= 27d567bc2d48d9f40e9ccc20ea370b15)
gjm comments on Rationality Quotes Thread September 2015 - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (482)
Physics seems like a weirdly low-level thing if we're thinking about whether or not animals' behavior could remain the same and their subjective experience could be a state of what we would call suffering in some possible world. I just don't think that you could make edits on that level and have changes that are so fine-tuned. Editing physics doesn't leave everything alone except subjective experience; editing physics breaks fire.
And if we have to settle for a physics that leaves at least most everything the same and life remains possible, then why wouldn't we expect reward mechanisms to evolve and for the default state of affairs to be one in which adaptive experiences 'feel good' and others 'feel bad'?
gjm called this a false dichotomy, but I think a better though perhaps more complex way of putting it is that you're mixing up your multi-level maps. Take free will and determinism as an example. Some people become fatalists because they think that determinism contradicts their idea of what possibility means. But they're contaminating their maps. You have a low-level map of physics where everything is lawful and there is no thing similar to what you might call 'possibility.' Then you have a high-level map of decision-making, and the fatalists take the lawfulness from their low-level physics map and draw it onto their high-level decision-making map, and say, "Well, the lawfulness overrules the possibility", and then they start making null decisions, but that's wrong. Possibility is a primitive notion in your high-level decision-making map and only in that map, just like the laws of physics are primitive notions in your low-level physics map and only in that map. In the territory, your brain runs on physics and your decision-making algorithm runs on physics, and your decision-making algorithm computing the output of logical nodes making decisions other than the one that it must make makes you feel possibility, and the whole process is lawful, and you have a bridge map between these two levels of maps; but the map is not the territory, and that goes both ways, and your high-level decision-making map is not the territory either, and sticking physics into it is like mixing apples and oranges. And even if that seems counterintuitive, I would say that any other policy is wrong, because the fatalists lose and the compatabilists win.
I agree with you about compatabilism, but I don't think that answers my question.
I am not saying that you could have physical situation exactly like the real world except with subjective experience reversed. I agree that in order to reverse subjective experience, you would need to make physical changes.
But with those physical changes, why couldn't you have a situation where adaptive experiences feel bad and non-adaptive ones feel good?
Maybe this is impossible. But if it is impossible, maybe that just shows that God is necessarily good.
How are you defining "bad" and "good", in this world whose creatures are so different from us?
I suggest: "feel good" means "have subjective properties that tend to make such creatures seek experiences that have them" and "feel bad" means "have subjective properties that tend to make such creatures avoid experiences that have them".
In which case, a world in which adaptive experiences feel worse than non-adaptive ones is a world in which creatures systematically pursue maladaptive goals and avoid adaptive ones. Evolution will get rid of such creatures pretty quickly, no?
Suppose I'm right about evolution (see above); how would you get from there to "God is necessarily good"?
I basically agree with your definitions. However:
Subjective properties that tend to make a creature seek something can only do that because those subjective properties correspond to some objective physical property, at least if you aren't supporting dualism.
Why do the subjective properties that, in fact, tend to make people seek such experiences, correspond to objective properties that lead to that particular physical result (i.e. tending to physically move toward those results.)
You can say this is a tautology, and it is, in a way. But if we are simply talking about the particular subjective property in itself (as opposed to the fact that we happen to be pointing to it using our tendency to seek it), it is not a tautology, even if it may be a physical necessity.
In other words: zombies are probably physically impossible. But if they were possible, and if they were actual, instead of real humans, then reality would be neutral, neither good nor evil, since there would be no experience. In a similar way, negative-humans (who tend to seek painful experiences and avoid pleasant ones) are likely physically impossible. But if they were possible and actual, then the universe would be evil, and nearly all experiences would be bad.
But as it is, most experiences are good, and the universe is good. Talking about God here is probably a distraction, but if the universe is good, then its cause or causes should also be good. If the facts that make the universe good are necessary, then its causes are necessarily good.
I suggest that things coming out of evolutionary biology are more like logical than physical necessities; we would expect the same sort of dynamics to apply even with very different physical laws. (On the other hand, the "necessity" involved is weaker, because evolution is a stochastic process and it's very possible for suboptimal bits of design to get fixed in a population.)
So, anyway, the situation seems to be something like this: Any organisms with anything we could reasonably identify as "feeling good" or "feeling bad" experiences will tend to seek the ones that "feel good" in preference to the ones that "feel bad", almost tautologously. This means that, doing whatever they do, they will tend to "feel better" than we can imagine them doing if (e.g.) they absurdly sought out "feeling bad" experiences in preference to "feeling good" ones.
