Total versus average happiness

Utilitarianism, prioritarianism and other varieties of consequentialism.
ReX
Posts: 2
Joined: Mon Jan 09, 2012 3:05 am

Total versus average happiness

Post by ReX » Mon Jan 09, 2012 3:35 am

I was contemplating the Total versus Average happiness problem.
It Total happiness is all that matters and you believe that a human always reaches minimum happiness (wellbeing, more so than say, a pig can ever hope to experience
it is better to be a human being dissatisfied than a pig satisfied; better to be Socrates dissatisfied than a fool satisfied
), and I do believe this is (provided you have the same amount of food and shelter Socrates had) the case. Up until a few years ago there was enough food in the world to feed everybody (it was just distributed unequally). Perhaps this has changed since the food riots (but the reason would then be ethanol subsidies and burning food instead of eating it). However, if you stick to the container view of life (cfr. Shelly Kagan lectures on death), supposing that all life has a minimal amount of worth making it a 'gift', then there's an absolute anyhow and either way (creating an absolute minimum of hedons in the world which didn't exist prior to the entity being [born]). We would thus be in a state where, not only would it be bad to kill anyone (even people in chronic pain/depression), we would have to create as many people as women (and men, but that takes less effort) would be able to bare.
The flipside, Average happiness, would require us to kill everybody who's below the average (or mean, because the average person usually doesn't exist) and eventually we'd have to kill the last person on earth.
Personally, I believe if the government has a duty, it would be to provide for every person on this planet in the basis necessities and those seem to be food and perhaps some shelter, education would be nice but indoctrination lurks at every corner and refusing a meal (say, hungerstrike) would be a right just as much as refusing to be confined to a certain space (vagabonding is no longer illegal in my country for decades now). Nevertheless, the option should be there. Assuming people will get these basic needs (through sharing, sympathy, stealing, charity, whatnot -I included economy and religion in the original list but on closer inspection that's no more realistic than expecting governments to do it-), we'd have the moral imperative to create extra people.
For a long time I didn't want to create extra people in this overpopulated world (nor kill, but let's postone the pacifism debate if possible). The reasoning was that famine is the flipside of population > food supply. Can we really postpone creating more people until there isn't a single person starving (or dying of cold/heat)? And if you're setting up the system, that is to say, if you can do the (moral) calculus of how much resources it would take for a certain amount of people; and you'd find that we've gone beyond our limits, is there an imperative to (not only stop creating new life, but also to) end human lives already in existence?
Can anybody suggest some reading on this topic? Thanks.

DanielLC
Posts: 707
Joined: Fri Oct 10, 2008 4:29 pm

Re: Total versus average happiness

Post by DanielLC » Mon Jan 09, 2012 5:45 am

but the reason would then be ethanol subsidies and burning food instead of eating it
The reason would be not putting enough resources into producing food to eat given the number of people we have. Whether we put those resources into burning plants or building big-screen TVs is irrelevant.
and you believe that a human always reaches minimum happiness
I find the idea that minimum happiness is positive very counter-intuitive. I would consider life in which you are actively and horribly tortured worse than death.

I've heard it suggested that the problem most people have with the Repugnant Conclusion is that they imagine "life barely worth living" to mean "life on the verge of suicide", rather than "life on the verge of neutrality", and so it is negative.

I'd suggest considering the question: Would you rather be twice as happy or live three times as long?

I lean towards total, but part of me figures that you're just a random person, so there's no way your experience can distinguish the number of people, and it shouldn't matter. Then there's that other part of me staring at both of them in confusion going "But those utility functions are incommensurate! How am I supposed to make a decision under uncertainty with undefined expected utility?"
Consequentialism: The belief that doing the right thing makes the world a better place.

