Further Thoughts on Utilitarianism

Originally I had planned on a post arguing that the best way to understand liberalism in all its various permutations is by equating it with utilitarianism.  But I thought it might be better to break it into two posts, to keep the length more manageable.  I’ll do the liberalism part next Sunday.  Keep in mind that today’s post is intended to be a defense of utilitarianism from the liberal perspective; it is not aimed at objections that non-liberals might have.  Thus I don’t address dogmatic libertarian arguments against progressive taxation.

1.  When I talk to people in the humanities, I almost always run into the complaint that utilitarianism focuses on aggregate utility, with no consideration of equality.  Perhaps the utilitarian policy criterion doesn’t explicitly mention equality, but utilitarianism is actually a radically egalitarian value system in two different senses.  The first is conceptual, utilitarianism treats the well-being of every human being equally (and Peter Singer would even include animals in this calculus.)  It doesn’t merely say “love thy neighbor,” it says that one should care just as much about the well-being of a stranger in a far away country, as one’s family or friends.  No favoritism is to be shown to people of different nationalities, races, genders, sexual preferences, or economic class.   Very few people, even very few liberals, are quite this unbiased.  For instance, in their proposals for domestic and international transfer programs, most liberals act as if they care far more about the poor in their own country than the poor in other countries.

The second argument is more pragmatic.  Although utilitarianism would theoretically allow for a highly unequal distribution of wealth, as a practical matter it almost certainly favors egalitarian distributions (other things equal), as an extra dollar is probably valued by a poor person much more highly than by a rich person.  Thus rather than viewing egalitarianism as a separate liberal value, I believe that it makes more sense to view it as an implication of utilitarianism.

Consider the policy implications of utilitarianism.  Because it is difficult to measure utility directly, and because most people assume that the marginal utility of money is higher for the poor than the rich, utilitarianism implies that no income inequality is justified under any grounds other than efficiency.  And yet when I talk to even very left wing people about income inequality, they almost always say that some differences are deserved.  Thus someone who studied 15 years to become a brain surgeon deserves more income than a kid working at McDonalds.  But in utilitarianism there is no “deserves,” there is only utility, and questions about how economic incentives affect total utility.  It is certainly easy to justify income inequality within utilitarianism, but only to the extent that “supply-side” effects are important.  Judged solely on the basis of who deserves what, utilitarianism is every bit as radically egalitarian as the most extreme Maoism.

Some have argued (I believe John Rawls, for instance) for a liberalism where public policies can only be adopted if they do not hurt the worst off members of society.  Although appealing at first glance, on closer inspection this principle is deeply flawed.  Virtually all policies have at least an infinitesimal effect on the lowest classes of society, for better or worse.  Thus in practice, a “maximin principle” would simply revert to a policy test where the well-being of the worst off was the only consideration that mattered””which is obviously at variance with utilitarianism, and also with common sense.

Consider a policy that had a tiny adverse effect on the lowest class (equivalent to the disutility of a mosquito bite), but made everyone else (including the slightly less poor) massively better off.  Isn’t that policy desirable?  If you want to say my “thought experiment” has no real world counterpart, fine, but then find me a real world example of a policy that would pass the utilitarian test, but be rejected on the maximin principle.  Liberals generally focus on the concerns of the lower classes, as they should under the utilitarian criterion (as there is more possibility of improving their well-being), however they do not follow the extreme maximin principle.

I’m not quite sure why so many intellectuals seem to view utilitarianism as a highly inegalitarian moral philosophy, when it’s clear (to me) that exactly the opposite is true.  Have they unthinkingly accepted some critique they learned in grad school?  Or do they associate utilitarianism with economic liberalism (Bentham, Mill, etc) and then assume that any value system that led to classical economics must be inherently unfair?  [I find their apparent unanimity on a question where they are so off base to be unnerving, like the pod people in Invasion of the Body Snatchers.]

2.  A second criticism is based on the fact that it would allow, in principle, for some societal set-ups that shock our moral sensibilities.  Yes, but only in principle.  Since I addressed this issue in a post last week, I won’t say anything more today.

