I've been struggling with some beliefs that are contrary to standard total utilitarian views. I'm sure there are standard responses to them, but I'm not sure where to look for them. As such, I would appreciate pointers at resources that elaborate or argue against the following view(s), or direct feedback if you have time.

I generally believe that it's morally valuable to improve the lives of existing beings and ensure the lives of future beings are good (e.g., animals are not born into lives of r-selected misery or into factory farms), and I tend to lean negative here as I'm not convinced making a happy person very happy is as important as making a suffering person not suffer (though perhaps I'm not holding the right picture in my mind when I try to imagine this). What I can't bring myself to endorse is the view that it's morally required to fill the universe with happy lives. To me, that seems morally neutral. While I suppose I would say that a word with 100 happy beings and 5 suffering beings is better than a world with 10 happy beings and 5 suffering beings. I wouldn't say it's a very good thing for 90 more happy beings to be created in the latter world, while I would say helping the 5 not to suffer so much is very good.

In short, I care a lot about consequences and want the best for existing and future beings but am not persuaded that the number of beings is all that important.

I know the above isn't perfectly articulated. I think there's a prioritarian aspect to it, but I'm not sure prioritarianism entirely captures my view.

Feedback appreciated!

It sounds like average utilitarianism: the idea that it's the well-being of the average person that matters. Is this correct?

A simple argument for total utilitarianism is this: Imagine you are totally selfish. Would you consider it to be ten times as good to live for a hundred years than ten? If so, you seem to think more life is better. If you care about everyone, then having 100 trillion person-years is ten times better than 10 trillion.

It's also possible that you care only about existing beings, although a lot of what you wrote doesn't seem to fit with that.

The problem with caring only about existing beings is that it's inconsistent. You might drink alcohol while pregnant, since the baby isn't sentient yet so its future life doesn't matter, but then once it's born you pay for treatment to get the baby to how he would have been, where the cost of the treatment is more than the value drinking the alcohol originally had.

If it's something else, I'm going to need you to clarify it more.

Thank you for the thoughts!

It's very possible I have a bunch of beliefs that don't hang together in a coherent way.

I don't think I'm an average utilitarian, because I don't believe the average is what matters. That is, I don't think creating more happy beings or increasing the happiness of already happy beings is as good as helping those who are suffering not suffer or preventing future suffering beings from existing even if the averages are maintained or improved in the former cases. This leads me to think I'm a prioritarian.

I do care about the quality of future beings' lives, but I don't think we have any special obligation to create happy lives. For those beings who will exist, I think their well being matters as much as those existing today, but again I don't think it's necessarily a good thing to create happy lives. That seems morally neutral to me. Utilitronium has zero appeal.

Maybe the way this all shapes up I shouldn't really call myself a consequentialist, but rather someone who takes consequences very seriously in tandem with other beliefs.

You sound negative-leaning. That is, you are more concerned with preventing sadness than creating happiness. Is this correct?

Do you believe that creating a happy person is as good as the combined effect of creating a sad person and elevating them to a happy person?

I'm definitely negative-leaning.

I'm not sure how to answer your question because it seems neutral to me whether one creates a happy person.

On my view, creating a sad person - knowingly - is bad, and helping a sad person to be happy is good. Creating a sad person unknowingly or unintentionally is unfortunate but not morally bad.

I'm not sure if this is a view I'm comfortable settling on, but the existence of suffering sentients bothers me much more than the idea of the extinction of sentients. That said, bringing about the extinction of sentients doesn't seem like a good thing to me, largely because so many of us want to continue existing and reproducing and because bringing about extinction would, besides running contrary to those desires would almost invariably cause a huge amount of suffering.

Negative-leaning and prioritarian views are consequentialist. You seem to be something like negative-leaning utilitarian or prioritarian, with the difference being (I think, though I'm not an expert) that prioritarians care more about those absolutely worst off, whereas NLUs just care about making the biggest difference, regardless of the absolute welfare of who's helped. You might also overlap a lot with David Benatar.

jason wrote: What I can't bring myself to endorse is the view that it's morally required to fill the universe with happy lives. To me, that seems morally neutral.

I hold the same view.

jason wrote:While I suppose I would say that a word with 100 happy beings and 5 suffering beings is better than a world with 10 happy beings and 5 suffering beings. I wouldn't say it's a very good thing for 90 more happy beings to be created in the latter world, while I would say helping the 5 not to suffer so much is very good.