User avatar
Ruairi
Posts: 385
Joined: Tue May 10, 2011 12:39 pm
Location: Ireland

Re: Total versus average happiness

Post by Ruairi » Mon Jan 09, 2012 7:31 pm

it seems to make sense to me, from a preferential util point of view, that a life is worth living (has positive utility) if the organism wants to continue living it. which then would seem to make the repungant conclusion repungant because of the verge of suicide thing, but why would a total viewpoint actually lead to this? just because a situation with many organisms on the verge of suicide would have barely positive utility doesnt mean that the most efficient use of our resources would be to make more. it would probably make much more happiness to improve the existing lives

by your system how do you define what has positive and negative utility ? i remember someone here suggesting that the experiences you want to be conscious for have positive utility and the ones you'd rather kinda just skip and not experience have negative utility.

is this what you're advocating?

on an off topic note i think the suicide thing is really interesting because maybe we could use it to see if wild animals have lives worth living, but maybe they dont understand suicide or maybe my system is wrong

DanielLC
Posts: 707
Joined: Fri Oct 10, 2008 4:29 pm

Re: Total versus average happiness

Post by DanielLC » Mon Jan 09, 2012 11:37 pm

that a life is worth living (has positive utility) if the organism wants to continue living it
It could be akrasia, or something like that. Perhaps you don't like living, but you have such a strong instinctual aversion to dying that you go against it. Also, you have to choose between living and dying, and if your instinct is that dying really sucks, you can't easily learn that this isn't the case.
i remember someone here suggesting that the experiences you want to be conscious for have positive utility and the ones you'd rather kinda just skip and not experience have negative utility.
That doesn't really seem to work. If someone will live long enough that time discounting makes them ignore the fact that they'll die eventually, they'll just want to skip over things if they think what comes next is better. Perhaps today is a good day, but tomorrow is Christmas.
Consequentialism: The belief that doing the right thing makes the world a better place.

User avatar
Brian Tomasik
Posts: 1107
Joined: Tue Oct 28, 2008 3:10 am
Location: USA
Contact:

Re: Total versus average happiness

Post by Brian Tomasik » Tue Jan 10, 2012 4:55 am

Ruairi wrote:on an off topic note i think the suicide thing is really interesting because maybe we could use it to see if wild animals have lives worth living, but maybe they dont understand suicide or maybe my system is wrong
I generally agree with your notion of when a life is worth living or not, but I'm wary of applying it to animals. Below is from a message I recently wrote to another friend.
1. It may be that most animals (except the smartest mammals and birds?), while conscious emotionally, don't understand death enough to realize that they could end their suffering by killing themselves. As an analogy, when I have a nightmare, I feel bad, but I'm not sufficiently in control of the situation that I can end the nightmare at will. I guess non-dreaming animals do have more control over their physical state than I do when asleep, but the point is that you can have emotions without being very smart.

2. Animals don't have painless ways to kill themselves. For short-lived animals, I think most of the total pain of their lives comes from dying. For example, most of the 1000 offspring of a beetle mother will die within a day or three of hatching. I think their lives up to the point of death might hover around being neutral between pain and happiness, so there's not much to be gained by early suicide.

3. Animals often don't do what's in their long-term hedonic interest, so even if suicide were optimal, they might not do it. (For example, when very nauseous, it may feel better to vomit immediately than to endure nausea for two hours going forward, but I can never muster up the courage to vomit.)

4. I think animals have a "will to live" that's partly separate from their hedonic well-being. However, I happen to care about their hedonic well-being rather than their will to live. Animals' behaviors are an integration of a huge number of signals and brain systems, so it's not surprising that some of these systems can act contrary to the hedonic-welfare-maximization systems.

5. If animals do kill themselves when their lives are not worth living, why don't we see more suicides on factory farms? Surely at least battery-cage hens would be better off killing themselves.

6. I think the animals that probably could contemplate suicide (chimpanzees?) likely do have lives worth living a good amount of the time.
I've heard rumors about octopus suicide, and no doubt some primates have killed themselves deliberately, but I'm not aware of further examples. (Are you?) If animals could commit suicide, surely they would do so at least as often as well-off humans. But I've never heard suggestions that the animal suicide rate is even near what it is for people. If that's right, then I suspect most animals can't kill themselves. (And if that's wrong, I'd love to learn more.)

User avatar
RyanCarey
Site Admin
Posts: 717
Joined: Sun Oct 05, 2008 1:01 am
Location: Melbourne, Australia

Re: Total versus average happiness

Post by RyanCarey » Tue Jan 10, 2012 6:13 am

Hi Rex,
the formal expression of this criticism of Total Utilitarianism is Derek Parfit's Repugnant Conclusion argument. You may find enjoy reading mine and Rehoot's wiki article re the Repugnant Conclusion.