3.   Some critics of utilitarianism advocate a deontological approach where concepts such as human freedom and dignity represent ends in themselves, not merely a means to an end.  If so, one cannot justify denying someone liberty on consequentialist grounds.  I don’t know if these “natural rights” really are “nonsense on stilts” (as Bentham argued), but I don’t believe many liberals actually believe in such abstractions.  Society always faces ethical trade-offs that require the (utilitarian) weighing of costs and benefits.  Thus most American liberals accepted the need for a military draft to help defeat the Nazis, despite the fact that FDR’s decision essentially sacrificed the freedom of young men for the greater good.  (And as far as dignity is concerned, picture the life of a draftee in boot camp.)  My point isn’t that the military draft is justified, but rather that it is not obviously unjustified when a country is fighting for it’s life, and that abstract principles like freedom and the rights of minorities are much less obvious when one is confronted with real world scenarios.  Next week I will discuss a whole range of public policies that liberals support on utilitarian grounds, which violate various “natural rights.”

4.   Robert Nozick argued that according to utilitarianism people should be willing to abandon reality and have their brain hooked up to a “happiness machine,” which fed in constant pleasure.  He argued that most people would actually prefer reality, even at the cost of some happiness.  I have already expressed my methodological objection to working through philosophical issues with one’s moral intuition.  Now I will provide some evidence for this skepticism.

A recent experiment described by Joshua Knobe on bloggingheads.tv showed that people do respond as Nozick expected when his hypothetical scenario was described to them.  However the experimenters then inverted the thought experiment and told people to imagine they had just woken up on a laboratory table surrounded by doctors who told them that what they remembered of their life up until now had been a dream.  They were asked to imagine being told that they were “really” just a regular person with a regular job in a totally different reality.  They were then asked if they wanted to stay in this new “real world” that they were unfamiliar with, or go back into their dream world surrounded by all their familiar family and friends.  Not surprisingly, in this case people preferred to go back to the dream world.  So Nozick’s intuition was not correct; it’s not that people prefer reality to dreams, it’s that they prefer the familiar to the unfamiliar.

[BTW, Knobe reported another experiment where individuals were asked whether people should be held responsible for their sins in a deterministic universe.  The answer was usually “no.”  But when given an example of a particularly egregious sin, people change their view, even with determinism still assumed.  As I argued with the earlier examples involving slavery and Nazis, moral intuitions involving highly emotional examples are not a reliable way to develop broad philosophical principles.]

5.  Some argue that there is much more to life than happiness, and that the utilitarian criterion is too narrow.  Nietzsche famously said something to the effect that “man does not seek happiness, only the Englishman does.”  But what is utility?  Is it happiness?  Subjective well-being?  Life satisfaction?  Peter Singer argues for a “preference utilitarianism”, which would aim at maximizing those preferences that individuals would prefer if fully informed and reflective.  Of course, there is the problem of who gets to decide which preferences are valid when confronting public policy issues.  Even so, preference utilitarianism has the virtue of addressing one major criticism of Benthamite (hedonic) utilitarianism””that people are complex and seek more than one objective in life.

A few years ago I read a book by Daniel Gilbert entitled Stumbling on Happiness.  It’s been a while, but I seem to recall that he had trouble defining happiness and at one point sort of threw up his hands and stated it was something like “that which we seek.”  Well I suppose that definition would make it easy to resolve the age-old dispute about whether people “really maximize utility.”  Another approach would be to look at how people actually use words.  Thus sometimes happiness is equated with hedonism, but just as often one might hear a phrase like the following:  “The only occasions Mother Theresa was truly happy was when she was helping the poor.”  I think it’s fair to say that hedonism doesn’t capture all the various meanings of the term ‘happiness.’  And yet if one defines utility too broadly (say the total well-being of society), then utilitarianism veers dangerously close to being a tautology.

To conclude, I do think that utilitarianism is a reasonable criterion for policy evaluation.  But even I have my doubts.  In my earlier post on political art I argued that the narrative arts often implicitly endorse liberal (utilitarian) values.  And even when you have conservative art that exalts something like patriotism, it generally does not offer a full blown assault on liberalism.  The only examples I can currently recall of what seemed like an explicit attack on liberalism/utilitarianism are the books A Clockwork Orange and Brave New World.  Because I read these more than thirty years ago I may have muddled the message, but I seem to recall that the first worried that the search for utility maximization would turn men into machines, while the second worried about turning men into children.  (Someone correct me if my memory failed me.)

My knowledge of Nietzsche’s writing is very superficial, but my hunch is that he intuited something deeply inhuman about a value system that ignores much of what it means to be human.  There are undoubted benefits from rising above our basest motives, but also costs. There are a few cases like mandatory seat belt laws where even I take the anti-utilitarian position for “dignity” reasons.

