world hundreds of years down the line will be completely different.
I agree small actions like driving to the store may have large actual consequences. However, I believe their expected consequences are very small. I think the probability of any given child being born will be practically the same regardless of whether one drives to the store or not. One could tell a story where driving to the store leads to A being born instead of B. However, one could tell a story practically as convincing where driving to the store leads to B being born instead of A. So one should practically stick to the prior that driving to the store does change the probability of A or B being born. Likewise, driving to the store could cause a given hurricane H, but is almost as likely to prevent it. So the probability of hurricane H is practically the same regardless of whether one drives to the store or not.
@Derek Shiller, I would be curious to know your thoughts on the above.
From your perspective now, given your uncertainty, you might have no reason to think that the drive changes things in a better or worse way, and so the expected value derived from taking the expected value of each possible consequence considered piecemeal and summed up might be neutral.
I think if you accept some degree of objective chance, this might completely resolve the problem. (I explore this idea, without direct connection to incomparability, here: https://philpapers.org/rec/SHICAT-11)
It is consistent with a purely subjective neutral expected value, however, that you might be quite sure that the results, if revealed in their full detail, would be incomparable.
This creates an intuitive tension. There is something called the "Principle of Reflection" -- it is a bit controversial, but very intuitive -- that says roughly that if you know you would believe something if you were put into a better epistemic position, you should believe it now. The challenge Bulldog raises relies on taking reflection seriously and thinking that we would almost surely see incomparability if we looked closely at all the details, so we should accept incomparability now.
Thanks for sharing your thoughts, Derek. I did not know about your paper.
One would ideally recognise the existing uncertainty, and become less uncertain with further research. However, I think it is very easy to underestimate uncertainty. So I can easily see further research making one more uncertain. Yet, this would only lead to incomparability if one is open to impresice probabilities (I am not)?
Executive summary: The author argues that if “incomparability” truly exists—meaning some goods or outcomes can’t be ranked as better, worse, or equal—then, combined with the vast ripple effects of our actions on future people, all actions become neither good nor bad, leading to moral nihilism; therefore, we should reject incomparability.
Key points:
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.
Crosspost of my blog.
Our actions have lots of unpredictable effects. If you drive to the store, you will delay everyone behind you in traffic. This will change when they next have sex, thus completely changing the identity of their future child. A different sperm and egg will fuse. This new child will go on to take a staggeringly large number of actions, each of which will change the identity of still more people. For this reason, because of your decision to drive to the store, the world hundreds of years down the line will be completely different.
It’s hard to overstate just how different. As a result of most of a 16th-century peasant’s random innocuous decisions, the world is stocked with completely different people from those who would have otherwise stocked it. Individual trips to the latrine are responsible for Hitler and Mao, as well as for staving off the monsters who’d have existed otherwise. The development of Democracy and humanity surviving the cold war was counterfactually dependent on random people’s decisions to have sex thousands of years ago.
This is pretty weird. It means that there are no unimportant actions. Everything we do has huge and unforeseen consequences (I’m going to call this the ripple effect thesis as a shorthand since just repeating it a bunch of times is cumbersome). This fact also poses serious challenges for various philosophical theories. Both deontology and certain kinds of risk aversion run into strong objections that depend on this thesis. Here, continuing the tradition of showing how the ripple effect thesis ruins theories, I’m going to show how it ruins incomparability.
I’m also probably going to write a paper on this, so don’t steal it or I will come to your house and epistemically wrong you—quite brutally, showing no restraint. You will start crying, saying, “oh no, I’ve been so epistemically wronged, how will I ever recover?” And I will laugh as I epistemically wrong the shit of you, to within an inch of your life.
Incomparability, for those who don’t know, is the idea that some states of affairs aren’t precisely comparable. If two states are incomparable then neither is better than the other, but they’re also not equal. You might value your kids incomparably—it’s not that you value them exactly equally, so that you’d save one over the other if given a dollar. Rather, it’s that they can’t be meaningfully compared.
The hallmark of the incomparable is that two incomparable goods stay incomparable even if one gets slightly better. This is different from goods that are equal. A ten dollar bill is equal in value to two five dollar bills. If you offer me a small payment to swap a ten dollar bill for two five dollar bills, then I’d take it. Incomparability doesn’t work this way. If two careers are incomparable, then even if you made one slightly better, they’d stay incomparable. If you’re uncertain about whether to be a doctor or lawyer, it’s not that you find them to be exactly equally good careers, but instead that they just can’t be precisely compared.