I may find some more reading for you later :)

Also, my general principle for dealing with this kind of problem is:
1. Use total utilitarianism
2. Remember that a life can get so bad that it's not worth living anymore
3. Many people in the world have lives that are not worth living.

Using this approach, we're not obligated to bring new people into the world or obligated to kill people (because in any realistic scenario, this would cause immense suffering)
You can read my personal blog here: CareyRyan.com

User avatar
Ruairi
Posts: 385
Joined: Tue May 10, 2011 12:39 pm
Location: Ireland

Re: Total versus average happiness

Post by Ruairi » Tue Jan 10, 2012 9:21 pm

i wonder if satisfaction has less to do with suicide than we realise, perhaps it has more to do with a sudden jolt in the amount of satisfaction one is experiencing compared with what is expected/used to. or something else less related to satisfaction/happiness than we realise? obviously it would seem the two are very related but maybe not exaclty how we expect. id say theres probably high impact research to be done in this area

suicide rates by country:

http://www.who.int/mental_health/preven ... index.html

http://en.wikipedia.org/wiki/List_of_co ... icide_rate (a lot disputed on its talk page)

if you do a search for life satisfaction the results seem to vary a lot by whos doing it.

anyway i was trying to say maybe suicide isnt a good marker of 0 utility, as in
I've heard it suggested that the problem most people have with the Repugnant Conclusion is that they imagine "life barely worth living" to mean "life on the verge of suicide", rather than "life on the verge of neutrality", and so it is negative.
although instinctively its exactly what i would expect, as i said a life worth living is one that wants to be lived
That doesn't really seem to work. If someone will live long enough that time discounting makes them ignore the fact that they'll die eventually, they'll just want to skip over things if they think what comes next is better. Perhaps today is a good day, but tomorrow is Christmas.
ok but what if they cant live forever?

@alan: cool thanks for that:)! i read a few stories about dogs and cats and a lion i think and stuff but thats it

User avatar
Brian Tomasik
Posts: 1107
Joined: Tue Oct 28, 2008 3:10 am
Location: USA
Contact:

Re: Total versus average happiness

Post by Brian Tomasik » Wed Jan 11, 2012 5:21 am

I agree with everything RyanCarey said.
Ruairi wrote: @alan: cool thanks for that:)! i read a few stories about dogs and cats and a lion i think and stuff but thats it
Fascinating. Do you remember where you read it?

User avatar
Ruairi
Posts: 385
Joined: Tue May 10, 2011 12:39 pm
Location: Ireland

Re: Total versus average happiness

Post by Ruairi » Wed Jan 11, 2012 11:54 am

this is the lion one, particularly interesting because it clearly shows so much emotion http://news.softpedia.com/news/Do-Anima ... 3441.shtml

two interesting discussions i just came across

http://fr.reddit.com/r/askscience/comme ... t_commits/

http://fr.reddit.com/r/askscience/comme ... an_commit/

:)

User avatar
RyanCarey
Site Admin
Posts: 717
Joined: Sun Oct 05, 2008 1:01 am
Location: Melbourne, Australia

Re: Total versus average happiness

Post by RyanCarey » Wed Jan 11, 2012 11:55 pm

It's worth noting that for better of for worse, the popular conception is that suicide is an irrational act. So this is not an argument that is going to uniformly work one way. e.g. animals cannot commit suicide therefore they suffer more than us. People would interject that the fact that animals do not commit suicide means they suffer less mental illness, so they are better off than us.

If you said that animals do commit suicide and therefore they suffer more than us, people would say that suicide is a mere consequence of mental illness and that an animal could be schizophrenic for example, it does not mean that animals are generally sad.

Since the issue is complex, it's not an easy one to use to bend public intuitions.
You can read my personal blog here: CareyRyan.com

User avatar
Brian Tomasik
Posts: 1107
Joined: Tue Oct 28, 2008 3:10 am
Location: USA
Contact:

Re: Total versus average happiness

Post by Brian Tomasik » Mon Jan 30, 2012 7:42 am

Thanks for the links, Ruairi. :)

The Softpedia reference has some interesting stories, but I wish it cited sources. At least half of the anecdotes could be explained in other ways because we know so little about the situation. For example, "But the female was indifferent to his advances, and one day the tomcat threw himself through the windows and died. The owner said because of the unshared love..." It seems plausible that the cat threw himself through the window for some reason (maybe just accident), and the owner attributed it to unshared love. Similarly, in the following story, the dog could have been just dumb: "Franz, a German shepherd dog, was laying on the railways line, near the railway station. Workers always chased away with stones the dog, but soon after the dog returned, and one day, the dog met with the Verona-Bologna train..."

