Toby Ord recently observed: “I have been surprised to see that some of my friends and acquaintances in the effective altruism community identify as Negative Utilitarians. Negative Utilitarianism (NU) is treated as a non-starter in mainstream philosophical circles, and to the best of my knowledge has never been supported by any mainstream philosopher, living or dead. This is quite an amazing lack of support: one can usually find philosophers who support any named position. On considering the theory in some detail, I cannot help but agree that the philosophical community has got this one right.”
16 Responses to “Why I’m Not A Negative Utilitarian”
April 26
Jacy AnthisIt worries me when EAs call certain moral frameworks ‘wrong’ or ‘non-starters.’ This moral realist approach is both unsubstantiated and alienating, even if it might be useful in short-term somewhat-coercive recruitment by telling people they’re “responsible” or “obliged” to do certain things as if there’s some godly moral high ground.
April 27
Wolf TivyLet’s put it this way: If someone claimed to follow a moral system where some group of people should wipe everyone else out and then not achieve anything for ever because they have no right to colonize beyond earth, then I think we would have no problem calling that moral system “wrong” or a “non-starter”. NU is worse than that one.
There is no absolute standard to appeal to, but what else are you going to call a value system that advocates mass genocide, other than “wrong”? “I don’t like your genocidal ambitions, but there is no absolute morality, so go ahead and advocate for my and my people’s death”?
It worries me when EAs use “there is no objective morality” to shut down criticism (this has happened more than once, I’m afraid). There is no objective morality, but there’s some important way in which it is productive to argue about value systems even if we can’t quite articulate it.
April 26
Michael WiebeI worry when nonphilosophers have strong views on metaethics.
April 26
Jacy AnthisMichael, I don’t think taking strong views on metaethics requires philosophical credentials, but regardless, I was speaking from a practical standpoint rather than a theoretical standpoint.
April 26
Jacy AnthisBen, it’s practically unsubstantiated in the sense that there’s no non-esoteric justification readily available. I’m just saying, it seems EA will be worse off the more we appeal to moral realism.
April 26
Caleb OntiverosIt’s not clear to me that statements like “a given normative theory is wrong” presuppose moral realism.
April 26
Jacy AnthisBen, in the way I meant it. sure, but “Quantum physics is unsubstantiated” sound clearly wrong because one reads “unsubstantiated” much differently in that context.
April 26
Oliver HabrykaI think it is definitely justifiable to say that the moral theory of negative utilitarianism does not correspond to reality, as in: it does not reflects true human values. Sure, some people might profess that their values correspond to negative utilitarianism, but that is about as important as someone professing their values correspond to the moral system of the bible.
The main problem I see with negative utilitarianism is that it is _simple_. Really simple. You have a single thing you care about, and that is suffering, that thing scales linearly and it is the ultimate goal to reduce it. But human values are not simple. We know from moral experiment and thousands of years of philosophical analysis that our values are damn complex and confusing, and they are probably not even consistent.
I think it is a justified belief to have, that human values can not be modeled by such a simple theory as NU. Normal Utilitarianism (at least the modern versions of it) allow for much more subtle value functions. They can care about aesthetics, diminishing returns on happiness, a context dependent weighing of pleasure versus suffering and etc. And I know that I care about these things and that I need these subtleties. I would be very surprised if this doesn’t stand for other human beings.
I reject negative utilitarianism in large parts because it is obvious that this theory is too simple to respond to the complexity of human value.
April 27
Kenneth MyersDarkly, I’m the outlying lunatic who actually wonders if it would be better to end sentient life to stop the suffering. I may be a negative utilitarian. At least on some mornings.
April 28
Jonas VollmerI don’t think this essay brings up any strong arguments against NU. It’s like saying “repugnant conclusion, ergo CU is wrong”.
April 28
Oliver HabrykaI think it is definitely justifiable to say that the moral theory of negative utilitarianism does not correspond to reality, as in: it does not reflects true human values. Sure, some people might profess that their values correspond to negative utilitarianism, but that is about as important as someone professing their values correspond to the moral system of the bible.
The main problem I see with negative utilitarianism is that it is _simple_. Really simple. You have a single thing you care about, and that is suffering, that thing scales linearly and it is the ultimate goal to reduce it. But human values are not simple. We know from moral experiment and thousands of years of philosophical analysis that our values are damn complex and confusing, and they are probably not even consistent.
I think it is a justified belief to have, that human values can not be modeled by such a simple theory as NU. Normal Utilitarianism frameworks (at least the modern versions of it) allow for much more subtle value functions. They can care about aesthetics, diminishing returns on happiness, a context dependent weighing of pleasure versus suffering and etc. And I know that I care about these things and that I need these subtleties. I would be very surprised if this doesn’t stand for other human beings.
I reject negative utilitarianism in large parts because it is obvious that this theory is too simple to respond to the complexity of human value.
But I agree with Jonas that the points raised in the article are not particularly strong. Negative utilitarianism is a consistent value function in the context of utility theory, the question is whether that value function actually corresponds to human values.
April 28
Tarn Somervell Fletcher“Jacy Anthis Ben, in the way I meant it. sure, but “Quantum physics is unsubstantiated” sound clearly wrong because one reads “unsubstantiated” much differently in that context.”
I read it in the same way as in the QM statement. I think many people would
April 28
Oscar RuedaDear Oliver, I have come to the same conclusion over and over again: it is very difficult to follow a negative utilitarian philosophy given “true” human values. Probably, the reason is that some of our values are very often contradictory (i.e. our behavior is not always rational) and don’t help us maximize our value function, whatever our value function is.
However simple the value function of negative utilitarianism is, reducing suffering, it provides a straightforward approach, for instance, to consider the interests of non–human animals.
It is arguable that negative utilitarianism, as other approaches of utilitarianism, may lead us to some form of “the repugnant conclusion”; yet, given the immense amount of easily preventable suffering in the world, I find negative utilitarianism a very effective way of leading one’s life.
April 29
Lukas Gloor«We know from moral experiment and thousands of years of philosophical analysis that our values are damn complex and confusing, and they are probably not even consistent.»
This is why you probably have to end up choosing just some intuitions and not all of them. There is no reason that your u-function needs to pay tribute to every intuition you have; in the end, it simply comes down to what you consider to be most fundamental. The tradition of academic ethics suggests that at least many humans are interested in some notion of “universalizability”, most major ethical theories seem to have this as an axiom. If evolution had made us sentient baby eaters, we might have complex values that also reflect the strong intuition to eat babies. If you’re uncomfortable with counterfactual-you having such strange values, then that’s an argument to stop acting as if the descriptive point that human derived or stated values are complex is reason that our chosen terminal values ought to be complex as well. Granted, you always need some input that is contingent: there are no universally compelling arguments. With NU, the input you need is altruism/universalizability, and regard for your well-being in the moment that you’re in.
April 29
Oliver HabrykaI agree that NU can sometimes be a useful heuristic to get close to my actual values. But it is also very obvious that our values will not be completely satisfied by such a simple value function like negative utiltarianism.
And this is not a question of choice, negative utilitarianism, as basically all simple moral theories go, leads to horrible conclusions and doesn’t work in the edge cases.
For a more in depth analysis of this problem see this paper by MIRI:
http://intelligence.org/files/ComplexValues.pdf
May 1
Philip GoetzBut Christian morality strongly emphasizes negative utilitarianism! Surely some philosopher in a largely Christian nation is a negative utilitarian.