Stop measuring happiness!

June 17, 2012 11:44 AM

Happyism: The Creepy New Economics of Pleasure. Economist Dierdre McCloskey, in the New Republic, digs into the mathematical underpinnings of the scientific study of happiness. Executive summary: she doesn't like what she finds.
posted by escabeche (26 comments total)

I have lost several IQ points and I am not happy.
posted by anigbrowl at 11:52 AM on June 17, 2012 [5 favorites]

If you don't know Deirdre, you should. Not only are the economics wonderful, but her "Crossing" is a beautiful memoir.
posted by chavenet at 12:22 PM on June 17, 2012 [3 favorites]

That sure is a long and meandering essay to couch such a simple thesis.
posted by Nomyte at 12:38 PM on June 17, 2012 [4 favorites]

Isn't this just Campbell's Law?
posted by erniepan at 12:57 PM on June 17, 2012 [2 favorites]

Thanks for saving me some time, Nomyte. I had to stop reading when she plunked down the little story of two guys playing Philosophy 101 word games ("How do you know I don't know?" "How do you know I don't know that you don't know?" etcetera ad nauseum), with no explanation of why she put it there, and I wondered if the rest of it was any better.
posted by Greg_Ace at 12:57 PM on June 17, 2012

Maybe the science of measuring happiness would work better if we measured unhappiness instead? I mean, if you measure happiness in terms of your purchasing power you immediately run into the fact that many of the basic requirements for happiness cannot be purchased, not to mention the law of diminishing returns.
It seems to me it would all be more practical if you looked at happiness as the natural basal state of a human being - which they seem to be doing with cats and goldfish, in this essay - and then figure out how comparitively not-unhappy humans are when you remove various bad things from their lives, like thirst, pain, mistrust etc.
I think you'd have a better shot at creating happiness if you made it possible for people to achieve these things and took the over-simplification of money out of it.
posted by Jane the Brown at 1:23 PM on June 17, 2012 [3 favorites]

Well, I liked it, and it was pretty clear what she was talking about by the end.
posted by anifinder at 1:28 PM on June 17, 2012

I think the premise--that trying to assign arbitrary numbers to the happiness of individuals and society is fraught with error (it isn't "science" and has no hopes of being one) and therefore any policies or programs implemented based on those ideas are doomed to fail--is sound and well-argued.

I note, as a non-academic, papers like this seem to feel obligated to touch as many established bases in the field as possible, possibly to forestall criticism? The penalty paid is that the reading is tedious to someone not into the subject; but the benefit is that the density of references provides a person who is fascinated by the subject numerous jumping-off points for further study.
posted by maxwelton at 1:47 PM on June 17, 2012 [1 favorite]

Good post. I've often questioned the value of what is studied in that peculiar cottage industry. Happiness is a state of being or even a pursuit (for some), and not something that can ever be entirely attained because our expectations of the good life are constantly changing and are relative to individual experience. Human misery, on the other hand, is much more a circumstance and is thus much easier to define and quantify. I'd rather see energy devoted to answerable questions like "what can we do specifically to eradicate misery?" than to "how can we make people happier?"
posted by runningdogofcapitalism at 1:51 PM on June 17, 2012 [1 favorite]

This essay is the most unreadable mess I've encountered in a long time, and I've read it twice now. Happiness is a good editor.

(Oh, and her response at the bottom to readers worrying about her dismissal of statistical significance is just like the essay itself: a muddle of pleas to authority in the guise of a friendly jumble of thoughts. The fact that a right-leaning libertarian has managed to convince today's Supreme Court to view statistical significance differently is not the ultimate validation of her ideas that she imagines it is.)
posted by chortly at 2:12 PM on June 17, 2012 [8 favorites]

@chortly Heh. You get a fave from me for this: "Happiness is a good editor."
posted by runningdogofcapitalism at 2:15 PM on June 17, 2012 [2 favorites]

Though that was a total mess, I do like ethi-nomics... and she referred to Mill as Jeremy Bentham's "young friend", which is bizarre but amusing.
posted by zinful at 2:33 PM on June 17, 2012

I've often questioned the value of what is studied in that peculiar cottage industry.

