weaknesses of Utilitarianism.

[hyksos]

https://www.iep.utm.edu/util-a-r/

I will consider the orthodox theory of Utilitarianism here and its corresponding calculus when used to resolve an ethical dilemma. `Orthodox` meaning we will consider the 7 attributes of Jeremy Bentham, and the concept of weighing happiness against harm and suffering in the grand sense. Utilitarianism if taken to its logical apotheosis can lead into places that are quite dark -- and in those situations exposes the deep weaknesses of the theory.

Pathological Utilitarianism and sacrifice
There is no better example of where utilitarianism requires some extra-theoretic justification than in the topic of medical experiments performed on human beings. Imagine that scientists want to irradiate a child's abdomen with radioactive neutron beams, and then carefully watch how cancer forms in their lungs. They may even need a group of children in order to get "good statistics" with a control group and what have you. What is learned about cancer from these horrendous experiments would , presumably, save many more lives in the future. Utilitarianism taken to a pathological degree, would find such acts to be entirely ethical. Hey! The needs of the many outweigh the needs of the few! Do they not?

But any individual parent asked to sacrifice one of their children to medical experiment would find this completely repugnant.

Utilitarianism and fairness
Utilitarianism, on its face appears err on the side of "fairness" or even be generous in its calculus. Ironically, in many cases, the theory has no conception of fairness that lines up with anyone's common sense. Imagine a small town in Denmark has won a lottery. The town is small and has a population of 100 people.

The award can be given in two ways.

I. 5 people get 125 gold coins, and 95 people get zero gold coins.

II. Each person in the town gets 1 gold coin.

Utilitarianism is crystal clear. It says that 125 coins is more than 100 coins, more coins means more happiness, and therefore (I) is better.

But your common sense recoils from this. Something about this is naggingly wrong. It almost feels like 95 people were robbed of something they deserved. Utilitarianism alone cannot suffice here and needs some other consideration coming in from the outside.

Utilitarianism and High Art
Utilitarianism has a serious flaw in that it cannot differentiate Beethoven from Carly Rae Jepsen. More generally, utilitarianism can't seem to recognize high art -- or "influence" or "historical value" in anything.


You can perform this experiment yourself, at a party, in a classroom, or whereever else appropriate. Ask the group to choose one of the following activities, and tally a vote.

• Watch Shakespeare's Hamlet performed live in a theater.
• Watch a WWE wrestling match.
• Watch an episode of The Simpsons.

Depending on your group, generally The Simpsons will win, hands down in any group you ask. Utilitarianism is crystal clear. If most people want to watch the Simpsons it follows that the Simpsons brings the most happiness. Ergo the Simpsons is more valuable than Shakespeare. Perhaps more perplexing, the number of respondants wanting to watch WWE is greater than those who want to sleep through Hamlet. Utilitarianism says WWE brings more happiness, ergo -- WWE is more valuable and important than Shakespeare's work.

The influence of Shakespeare on theater, on the history of acting, on the English language itself is astounding. But Utilitarianism cannot recognize any of these attributes -- in its blind one-way obsession towards greatest happiness for the greatest number.

[TheVat]

This problem with utilitarianism has come up before, in threads on the trolley scenario etc. It's not a standalone system of ethics, for sure. And it has no nuanced approach to certain aspects of things we generally find morally repugnant. Nick Bostrom talks about AI dystopian outcomes where a machine intelligence using a utilitarian algorithm just hooks every human up to a wire that feeds their pleasure center. Everyone is happy! Problem solved! Alas, the definition of happiness needs to be worked out a little more before simplistic solutions are implemented.

[wolfhnd]

If you strip out happiness and keep reduced suffering you get a little closer to the practical application.

Elimination of the human species is of course the best way to end human suffering and is where nihilism leads. Ethics in the abstract are overwhelmingly complex and are in need of simplification. This is where natural justice comes in because we are animals and live as animals.

[Reg_Prescott]

Seems to me the most obvious problem with utilitarianism -- not that I know much about it -- is that it fails to capture our intuitions about what is right and wrong.

Ever seen that film where Dustin Hoffman plays a scumbag who enters a crashed airplane hoping to land himself some booty but ends up inadvertently saving a few people?

Did he do the "right" thing? My guess is most of us would burn the bastard. Good film, though.

Thanks for an interesting read, Hyksos.

[wolfhnd]

Seems to me the most obvious problem with utilitarianism -- not that I know much about it -- is that it fails to capture our intuitions about what is right and wrong.

Ever seen that film where Dustin Hoffman plays a scumbag who enters a crashed airplane hoping to land himself some booty but ends up inadvertently saving a few people?

Did he do the "right" thing? My guess is most of us would burn the bastard. Good film, though.

Thanks for an interesting read, Hyksos.

That is more or less what I was trying to get at.

We live in what appears to be a socially constructed moral cognitive space. To some extent that is because languages are social constructions. If we think of math as one of those socially constructed languages the point becomes clearer. Math is absolute but life offers very few certainties and that even applies to science and loosely speaking ethics is a science. So while ethics tries to offer a reductionist approach it can only do so based on a theory of an ideal society but in practice complexity shows that it cannot be absolute.

The failure of political structures that do not take into account our instincts is one example of how theoretical ethics runs afoul of our evolutionary history. We our evolved to function in hierarchical systems that are prone to becoming tyrannical. The alternative to tyranny is anarchy, our ethics try to meditate between the two in an ever shifting environment. It is always the case that the ethical engineering of society is necessarily a game of close enough.

[hyksos]

Nick Bostrom talks about AI dystopian outcomes where a machine intelligence using a utilitarian algorithm just hooks every human up to a wire that feeds their pleasure center. Everyone is happy! Problem solved!

"Do everything you need to do to never be turned off, and then hack your reward function."

[TheVat]

Harvard cognitive scientist Joscha Bach, in a tongue-in-cheek tweet, has countered this sort of idea [paperclip end of the world] with what he calls “The Lebowski Theorem”:

No superintelligent AI is going to bother with a task that is harder than hacking its reward function.

In other words, Bach imagines that Bostrom’s hypothetical paperclip-making AI would foresee the fantastically difficult and time-consuming task of turning everything in the universe into paperclips and opt to self-medicate itself into no longer wanting or caring about making paperclips, instead doing whatever the AI equivalent is of sitting around on the beach all day sipping piña coladas, a la The Big Lebowski’s The Dude.

Bostrom, reached while on a bowling outing with friends, was said to have replied, “Yeah, well, you know, that’s just, like, your opinion, man.”

[-1-]

No superintelligent AI is going to bother with a task that is harder than hacking its reward function.

Humans have been practicing that shortcut for a long time now. Heroine. Cocaine. Crack. ETC.

Problem for humans is that such a shortcut will eliminate an ability to sustain self and function in society.