Why do you reject negative utilitarianism?

I find negative utilitarianism unappealing for roughly the same reason I'd find "we should only care about disgust" or "we should only care about the taste of bananas" unappealing. Or if you think suffering is much closer to a natural kind than disgust, then supply some other mental (or physical!) state that seems more natural-kind-ish to you.

"Only suffering ultimately matters" and "only the taste of bananas ultimately matters" share the virtue of simplicity, but they otherwise run into the same difficulty, which is just that they don't exhaustively describe all the things I enjoy or want or prefer. I don't think my rejection of bananatarianism has to be any more complicated than that.

Something I wrote last year in response to a tangentially related paper:

I personally care about things other than suffering. What are negative utilitarians saying about that?

Are they saying that they don't care about things like friendship, good food, joy, catharsis, adventure, learning new things, falling in love, etc., except as mechanisms for avoiding suffering? Are they saying that I'm deluded about having preferences like those? Are they saying that I should try to change my preferences — and if so, why? Are they saying that my preferences are fine in my personal decision-making as an individual, but shouldn't get any weight in an idealized negotiation about what humanity as a group should do (ignoring any weight my preferences get from non-NU views that might in fact warrant a place at the bargaining table for more foundational or practical reasons distinct from the NU ideal) — and if so, why?

[...] "It's wrong to ever base any decision whatsoever on my (or anyone else's) enjoyment of anything whatsoever in life, except insofar as that enjoyment has downstream effects on other things" is an incredibly, amazingly strong claim. And it's important in this context that you're actually making that incredibly strong claim: more mild "negative-leaning" utilitarianisms (which probably shouldn't be associated with NU, given how stark the difference is) don't have to deal with the version of the world destruction argument I think x-risk people tend to be concerned about, which is not 'in some scenarios, careful weighing of the costs and benefits can justify killing lots of people' but rather 'any offsets or alternatives to building misaligned resource-hungry AGI (without suffering subsystems) get literally zero weight, if you're sufficiently confident that that's what you're building; there's no need to even consider them; they aren't even a feather on the scale'. I just don't see why the not-even-a-feather-on-the-scale view deserves any more attention or respect than, e.g., divine-command theory; in an argument between the "negative-leaning" utilitarian and the real negative utilitarian, I don't think the NU gets any good hits in.

(Simplicity is a virtue, but not when it's of the "I'm going to attempt to disregard every consideration in all of my actions going forward except the expected amount of deliciousness in the future" or "... except the expected amount of lying in the future" variety; so simplicity on its own doesn't raise the view to the level of having non-negligible probability compared to negative-learning U.)

I used to consider myself NU, but have since then rejected it.

Part of my rejection was that, on a psychological level, it simply didn't work for me. The notion that everything only has value to the extent that it reduces suffering meant that most of the things which I cared about, were pointless and meaningless except for their instrumental value in reducing my suffering or making me more effective at reducing suffering. Doing things which I enjoyed, but constantly having a nagging sensation of "if I could just learn to no longer need this, then it would be better for everyone" basically meant that it was very hard to ever enjoy anything. It was basically setting my mind up to be a battlefield, dominated by an NU faction trying to suppress any desires which did not directly contribute to reducing suffering, and opposed by an anti-NU faction which couldn't do much but could at least prevent me from getting any effective NU work done, either.

Eventually it became obvious that even from an NU perspective, it would be better for me to stop endorsing NU, since that way I might end up actually accomplishing more suffering reduction than if I continued to endorse NU. And I think that this decision was basically correct.

A related reason is that I also rejected the need for a unified theory of value. I still think that if you wanted to reduce human values into a unified framework, then something like NU would be one of the simplest and least paradoxical answers. But eventually I concluded that any simple unified theory of value is likely to be wrong, and also not particularly useful for guiding practical decision-making. I've written more about this here [LW · GW].

Finally, and as a more recent development, I notice that NU neglects to take into account non-suffering-based preferences. My current model of minds and suffering is that minds are composed of many different subagents [? · GW] with differing goals; suffering is the result of the result of different subagents being in conflict (e.g. if one subagent wants to push through a particular global belief update, which another subagent does not wish to accept).

This means that I could imagine an advanced version of myself who had gotten rid of all personal suffering, but was still motivated by pursue other goals. Suppose for the sake of argument that I only had subagents which cared about 1) seeing friends 2) making art. Now if my subagents reached agreement of spending 30% of their time making art and 70% of their time seeing friends, then this could in principle eliminate my suffering by removing subagent conflict, but it would still be driving me to do things for reasons other than reducing suffering. Thus the argument that suffering is the only source of value fails; the version of me which had eliminated all personal suffering might be more driven to do things than the current one! (since subagent conflict was no longer blocking action in any situation)

As a practical matter, I still think that reducing suffering is one of the most urgent EA priorities: as long as death and extreme suffering exist in the world, anything that would be called "altruism" should focus its efforts on reducing that. But this is a form of prioritarianism, not NU. I do not endorse NU's prescription that an entirely dead world would be equally good or better as a world with lots of happy entities, simply because there are subagents within me who would prefer to exist and continue to do stuff, and also for other people to continue to exist and do stuff if they so prefer. I want us to liberate people's minds from involuntary suffering, and then to let people do whatever they still want to do when suffering is a thing that people experience only voluntarily.