So far, so good. But to describe this by saying that "most experiences are good, and the universe is good" seems wrong to me, and I don't see how you get from "the facts that make the universe good are necessary" to "its causes are necessarily good". To expand on those points:
First: yes, living things will do nicer things in preference to nastier. But those are comparative terms. If everything in (say) human experience were made substantially more or substantially less pleasant uniformly it would have little impact on our decisions, but surely it could change whether "most experiences are good". My actual impression is that most experiences are neutral and the amounts of good and bad aren't terribly different. We have both orgasms and headaches; both enlightenment and boredom; both the satisfaction of a job well done and the frustration of failing to achieve what we should. I don't see any reason to think other animals' balance is very different.
Second: if "the universe is good" is meant to be just another way of saying "most experience is good" then I don't think passing from one to the other does anything other than invite confusion. If it's not, then I think you need some further justification for saying that the universe is good.
Third: let's suppose, e.g., that some simple argument in terms of evolution does lead to the conclusion that, largely independent of the details of physical law etc., most experiences of most living things will be experienced as good rather than bad. How do you get from there to "the causes of the universe are necessarily good"? I just don't see it. There certainly isn't any generally valid inference from "X has property P" to "the causes of X have property P"; consider e.g. the case where X is the physical universe and P is "is made of matter and energy" or "is made of quantum fields" or something of the kind; if the physical universe has causes at all then they might be mathematical laws, or gods, or happenings in some other universe with different laws; they certainly needn't be made of the same stuff as our universe.
I think I disagree with the first part (that we can't understand the idea of something doing things that feel bad rather than feel good, or that there are obvious reasons why this is impossible in principle), but I won't argue that for two reasons. First because that would probably end up being a very time consuming discussion, and second because, as I said, I think I can make my argument even if I grant your point.
Regarding your three points:
It sounds like you are saying that there is a balance of good and evil in human life which is basically equal, so that overall life is neutral. If this is the case, it would certainly make sense to say that the universe is neutral as well. But I disagree that the overall balance is neutral, and I think that most people disagree. Most people have a strong preference to live, rather than to die, even in a painless way. This implies that they think the value of the rest of their life is significantly higher than zero. This suggests that you are undervaluing positive experiences and overemphasizing negative experiences, compared to the values that most people place on those. And I doubt that in practice you personally think your life is worthless on balance.
When I say that the universe is good, I do not just mean that life or experience is overall good. If that needs additional justification, I will do that in addressing the third point.
I agree that a cause does not always have the properties of the effect. But goodness has a particular meaning which affects the issue. Without trying to give a formal definition, saying that something is good certainly means something like "this is a desirable state of affairs," and likewise saying that something is bad would mean something undesirable. But then if a cause brings about a good effect, it brings about a desirable state of affairs. And if the thing was desirable, bringing it about was desirable, and so there is reason to call the cause good as well, even though it still doesn't mean exactly the same thing (to call the thing good and to call its cause good.)
I don't think my first part says quite what you say it says, but never mind that since you've agreed not to contest it :-).
On good versus neutral versus bad, and wanting to live: wanting and liking are different things, and I think there are (alas) plenty of people who feel that their life contains more bad than good but still have no wish to die. (And, also alas, some who admit that their life contains more good than bad but do want to die, at least some of the time.) No, I don't think my life is worthless on balance. I think I have an unusually good life. [EDITED to add: And I expect a lot of other people here on LW have unusually good lives too. It's a group selected for high intelligence (almost necessary to find much LW material interesting), reasonable amounts of leisure time (else we'd be doing other things), not being overwhelmed with other concerns (else again we'd be doing other things), easy access to the internet (since that's where LW is), membership in a somewhat-dominant culture (because LW is anglophone and founded by people in the US), etc. None of these things is universal here, none of them is either necessary or sufficient for a good life, but they all tend to be characteristic of LW participants and they all tend to go along with a more pleasant life.]
On causes and effects: if by "good" you mean only "bringing about more desirable than undesirable effects, on balance" then I am, at least provisionally, prepared to agree that if an effect is good then that's evidence for the goodness of each of its causes. But it seems to me that this is a much weaker sense of "good" than the one usually intended by people who say that God is "good". Suppose I have a child (as in fact I do) and that I alternate between treating her kindly and beating her (as in fact I do not). If I do her more good than harm, overall, does this justify calling me "good"? Of course not.