Then what do you mean by "better" in the first sentence? Is it just a lexical preference that gets overiden as soon as you're also able to affect the total amount of suffering?

jason wrote:Utilitronium has zero appeal.

jason wrote:That said, bringing about the extinction of sentients doesn't seem like a good thing to me, largely because so many of us want to continue existing and reproducing and because bringing about extinction would, besides running contrary to those desires would almost invariably cause a huge amount of suffering.

If you think a preference to go on existing being thwarted is negative in itself, then this would likely be some sort of preference utilitarianism. I think negative preference utilitarianism is a consistent position that gives very intuitive (for population ethical standards anyway) conclusions. The most counterintuitive aspect of it, apart from the general arguments against why preferences matter as opposed to experiential states, is that the creation of an almost perfect life is negative to the extent it still has some thwarted preferences in it. This doesn't seem more counterintuitive than the repugnant and very repugnant conclusions, and on top of that, the negative population ethics seem to fit very well with preference utilitarianism from a more top-down/theoretical perspective. Consider the individual case: Is it morally urgent to add a new preference to an individual that already has preferences, all else being equal (i.e. none of the pre-exisitng preferences becoming more violated or more fulfilled)? It seems not, not even if the new preference is completely fulfilled. The intuition behind the negative preference view can be translated into "solving problems" as opposed to "creating solved problems". Furthemore, consider the odd implications of classical preference utilitarianism: you would want to maximize the surplus of satisfied preferences over unsatisfied ones, but since there is no content-requirement for preferences, you could just tile the universe with beings who have very easily satisfiable preferences (very unlike humans), which would be an even more pointless endeavor than the creation of utilitronium. (Prference utiltiarianism often also comes in prior-existence varieties inspired by Peter Singer, but I'm pretty sure such views are inconsistent because of transitivity issues.)

Alternatively, if you think that only experiential states matter, your worry about people's preferences to go on existing would, it seems, just reflect a concession that the view you're proposing seems very counterintuitive. I agree with some of the points made above that the position you advocate does seem like it might be inconsistent once you flesh it out in detail. I think such inconsistency is hard or impossible to avoid if you use to common framework of "positive and negative welfare". This is sometimes called welfarist axiologoy, axiology being your "theory of what matters" or, in consequentialist terms, your definition/axiom for "utility". If you say happiness is positive (i.e. ethically preferable over non-existence) for existing beings but not for beings that don't exist, you're introducing the category "existing being" as an ethically relevant entity. On a reductionist account of personal identity, it becomes questionable whether you can keep up this category and its ethical relevance. The reductionist account would imply that every split-second, a "new" person is coming into existence, as opposed to a numerically identical person going on existing over time (what could that possibly mean?). It seems that, given that the reductionist account of personal identity is correct, all of ethics turns into population ethics. Not killing a being (by omission) becomes in consequentialist terms ethically equivalent to creating a new being (by action). And if the happiness of existing beings can make up for their suffering, why not also the happiness of "new" beings?

The reductionist account at first exerts a strong pull towards classical utilitarianism. However, I think it has been overlooked that you can bite the bullet the other way as well, and possibly even more elegantly so: When people are imagining whether they would accept some suffering (e.g. walking over hot sand) in order to be happy later (e.g. swimming in the ocean), they think that the future person will still be them, and the choice they are making is egoistic rather than altruistic. Now, I see consequentialist ethics as playing the game of figuring out what some intuitively plausible notion of "systematized altruism" would imply. And interestingly, when people are asked the same question in an altruistic framing, the answer turns out to be much less classically-inclined! If you asked people whether they would want to simulate some painful hot sand states over here, and then simulate some happy ocean states elsewhere over there, a lot of people would reply something like "Why would I want to create these happy states? And no, you shouldn't create suffering states!" It seems that the egoistic case is dominated by an impression of personal identity and evolutionary tendency to want to go on living and experience cool stuff. When we just look at the experiences themselves, in the moment in isolation, it becomes much less clear whether the same "exchange rate" should apply also in cases characterized as altruistic.