The happiness business? Well, it sells lots of books to people who would be ashamed to shop in the Self-Help section but are proud to shop in the SCIENCE! section. They buy the supposed answer to "what makes [other] people happy?" because they want to find an easy way to make themselves happy (or at least explain why they are unhappy). The same goes for all those scientificish books about learning and creativity.

And I guess a lot of those same people form public policy, so happiness stats will be worked into official data.
posted by pracowity at 2:39 PM on June 17, 2012

I feel like I'm either reading the wrong sorts of media or the very right sorts, because I was lost for context on this essay. I get what she's saying, but the problem she's attacking isn't one that I've encountered much of.
posted by immlass at 3:14 PM on June 17, 2012

I couldn't get through this horribly written essay so I skimmed and it's possible that I missed what I was seeking: reckoning with the fact that the happiness data doesn't argue against material improvement, but rather against insane greed.

Basically, happiness tracks inequality inversely. Places with high happiness also have low inequality and high longevity. Sure, measurement of any social quality is inexact. But this stuff repeatedly converges against extremes of inequality and she seems to think that its stance against materialism is Puritanical and snobby, rather than about inequality.

It's actually the reverse: the idea is basically that if you allow markets to rule in such a way that you create an infinite arms race of status goods, *everyone* is worse off, not just the poor. So it's not about taking from the rich so that everyone has more "spiritual" or artistic types of pleasure.

When you have less extreme inequality, the rich are healthier too because basically, it's healthier not to have to worry about the peasants with pitchforks while living in an enclosed dome protected by armed guards and also healthier not to have to always be competing against the rest of the elite every second for fear of falling into dire poverty.

Less status stress translates into better happiness and health for everyone (even baboons)— it's not about taking away material pleasures because they are "bad," it's about making things more equal so that everyone can be happier. Like, say, a ratio of 1000 to one CEO salary to worker, not 10,000. That's hardly communism or perfect equality or some puritan anti-material pleasure situation. But it's a happier one and from what I can see, she misses this aspect of the question, which to me is why the measurement of happiness actually tells us some important things.
posted by Maias at 3:17 PM on June 17, 2012 [17 favorites]

In the first half of the article, McCloskey argues that we can't measure happiness. She presents the standard objection to a Benthamite felicific calculus, that my love of apples can't be measured against your love of oranges, and her argument is perfectly sound though not particularly original. But the second half of the article takes a more interesting turn. Here she argues that we can measure happiness. Happiness, it turns out, is closely related to material prosperity. The higher our income, the more opportunities there are for personal fulfilment and the cultivation of the good life. It's obvious, for example, that we are happier now than we were back in the 1950s, because -- well, just because:

One of the proponents of happiness studies, the eminent British economist Richard Layard, is fond of noting that “happiness has not risen since the ’50s in the U.S. or Britain or (over a shorter period) in western Germany.” Such an allegation casts doubt on the relevance of the “happiness” so measured. No one who lived in the United States or Britain in the ’50s (I leave judgments on West Germany in the ’70s to others) could possibly believe that the age of Catcher in the Rye or The Loneliness of the Long-Distance Runner was more fulfilling than recent times.

It also follows from her argument that we should embrace the economic system that has blessed us with all this material prosperity. McCloskey is thus a cheerleader for capitalism (or, as she prefers to call it, 'bourgeois virtue'), as against those critics like Terry Eagleton ('a superb if determinedly left-wing professor' -- McCloskey likes to reassure her opponents that they are 'brilliant' and 'superb' despite being utterly misguided) who have dared to suggest that bourgeois accumulation might not lead to universal happiness. Anyone who questions the benefits of economic growth is ridiculed as a 'puritan' who doesn't like the sight of other people enjoying themselves.