What does your argument from goodness of experience really come down to? Only this, I think: "The universe has in it more good experiences than bad, perhaps just because of the logic of evolution. So whatever, if anything, caused our universe to exist may have the property of tending to produce more good experiences than bad. The god in whom I believe is supposed to have that property, so this imbalance between good and bad experiences is evidence for that god". I hope it's clear that if so it's very weak evidence.
I basically agree with what you said about wanting and liking. That's one reason why it's easy for me to imagine a situation where I constantly do things that feel awful and avoid the pleasant ones. Because to some extent that already happens. But life is not even close to being like that as a whole.
I also agree that people on LW are likely to have better lives on average than people in general, and also that at least some people would say that overall there is more bad than good in their lives, while still not wanting to commit suicide or anything like that. But I very much doubt that this is even close to a majority of people, even in very poor countries or in a historical sense. I realize I could be wrong about this and that (if so) someone could prove it with the proper statistics. But this is my current sense of the situation.
Regarding calling the causes good, that is one reason why I said it was a distraction (relative to this argument) to talk about God. Because normally when people say "God," they think of a person, and in fact a particular person, with the result that instead of talking about some particular point, people are actually discussing whether or not the doctrines of someone's religion are true or false as a whole.
In any case, as you point out, saying that someone is a "good person," has a particular meaning, certainly more than saying that overall the person causes more good than bad. But in that way there is no contradiction in someone being a good thing but a bad person, because "good thing" and "good person" mean two different things.
I agree with your summary of the argument, at least as far as it goes.
Well, if you detach your argumentation about good versus bad experience from talk of "God", then all you're left with is: People have more good than bad experiences on the whole, so if the universe has causes then they are more likely to be such as to generally produce (ultimately) more good than bad experiences, than to be such as to generally produce (ultimately) more bad than good experiences.
Which is fair enough, but I'm sure originally this was being proposed as evidence for belief in a good god, and it really doesn't seem to me to offer more than a tiny amount (and, further, in so far as it does the highly mixed character of human experience seems to provide at least as much evidence against the perfectly good gods of various religions).
Actually, that wasn't how I meant it even originally (even if in fact it is weak evidence for that). I suppose it wasn't unreasonable for you to understand it that way because I used the word "benevolence," and in reply to a comment about whether God is good or not, but in fact I was only objecting to the implied claim that the laws of physics (and therefore whatever causes them) are overall absolutely neutral, and I used the word benevolence because you used it.
I agree with you that there is good evidence against the idea of God as a perfectly good person, in a literal sense, especially if you understand that in the sense that God is someone who is supposed to look at every particular thing which is happening and decide whether or not it's a good idea to allow it to happen or not. It seems very likely that no such thing is happening.
Actually, you're affirming the consequent. If the cause of the universe is good, then the universe will be good. But if we observe that the universe is good, then we cannot infer that the cause of the universe is necessarily good.
Classic example:
Affirming the consequent is valid in the terms of the evidence relation. In other words, the fact that I have a sore throat is evidence that I have the flu, although not proof.
In the same way, the universe being good is evidence that the cause is good. If up to that point, we assumed that the cause was entirely neutral, our best estimate will now be that the cause is good.
That's true, but there are other equally explanatory hypotheses that don't involve good causes, so at the very least, the fact that the universe is good (although I would contest or at least qualify that claim) is not evidence either way. It doesn't help you figure out which possible world you're in.
To be clear, I didn't think you were proposing any sort of p-zombie-like hypothesis, where mental states are epiphenomenal or otherwise.
I think this is the same sort of error that a lot of people make when they ask "Why am I who I am instead of someone else?" They think that their identity exists primitively in the territory or even that it existed before their body; but it's a wrong question because they have the causality reversed. The question would make sense if you were some ghost-in-the-machine and you or someone else picked some physical body, but that's not how it works. Your body caused your mind, so you always are who you are and the question is confused. (I also don't mean to exclude the possibility of anthropic reasoning in this example.) The good and the bad don't exist primitively in the territory, they are caused by evolutionary processes that develop organisms with reward mechanisms, and we identify reward and punishment, among other things, with these high-level concepts of 'good' and 'bad' that didn't exist before us, so good is always good and bad is always bad and the question is confused.
Although with what I've seen in neuroscience, some creature like what you describe doesn't necessarily seem outside of the realm of physical possibility to me. If people can simultaneously observe that they are paralyzed and come up with endless excuses for why they aren't, and the cognitive processes we're talking about in this case aren't too intertwined, then I can perhaps conceive of a creature that continues to perform adaptive behaviors but experiences suffering and happiness in the reverse. But I would expect someone to have to construct it, not for it to evolve. Or maybe an evolved creature with the most horrible sort of oddly complex neurological lesion.