However, others -- if true -- are more suggestive of suicidal intent. E.g.:
But this attachment of the dogs is also applied to their human masters. This story occurred in Rome: the owner of Shastra, a Spanish cockerel, died. When the corpse was pulled out of the house, the dog tossed itself from the third level. It just broke one leg. It was brought to the veterinarian but once again home, the pulled out itself from the leash and threw itself again.

This time it died. Perhaps, the places in which the dog played so many times with its master could have recalled the dog such painful records that it could not resist and suicided.
The article also confirms my suspicion: "Somehow, this behavior is linked to self-conscience, thus only 'brainy' animals, like mammals and some birds, consciously commit suicide." Of course, this claim is subject to reporting bias because it's easier to see such behavior in big animals with whom humans have regular interaction.
RyanCarey wrote:It's worth noting that for better of for worse, the popular conception is that suicide is an irrational act.
I think it sometimes is and sometimes isn't irrational in humans. However, the original claim was that at least when it is rational for animals to kill themselves, they will do so (whether or not they also do so when it's not rational).

I argued against that claim on the basis that most animals aren't capable of deliberate suicide, and in any event, there are other evolutionary impulses that may irrationally impede self-killing. For that matter, many humans are irrational not to kill themselves as well. (Unless they're like Hamlet and fear punishment in the afterlife for doing so.)

User avatar
Ruairi
Posts: 385
Joined: Tue May 10, 2011 12:39 pm
Location: Ireland

Re: Total versus average happiness

Post by Ruairi » Mon Jan 30, 2012 7:53 pm

ah yeah but then if we say that animals in nature generally have bad lives but generally dont commit suicide because they're not acting in a rational preference-optimization way then how can we trust what they try to achieve as what actually makes them happy?

as in surely theres an argument there that how can you then know what non-human animals want if we cant trust the things they try to achieve as what they want?

ugh sorry, i mean if we currently interpret "the tiger tries to eat the antelope" as "the tiger wants to eat the antelope"

then surely it follows that "the animal is not commiting suicide" as "the animal doesn not want to commit suicide"

and if we say maybe they would prefer suicide then maybe they'd prefer not to eat antelope too?

User avatar
Hedonic Treader
Posts: 328
Joined: Sun Apr 17, 2011 11:06 am

Re: Total versus average happiness

Post by Hedonic Treader » Mon Jan 30, 2012 8:45 pm

Ruairi wrote:ugh sorry, i mean if we currently interpret "the tiger tries to eat the antelope" as "the tiger wants to eat the antelope"
That doesn't win you much. Neither are equal to "The tiger has performed a rational expected utility consideration and concluded that his total future experiences will be best if he now eats the antelope". You can trigger feasting behavior patterns by stimulating parts of the brain, I think it was in the hypothalamus. You can do it both in humans and in other animals. Ironically, it was originally confused with aggression because of the non-human animals' resulting predatory behavior.

Now of course, eating feels good, but there's probably no rational calculation of indirect effects involved, like prevented suffering from earlier death caused by not feeding now. Humans can do this, but it's doubtful for most other animals. Some of them do prepare for the future (e.g. birds stashing food) and think strategically in some sense, but on the other hand, you have behavior like chimps predictably grabbing the greater visible food reward even if they know from experience that it will result in a smaller total food reward being given to them. Sapolsky attributed this to a lack of impulse control.
"The abolishment of pain in surgery is a chimera. It is absurd to go on seeking it... Knife and pain are two words in surgery that must forever be associated in the consciousness of the patient."

- Dr. Alfred Velpeau (1839), French surgeon

DanielLC
Posts: 707
Joined: Fri Oct 10, 2008 4:29 pm

Re: Total versus average happiness

Post by DanielLC » Tue Jan 31, 2012 3:16 am

as in surely theres an argument there that how can you then know what non-human animals want if we cant trust the things they try to achieve as what they want?
You can't trust whether or not someone wants something they've never experienced, like death, no matter what they do. You can only tell if they want something by how their behavior changes once they get it.
Consequentialism: The belief that doing the right thing makes the world a better place.