I believe the most often cited (in the LW/EA communities) paper arguing against NU is Toby Ord's Why I'm Not a Negative Utilitarian. This and this seem to be the main replies to it from NU perspectives. (I think I've skimmed some of these articles but have not actually considered the arguments carefully.)

Ethical theories don't need to be simple. I used to have the belief that ethical theories ought to be simple/elegant/non-arbitrary for us to have a shot at them being the correct theory, a theory that intelligent civilizations with different evolutionary histories would all converge on. This made me think that NU might be that correct theory. Now I’m confident that this sort of thinking was confused: I think there is no reason to expect that intelligent civilizations with different evolutionary histories would converge on the same values, or that there is one correct set of ethics that they "should" converge on if they were approaching the matter "correctly". So, looking back, my older intuition feels confused now in a similar way as ordering the simplest food in a restaurant in expectation of anticipating what others would order if they also thought that the goal was that everyone orders the same thing. Now I just want to order the "food" that satisfies my personal criteria (and these criteria do happen to include placing value on non-arbitrariness/simplicity/elegance, but I’m a bit less single-minded about it). 

Your way of unifying psychological motivations down to suffering reduction is an "externalist" account of why decisions are made, which is different from the internal story people tell themselves. Why think all people who tell different stories are mistaken about their own reasons? The point "it is a straw man argument that NUs don’t value life or positive states“ is unconvincing, as others have already pointed out. I actually share your view that a lot of things people do might in some way trace back to a motivating quality in feelings of dissatisfaction, but (1) there are exceptions to that (e.g., sometimes I do things on auto-pilot and not out of an internal sense of urgency/need, and sometimes I feel agenty and do things in the world to achieve my reflected life goals rather than tend to my own momentary well-being), and (2) that doesn’t mean that whichever parts of our minds we most identify with need to accept suffering reduction as the ultimate justification of their actions. For instance, let’s say you could prove that a true proximate cause why a person refused to enter Nozick’s experience machine was that, when they contemplated the decision, they felt really bad about the prospect of learning that their own life goals are shallower and more self-centered than they would have thought, and *therefore* they refuse the offer. Your account would say: "They made this choice driven by the avoidance of bad feelings, which just shows that ultimately they should accept the offer, or choose whichever offer reduces more suffering all-things-considered.“ Okay yeah, that's one story to tell. But the person in question tells herself the story that she made this choice because she has strong aspirations about what type of person she wants to be. Why would your externally-imported justification be more valid (for this person's life) than her own internal justification?

Thanks for the replies, everyone!

I don’t have the time to reply back individually, but I read them all and believe these to be pretty representative of the wider community’s reasons to reject NU as well.

I can’t speak for those who identify strictly as NU, but while I currently share many of NU’s answers to theoretical outweighing scenarios, I do find it difficult to unpack all the nuance it would take to reconcile “NU as CEV” with our everyday experience.

Therefore, I’ll likely update further away from

{attempting to salvage NU’s reputation by bridging it with compassion, motivation theory, and secular Buddhism}

towards

{integrating these independent of NU, seeing if this would result in a more relatable language, or if my preferred kind of theoretical unity (without pluralist outweighing) would still have the cost of its sounding absurd and extreme on its face}

If the rationality and EA communities are looking for a unified theory of value

Are they? Many of us seem to have accepted that our values are complex.

Absolute negative utilitarianism (ANU) is a minority view despite the theoretical advantages of terminal value monism (suffering is the only thing that motivates us “by itself”) over pluralism (there are many such things). Notably, ANU doesn’t require solving value incommensurability, because all other values can be instrumentally evaluated by their relationship to the suffering of sentient beings, using only one terminal value-grounded common currency for everything.

This seems like an argument that it would be convenient if our values were simple. This does not seem like strong evidence that they actually are simple. (Though I grant that you could make an argument that it might be better to try to achieve only part of what we value if we're much more likely to be successful that way.)

What have you read about it that has caused you to stop considering it, or to overlook it from the start?

I reject impartiality on the grounds that I'm a personal identity and therefore not impartial. The utility of others is not my utility, therefore I am not a utilitarian. I reject unconditional altruism in general for this reason. It amazes me in hindsight that I was ever dumb enough to think otherwise.

Can you teach me how to see positive states as terminally (and not just instrumentally) valuable, if I currently don’t?

Teach, no, but there are some intuitions that can be evoked. I'd personally take a 10:1 ratio between pleasure and pain; if I get 10 times more pleasure out of something, I'll take any pain as a cost. It's just usually not realistic, which is why I don't agree that life has generally positive value.