You might be seeing the faint outline of the picture of how incomparability combined with the ripple effect thesis makes none of our actions matter. Let’s start with a simple case.
Jon and Fred. Currently a machine is going to create Jon. However, if you drive to the store (which would benefit you a small amount), it will create Fred. Jon and Fred are incomparable—wildly different kinds of people, neither better nor worse than the other. Should you drive to the store?
If incomparability exists, then driving to the store is neither good nor bad. Remember the hallmark of the incomparable: goods A and B are incomparable if neither is better than the other, and neither would continue to be better if either one got slightly better. Jon and Fred are incomparable. Driving to the store is an improvement to the creation of Fred. Adding that to the Fred side of the ledger just is slightly improving the creation of Fred. But by definition, that’s not enough to break incomparability.
Okay, now let’s change the scenario slightly.
Two Groups: Currently a machine is going to create one group of people (group A). If you drive to the store, which would benefit you, then a different group of people would be created (group B). The groups will have the same expected number of people. Many of the people in group A are better than many of the people in group B. Many in group B are better than many of the people in group A. Many others are incomparable. Groups A and B will have very different effects on the world but these effects are equal in expectation. Should you drive to the store?
It seems like, once again, driving to the store is neutral. This scenario is like the last one, except we’ve just added more incomparability. We’ve also added non-incomparable goods that are neutral in expectation. But adding something that’s expectedly neutral and more incomparability doesn’t make the situation better. In fact, it makes it worse!
For it to be good to replace good A with good B, assuming they’re incomparable, you’ll need to add something really good on good A’s side of the ledger. If you’re currently planning on creating Jon, for it to be better to create Fred, assuming they’re incomparable, there must be some other very large associated good.
For this reason, to justify replacing a very large number of goods with other incomparable goods, you’ll need some very large benefit to accrue from this replacement. Crucially, the size of the needed benefit will be proportional to the number of goods you’re replacing. If we assume that each replacement needs 50 goodness points worth of benefit to be worth doing, then you’ll need 50 million goodness points for it to be good to swap out a million incomparable goods with a million other incomparable goods. Otherwise it would just be neutral. The resultant state of affairs would be incomparable with the original state of affairs, neither better nor worse.
So what have we learned? Well, we’ve learned two things:
Here is the bad news: every time you drive to the store, you do all the stuff mentioned in 2. So if 2 is correct, then driving to the store is never good nor bad. It’s always incomparable with staying home.
When you drive to the store, because of ripple effects, you change the identities of staggeringly large numbers of people. In total, you change the identities of many billions of people, if not more. In many cases, you change people out for others they are incomparable with. And by changing the world, many goods are replaced with other incomparable goods. You also affect large quantities of non-incomparable goods, but do so in a way that’s neutral in expectation, so that’s irrelevant to decision-making.
Thus, if incomparability exists, driving to the store at any time rather than another is always neither good nor bad. Small actions are never good or bad. The drops are too small to perturb the astronomical incomparable wave. Your actions aren’t neutral exactly—it’s not that they keep the world’s value the same. Instead, the resultant value is incomparable with the initial value. Things are neither better nor worse nor the same.
This is a bad enough result. It seems obvious that some actions we take are good. It’s a good thing to take a walk if doing so makes you happier. It’s a good thing to drive to the store, if doing so allows you to get some valuable goods. Incomparability implies that this isn’t good or bad. That’s a bit hard to believe.
But things get worse. It isn’t just that small actions are neither good nor bad. Big ones are too. Consider an action like saving a life. That seems like a good thing! But if there are incomparable goods, then it’s not. After all, by saving a life, you affect billions of incomparable goods. Merely saving a life isn’t enough to outweigh the huge ocean of incomparability that you bring about. If action A brings about a billion happy people, and action B brings about a billion incomparable happy people and saves a life, then the two actions are incomparable.
As we’ve said, in order for it to be good to swap out one set of goods with another incomparable set, there will have to be some associated good proportional to the size of the set. Because the set of incomparable goods that you affect constantly is so large, the needed premium would have to be much larger than the expected effects of any of our actions. In fact, as the future gets larger and larger, you perturb more incomparable goods. So for any action, no matter how good, if the future is large enough, it would be neither good nor bad. Incomparability eats up all.
There’s one objection you might be thinking of. Perhaps, when you normally act, because you don’t know which incomparable goods you are affecting, you get to ignore them. Maybe if an action produces a small benefit and affects the distribution of lots of incomparable goods, assuming its effect on the distribution of incomparable goods is neutral in expectation, the action is good overall. So then, because you have no reason to think you’re worsening the world’s incomparable goods, you can go about your normal business without being mugged by incomparability.