User avatar
Brian Tomasik
Posts: 1107
Joined: Tue Oct 28, 2008 3:10 am
Location: USA
Contact:

Re: Total versus average happiness

Post by Brian Tomasik » Wed Feb 01, 2012 8:57 am

Hedonic Treader wrote:
Ruairi wrote:ugh sorry, i mean if we currently interpret "the tiger tries to eat the antelope" as "the tiger wants to eat the antelope"
That doesn't win you much. Neither are equal to "The tiger has performed a rational expected utility consideration and concluded that his total future experiences will be best if he now eats the antelope".
Yeah, I think HT is right. A lot of behaviors are reflexive and not controlled by the conscious brain. Moreover, we often do things that we regret (like HT's chimp example) because we can't do what's best for the long term. For example, imagine that you're a sickly mouse wondering whether you should jump from a 1,000-meter cliff to kill yourself. Rationally, maybe, but you might not be able to overcome your short-term fear in order to carry it out. I probably wouldn't be able to do it even as a human.

Finally, I would guess that animals like frogs, minnows, and ants aren't smart enough to understand suicide. Rational choice doesn't tell us about options which aren't understood. If I were smarter, I could win $1 million by solving one of the remaining Clay Mathematics Institute's Millennium Prize Problems. The reason I don't is not because I'm making an irrational hedonic choice.

Ubuntu
Posts: 171
Joined: Tue Sep 07, 2010 1:30 am

Re: Total versus average happiness

Post by Ubuntu » Sun Mar 04, 2012 5:43 pm

ReX wrote:I was contemplating the Total versus Average happiness problem.
It Total happiness is all that matters and you believe that a human always reaches minimum happiness (wellbeing, more so than say, a pig can ever hope to experience
it is better to be a human being dissatisfied than a pig satisfied; better to be Socrates dissatisfied than a fool satisfied
), and I do believe this is (provided you have the same amount of food and shelter Socrates had) the case. Up until a few years ago there was enough food in the world to feed everybody (it was just distributed unequally). Perhaps this has changed since the food riots (but the reason would then be ethanol subsidies and burning food instead of eating it). However, if you stick to the container view of life (cfr. Shelly Kagan lectures on death), supposing that all life has a minimal amount of worth making it a 'gift', then there's an absolute anyhow and either way (creating an absolute minimum of hedons in the world which didn't exist prior to the entity being [born]). We would thus be in a state where, not only would it be bad to kill anyone (even people in chronic pain/depression), we would have to create as many people as women (and men, but that takes less effort) would be able to bare.
If happiness and suffering are all that matter then life has no 'minimal amount of worth' beyond it's being instrumentally beneficial in maximizing happiness and/or minimizing suffering (for the subject of that life as well as for other sentient beings in general). It would not be wrong (according to HU) to kill people who are in chronic pain if there is no possible way to improve their standard of living to the point of it being an overall good experience and there was no alternative that would lead to the same, or greater, benefit with less cost. There's no moral value in creating sentient life just for the sake of creating more sentient life if doing so doesn't raise the 'net' balance of pleasure over pain (I'm still not sure what I think about aggregation but for now, I might be leaning toward it). Resources set aside for not yet existing people could benefit existing humans (and other sentient beings) instead and considering humanity's current collective carbon footprint and it's ecological consequences, I think lowering the human population (through not reproducing, not killing anyone or even allowing them to die through disease, poverty or natural disaster), to less than one billion, would probably increase the net balance of pleasure over pain compared to creating more people in an overpopulated world (and the world is 'overpopulated' when you consider our impact on the environment if not in terms of actual available land and resources).

-Edit : I think I misunderstood what you meant by all lives (when basic necessities aren't an issue) having a necessary 'minimal amount of worth' but either way, I don't see how increasing the population necessarily raises the sum total of happiness even if there is a guaranteed minimal amount of happiness that all fed and relatively healthy humans have, and I don't believe there is.