The problem with McCloskey's celebration of 'bourgeois virtue' is that it is essentially an exercise in cherry-picking in which she has selected all the nice bits of capitalism and discarded all the nasty bits. Her ideal type of the Good Bourgeois is undeniably attractive -- but then it would be, wouldn't it? Whether it really bears much resemblance to the real lives of eighteenth-century sugar-planters or nineteenth-century factory-owners is another matter. Nor does McCloskey address the nagging suspicion that our wealth, our happiness, our personal fulfilment as children of the modern West may have been purchased at the cost of other people's misery.
posted by verstegan at 4:20 PM on June 17, 2012 [12 favorites]

So, an economist who has reservations about applying a scientific approach to a complex, human based phenomenon?
posted by memebake at 4:32 PM on June 17, 2012 [4 favorites]

Tried three times and just couldn't get through the essay.
posted by meinvt at 5:35 PM on June 17, 2012

If you want to learn about happiness, seek the answer from your dog.
posted by caddis at 6:18 PM on June 17, 2012 [1 favorite]

Trust me, you poors don't want ice cream and pastrami anyway.
posted by fleetmouse at 6:24 PM on June 17, 2012

I came looking for equations, but left disappointed.
posted by bessel functions seem unnecessarily complicated at 9:00 PM on June 17, 2012

Yeah it seems like there is a real divergence in what people like to read. this FPP linked to another 'long form' story that seriously buried the lead as well. Some people want to be enveloped in a story, they want to enjoy the act of reading, rather then just getting the facts.

I really don't. I don't need a history lesson before you get to the point. If the argument is that there is something 'wrong' with the way we think about happiness today, why are you spending paragraph after paragraph telling me how people thought about it in the 1800s or ancient China?

I mean, come on look at this:

_{It’s not science. At the most lofty level of scientific method, the hedonicists cheekily and with foreknowledge mix up a “non-interval” scale with an “interval” scale. If you like the temperature in Chicago today better than the one on January 15, you might be induced by the interviewer to assign 2.76 to today and a 1.45 to January 15. But such an assignment is of course arbitrary in God’s eyes. It is not a measure in her view of the difference even in your heart (since to her all hearts are open) between a nice day and a cold day. By contrast, an interval scale, such as Fahrenheit or Celsius temperature on the two days in question, does measure, 1-2-3. God doesn’t care which scale you use for hedonics as long as it’s an interval scale. Non-interval scales merely rank (and classifications merely arrange). We couldn’t base a physics on asking people whether today was “hot, nice, or cold” and expect to get anything quantitative out of it.}

Like, does she actually think her readers are so stupid they need a 168 word paragraph to with an illustrative example to explain an interval scale?

And anyway, how is this paragraph even correct? Yes, asking a person if it was "hot, nice, or cold" wouldn't be a very good way to measure temperature. but if you asked thousands of people, and correlated their answers with real temperature data for that area you might actually be able to come up with a fairly narrow range for what the actual temperature was Like: in San Diego in June 20% 'hot' and 70% 'nice' could mean around 75-85 degrees, but but in Boston it might mean 65-75.

Or who knows. The point is, the fact that people can only give you a subjective evaluation of comfort with relation to temperature doesn't mean that temperature isn't an intrinsically quantifiable thing that really exists. Obviously there is some kind of chemical situation going on in the brain that produces a sensation of happiness. Unlike temperature, it can't be measured easily with instruments, but like temperature people can tell us their subjective evaluation of it.
posted by delmoi at 2:24 AM on June 18, 2012 [2 favorites]

It is like the pre-election surveys that ask people now to “vote” for or against President Obama in 2012. The sampling “error” is always 2 or 3 percent (because the sample size is always about 1,000 or 1,500 and the probability of a “yes” is about 50 percent: The math of the binomial calculation is left as an exercise for the reader), which is carefully reported by the journalists as though it were the error in predicting the actual election. But the error in the important prediction has nothing to do with mere sampling error. If unemployment gets below 8 percent, Obama will win, whatever the “significance” of the survey results now. If he is discovered to practice adultery, he will lose.