I think that strict negative hedonistic utilitarianism becomes consistent and much more intuitive if you adopt a different axiology than classical utilitarians use. Negative utilitarianism requires a Buddhist view on suffering and contentment. This view claims that happiness is not in the same way morally urgent as the prevention of suffering. Happiness according to this view is ethically equivalent to states that are absent of any cravings or longings. There seems to be nothing wrong whatsoever with hedonistically neutral flow states with a low level of self-awareness and time being experienced as flying. Likewise, a Buddhist in a meditative state absent of an cravings seems to be a perfectly fine thing as well, and it is hard to see why there should be moral urgency to turn such a state into orgasm. This Buddhist view focuses on the moments itself, not on how much we desire certain moments when we imagine them from the outside. If in the moment you're in, you have no desire to get out of your state of consciousness or change something about it (this is what constitutes suffering according to this view!), then everything is perfectly fine also in ethical terms, because the ethical "ought" would correspond to the aggregated internally felt "wants" of all consciousness-moments.

Adriano Mannino and I have written drafts for papers on both negative hedonistic and negative preference utilitarianism. Please (for anyone) PM me on facebook (Lukas Gloor) if you're interested in reading the drafts.

I should add that technically it is also possible to have a negative view on population ethics and be a prioritarian at the same time, but I suspect that a lot of the appeal of prioritarianism is also present in negative utilitarianism. And prioritarianism has the disadvantage (if one cares about this) that there are infinite prioritarian weighting functions that all seem equally plausible on the face of it.

Thank you, Brian and Lukas. You've both given me a lot to mull over and look into.

Lukas, I'm curious about this statement of yours:

The reductionist account would imply that every split-second, a "new" person is coming into existence, as opposed to a numerically identical person going on existing over time (what could that possibly mean?). It seems that, given that the reductionist account of personal identity is correct, all of ethics turns into population ethics.

Is there a separate thread on identity? Or is there something you can point me to that makes the case that such a radically reductionist view is correct? I've not looked into the issue much, but have previously found Parfit's idea of overlapping chains of psychological connectedness to be pretty appealing.

Yeah, I've debated this with Lukas. I think there are coherent ways to care about personal identity, just like there are coherent ways to care about consciousness even though there are no hard boundaries in that domain either. To be sure, I don't think we should care about personal identity ethically, but I think doing so is just as coherent as caring about consciousness.

Jason, I don't know of any specific thread on personal identity here, but I know that it has come up a couple of times. Many members here adhere to the reductionist view. DanielLC might actually have been the first person who referred me to it initially. There is a section on it in our draft for a paper on negative utiltiarianism, but if you're already familiar with Parfit, there might not be too much new stuff.

Brian Tomasik wrote:Yeah, I've debated this with Lukas. I think there are coherent ways to care about personal identity, just like there are coherent ways to care about consciousness even though there are no hard boundaries in that domain either. To be sure, I don't think we should care about personal identity ethically, but I think doing so is just as coherent as caring about consciousness.

1) Your claim about there being no hard boundaries for consciousness seems insane to me. Granted, this is not strong evidence against it because I'm not aware of any views on philosophy of mind that don't seem insane to me, but I'd still be cautious to use this particular example for the point you're making.

2) I agree that such positions are consistent. But there are also coherent ways to be speciesist, racist or nepotist in the way common sense morality advocates it, or to attribute ethical importance to these categories. Ask yourself why you reject drawing lines there and I predict that for most consequentialists, the same reason should also carry weight in the case of personal identity.

1) My view on consciousness is shared by almost everyone on LessWrong.

2) Yes, I agree. Speciesism is also a coherent way to differentiate moral worth, but we don't agree with it. I feel the same way about personal identity.

ExtendedCircle wrote:Jason, I don't know of any specific thread on personal identity here, but I know that it has come up a couple of times. Many members here adhere to the reductionist view. DanielLC might actually have been the first person who referred me to it initially. There is a section on it in our draft for a paper on negative utiltiarianism, but if you're already familiar with Parfit, there might not be too much new stuff.

What interested me most about what you said, and maybe this is reading too much into it, is the idea that a "new" person exists every second. Perhaps where I went astray was thinking you were suggesting that there's no personal identity across seconds within, say, normal human adults. Is that what you were advocating? If so, i would be interested in your paper, as it's much more radical of a view than I'm prepared to accept currently.

jason wrote:What interested me most about what you said, and maybe this is reading too much into it, is the idea that a "new" person exists every second. Perhaps where I went astray was thinking you were suggesting that there's no personal identity across seconds within, say, normal human adults. Is that what you were advocating?

Yes, that's what I meant. I'm sending you a PM.