Robin Hanson recently argued that economists should be honest brokers on questions of efficiency.  So I guess if I was an economist advising the Governor I would tell him that a mandatory seat belt law would make Massachusetts a happier place.  And then as a voter I’d go out and vote in favor of a referendum to repeal such a law in Massachusetts.

[I believe we are the only state where people can’t be pulled over for not wearing a seat-belt, and we have the lowest rate of seat-belt usage in the country.]

PS:  My lukewarm endorsement of utilitarianism should not be construed as endorsing the silly view that some economists have about people being selfish.  As I argued in an earlier post, a world of selfish people would be too nightmarish to contemplate.


Tags:

 
 
 

12 Responses to “Further Thoughts on Utilitarianism”

  1. Gravatar of TGGP TGGP
    19. April 2009 at 12:46

    I haven’t read Nozick’s book, but if I understand the thought experiment correctly, I’d gladly enter the machine. I find it plausible that I am in such a machine right now.

    The bit about preferences being sufficiently well-informed & reflective sounds suspiciously like it could be used for smuggling in “false consciousness” (an idea I have a very low opinion of) through the back-door.

  2. Gravatar of Philo Philo
    19. April 2009 at 13:08

    A few comments on your reflections on utilitarianism:

    “Peter Singer would even include animals in th[e utilitarian] calculus,” and his argument for doing so–for rejecting “speciesism”–is irresistible. It is merely self-serving prejudice to assign value to the lives of members of our species while not counting the lives of others *at all*.

    But the following can be accepted only with major reservations: “it says that one should care just as much about the well-being of a stranger in a far away country, as one’s family or friends”? At the level of abstract principle, yes, but (obviously) not in everyday practice. There is a good utilitarian reason why I (in a sense) *care more* about my own well-being than about that of any distant stranger: I know my own circumstances, I know what I want, I am naturally motivated (without a lot of psychic strain) to act so as to satisfy my wants, and *I am in a relatively good position* to succeed. With a distant stranger none of these conditions hold. (This line of thought does not justify *complete selfishness*, but it does justify *somewhat* self-centered “caring,” in the ordinary sense of the term.)

    “Charity begins at home.” If I tried to give equal attention to satisfying the wants of *every sentient being now alive*, not to mention future such beings, I would fail to satisfy my own wants. I would quickly expire, and thenceforth be incapable of *any* action–not a satisfactory outcome from the utilitarian perspective.

    Once *speciesism* has been abandoned, the principle that “the marginal utility of money is higher for the poor than the rich” must go, too. An extra dollar devoted to improving the lot of, say, an ant may well buy less utility/happiness/pleasure than an extra dollar devoted to a human being, even a rich one. (It’s not “speciesist” to hold that the average human being has a better, richer life than the average member of any lower species now present on earth: it’s just an empirical fact.) Even in limited form–limited to human beings–the Diminishing Marginal Utility principle is only *very roughly* true. And (I should say) its importance is overwhelmed by dynamic considerations. Utilitarianism demands a long-run view, rather than a short-sighted concentration on the present (i.e., on the *very near* future). Allocating resources for future production must be more important than equally distributing resources among those human beings who happen to be alive now.

    “I’m not quite sure why so many intellectuals seem to view utilitarianism as a highly inegalitarian moral philosophy, when it’s clear (to me) that exactly the opposite is true.” I think both you and the “many intellectuals” are losing sight of the extreme *abstraction* of utilitarianism. To get *any* practical conclusion from the theory you need to add massive amounts of empirical information. Accepting utilitarianism does not, *in and of itself*, carry one very far in politics.

    Your reservations about relying on “moral intuition” are spot on.

  3. Gravatar of Philo Philo
    19. April 2009 at 15:48

    A few comments on your reflections on utilitarianism:

    “Peter Singer would even include animals in th[e utilitarian] calculus,” and his argument for doing so-for rejecting “speciesism”-is irresistible. It is merely self-serving prejudice to assign value to the lives of members of our species while not counting the lives of others *at all*.