Sadly, I don’t think this really works.
A first problem is: this doesn’t really seem to be how incomparability works. If you knew that by pressing a button, there’d be a huge switch in the distribution of incomparable goods, and a child would get a lollipop, seems like the action is neither good nor bad. After all, the lollipop shouldn’t be the tiebreaker.
A second, bigger problem: this view implies that every time you take an action, if you were told about one incomparable good it affected, then you should regard the action as neither good nor bad. But this is pretty weird. It shouldn’t be that you regard actions as good or bad even though you know that if you knew a bit more you’d regard them as neither good nor bad.
Imagine, for instance, that you can walk to the store at 2 pm. Doing so would benefit you. Currently you are planning on doing it. It’s 1 pm right now. However, at 1:30 you’ll be told one of the incomparable goods that your action would affect (e.g. Steve the artist wouldn’t exist and Mary the neuroscientist would exist in his place). On this view, even though you now endorse the act, you know with certainty that you won’t endorse it any longer at 1:30. That’s pretty weird. You shouldn’t regard acts as good if you know, with certainty, that if you knew more you’d consider them neutral.
Third, the ocean of incomparability is large enough that all our acts end up being neither good nor bad if incomparable goods exist. Thus, even if somehow this view can hold on to the notion that normal actions are good in expectation, it faces the troubling result that none of them end up being good. This is a bad result on its own, and it’s pretty weird to consider an act good in expectation if you know it isn’t good!
Fourth, this idea depends on a distinction between incomparable goods you know about and ones you don’t know about. If you knew driving to the store would replace Bob with Fred, then you shouldn’t do it, but if you don’t then you should. But knowing isn’t a precise category and comes in degrees. Is it enough to know the first letter of Bob’s name? This view requires locating some terribly arbitrary line (see Hare’s paper for more detail on this challenge applied to a slightly different view).
Thus, this way out of the problem is riddled with hideous structural problems. I think it’s unsalvageable.
Now, you might be thinking to yourself: “this all sounds very well if I’m a consequentialist, but I’m not. I’m a deontologist. So can’t I just ignore this and go on my merry way believing incomparability.”
No.
First of all, every ethical view that makes any sense says that consequences are among the things that matter. It’s good to save lives. It’s good to make the world better. As Rawls says, a view that rejected that would be crazy. And yet incomparability seems to require we reject that. Nothing we do makes the world better if there are incomparable goods.
Second, the argument doesn’t establish that all our actions have neutral consequences. It establishes that all our actions have incomparable consequences. Neutrality can be easily outweighed. A giant ocean of incomparability can’t be.
Take an example of an action that you take for non-consequentialist reasons. Maybe you call your mother because you love her very much and owe it to her. Well, lovely as your mother no-doubt is, here is an action that would be much better: saving ten people, while having the same ripple-effects as calling your mother (by this I mean that the effects on the identity of future people would be the same across the two cases). But the argument establishes that saving ten people wouldn’t be good nor bad. Calling your mother is strictly worse than this. So incomparability doesn’t just infect consequentialist reasons. It infects all the reasons. If you believe in incomparability, you have no reason to call your mother!
I think this is a strong argument against incomparability. But let me be clear on the kinds of incomparability it doesn’t threaten. You can still believe in incomparability, without thinking it implies nihilism, if the incomparable goods aren’t affected by our actions. For instance, it could be that infinite sets of well-off people are incomparable. This wouldn’t imply nihilism because your actions don’t affect the creation of infinite sets of well-off people.1 Similarly, you could think that very different kinds of worlds that God might create would be incomparable.
But if you think that any goods that humans affect are incomparable, then you get this problem. So this rules out thinking:
There are other good reasons to reject incomparability. But this seems like the most severe problem. If there are incomparable goods, as many people suggest, then nothing we do could possibly matter in a world where our actions have unpredictable consequences. That is a very implausible result, and so we should give up belief in the incomparable.
I'm not well-learned enough on this, but it seems that you would appreciate this very cool recent post which (badly summarized) kind of explains how one can still have forms of moral action-guidance even if some sound moral theories (in your post, a form of impartial consequentialism?) imply that "none of our actions matter" : Resolving radical cluelessness with metanormative bracketing.
Thanks Jo! Yeah, the perspective I defend in that post in a nutshell is:
(Similarly, the decision theory of "bracketing" might also resolve incomparability within consequentialism, but see here for some challenges.)
Re: the first link, what do you think of Dynamic Strong Maximality, which avoids money pumps while allowing for incomparability?