Also, although I would argue otherwise, an HU could claim that, for whatever reasons, most humans have the potential to feel a greater amount of pleasure than pigs are capable of but an HU can't consistently argue that it would be better to be a 'human dissatisfied than a pig satisfied'. The pleasure that all sentient beings feel is qualitatively equal, if not quantitatively.
The flipside, Average happiness, would require us to kill everybody who's below the average (or mean, because the average person usually doesn't exist) and eventually we'd have to kill the last person on earth.
Personally, I believe if the government has a duty, it would be to provide for every person on this planet in the basis necessities and those seem to be food and perhaps some shelter, education would be nice but indoctrination lurks at every corner and refusing a meal (say, hungerstrike) would be a right just as much as refusing to be confined to a certain space (vagabonding is no longer illegal in my country for decades now). Nevertheless, the option should be there. Assuming people will get these basic needs (through sharing, sympathy, stealing, charity, whatnot -I included economy and religion in the original list but on closer inspection that's no more realistic than expecting governments to do it-), we'd have the moral imperative to create extra people.
For a long time I didn't want to create extra people in this overpopulated world (nor kill, but let's postone the pacifism debate if possible). The reasoning was that famine is the flipside of population > food supply. Can we really postpone creating more people until there isn't a single person starving (or dying of cold/heat)? And if you're setting up the system, that is to say, if you can do the (moral) calculus of how much resources it would take for a certain amount of people; and you'd find that we've gone beyond our limits, is there an imperative to (not only stop creating new life, but also to) end human lives already in existence?
Can anybody suggest some reading on this topic? Thanks.
Total happiness makes sense to me because it refers to happiness that actually exists. The 'average' anything of a population is just a statistical convenience we use to understand what most people have or how most people are, there are no 2.4 children walking around.

DanielLC
Posts: 707
Joined: Fri Oct 10, 2008 4:29 pm

Re: Total versus average happiness

Post by DanielLC » Sun Mar 04, 2012 6:45 pm

and the world is 'overpopulated' when you consider our impact on the environment
You weren't considering our impact on the environment. You were considering the impact of the marginal population on the rest of the population. The impact on the environment has nothing to do with if people would be happier overall if there were few of them.

Also, unless billions of lives aren't worth living, I find it very unlikely that getting rid of five sixths of them would raise the happiness of the remainder by a factor of six.
Total happiness makes sense to me because it refers to happiness that actually exists. The 'average' anything of a population is just a statistical convenience we use to understand what most people have or how most people are, there are no 2.4 children walking around.
I figure that "you" are randomly one person from the population. As such, your expected happiness is the average happiness. What's the point of having ten times as many people if that just means you're a tenth as likely to be any given one?

I don't entirely buy this argument, but I do find it fairly convincing.

I think the biggest problem with average utilitarianism is that the amount of utility in a universe with no people is undefined. We're not in such a universe, but I still don't like the idea of a moral system with holes in it.
Consequentialism: The belief that doing the right thing makes the world a better place.

Ubuntu
Posts: 171
Joined: Tue Sep 07, 2010 1:30 am

Re: Total versus average happiness

Post by Ubuntu » Sun Mar 04, 2012 7:59 pm

and the world is 'overpopulated' when you consider our impact on the environment

You weren't considering our impact on the environment. You were considering the impact of the marginal population on the rest of the population. The impact on the environment has nothing to do with if people would be happier overall if there were few of them.
You've probably heard that we would need 5-6 Earths to sustain us if every one lived the typical American lifestyle. Around 80% of all human beings live on less than 10$ a day which suggests that the standard of living for most people is nowhere near the standard of living for most people in developed countries (although people who don't live in absolute poverty can still live relatively happy lives, I'm sure). If poverty, disease and loss of life due to natural disasters is eradicated, which it should be, the overall human carbon footprint would rise drastically, many researchers agree that the world will be unrecognizable by the 22nd century even if we were to stop emitting green house gases today, and this would lead to an increase in tropical diseases, poor crops, tsunamis, the extinction of entire species etc. Lowering the (human) population will help prevent even further global warming which will result in a higher standard of living for a smaller population of humans than would exist for a larger one, I think. Besides ecological reasons, there are already countless orphans who would be better off if resources set aside for non-existing children were given to them instead. Why bring 1 happy child into the world when you could help a less advantaged child which would increase the same benefit while also reducing harm?
Also, unless billions of lives aren't worth living, I find it very unlikely that getting rid of five sixths of them would raise the happiness of the remainder by a factor of six.
I'm not saying that anyone who exists should be 'gotten rid of', only that it would be better if most people stopped having children or at least limited themselves to one. Most people live incredibly miserable lives, I think, so the idea that drastically lowering the population would increase the net balance of pleasure over pain isn't far fetched to me.
Total happiness makes sense to me because it refers to happiness that actually exists. The 'average' anything of a population is just a statistical convenience we use to understand what most people have or how most people are, there are no 2.4 children walking around.
I figure that "you" are randomly one person from the population. As such, your expected happiness is the average happiness. What's the point of having ten times as many people if that just means you're a tenth as likely to be any given one?
My 'expected' happiness is non-existent, it has nothing to do with what my actual happiness will be. The fact that it's hard to calculate what someone's actual happiness will be is besides the point. Sometimes I get the impression that many utilitarians don't seem to consider that there are actually right and wrong decisions when it comes to producing the greatest surplus of pleasure over pain regardless of what's probable or expected, that might be off on a tangent.