    But the following can be accepted only with major reservations: “it says that one should care just as much about the well-being of a stranger in a far away country, as one’s family or friends”? At the level of abstract principle, yes, but (obviously) not in everyday practice. There is a good utilitarian reason why I (in a sense) *care more* about my own well-being than about that of any distant stranger: I know my own circumstances, I know what I want, I am naturally motivated (without a lot of psychic strain) to act so as to satisfy my wants, and *I am in a relatively good position* to succeed. With a distant stranger none of these conditions hold. (This line of thought does not justify *complete selfishness*, but it does justify *somewhat* self-centered “caring,” in the ordinary sense of the term.)

    “Charity begins at home.” If I tried to give equal attention to satisfying the wants of *every sentient being now alive*, not to mention future such beings, I would fail to satisfy my own wants. I would quickly expire, and thenceforth be incapable of *any* action-not a satisfactory outcome from the utilitarian perspective.

    Once *speciesism* has been abandoned, the principle that “the marginal utility of money is higher for the poor than the rich” must go, too. An extra dollar devoted to improving the lot of, say, an ant may well buy less utility/happiness/pleasure than an extra dollar devoted to a human being, even a rich one. (It’s not “speciesist” to hold that the average human being has a better, richer life than the average member of any lower species now present on earth: it’s just an empirical fact.) Even in limited form-limited to human beings-the Diminishing Marginal Utility principle is only *very roughly* true. And (I should say) its importance is overwhelmed by dynamic considerations. Utilitarianism demands a long-run view, rather than a short-sighted concentration on the present (i.e., on the *very near* future). Allocating resources for future production must be more important than equally distributing resources among those human beings who happen to be alive now.

    “I’m not quite sure why so many intellectuals seem to view utilitarianism as a highly inegalitarian moral philosophy, when it’s clear (to me) that exactly the opposite is true.” I think both you and the “many intellectuals” are losing sight of the extreme *abstraction* of utilitarianism. To get *any* practical conclusion from the theory you need to add massive amounts of empirical information. Accepting utilitarianism does not, *in and of itself*, carry one very far in politics.

    Your reservations about relying on “moral intuition” are spot on.

  4. Gravatar of ssumner ssumner
    19. April 2009 at 17:38

    Dear TGGP, Yes, I agree with both of your points. Singer’s point probably makes sense for someone who is clearly insane (but even that is a tricky issue), but otherwise we are probably generally better off letting people decide for themselves what is best for them. BTW, couldn’t one view anti-depressants as a sort of “happiness machine?”

    Philo, I also agree with much of what you say. My point about strangers on the other side of the world was only in principle, I agree that we are much better placed to help those close to us. Indeed, I think we often overestimate the extent to which we are even able to help those close to us. I’m not sure Washington bureaucrats can do much for those living in Cuidad Juaraz, but I’m also not convinced they can do much for those living across the river in El Paso. And I still think my “double standard” view of Western liberals is roughly correct. Even if we don’t know a lot about people in Bangladesh, does anyone serious doubt that just flying over that country with a Boeing 747 dumping Federal Reserve notes out the window would do more in strictly utilitarian terms than most domestic programs? Does anyone think nationalism is NOT the reason why we would never consider replacing our current unemployment comp. system with this alternative?

    I am skeptical of programs to massively redistribute income, although I do think we would be better off replacing some existing government programs (like the minimum wage, welfare, food stamps, unemployment comp., etc) with an hourly wage subsidy for low wage workers. My skepticism about income redistribution is not based on the conservative “undeserving poor” concept, but rather on supply-side considerations. Thus I have the same utilitarian value system as many American liberals, just a somewhat different view about the incentive effects of high implicit marginal tax rates. I also think that much of what is viewed as “needs” could actually be met by a Singapore-style forced saving program. I read that they only spent about 1% of GDP on transfers. Much of social spending in the Nordic countries merely goes in a circle from taxpayers to the government and back to the very same taxpayers. On the other hand I don’t have the ideological opposition to redistribution that many libertarians have–if it can work we should do it. The idea that we each earned our income “all by ourselves” is nonsense–without the cooperation of those around us we’d all be living in caves.

    Utilitarianism by itself doesn’t begin to solve complex societal problems, rather it simply provides a framework for evaluating cost/benefit estimates from economic studies. One can be a utilitarian and be a Maoist, or a radical libertarian–it all depends on one’s estimates of elasticities. Guess wrong and you get the Great Leap Forward.

    I don’t have strong views on other species, as I have no idea how their minds work. I do think, however, that if we start treating each other with more consideration, it is almost inevitable that we will at least try to treat animals better as well. I’ll leave it to others to figure out why these two trends are related.