What is this person talking about? Everyone understands that polls don't predict election results given the possibility of unexpected events, like Obama having an affair. The polls measure how people feel at a given point in time. And they work pretty well. Baring exogenous events they actually do a pretty good job telling you who is going to win an election.

Why is this economist lecturing people on what is and is not science, by the way?

And the literature simply strides past the problem evident since the first criticisms of Bentham—that utils cannot be compared among people. This is not merely a technical matter. It is also a matter of human dignity. Suppose that Connor and Lily have the same happinessproducing circumstances of income and health and so forth. Yet Lily’s reported “very happy” value of 2.8 exceeds Connor’s meager 1.4. Well, then: Lily must be a better machine for making happiness than Connor is. Never mind that Lily-utils have nothing to do with Connor-utils. We are in the realm of “Why is a mouse when it spins?” (The answer, of course, is, “The higher the fewer.”)

There is in fact substantial biological and psychological evidence that some people are more cheerful than others—just in case you didn’t already know that. So, let’s see: What are the “policy implications?” Shoot the Connors? Give all the money to the Lilys? “A Brahmin,” noted a Hindu lawyer, “is entitled to exactly five-and-twenty times as much happiness as everyone else.” If it would please 99 people to feed on the hundredth, then util-maximization says, get out the cooking pot.

So the last part of that is a completely brainless attack on a strawman version of utilitarianism. I think you also include an idea of justice or fairness as well as pure utility maximization.

Anyway, other problem is that while it's true people don't all have the same 'calibration' as far as self-reported comfort, again, people wouldn't all give you the same subjective temperature evaluations either. But that doesn't mean temperature doesn't exist.

Generally, people are trying to figure out what would increase or decrease happiness. So it's just like measuring salt with your tongue. Maybe it's "too salty" and you add more water/ingredients. Maybe it's not salty enough so you add more salt. You don't need to know the exact salinity. You don't need to the exact happiness measure to tell if someone is more happy or less happy
posted by delmoi at 2:42 AM on June 18, 2012 [2 favorites]

The knock-down argument against the 1-2-3 studies of happiness comes from the philosopher’s (and the physicist’s) toolbox: a thought experiment. “Happiness” viewed as a self-reported mood is surely not the purpose of a fully human life, because, if you were given, in some brave new world, a drug like Aldous Huxley’s imagined “soma,” you would report a happiness of 3.0 to the researcher every time. Dopamine, an aptly named neurotransmitter in the brain, makes one “happy.” Get more of it, right? Something is deeply awry.

*sigh*. How dopamine is used in the brain is actually fairly complicated, but my understanding is that it's more like the the sensation of desire or drive for a particular thing. So if you're hungry and you see a delicious cake: the dopamine fires when you see it and your desire builds up, rather then when you actually eat it.

Anyway, this article is a total slog and it's hard to see what exactly the problem is. She thinks these people are incorrect to try to increase people's subjective happiness, but it isn't at all clear what the problem is with them trying to do it. She seems to be saying that people's lives are pretty good today compared to a couple hundred years ago, and that's undoubtedly true. But it's not like trying to boost happiness would somehow cause us to backslide. A lot of the increase comes from science: modern medicine, computers, etc not necessarily -- whatever it is she's talking about. People seeking to boost subjective happiness levels would undoubtedly try to continue funding science.
posted by delmoi at 3:06 AM on June 18, 2012 [1 favorite]

Why is this economist lecturing people on what is and is not science, by the way?

A perfectly rational person would always accept an economist's definition of a problem as reflecting whatever domain they are talking about. Economists are so damn taut n logical.
posted by srboisvert at 3:12 AM on June 18, 2012

Metafilter: That sure is a long and meandering essay to couch such a simple thesis.
posted by symbioid at 11:35 AM on June 18, 2012