If I'm understanding you correctly (what's the point of having ten times as many people if it means being only a tenth as likely to be given one), the objective of HU isn't to increase the likeliness that I will experience happiness, it's to produce the greatest possible balance of happiness over suffering as is possible.


I don't entirely buy this argument, but I do find it fairly convincing.

I think the biggest problem with average utilitarianism is that the amount of utility in a universe with no people is undefined. We're not in such a universe, but I still don't like the idea of a moral system with holes in it.
I don't understand the logic behind average utilitarianism and I'm not sure how to point out how clearly flawed I think it is (although my reasoning could be flawed or I may be misunderstanding it). In universe A, 8 people feel 90 points of pleasure each but 2 people with 200 hundred 'hedons' raise the average to 112 even though the actual sum pleasure in this universe, if aggregation makes sense, is 1120. There isn't 112 points of pleasure in this universe, there's 1120. The objective of HU is to maximize actual pleasure because actual pleasure is what has actual value. The 112 gives us an idea of how happy most people are, it doesn't tell us how much happiness exists in this universe which is what HU is concerned with, not how happy most people are. Killing less happy people to raise the average level of happiness is bizarre since it won't actually raise happiness, no one will actually be happier as a result. It might increase your chances of being happy but not in anyway that will actually make you happier. For example, you can say that there's a 60% chance that X will occur and then you find out some information that pushes the likeliness of X occurring up to 75%, but X is or isn't going to occur regardless of what you think the "chances" of X occurring are. X was already determined over 13 billion years ago, saying that it's likely or unlikely to occur is just making a prediction based on the knowledge we have.

User avatar
Arepo
Site Admin
Posts: 1097
Joined: Sun Oct 05, 2008 10:49 am

Re: Total versus average happiness

Post by Arepo » Tue Mar 06, 2012 1:40 pm

Average utilitarianism can give you a situation with multiple happy people in which killing one of the less happy people, making everyone marginally less happy in the process (but one of them only momentarily so, before he dies), gives you a greater average.
"These were my only good shoes."
"You ought to have put on an old pair, if you wished to go a-diving," said Professor Graham, who had not studied moral philosophy in vain.

User avatar
Pablo Stafforini
Posts: 174
Joined: Thu Dec 31, 2009 2:07 am
Location: Oxford
Contact:

Re: Total versus average happiness

Post by Pablo Stafforini » Tue Mar 06, 2012 8:35 pm

Arepo wrote:Average utilitarianism can give you a situation with multiple happy people in which killing one of the less happy people, making everyone marginally less happy in the process (but one of them only momentarily so, before he dies), gives you a greater average.
Average utilitarianism has even more implausible implications. For instance, consider a world A in which people experience nothing but agonizing pain. Consider next a different world B which contains all the people in A, plus arbitrarily many people all experiencing pain only slightly less intense. Since the average pain in B is less than the average pain in A, average utilitarianism implies that B is better than A. This is clearly absurd, since B differs from A only in containing a surplus of agony.

A while ago Brian mentioned that Eliezer Yudkowsky endorses average utilitarianism. I wonder if he is aware of these difficulties.
Last edited by Pablo Stafforini on Wed Mar 07, 2012 3:11 am, edited 1 time in total.

Post Reply