  5. Gravatar of Giedrius Giedrius
    20. April 2009 at 00:25

    Great post. A couple of things that came to my mind:
    1. What about interpersonal utility comparison?
    2. I am not an expert in this area, but did someone prove that Sam Peltzman was wrong regarding seat belts?

  6. Gravatar of Brandon Brandon
    20. April 2009 at 04:11

    But in utilitarianism there is no “deserves,” there is only utility, and questions about how economic incentives affect total utility.

    Unless you are dealing with a very deliberately limited selection of utilitarianisms, this is certainly false; there are plenty of utilitarianisms where there is a ‘deserves’, because there are utilitarians think that desert can be analyzed in terms of utility. This is my worry throughout your entire argument; you keep talking as if ‘utilitarianism’ were a unified position (with the single exception of distinguishing preference and hedonic forms), whereas, in fact, there are lots of different kinds, and one cannot always talk about them as if they agreed even on fundamentals. There are almost certainly utilitarianisms that imply liberalism — I think Mill’s is deliberately set up to do so, for instance — but there are others for which this claim seems very implausible. And of course the major research puzzle for utilitarians since the nineteenth century has been finding a utilitarianism that maximizes the advantages of all the very different kinds of utilitarianism, so utilitarianism has always been in a state of change and development. Thus everyone, for and against, should be taking the trouble to specify which kinds of utilitarianism are really in view in any given argument.

  7. Gravatar of Asymptosis » Win-Win Utilitarianism: The Greatest Good For the Greatest Number Asymptosis » Win-Win Utilitarianism: The Greatest Good For the Greatest Number
    20. April 2009 at 10:23

    […] Sumner at The Money Illusion has a very interesting (followup) post on Utilitarianism–the doctrine (as here described on Wikipedia) that “the moral worth of an action is […]

  8. Gravatar of Steve Roth Steve Roth
    20. April 2009 at 10:26

    Scott, lots of good thinking here.

    I’d like to propose an alternative to (the obviously problematic) pareto-efficient utilitarianism, which I fondly call win-win utilitarianism. I needed links and formatting to write it up properly, so I’ve posted it here:

    http://www.asymptosis.com/win-win-utilitarianism-the-greatest-good-for-the-greatest-number.html

  9. Gravatar of ssumner ssumner
    20. April 2009 at 13:43

    Giedrius, In principle interpersonal comparisons matter, but I argue that policymakers cannot easily make those comparisons. At best they can assume that a dollar is valued more by the poor than the rich. (This assumption has some support from the tendency of people to buy life insurance.)

    Recently I haven’t heard much about the Peltzmen study, which makes me think it’s findings haven’t held up. But obviously I hope I am wrong, it would provide extra support for my opposition to seat-belt laws.

    Brandon, By utilitarianism I meant the value system that advocates maximizing total utility. I did spell out some of the difficulties (should animals be included) but you’re right that in my discussion of economic policy I ignored many of these complexities, and basically focused on a simple version that is roughly maximizing total human happiness. Despite my defense of utilitarianism, I can’t say I’m very impressed with the idea. It’s just that I don’t find many anti-utilitarian arguments to be particularly persuasive either. That was my main point.

    Steve, That’s an interesting web site, but I do have doubts about your argument that more equality promotes economic growth. All of these variables are highly endogenous, and I argued earlier (in my Denmark post) that cultural values may cause both equality and growth. I doubt whether, holding culture constant, progressive taxes would cause more growth. Also, the graph on wealth equality has only 7 countries, and Sweden shows up as the most unequal. That’s not going to convince many people. What is the implication; that we should be more like Sweden, or as different from Sweden as possible?

    I hope this doesn’t sound too negative, I do think that if cultural attitudes become more egalitarian, that would strongly promote economic growth. So I do think you are on to something, I just see culture rather than tax rates as being the key force.

  10. Gravatar of nerdbound nerdbound
    21. April 2009 at 01:13

    Watchmen is another great anti-Utilitarian work.

  11. Gravatar of ssumner ssumner
    22. April 2009 at 16:40

    Thanks nerdbound, I’ll try to catch it.

  12. Gravatar of Winton Bates Winton Bates
    25. April 2009 at 20:00

    I think there is a lot to be said for the views of people like Henry Hazlitt and Leland Yeager on utilitarianism i.e. that the appropriate objective is to foster peaceful cooperation among individuals pursuing their own diverse goals in life.

Leave a Reply