Thanks very much for the Trout reference–his paper looks very interesting. While I do incline to the view that people have gotten less happy in recent decades, particularly in relation to stress, this is only a suspicion that could well prove wrong, and I wouldn’t stake a lot on it at this point. (I’ll be circulating a paper extending the ideas in that paper later this spring.) Glad you’re pushing us to take a more optimistic view seriously, since we need to have a real debate on these matters and not let our prejudices drive the inquiry!

I’m also inclined to agree with your points about physiological measures; while there’ll be issues figuring out what to say when they diverge from self-reports, I expect we’ll learn a lot from them.

I’m glad you found my blog! Like I said in the comments, and a previous post, I really like this paper. It is far and away the best thing I’ve found regarding the problems of self-reports. Do you know J.D. Trout at Loyola Chicago? He’s working on a book on happiness, and as far as I can tell, he’s not very skeptical of self-reports. I’d love to see the two duke it out on this score.

I realize (especially after this comments discussion) you were making a pretty modest claim about paternalism. I’ve got a larger project in the works on what a good consequentialist argument for paternalism based on cognitive limitation would have to look like, and the core of it, as you can see, is an empirical comparison of the means-ends reliability of government institutions against individual decision making. I can’t blame you (or anyone) for not stepping carefully around one of my intellectual pet peeves.(But on my blog, I fixate on them!) The whole point of the project is to help people see that cognitive limitations have no particular implications for paternalism absent actual evidence for the reliability of paternalistic policy. I think it really is fair to expect symmetrical treatment of minds and institutions, and fair to point it out if there is an undefended assumption about the relative merits of one over the other.

Here’s something I would love to talk to you about: I think the survery instruments fail to track a very likely increase in real happiness. You seem to think the reverse. In this post, I argue that we’ll have to get good and reductive, measuring the physical correlates of good and bad feelings, in order to know for sure.

You are also right that I do not, in that paper, subject governments to much scrutiny. But that’s not necessary for my purposes there, and it’s a major reason for the qualifications in my argument (which Blar noted). My point was simply to illustrate why we should be interested in the idea that people may systematically botch judgments regarding their happiness. And one reason is that such evidence can weaken–not defeat, but weaken–consequentialist arguments for liberal strictures on paternalism. Those arguments will be weakened insofar as premises about governmental incompetence have to bear greater weight than they otherwise would. Maybe this won’t amount to much if governments are *obviously* completely inept in this realm, but that’s an extremely strong assumption, one that seems to me implausible.

We can agree that in most personal decisions the individual is better placed to decide how to promote her own interests. It is doubtful that the public welfare would be served by a Ministry of Happiness with meddlesome case workers who must sign off on our marriage proposals, choice of occupation, etc. But most liberals (myself included) want to prohibit certain sorts of intervention altogether, save perhaps in extreme cases. I think the sorts of considerations discussed in my paper should weaken our confidence in the idea that such interventions could never be effective in promoting well-being.

My target is not liberalism, which I take as given (though not on consequentialist grounds!). People are entitled to be treated with respect, and not like children, whether that makes them happier or not. I suspect that some forms of paternalism–there are many, of which classic “gun to head” coercion is just one–would be both effective and acceptable means of promoting happiness. But if a proposed form of paternalism violates principles of respect for persons, then it is impermissible. And if that means we are doomed to unhappiness, then so be it. (I fear we may be in a bit of a bind: perhaps human beings are morally entitled to broad freedoms, but psychically ill-equipped to fashion good lives for themselves under such conditions.)

Regarding John Brothers’ comment: I don’t see the problem here. If most people believe themselves to be happy, yet most are not, and the state knows this, then there is a perfectly clear sense in which the state knows more about how happy people are than they do. They would enjoy an important epistemic advantage over the public in such a case. I suspect you were taking me to say that the government could more reliably judge how happy any given individual is than she herself could. But that is a stronger claim than what I asserted.

Will, Sir: Thank you for having a great blog. Reading it after having not done so in some months makes me wonder why I don’t read it more regularly.

As for presumptions, I cannot see the point in attempting to establish them one way or the other in general. Political philosophy, even informed by psychological and institutional research probably can’t give us useful generalizations in this regard. (That’s my opinion.) The intricacies of particular policy questions are so complex that focusing on given cases seems much more productive. I have not posted on your “What is Philosophy Good For?” thread but that is an intersting topic and I’m trying to get to it through this discussion and the one on the moral significance of growing wealth.

If it could be shown that a realistically feasible government meeting widely accepted conditions for legitimacy could successfully implement policies that would significantly improve the rate at which individuals successfully fitted their actions to their ends, THEN the liberal presumption against paternalism would be weakened.

I would have been quite happy. But he did not say that. He said, people make systematic errors in prudential reasoning. And therefore the presumption against paternalism is weakenend. But—and this was my point—that conclusion is a non-sequitur absent any consideration of the reliability of government action in this kind of domain.

I guess the important thing, whatever you think Haybron is saying, is that we agree that this debate is worth having. I got the feeling from the way you called “fallacy” on Haybron that you were trying to nip it in the bud, but from your other posts I know that’s not so. Human cognition and government institutions are both non-ideal, and figuring out what to do about that is an important, complex, and unresolved question, one about which we have different priors but are still open to evidence. I think that Haybron’s passage suggests that he’s in the same boat, but I doubt that it’s worth having a big debate to try to parse the meaning of his words.

Simple rights-protecting machines could conceivably promote individual and collective well-being better than anything else. I guess I meant (2) to deny that participation in goverment was a good in itself, part of — though not the whole of — cooperation with one’s society.

Also, unlike Rousseau, I think that you can have a defective general will in which lots of individuals and/or groups get the shaft. For R that would be the “will of all”, the honorary title general will just going to a good union of wills. I’d say the question of (a) whether you have a unity of individual interests that is different from the aggregation of those interests is analytically distinct from (b) whether you’ve got one where each “remains as free as before”.

Arrow pretty well showed that you can’t get a unique social choice from aggregating individual interest but, importantly, did not show that there can be no general will where each “remains as free as before” (or, for that matter, where each does not but perhaps thinks he does). In fact, Aarow was I’m told something of a Rousseauian.

That’s simply not what he’s saying. He’s saying that people don’t know what’s good for them, so the argument for paternalism is strengthened. Notice that he doesn’t even consider that the means-ends unreliability of government might be a problem. As you note, he says there are reasons based in the value of autonomy for still resisting paternalism. Which is not to say that he doesn’t think paternalism wouldn’t work.

I was approving of Haybron’s arguments that we don’t necessarily know what makes us happy. Not his contention that the government may have a comparative epistemic and practical advantage.