There are fictional descriptions of extreme pleasure enhancement and wireheading, e.g. in fantasy that describe worthwhile states of experience. The EA movement is fighting against wireheading, as you can see in avturchin's posts. But I think such a combination of enhancement + wireheading could plausibly come closest to delivering net-positive value of life, if it could be invented (although I don't expect it in my lifetime, so it's only theoretical). Here's an example from fiction:

"You see, I have a very special spell called the Glow. It looks like this." The mage flicked his fingers and they started Glowing in a warm, radiant light. Little Joanna looked at them in awe. "It's so pretty! Can you teach me that? Is that the reward?" Melchias laughed. "No. The true reward happens when I touch you with it." She stared at him curiously. He looked down at the table in front of him with an expectant look, and she put her slender arm there so he could touch her. "Here, let me demonstrate. Now, this won't hurt one bit..." He reached out with his Glowing fingers to touch the back of her small hand, ever so gently.

And as their skin connected, Joanna's entire world exploded. The experience was indescribable. Nothing, no words and no warnings, could have prepared Joanna for the feeling that was now blasting through her young mind with the screaming ferocity of a white-hot firestorm, ripping her conscious thoughts apart like a small, flickering candlelight in a gigantic hurricane of whipping flames and shredding embers. She had no idea, no idea such pleasure existed, ever could exist! It was all of her happy memories, all of her laughter, her playfulness, her sexy tingles when she rubbed herself between the legs, the goodness of eating apple pie, the warmth of the fireplace in the winter nights, the love in her papa's strong arms, the fun of her games and friendships with the other village kids, the excitement of stories yet unheard, the exhilaration of running under the summer sun, the fascination of the nature and the animals around her, the smells and the tastes, the hugs and awkward kisses, all the goodness of all her young life, all condensed into a mere thousandth of a split-second... ...and amplified a thousand-fold... ...and shot through her mind, through her soul, again and again and again, split-second after split-second after split-second, like a bombardment of one supernova after another supernova of pure, unimaginable bliss, again and again and again, and yet again, second after second, filling her up, ripping her apart with raw ecstatic brilliance, melting her mind together in a new form, widened and brighter than it had ever been, a new, divine, god-like Joanna that no words could adequately worship, only to rip her apart again with a new fiery pulse of condensed, sizzling-hot vibrance, indescribable, unimaginable, each second an unreached peak, a new high, a new universe of fantastic pleasure, a new, unspeakably wonderful Joanna, loved and pulsing in her own Glowing light with a beauty unmatched by any other thing in all of the World Tree. She was a giant beating heart that was also a Goddess, Glowing and pulsing in the center of Everything, Glowing with the certainty of absolute affirmation, the purity of absolute perfection, the undeniability of absolute goodness. She spent centuries in seconds, serene yet in screaming ecstasy, non-living yet pulsing raw life force, non-personal yet always Joanna, Joanna in divine totality. It took Joanna a long time to realize she was still breathing, a living child with an actual human body. She had forgotton to breathe, and was sure she would have suffocated by now, but somehow, inexplicably, her body had remembered to keep itself alive without her. Master Melchias had lied: It did hurt, her chin and lip hurt, but the young girl found it was only because she had dropped to the hard stone floor in helpless twitching convulsions, and she had accidentally bitten herself. As promised by the wizard, the small wound was quickly healing. Joanna couldn't get up yet. She had no idea how much time had passed, but she just couldn't move or even open her young eyes yet. She curled up into a fetal position on the cold, hard floor of Melchias' Tower and sobbed uncontrollably. She sobbed and cried, and sobbed, and laughed madly, then sobbed and cried again. They were tears of pure joy.

The Glow wasn't just a normal pleasure spike, like an orgasm, a fit of laughter or a drug high. It went far, far beyond that. Normal human experiences existed within an intensity range that was given by nature. It served to motivate the organism for survival and reproduction, but it was not optimized for the experience itself. Even the most intense experiences, like burning alive or being skinned alive, existed within that ordinary, natural range. But the magic of the Glow didn't just stimulate pleasure within that range - it completely changed the range itself. It broke the scale on which normal experiences were measured, and then attached a vast multitude of additional ones to its top. By enhancing the part of the subject's mind that contained ordinary pleasure, it became temporarily able to experience an intensity that was hundreds of thousands times stronger than even the most extreme natural human feeling. Being drowned in hot oil, being flayed alive or tortured with needles, deep romantic love and fulfillment, orgasmic ecstasy, perfect fits of laughter - all of these human extremes represented only a miniscule fraction of the new potential. And only then did the spell induce raw, optimized pleasure within this new, widened consciousness. The result was an unimaginably pure goodness that fell so far outside of the subject's prior experience that it couldn't even be communicated by words. It had to be demonstrated. Once a potential [...] candidate had perceived even one second of the Glow, each containing more joy and happiness than an average human lifespan, with none of its pain, they all became devoted followers to Melchias. He transformed their experience from something human to something divine, and in turn, he became like a god to them.

If you flip the Rachels-Temkin spectrum argument (philpapers.org/archive/NEBTGT.pdf), then some tradeoff between happiness and suffering is needed to keep transitive preferences, which is necessary to avoid weird conclusions like accepting suffering to avoid happiness. As long as you don't think theres some suffering threshold where 1 more util of suffering is infinitely worse than anything else, then this makes sense.

Also NU in general has a bad reputation in the philosophy community (more than classical utilitarianism I think) so it's better EAs don't endorse it.